Flotation process prediction method and device based on mechanism and data hybrid drive

Abstract

The invention discloses a flotation process prediction method and device based on mechanism and data hybrid drive, and relates to the field of beneficiation, theoretical reliability is guaranteed through mechanism models such as foam generation and ore particle adsorption, and full-process system coverage from bubble characteristics to final grade is achieved by combining with field data to adapt to dynamic working conditions, so that the flotation process prediction method and device based on mechanism and data hybrid drive are provided. The method has the advantages that errors caused by simplified assumption of a pure mechanism model are made up, key indexes such as bubble parameters and mineral floating quantity are accurately quantified to support operation optimization, the method can also naturally adapt to complex industrial flotation processes such as rough flotation, fine flotation and scavenging in which multiple grooves are connected in series, the defects of the pure mechanism or pure data model are effectively overcome, and the prediction and control accuracy of the flotation process is improved. And core production indexes such as flotation quality, tailing concentration, concentrate grade and tailing grade of the obtained useful minerals / gangue can help operators to timely optimize the process, reduce the loss of the useful minerals and improve the concentrate quality, so that the flotation efficiency and the resource recovery rate are improved.

Description

Technical Field

[0001] This application relates to the field of mineral processing technology, and in particular to a method and apparatus for predicting flotation processes based on a hybrid mechanism and data-driven approach. Background Technology

[0002] As the mineral processing industry continues to develop towards intelligence and precision, accurate simulation, prediction, and optimization control of production processes have become a core requirement. Flotation, as one of the most important mineral separation processes, directly impacts process design, operational optimization, and key indicators such as final concentrate grade and recovery rate through the accuracy of its process model.

[0003] In related technologies, research on flotation process modeling mostly focuses on a specific process in single-cell flotation, such as the rising velocity of froth in the slurry, the probability of mineral particles being adsorbed by froth, and the relationship between the amounts of substances entering and leaving the flotation cell. However, the applicant recognizes that existing technologies are mostly focused on modeling single-cell or local processes, and are based on data-driven modeling, making it difficult to balance the accuracy of physical mechanisms with adaptability to complex operating conditions, and thus failing to provide comprehensive and accurate model support for the optimized control of the entire flotation process. Summary of the Invention

[0004] In view of this, this application provides a flotation process prediction method and apparatus based on a hybrid mechanism and data-driven approach. The main purpose is to solve the problem that existing technologies mostly focus on modeling single cells or local processes, and are based on data-driven modeling, which makes it difficult to balance the accuracy of physical mechanisms with the adaptability to complex working conditions, and cannot provide comprehensive and accurate model support for the optimization control of the entire flotation process.

[0005] According to a first aspect of this application, a flotation process prediction method based on a hybrid mechanism and data-driven approach is provided, the method comprising:

[0006] Obtain the flotation pulp foam diameter model, flotation pulp foam rise velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell;

[0007] The bubble diameter and bubble rise velocity of the current flotation cell are calculated based on the flotation pulp foam diameter model and flotation pulp foam rise velocity model corresponding to the current flotation cell.

[0008] Read the tailings flow rate of the previous flotation cell, and input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate and concentrate flow rate of the current flotation cell.

[0009] Based on the flotation pulp particle adsorption model corresponding to the current flotation cell, the useful mineral flotation constant and gangue flotation constant of the current flotation cell are calculated using the bubble diameter of the current flotation cell.

[0010] Based on the flotation model corresponding to the current flotation cell, the useful mineral flotation constant and gangue flotation constant of the current flotation cell are used to calculate the useful mineral flotation mass and gangue flotation mass of the current flotation cell.

[0011] The tailings concentration and tailings grade of the previous flotation cell are read, and the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of useful minerals floated out and the mass of gangue floated out of the current flotation cell are used to calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell.

[0012] According to a second aspect of this application, a flotation process prediction device based on a hybrid mechanism and data-driven approach is provided, the device comprising:

[0013] The model acquisition module is used to acquire the flotation pulp foam diameter model, flotation pulp foam rising velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell.

[0014] The bubble parameter calculation module is used to calculate the bubble diameter and bubble rising speed of the current flotation cell based on the flotation pulp foam diameter model and flotation pulp foam rising speed model corresponding to the current flotation cell.

[0015] The flow distribution calculation module is used to read the tailings flow rate of the previous flotation cell, input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single cell slurry flow rate model corresponding to the current flotation cell, and obtain the tailings flow rate and concentrate flow rate of the current flotation cell.

[0016] The flotation constant calculation module is used to calculate the useful mineral flotation constant and gangue flotation constant of the current flotation cell based on the flotation pulp particle adsorption model corresponding to the current flotation cell and using the bubble diameter of the current flotation cell.

[0017] The mineral floatation calculation module is used to calculate the useful mineral floatation mass and gangue floatation mass of the current flotation cell based on the flotation model corresponding to the current flotation cell and using the useful mineral flotation constant and gangue flotation constant of the current flotation cell.

[0018] The grade and concentration calculation module is used to read the tailings concentration and tailings grade of the previous flotation cell, and calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell using the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, and the floated mass of useful minerals and gangue in the current flotation cell.

[0019] According to a third aspect of this application, an apparatus is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described in any of the first aspects above.

[0020] According to a fourth aspect of this application, a medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0021] By employing the above technical solutions, the technical solutions provided in the embodiments of this application have at least the following advantages:

[0022] This application provides a flotation process prediction method and apparatus based on a hybrid mechanism and data-driven approach. The method first obtains the flotation pulp foam diameter, foam rising velocity, single-cell pulp flow rate, pulp particle adsorption, and flotation model corresponding to the current flotation cell. Then, it calculates the bubble diameter and rising velocity sequentially using the foam model. Combining the tailings flow rate and bubble parameters from the previous cell, it obtains the current cell tailings and concentrate flow rates from the single-cell flow model. Based on the particle adsorption model, it calculates the flotation constants for useful minerals and gangue from the bubble diameter. Using the flotation model, it calculates the mineral float mass from the flotation constants. Finally, it calculates the current cell tailings concentration, concentrate grade, and tailings grade by combining the previous cell tailings concentration and grade with the current cell flow rate and mineral float mass. This system ensures theoretical reliability through mechanistic models of foam generation and mineral particle adsorption, while also adapting to dynamic operating conditions using field data. It achieves full-process system coverage from bubble characteristics to final grade, compensating for errors caused by simplistic assumptions in pure mechanistic models. It accurately quantifies key indicators such as bubble parameters and mineral flotation volume to support operational optimization. Furthermore, it naturally adapts to complex industrial flotation processes involving multiple cells in series, including roughing, cleaning, and scavenging, effectively addressing the shortcomings of pure mechanistic or pure data models and improving the accuracy of flotation process prediction and control. This application forms a complete closed-loop process through bubble parameter calculation, slurry flow distribution, flotation constant derivation, mineral flotation quality prediction, and ultimately, tailings / concentrate grade calculation. It obtains core production indicators such as useful mineral / gangue flotation quality, tailings concentration, concentrate grade, and tailings grade, helping operators optimize processes in a timely manner, reduce useful mineral loss, improve concentrate quality, and thus increase flotation efficiency and resource recovery rate.

[0023] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0024] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0025] Figure 1 This illustration shows a flowchart of a method for predicting flotation processes based on a hybrid mechanism and data-driven approach, as provided in an embodiment of this application.

[0026] Figure 2A This illustration shows a schematic flowchart of another method for predicting flotation processes based on a hybrid mechanism and data-driven approach, as provided in an embodiment of this application.

[0027] Figure 2B This paper illustrates a schematic diagram of a copper ore flotation simulation process architecture provided in an embodiment of this application.

[0028] Figure 3A This illustration shows a structural diagram of a flotation process prediction based on a hybrid mechanism and data-driven approach, as provided in an embodiment of this application.

[0029] Figure 3B This illustration shows another structural diagram of flotation process prediction based on a hybrid mechanism and data-driven approach provided in an embodiment of this application.

[0030] Figure 4 A schematic diagram of the device structure of an embodiment of this application is shown. Detailed Implementation

[0031] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0032] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0033] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0034] This application proposes a flotation process prediction method based on a hybrid mechanism and data-driven approach. It constructs a flotation process model framework using an overall equilibrium model; establishes a flotation pulp foam diameter model based on the flotation pulp foam generation mechanism; establishes a flotation pulp foam rise velocity model based on the flotation pulp flow mechanism; and establishes a flotation pulp particle adsorption model based on the collision and adhesion mechanism between mineral particles and bubbles. Combining the pulp foam diameter model, rise velocity model, and particle adsorption model, a single-cell flotation model is established. Based on this, a multi-cell continuous flotation model is established, thus forming a flotation process model. The implementing entity of this application can be a flotation process prediction system. This system relies on the computing power of a server to provide services to users. The server can be an independent server or a server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0035] This application provides a flotation process prediction method based on a hybrid mechanism and data-driven approach, such as... Figure 1 As shown, the method includes:

[0036] 101. Obtain the flotation pulp foam diameter model, flotation pulp foam rising velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell.

[0037] In this embodiment, the flotation pulp foam diameter model is constructed based on the current flotation cell operating data and the flotation pulp foam generation mechanism; the flotation pulp foam rise velocity model is constructed based on the current flotation cell operating data and the flotation pulp flow mechanism; the single-cell pulp flow rate model is constructed based on the current flotation cell operating data and the overall balance mechanism; and the flotation pulp particle adsorption model is constructed based on the current flotation cell operating data and the collision and adhesion mechanism between particles and bubbles. These models reflect the essential laws of the flotation process, possess solid theoretical interpretability, and avoid the models deviating from the actual physicochemical behavior of flotation. Furthermore, by incorporating actual operating data, the models are adapted to the specific operating conditions of the current flotation cell, such as reagent dosage and aeration parameters. This solves the problems of pure mechanistic models being disconnected from the actual situation due to simplified assumptions, and pure data models lacking generalization ability due to changes in operating conditions, significantly improving the models' accuracy and adaptability.

[0038] 102. Calculate the bubble diameter and bubble rise velocity of the current flotation cell based on the flotation pulp foam diameter model and flotation pulp foam rise velocity model corresponding to the current flotation cell.

[0039] In this embodiment, the flotation pulp foam diameter model is used to characterize the generation law of foam diameter and correlate pulp characteristics with foam size. The flotation pulp foam rising speed model is used to describe the dynamic characteristics of bubbles rising in the pulp and reflect the influence of buoyancy, drag, etc. on the rising speed. The bubble diameter directly affects the adsorption efficiency between mineral particles and bubbles, and the bubble rising speed determines how fast the mineral particles float with the bubbles. These two parameters can accurately capture the key characteristics of bubble behavior, providing a basis for subsequent flotation process analysis and ensuring the accuracy of flotation process modeling.

[0040] 103. Read the tailings flow rate of the previous flotation cell, and input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate and concentrate flow rate of the current flotation cell.

[0041] In this embodiment, the flotation process is continuous, with tailings from the previous flotation cell entering the current cell. The tailings flow rate from the previous cell considers the dynamic correlation of the slurry in each cell during the flotation process, avoiding the one-sidedness of calculating the flow rate of a single cell in isolation, and making the results more consistent with the logic of actual continuous production. The diameter of the bubbles in the current flotation cell affects the adsorption efficiency between the mineral particles and the bubbles, and the rising speed of the bubbles affects the rate at which the mineral particles float with the bubbles. Therefore, these factors directly affect the flotation separation process of the mineral particles. The final tailings flow rate and concentrate flow rate of the current flotation cell are calculated using a single-cell slurry flow rate model. By combining bubble diameter, rising speed, and flow rate distribution, the flow rate distribution law is explained from the perspective of the flotation mechanism, making the calculation of tailings and concentrate flow rates more accurate, rather than simply relying on experience or simplified assumptions.

[0042] 104. Based on the flotation pulp particle adsorption model corresponding to the current flotation cell, calculate the useful mineral flotation constant and gangue flotation constant of the current flotation cell using the bubble diameter of the current flotation cell.

[0043] In this embodiment, the flotation pulp particle adsorption model is established based on the adsorption mechanism between mineral particles (i.e., valuable minerals and gangue) and air bubbles, describing the adsorption law of mineral particles by air bubbles. Since the bubble diameter directly affects the adsorption efficiency between mineral particles and air bubbles, the flotation constants for valuable minerals and gangue can be calculated based on the flotation pulp particle adsorption model and the actual bubble diameter of the current flotation cell. The flotation constant for valuable minerals reflects the ease or rate of flotation of valuable minerals, while the flotation constant for gangue reflects the ease or rate of flotation of gangue. Because valuable minerals and gangue have different surface physicochemical properties and exhibit different adsorption behaviors towards air bubbles, calculating their respective flotation constants accurately characterizes the differences in flotation performance under current operating conditions, making the calculation of flotation constants more consistent with the complexities of industrial settings and improving parameter reliability.

[0044] 105. Based on the flotation model corresponding to the current flotation cell, calculate the floatation mass of the useful mineral and the floatation mass of the gangue in the current flotation cell using the useful mineral flotation constant and the gangue flotation constant of the current flotation cell.

[0045] In the embodiments of this application, the flotation model is constructed based on the laws of mineral particle-bubble interaction and flotation kinetics. By inputting two types of flotation constants into the model, the mass of useful minerals carried out by bubbles and the mass of gangue floated out in the current flotation cell can be calculated. This makes the flotation mass of useful minerals and gangue more accurately reflect the actual flotation separation results, ensuring both theoretical scientificity and close alignment with the actual needs of industrial production.

[0046] 106. Read the tailings concentration and tailings grade of the previous flotation cell, and use the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of useful minerals floated out of the current flotation cell and the mass of gangue floated out of the current flotation cell to calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell.

[0047] In this embodiment, flotation is a multi-cell continuous operation process. By fully considering the dynamic correlation between the output of the previous cell and the input of the current cell, the calculation results are more consistent with the continuous reality of industrial production, avoiding the one-sidedness of isolated analysis of a single cell. Moreover, tailings concentration, concentrate grade, and tailings grade are core indicators for evaluating flotation effectiveness. Concentrate grade determines product quality, tailings grade reflects the loss rate of valuable minerals, and tailings concentration is related to subsequent tailings treatment costs. These indicators provide operators with intuitive information for adjusting process parameters such as aeration rate, reagent addition, and slurry flow rate, thereby optimizing flotation efficiency and improving resource recovery rate.

[0048] This application provides a flotation process prediction method based on a hybrid mechanism and data-driven approach. Compared with the prior art, this application first obtains the flotation pulp foam diameter, foam rising velocity, single-cell pulp flow rate, pulp particle adsorption, and flotation model corresponding to the current flotation cell. Then, it calculates the bubble diameter and rising velocity sequentially through the foam model. Combining the tailings flow rate and bubble parameters of the previous cell, it obtains the tailings and concentrate flow rates of the current cell from the single-cell flow model. Based on the particle adsorption model, it calculates the flotation constants of useful minerals and gangue from the bubble diameter. Through the flotation model, it calculates the mineral float-out mass from the flotation constants. Finally, it calculates the tailings concentration, concentrate grade, and tailings grade of the current cell by combining the tailings concentration and grade of the previous cell with the current cell flow rate and mineral float-out mass. This system ensures theoretical reliability through mechanistic models of foam generation and mineral particle adsorption, while also adapting to dynamic operating conditions using field data. It achieves full-process system coverage from bubble characteristics to final grade, compensating for errors caused by simplistic assumptions in pure mechanistic models. It accurately quantifies key indicators such as bubble parameters and mineral flotation volume to support operational optimization. Furthermore, it naturally adapts to complex industrial flotation processes involving multiple cells in series, including roughing, cleaning, and scavenging, effectively addressing the shortcomings of pure mechanistic or pure data models and improving the accuracy of flotation process prediction and control. This application forms a complete closed-loop process through bubble parameter calculation, slurry flow distribution, flotation constant derivation, mineral flotation quality prediction, and ultimately, tailings / concentrate grade calculation. It obtains core production indicators such as useful mineral / gangue flotation quality, tailings concentration, concentrate grade, and tailings grade, helping operators optimize processes in a timely manner, reduce useful mineral loss, improve concentrate quality, and thus increase flotation efficiency and resource recovery rate.

[0049] Furthermore, as a refinement and extension of the specific implementation methods of the above embodiments, and in order to fully illustrate the specific implementation process of this embodiment, this application provides another flotation process prediction method based on a hybrid mechanism and data-driven approach, such as... Figure 2A As shown, the method includes:

[0050] 201. Obtain the flotation pulp foam diameter model, flotation pulp foam rising velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell.

[0051] In this embodiment, the flotation pulp foam diameter model describes the generation law of foam diameter in the pulp of the current flotation cell, the relationship between influencing factors and foam diameter, and calculates the foam diameter; the flotation pulp foam rising speed model describes the variation law of foam rising speed in the pulp of the current flotation cell, and determines the rising rate of foam in the pulp by combining principles such as fluid mechanics; the single-cell pulp flow rate model establishes the flow distribution relationship of pulp in the cell based on the mechanism of mass conservation, focusing on the current flotation cell; the flotation pulp particle adsorption model describes the law of adsorption behavior such as collision and adhesion between particles and bubbles in the current flotation cell, providing a basis for subsequent mineral flotation characteristic analysis; the flotation model integrates the above models or covers a more comprehensive range of flotation process laws, and is used to describe the complete process from pulp flow and bubble behavior to mineral separation in the current flotation cell, so as to calculate key indicators such as mineral float-out amount.

[0052] By acquiring on-site production data from the flotation cell, including raw ore grade, concentrate grade, yield, slurry flow rate, slurry concentration, flotation cell aeration rate, liquid level, impeller speed, and reagents (including collectors, frothers, lime, etc.), time-series alignment was performed on each measured variable in the on-site production data, using the variable with the longest sampling interval as the benchmark. Grade had the longest sampling time; within the grade sampling interval, other variables were aligned with the grade time-series by averaging their values. Then, combining theories such as the flotation slurry foam generation mechanism and flotation slurry flow mechanism, models for flotation slurry foam diameter, flotation slurry foam rise velocity, single-cell slurry flow rate, flotation slurry particle adsorption, and flotation were constructed. These models accurately match the actual operating conditions of the cell, avoiding errors caused by differences in operating conditions in general models, and making subsequent calculations more closely reflect the actual flotation process in industrial settings.

[0053] 202. Calculate the bubble diameter of the current flotation cell based on the flotation pulp foam diameter model corresponding to the current flotation cell.

[0054] In this embodiment, the bubble diameter of the current flotation cell is calculated based on the flotation pulp foam diameter model corresponding to the current flotation cell, and the calculation formula is as follows: Formula 1:

[0055] Formula 1:

[0056] γ=γ 水 [1-aln(1+c / b)],

[0057]

[0058] Among them, D b ρ represents the diameter of the bubble in the current flotation cell. fγ represents the apparent density of the pulp in the current flotation cell, and γ represents the gas-liquid interfacial tension in the current flotation cell, which is affected by flotation reagents, so it is calculated using the Shishkovsky empirical formula. ε represents the average energy dissipation rate of the current flotation cell. 水 The gas-water interfacial tension is represented by 'a', the coefficient of the first model is represented by 'b', the coefficient of the second model is represented by 'c', the frother concentration of the current flotation cell is represented by 'n', and the impeller speed of the current flotation cell is represented by 'n'. min n represents the minimum impeller speed in the current flotation cell. max This represents the maximum impeller speed of the current flotation cell, and α represents the third model coefficient. The flotation pulp foam diameter model is based on the flotation pulp foam generation mechanism. Here, a and b are determined according to the type of frother in the current flotation cell, and α is determined according to the impeller structure of the current flotation cell. This model can accurately match flotation scenarios with different reagent systems and equipment structures, significantly improving the accuracy of bubble diameter calculations under different operating conditions. For example, the apparent density ρ of the pulp... f Take 1.24 g / cm 3 gas-water interfacial tension γ 水 Take 72, and the foaming agent concentration c is 3 × 10. -4 %, with coefficients a and b taken as 0.17 and 3.9 × 10, respectively. -4 Maximum impeller speed n max Given 500 r / m, the minimum impeller speed n min The impeller speed n is set to 400 r / m, and the coefficient α is set to 1. The flotation pulp foam diameter model is based on the flotation pulp foam generation mechanism as its theoretical core. At the same time, it introduces the Shishkowski empirical formula and equipment / reagent correlation parameters to ensure the scientific nature of the theory and adapt to the complex and variable pulp, reagent and equipment conditions in industrial sites, making the bubble diameter calculation more in line with actual production.

[0059] 203. Calculate the bubble rising speed of the current flotation cell based on the flotation pulp foam rising speed model corresponding to the current flotation cell.

[0060] In this embodiment of the application, the bubble rising velocity of the current flotation cell is calculated based on the flotation pulp foam rising velocity model corresponding to the current flotation cell, and the calculation formula is as follows: Formula 2:

[0061] Formula 2:

[0062] Among them, U b ρ represents the current bubble rising velocity in the flotation cell. f ρ represents the apparent density of the slurry in the current flotation cell. gLet ρ represent air density, g represent gravitational acceleration, and μ represent the current pulp viscosity in the flotation cell. The flotation pulp foam rise velocity model is constructed based on the flotation pulp flow mechanism, integrating key physical parameters such as pulp apparent density, pulp viscosity, and air density. It accurately describes the rise characteristics of bubbles in the pulp from a fluid dynamics perspective, making the calculation results more consistent with the actual bubble motion behavior during flotation. For example, the pulp viscosity μ is taken as 0.0016, and the air density ρ... g Take 0.0012 g / cm 3 .

[0063] 204. Read the tailings flow rate of the previous flotation cell, and input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate and concentrate flow rate of the current flotation cell.

[0064] In this embodiment, the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell are input into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate of the current flotation cell. The calculation formula is as follows: Formula 3:

[0065] Formula 3:

[0066] Among them, Q i Q represents the current tailings flow rate in the flotation cell. i-1 q represents the tailings flow rate of the previous flotation cell. i The current aeration rate of the flotation cell is represented by t, the current bubble rise time of the flotation cell is represented by r, and the current bubble radius of the flotation cell is represented by D. b d represents the current bubble diameter in the flotation cell, d represents the current bubble thickness in the flotation cell, H represents the current flotation cell height, and U represents the current bubble height in the flotation cell. b This indicates the current bubble rising speed in the flotation cell. For example, the flotation cell airflow rate q. i Take 0.8 m³ / s, flotation cell height H as 4.8 m, and bubble thickness d as 10. -4 m. By introducing the tailings flow rate of the previous flotation cell as input, the characteristics of continuous multi-cell flotation operation are fully considered, so that the current cell flow rate is closely related to the material state of the previous cell, which is more in line with the dynamic continuity of industrial production and avoids the one-sidedness of calculating the flow rate of a single cell in isolation.

[0067] Then, the difference between the tailings flow rate of the previous flotation cell and the tailings flow rate of the current flotation cell is used as the concentrate flow rate of the current flotation cell. The bubble diameter and bubble rising speed of the current cell are combined with the flow model. Starting from the mechanism logic of bubble movement, mineral particle flotation and flow distribution, the tailings flow rate of the current cell is accurately calculated. Then, the concentrate flow rate is obtained by the difference between the tailings flow rates of the previous cell and the current cell, so that the flow distribution can better reflect the actual separation effect.

[0068] Optionally, the single-tank slurry flow rate model is established using an overall equilibrium model. To improve calculation accuracy, a correction hyperparameter can be added to the overall equilibrium model, and the calculation formula is shown in Formula 4 below:

[0069] Formula 4:

[0070] Among them, Q i Q represents the current tailings flow rate in the flotation cell. i-1 q represents the tailings flow rate of the previous flotation cell. i The current aeration rate of the flotation cell is represented by t, the current bubble rise time of the flotation cell is represented by r, and the current bubble radius of the flotation cell is represented by D. b d represents the current bubble diameter in the flotation cell, d represents the current bubble thickness in the flotation cell, H represents the current flotation cell height, and U represents the current bubble height in the flotation cell. b θ represents the current bubble rising speed in the flotation cell, and θ represents the flow correction hyperparameter.

[0071] 205. Construct the flotation pulp particle adsorption model corresponding to the current flotation cell.

[0072] In this embodiment, the induction time model corresponding to the current flotation cell is obtained. The apparent density of the pulp, the bubble diameter, and the bubble rising speed of the current flotation cell are input into the induction time model corresponding to the current flotation cell to obtain the induction time of the current flotation cell. The calculation formula is as follows: Formula 5:

[0073] Formula 5:

[0074]

[0075] φ cr =0.5φ c ,

[0076]

[0077] N = -0.437Re 1.0562 ,

[0078]

[0079] Where T represents the current sensing time of the flotation cell, and D... p D represents the diameter of the mineral particles in the current flotation cell. b U represents the current bubble diameter in the flotation cell. b This represents the current bubble rising velocity in the flotation cell, A represents the first induction time model coefficient, B represents the second induction time model coefficient, and φ represents the... c φ represents the collision angle. crIndicates the capture angle, V p The following parameters represent the current settling velocity of the mineral particles in the flotation cell: X represents the first hydrodynamic coefficient, Y represents the second hydrodynamic coefficient, Re represents the bubble Reynolds number, M represents the third hydrodynamic coefficient, N represents the fourth hydrodynamic coefficient, C represents the comprehensive coefficient, and ρ represents the overall coefficient. f ρ represents the apparent density of the slurry in the current flotation cell. g Let r represent the air density, r represent the bubble radius in the current flotation cell, and g represent the acceleration due to gravity. This represents the fluid dynamic viscosity coefficient.

[0080] Next, obtain the probability model corresponding to the current flotation cell. Input the induction time, bubble diameter, and bubble rising speed of the current flotation cell into the probability model corresponding to the current flotation cell to obtain the adhesion success probability of the current flotation cell. The calculation formula is as follows: Formula 6:

[0081] Formula 6:

[0082] Among them, P a The value represents the probability of successful adhesion in the current flotation cell, δ represents the probability model coefficient, Re represents the bubble Reynolds number, and U represents the bubble number. b D represents the current bubble rising speed in the flotation cell, T represents the current induction time in the flotation cell, and D represents the current bubble rising speed in the flotation cell. b D represents the diameter of the bubble in the current flotation cell. p This indicates the diameter of the mineral particles in the current flotation cell.

[0083] Subsequently, based on the collision and adhesion mechanism between mineral particles and bubbles, an adsorption model for the flotation pulp particles corresponding to the current flotation cell is constructed using the adhesion success probability of the current flotation cell, the particle diameter of the current flotation cell, the apparent gas velocity of the current flotation cell, the bubble diameter of the current flotation cell, and the bubble Reynolds number, as shown in Formula 7 below:

[0084] Formula 7:

[0085] Where K represents the flotation constant of the current flotation cell, P a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g Re represents the apparent gas velocity in the current flotation cell, Re represents the bubble Reynolds number, and D represents the apparent gas velocity in the flotation cell. b This indicates the diameter of the bubbles in the current flotation cell. For example, the particle diameter D. p Take 1.5 × 10 -4 m, bubble Reynolds number Re is 76.5, apparent gas velocity V g Take 0.8, and the gravitational acceleration g is taken as 9.8 m / s². 2 Hydrodynamic viscosity coefficient Take 0.0016.

[0086] The flotation slurry particle adsorption model revolves around the microscopic logic of particle-bubble collision, adhesion, and flotation. It quantifies the time characteristics from particle-bubble encounter to adhesion through a time-sensing model, quantifies the probability of successful adhesion after collision through a probabilistic model, and finally constructs a flotation constant model based on the collision-adhesion mechanism. This model fully recreates the core microscopic process of flotation, with each step relying on theories of fluid mechanics and particle interaction, laying a solid scientific foundation for the quantitative analysis of the flotation process. Furthermore, the model integrates multi-dimensional parameters such as slurry apparent density, bubble diameter / rise velocity, particle diameter, apparent gas velocity, and Reynolds number, comprehensively characterizing adsorption behavior from the perspectives of slurry, bubbles, particles, and fluid dynamics. This allows for precise adaptation to the complex and variable conditions of industrial settings, significantly improving the accuracy of parameter calculations.

[0087] 206. Based on the flotation pulp particle adsorption model corresponding to the current flotation cell, calculate the useful mineral flotation constant and gangue flotation constant of the current flotation cell using the bubble diameter of the current flotation cell.

[0088] In this embodiment, for the flotation process of useful minerals, the useful mineral correction hyperparameter is obtained, and the useful mineral correction hyperparameter is used to adjust the flotation pulp particle adsorption model corresponding to the current flotation cell to obtain the useful mineral flotation pulp particle adsorption model corresponding to the current flotation cell. The bubble diameter of the current flotation cell is input into the useful mineral flotation pulp particle adsorption model corresponding to the current flotation cell to obtain the useful mineral flotation constant of the current flotation cell. The calculation formula is as follows: Formula 8:

[0089] Formula 8:

[0090] Among them, K n P represents the flotation constant of the useful mineral in the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g Re represents the apparent gas velocity in the current flotation cell, Re represents the bubble Reynolds number, and D represents the apparent gas velocity in the flotation cell. b The value represents the bubble diameter in the current flotation cell, and α represents the useful mineral correction hyperparameter. α is calculated by inputting the collector dosage, frother dosage, lime dosage, and current aeration volume of the flotation cell into the BP neural network. This ensures the theoretical rigor of the flotation law and adapts to the dynamic changes of on-site process parameters, making the model more consistent with complex industrial flotation conditions.

[0091] For the gangue flotation process, the gangue correction hyperparameter is obtained, and the gangue pulp particle adsorption model corresponding to the current flotation cell is adjusted using the gangue correction hyperparameter to obtain the gangue flotation pulp particle adsorption model corresponding to the current flotation cell. The bubble diameter of the current flotation cell is input into the gangue flotation pulp particle adsorption model corresponding to the current flotation cell to obtain the gangue flotation constant of the current flotation cell. The calculation formula is as follows: Formula 9:

[0092] Formula 9:

[0093] Among them, K m P represents the gangue flotation constant of the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g Re represents the apparent gas velocity in the current flotation cell, Re represents the bubble Reynolds number, and D represents the apparent gas velocity in the flotation cell. b β represents the bubble diameter in the current flotation cell, and β represents the gangue correction hyperparameter. β is calculated by inputting the collector dosage, frother dosage, lime dosage, and current flotation cell aeration rate into a BP neural network.

[0094] It should be noted that α and β are functions of the amounts of collector, frother, lime, and aeration, and θ is a function of the amounts of frother and aeration, as shown in Formula 10 below:

[0095] Formula 9: α = f j (Dosage of collector, dosage of frother, dosage of lime, and aeration amount)

[0096] β=f k (Dosage of collector, dosage of frother, dosage of lime, and aeration amount)

[0097] θ = g (amount of foaming agent, amount of air),

[0098] Wherein, α represents the useful mineral correction hyperparameter, which is a parameter related to the flotation behavior of the useful mineral. Its value is determined by the amount of collector, frother, lime, and aeration. β represents the gangue correction hyperparameter, which is a parameter related to the flotation behavior of gangue minerals. Its value is affected by the amount of collector, frother, lime, and aeration. θ represents the flow rate correction hyperparameter, which is a parameter related to the characteristics of the foam itself and is determined only by the amount of frother and aeration.

[0099] Optionally, the correction hyperparameters α, β, and θ are calculated using the obtained field production data, and then a BP neural network is used to identify and model the correction hyperparameters α, β, and θ. Specifically, a three-layer BP network is used for modeling. For the correction hyperparameters α and β, the number of nodes in each layer of the network is 4-7-2, and the inputs are the amount of collector, the amount of foaming agent, the amount of lime, and the amount of aeration. For the correction hyperparameter θ, the number of nodes in each layer of the network is 2-6-1, and the inputs are the amount of foaming agent and the amount of aeration. The training of the network can be completed using relevant computing software.

[0100] 207. Based on the flotation model corresponding to the current flotation cell, calculate the floatation mass of the useful mineral and the floatation mass of the gangue in the current flotation cell using the useful mineral flotation constant and the gangue flotation constant of the current flotation cell.

[0101] In this embodiment, a useful mineral flotation model is obtained from the flotation model corresponding to the current flotation cell. The useful mineral flotation constant and the average pulp concentration in the current flotation cell are input into the useful mineral flotation model to obtain the useful mineral floatation mass of the current flotation cell. The calculation formula is as follows: Formula 11:

[0102] Formula 11: M n =K n V i C ri ,

[0103] Among them, M n K represents the mass of useful minerals that float to the surface in the current flotation cell. n V represents the flotation constant of the useful mineral in the current flotation cell. i C represents the volume of the current flotation cell. ri This indicates the average concentration of the pulp in the current flotation cell.

[0104] Obtain the gangue flotation model from the flotation model corresponding to the current flotation cell. Input the gangue flotation constant and the average pulp concentration in the current flotation cell into the gangue flotation model to obtain the gangue float mass of the current flotation cell. The calculation formula is as follows: Formula 12:

[0105] Formula 12: M m =K m V i C mi ,

[0106] Among them, M m K represents the current gangue floating quality in the flotation cell. m V represents the gangue flotation constant of the current flotation cell. i C represents the volume of the current flotation cell. ri This indicates the average concentration of the pulp in the current flotation cell.

[0107] 208. Read the tailings concentration and tailings grade of the previous flotation cell, and use the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of useful minerals floated out of the current flotation cell and the mass of gangue floated out of the current flotation cell to calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell.

[0108] In this embodiment, the concentrate grade of the current flotation cell is calculated using the mass of useful minerals and gangue floating in the current flotation cell, as shown in Formula 13 below:

[0109] Formula 13:

[0110] Among them, G j M represents the concentrate grade in the current flotation cell. j M represents the mass of useful minerals that float to the surface in the current flotation cell. m This indicates the quality of the gangue flotation in the current flotation cell.

[0111] Next, the tailings grade of the current flotation cell is calculated using the tailings flow rate, tailings concentration, and tailings grade of the previous flotation cell, as well as the mass of valuable minerals and gangue floated out of the current flotation cell. The calculation formula is shown in Formula 14 below:

[0112] Formula 14:

[0113] Among them, G i M represents the tailings grade in the current flotation cell. j M represents the mass of useful minerals that float to the surface in the current flotation cell. m Q represents the current quality of gangue floating in the flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1 G represents the tailings concentration from the previous flotation cell. i-1 This indicates the grade of the tailings from the previous flotation cell.

[0114] Subsequently, the tailings concentration of the current flotation cell is calculated using the tailings flow rate of the previous flotation cell, the tailings concentration of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of valuable minerals floated out of the current flotation cell, and the mass of gangue floated out of the current flotation cell. The calculation formula is shown in Formula 15 below:

[0115] Formula 15:

[0116] Among them, C i Q represents the current tailings concentration in the flotation cell. i Q represents the current tailings flow rate in the flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1M represents the tailings concentration from the previous flotation cell. j M represents the mass of useful minerals that float to the surface in the current flotation cell. m This indicates the quality of the gangue flotation in the current flotation cell.

[0117] In summary, a single-cell flotation model can be constructed based on steps 201-207. On this basis, multiple single-cell flotation models are connected in series to establish a multi-cell continuous flotation model, thereby forming a complete flotation process model.

[0118] Optionally, for a flotation sub-process consisting of multiple flotation cells connected in series, the output tailings of the previous flotation cell are used as the input slurry of the current flotation cell. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of each flotation cell are calculated sequentially. The output tailings of the last flotation cell in the series are used as the output tailings of the flotation sub-process. The multiple flotation sub-processes are connected in series to obtain the full flotation process. The output tailings of the previous flotation process are used as the input slurry of the current flotation sub-process. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of each flotation sub-process are calculated sequentially. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of the last flotation sub-process in the series are used as the flotation process prediction results for the full flotation process.

[0119] Optionally, for flotation processes with return material, an iterative method is used to simulate the output concentrate and tailings. Steps 201-207 are used to calculate the grade, flow rate, concentration, and other attributes of the sub-process concentrate and tailings. The mineral grade is calculated as the ratio of the amount of useful minerals floated to the total amount of ore floated. The concentrate grade and flow rate of the previous flotation cell are obtained, and the flow rate difference between the current flotation cell's concentrate flow rate and the previous flotation cell's concentrate flow rate, as well as the grade difference between the current flotation cell's concentrate grade and the previous flotation cell's concentrate grade, are calculated. Obtain preset flow difference thresholds e1 and e2. If the absolute value of the flow difference is greater than or equal to the preset flow difference thresholds and the absolute value of the grade difference is greater than or equal to the preset grade difference thresholds, then use the tailings from the current flotation cell and the tailings from the previous flotation cell as the input slurry for the current flotation cell to recalculate the tailings flow rate, concentrate flow rate, valuable mineral floated mass, gangue floated mass, tailings concentration, concentrate grade, and tailings grade of the current flotation cell. Here, e1 is set to 0.1% and e2 is set to 0.01%. If the absolute value of the flow difference is less than the preset flow difference thresholds and the absolute value of the grade difference is less than the preset grade difference thresholds, then use the results of this calculation as the output results of the concentrate and tailings for this sub-process.

[0120] Based on the above process, the schematic diagram of a copper ore flotation simulation process architecture proposed in this application embodiment is as follows:

[0121] like Figure 2B As shown, the flotation process in this flowchart (including multiple tanks connected in series such as roughing, cleaning, and scavenging) can be modeled and its indicators calculated using a flotation process prediction method driven by a hybrid mechanism and data: After pretreatment such as grinding, the raw ore enters each flotation cell. For each "current flotation cell" (such as roughing cell one, column cleaning cell one, sulfur cleaning cell, etc.), the apparent density of the slurry in the cell is first collected, and then input into the flotation slurry foam diameter model and the flotation slurry foam rising velocity model to obtain the bubble diameter and bubble rising velocity of the cell; then, the tailings flow rate of the "previous flotation cell" (such as the previous grinding output cell, roughing cell two, etc.) is obtained, and combined with the bubble diameter and bubble rising velocity of the current flotation cell, it is input into the single-cell slurry flow rate model to calculate the tailings flow rate and concentrate flow rate of the current flotation cell, realizing the transfer and distribution of flow rates between cells; based on the flotation slurry particle adsorption model, the current flotation... The bubble diameter in the flotation cell is used to calculate the flotation constants for valuable minerals (such as copper minerals) and gangue within the cell, reflecting the differences in flotation characteristics among different minerals. Next, the average pulp concentration of the current flotation cell is collected. Combined with the flotation constants for valuable minerals, gangue, and the average pulp concentration, this data is input into the flotation model to obtain the floating mass of valuable minerals and gangue in the current flotation cell, quantifying the mineral separation effect within a single cell. Finally, the tailings concentration and grade of the previous flotation cell are obtained. Combined with the tailings flow rate, floating mass of valuable minerals, and floating mass of gangue in the current flotation cell, the tailings concentration, concentrate grade (such as the grade of copper concentrate or sulfur concentrate), and tailings grade of the current flotation cell are calculated. This systematically characterizes the flotation process from roughing to cleaning and scavenging in each cell section, ultimately obtaining copper concentrate, sulfur concentrate products, and tailings that meet the specified standards, providing precise model support for the optimized control of the entire flotation process.

[0122] This application provides a flotation process prediction method based on a hybrid mechanism and data-driven approach. Compared with the prior art, this application first obtains the flotation pulp foam diameter, foam rising velocity, single-cell pulp flow rate, pulp particle adsorption, and flotation model corresponding to the current flotation cell. Then, it calculates the bubble diameter and rising velocity sequentially through the foam model. Combining the tailings flow rate and bubble parameters of the previous cell, it obtains the tailings and concentrate flow rates of the current cell from the single-cell flow model. Based on the particle adsorption model, it calculates the flotation constants of useful minerals and gangue from the bubble diameter. Through the flotation model, it calculates the mineral float-out mass from the flotation constants. Finally, it calculates the tailings concentration, concentrate grade, and tailings grade of the current cell by combining the tailings concentration and grade of the previous cell with the current cell flow rate and mineral float-out mass. This system ensures theoretical reliability through mechanistic models of foam generation and mineral particle adsorption, while also adapting to dynamic operating conditions using field data. It achieves full-process system coverage from bubble characteristics to final grade, compensating for errors caused by simplistic assumptions in pure mechanistic models. It accurately quantifies key indicators such as bubble parameters and mineral flotation volume to support operational optimization. Furthermore, it naturally adapts to complex industrial flotation processes involving multiple cells in series, including roughing, cleaning, and scavenging, effectively addressing the shortcomings of pure mechanistic or pure data models and improving the accuracy of flotation process prediction and control. This application forms a complete closed-loop process through bubble parameter calculation, slurry flow distribution, flotation constant derivation, mineral flotation quality prediction, and ultimately, tailings / concentrate grade calculation. It obtains core production indicators such as useful mineral / gangue flotation quality, tailings concentration, concentrate grade, and tailings grade, helping operators optimize processes in a timely manner, reduce useful mineral loss, improve concentrate quality, and thus increase flotation efficiency and resource recovery rate.

[0123] Furthermore, as Figure 1 In a specific implementation of the method, this application provides a flotation process prediction device based on a hybrid mechanism and data-driven approach, such as... Figure 3A As shown, the device includes: a model acquisition module 301, a bubble parameter calculation module 302, a flow distribution calculation module 303, a flotation constant calculation module 304, a mineral floatation amount calculation module 305, and a grade and concentration calculation module 306.

[0124] The model acquisition module 301 is used to acquire the flotation pulp foam diameter model, flotation pulp foam rising velocity model, single cell pulp flow rate model, flotation pulp particle adsorption model and flotation model corresponding to the current flotation cell.

[0125] The bubble parameter calculation module 302 is used to calculate the bubble diameter and bubble rising speed of the current flotation cell based on the flotation pulp foam diameter model and the flotation pulp foam rising speed model corresponding to the current flotation cell.

[0126] The flow distribution calculation module 303 is used to read the tailings flow rate of the previous flotation cell, input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single cell slurry flow rate model corresponding to the current flotation cell, and obtain the tailings flow rate and concentrate flow rate of the current flotation cell.

[0127] The flotation constant calculation module 304 is used to calculate the useful mineral flotation constant and gangue flotation constant of the current flotation cell based on the flotation pulp particle adsorption model corresponding to the current flotation cell and using the bubble diameter of the current flotation cell.

[0128] The mineral floatation calculation module 305 is used to calculate the useful mineral floatation mass and gangue floatation mass of the current flotation cell based on the flotation model corresponding to the current flotation cell and using the useful mineral flotation constant and gangue flotation constant of the current flotation cell.

[0129] The grade and concentration calculation module 306 is used to read the tailings concentration and tailings grade of the previous flotation cell, and calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell using the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, and the floated mass of useful minerals and gangue in the current flotation cell.

[0130] In specific application scenarios, the bubble parameter calculation module 302 is used to calculate the bubble diameter of the current flotation cell based on the flotation pulp foam diameter model corresponding to the current flotation cell.

[0131]

[0132]

[0133]

[0134] Among them, D b ρ represents the diameter of the bubble in the current flotation cell. f γ represents the apparent density of the pulp in the current flotation cell, γ represents the gas-liquid interfacial tension in the current flotation cell, and ε represents the average energy dissipation rate of the current flotation cell. 水 The gas-water interfacial tension is represented by 'a', the first model coefficient is represented by 'b', the second model coefficient is represented by 'c', the frother concentration in the current flotation cell is represented by 'n', and the impeller speed in the current flotation cell is represented by 'n'. min n represents the minimum impeller speed of the current flotation cell. maxThe maximum impeller rotation speed of the current flotation cell is represented by α, and the third model coefficient is represented by a and b, where a and b are determined based on the type of frother in the current flotation cell, and α is determined based on the impeller structure of the current flotation cell. The flotation pulp foam diameter model is constructed based on the operating data of the current flotation cell and the flotation pulp foam generation mechanism. The bubble rising velocity of the current flotation cell is calculated based on the flotation pulp foam rising velocity model corresponding to the current flotation cell.

[0135]

[0136] Among them, U b ρ represents the rising velocity of the bubbles in the current flotation cell. f ρ represents the apparent density of the slurry in the current flotation cell. g denoted by air density, g represents gravitational acceleration, and μ represents the viscosity of the slurry in the current flotation cell. The flotation slurry foam rise velocity model is constructed based on the operating data of the current flotation cell and the flotation slurry flow mechanism.

[0137] In specific application scenarios, the flow distribution calculation module 303 is used to input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single-cell slurry flow rate model corresponding to the current flotation cell, so as to obtain the tailings flow rate of the current flotation cell.

[0138]

[0139] Among them, Q i Q represents the tailings flow rate of the current flotation cell. i-1 q represents the tailings flow rate of the previous flotation cell. i The aeration rate of the current flotation cell is represented by t, the bubble rise time of the current flotation cell is represented by r, and the bubble radius of the current flotation cell is represented by D. b The bubble diameter of the current flotation cell is represented by d, the bubble thickness of the current flotation cell is represented by H, and the height of the current flotation cell is represented by U. b The bubble rising speed of the current flotation cell is represented by the single-cell slurry flow rate model, which is constructed based on the operating data and overall balance mechanism of the current flotation cell. The difference between the tailings flow rate of the previous flotation cell and the tailings flow rate of the current flotation cell is taken as the concentrate flow rate of the current flotation cell.

[0140] In specific application scenarios, the flotation constant calculation module 304 is used to obtain the flotation pulp particle adsorption model corresponding to the current flotation cell.

[0141]

[0142] Where K represents the flotation constant of the current flotation cell, and P a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, where Re represents the bubble Reynolds number, and D represents the apparent gas velocity in the current flotation cell. b This represents the bubble diameter of the current flotation cell. The flotation pulp particle adsorption model is constructed based on the operating data of the current flotation cell and the collision and adhesion mechanism between particles and bubbles. For the flotation process of useful minerals, useful mineral correction hyperparameters are obtained. These useful mineral correction hyperparameters are used to adjust the flotation pulp particle adsorption model corresponding to the current flotation cell, resulting in the useful mineral flotation pulp particle adsorption model for the current flotation cell. The bubble diameter of the current flotation cell is input into the useful mineral flotation pulp particle adsorption model for the current flotation cell to obtain the useful mineral flotation constant for the current flotation cell.

[0143]

[0144] Among them, K n P represents the useful mineral flotation constant of the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, and the bubble Reynolds number is represented by D. b The bubble diameter of the current flotation cell is represented by α, and the useful mineral correction hyperparameter is represented by α. α is calculated by inputting the collector dosage, frother dosage, lime dosage, and aeration rate of the current flotation cell into a BP neural network. For the gangue flotation process, the gangue correction hyperparameter is obtained, and the gangue pulp particle adsorption model corresponding to the current flotation cell is adjusted using the gangue correction hyperparameter to obtain the gangue flotation pulp particle adsorption model corresponding to the current flotation cell. The bubble diameter of the current flotation cell is input into the gangue flotation pulp particle adsorption model corresponding to the current flotation cell to obtain the gangue flotation constant of the current flotation cell.

[0145]

[0146] Among them, K m P represents the gangue flotation constant of the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, and the bubble Reynolds number is represented by D. bβ represents the bubble diameter of the current flotation cell, and β represents the gangue correction hyperparameter. β is calculated by inputting the collector dosage, the frother dosage, the lime dosage, and the current aeration volume of the flotation cell into the BP neural network.

[0147] In specific application scenarios, the flotation constant calculation module 304 is used to obtain the induction time model corresponding to the current flotation cell. The bubble diameter and bubble rising velocity of the current flotation cell are input into the induction time model to obtain the induction time of the current flotation cell.

[0148]

[0149]

[0150] φ cr =0.5φ c ,

[0151]

[0152] N = -0.437Re 1.0562 ,

[0153]

[0154] Where T represents the sensing time of the current flotation cell, and D p D represents the diameter of the mineral particles in the current flotation cell. b U represents the diameter of the bubble in the current flotation cell. b The value represents the rising velocity of the bubbles in the current flotation cell, A represents the first induction time model coefficient, B represents the second induction time model coefficient, and φ represents the rising velocity of the bubbles in the current flotation cell. c φ represents the collision angle. cr Indicates the capture angle, V p The following values ​​represent the settling velocity of the mineral particles in the current flotation cell: X represents the first hydrodynamic coefficient, Y represents the second hydrodynamic coefficient, Re represents the bubble Reynolds number, M represents the third hydrodynamic coefficient, N represents the fourth hydrodynamic coefficient, C represents the comprehensive coefficient, and ρ represents the total settling velocity. f ρ represents the apparent density of the slurry in the current flotation cell. g Let represent air density, r represent the bubble radius of the current flotation cell, and g represent gravitational acceleration. The fluid dynamic viscosity coefficient is represented; the probability model corresponding to the current flotation cell is obtained by inputting the induction time, bubble diameter, and bubble rising velocity of the current flotation cell into the probability model corresponding to the current flotation cell to obtain the adhesion success probability of the current flotation cell.

[0155]

[0156] Among them, P a The value represents the probability of successful adhesion in the current flotation cell, δ represents the probability model coefficient, Re represents the bubble Reynolds number, and U represents the bubble number. b D represents the rising speed of the bubbles in the current flotation cell, T represents the induction time of the current flotation cell, and D represents the rising speed of the bubbles in the current flotation cell. b D represents the diameter of the bubble in the current flotation cell. p The particle diameter of the current flotation cell is represented. Based on the collision and adhesion mechanism between the particles and bubbles, the adsorption model of the flotation pulp particles corresponding to the current flotation cell is constructed using the adhesion success probability of the current flotation cell, the particle diameter of the current flotation cell, the apparent gas velocity of the current flotation cell, the bubble diameter of the current flotation cell, and the bubble Reynolds number.

[0157] In specific application scenarios, the mineral flotation calculation module 305 is used to obtain the useful mineral flotation model from the flotation model corresponding to the current flotation cell, input the useful mineral flotation constant of the current flotation cell into the useful mineral flotation model, and obtain the useful mineral flotation mass of the current flotation cell.

[0158] M n =K n V i C ri ,

[0159] Among them, M n K represents the mass of useful minerals floated out of the current flotation cell. n V represents the useful mineral flotation constant of the current flotation cell. i C represents the volume of the current flotation cell. ri This represents the average pulp concentration of the current flotation cell; the gangue flotation model is obtained from the flotation model corresponding to the current flotation cell, and the gangue flotation constant of the current flotation cell is input into the gangue flotation model to obtain the gangue float mass of the current flotation cell.

[0160] M m =K m V i C mi ,

[0161] Among them, M m K represents the mass of gangue floated out of the current flotation cell. m V represents the gangue flotation constant of the current flotation cell. i C represents the volume of the current flotation cell. ri This indicates the average concentration of the pulp in the current flotation cell.

[0162] In specific application scenarios, the grade and concentration calculation module 306 is used to calculate the concentrate grade of the current flotation cell using the floated mass of useful minerals and the floated mass of gangue in the current flotation cell.

[0163]

[0164] Among them, G j M represents the concentrate grade of the current flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m This indicates the gangue flotation mass of the current flotation cell; the tailings grade of the current flotation cell is calculated using the tailings flow rate, tailings concentration, and tailings grade of the previous flotation cell, as well as the useful mineral flotation mass and gangue flotation mass of the current flotation cell.

[0165]

[0166] Among them, G i M represents the tailings grade of the current flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m Q represents the mass of gangue floated out of the current flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1 G represents the tailings concentration of the previous flotation cell. i-1 This indicates the tailings grade of the previous flotation cell; the tailings concentration of the current flotation cell is calculated using the tailings flow rate of the previous flotation cell, the tailings concentration of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of valuable minerals floated out of the current flotation cell, and the mass of gangue floated out of the current flotation cell.

[0167]

[0168] Among them, C i Q represents the tailings concentration in the current flotation cell. i Q represents the tailings flow rate of the current flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1 M represents the tailings concentration of the previous flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m This indicates the quality of the gangue flotation in the current flotation cell.

[0169] In specific application scenarios, such as Figure 3B As shown, the device also includes: a return material flotation process calculation module 307 and a flotation full process calculation module 308.

[0170] The return material flotation process calculation module 307 is used for flotation processes with return material to obtain the concentrate grade and concentrate flow rate of the previous flotation cell, calculate the flow rate difference between the concentrate flow rate of the current flotation cell and the concentrate flow rate of the previous flotation cell, and the grade difference between the concentrate grade of the current flotation cell and the concentrate grade of the previous flotation cell; obtain preset flow rate difference thresholds and preset grade difference thresholds; if the absolute value of the flow rate difference is greater than or equal to the preset flow rate difference threshold and the absolute value of the grade difference is greater than or equal to the preset grade difference threshold, then the tailings of the current flotation cell and the tailings of the previous flotation cell are used as the input slurry of the current flotation cell to recalculate the tailings flow rate, concentrate flow rate, useful mineral floated mass, gangue floated mass, tailings concentration, concentrate grade and tailings grade of the current flotation cell.

[0171] The flotation process calculation module 308 is used to calculate the tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of each flotation cell in a series of flotation sub-processes, taking the output tailings of the previous flotation cell as the input slurry of the current flotation cell. The output tailings of the last flotation cell in the series are used as the output tailings of the flotation sub-process. The multiple flotation processes are then connected in series to obtain the flotation process... For the entire process, the output tailings of the previous flotation sub-process are used as the input slurry for the current flotation sub-process. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of each flotation sub-process are calculated sequentially. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of the last flotation sub-process in the series are used as the flotation process prediction result for the entire flotation process.

[0172] This application provides an apparatus that, compared with the prior art, first obtains the flotation pulp foam diameter, foam rising speed, single-cell pulp flow rate, pulp particle adsorption, and flotation model corresponding to the current flotation cell. Then, it calculates the bubble diameter and rising speed sequentially through the foam model. Combining the tailings flow rate and bubble parameters of the previous cell, it obtains the tailings and concentrate flow rates of the current cell from the single-cell flow model. Based on the particle adsorption model, it calculates the flotation constants of useful minerals and gangue from the bubble diameter. Through the flotation model, it calculates the mineral float-out mass from the flotation constants. Finally, it calculates the tailings concentration, concentrate grade, and tailings grade of the current cell by combining the tailings concentration and grade of the previous cell with the current cell flow rate and mineral float-out mass. This system ensures theoretical reliability through mechanistic models of foam generation and mineral particle adsorption, while also adapting to dynamic operating conditions using field data. It achieves full-process system coverage from bubble characteristics to final grade, compensating for errors caused by simplistic assumptions in pure mechanistic models. It accurately quantifies key indicators such as bubble parameters and mineral flotation volume to support operational optimization. Furthermore, it naturally adapts to complex industrial flotation processes involving multiple cells in series, including roughing, cleaning, and scavenging, effectively addressing the shortcomings of pure mechanistic or pure data models and improving the accuracy of flotation process prediction and control. This application forms a complete closed-loop process through bubble parameter calculation, slurry flow distribution, flotation constant derivation, mineral flotation quality prediction, and ultimately, tailings / concentrate grade calculation. It obtains core production indicators such as useful mineral / gangue flotation quality, tailings concentration, concentrate grade, and tailings grade, helping operators optimize processes in a timely manner, reduce useful mineral loss, improve concentrate quality, and thus increase flotation efficiency and resource recovery rate.

[0173] It should be noted that other corresponding descriptions of the functional units involved in the flotation process prediction device based on mechanism and data hybrid driving provided in this application embodiment can be found in the following references. Figure 1 and Figures 2A to 2B The corresponding descriptions in [the document] will not be repeated here.

[0174] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0175] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0176] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

[0177] In an exemplary embodiment, see Figure 4 The invention also provides a device comprising a bus, a processor, a memory, and a communication interface, and may further include an input / output interface and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores a computer program, and the processor executes the program stored in the memory to perform the mechanism- and data-driven flotation process prediction method described in the above embodiments.

[0178] A medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the flotation process prediction method based on a hybrid mechanism and data-driven approach.

[0179] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented in hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0180] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0181] Those skilled in the art will understand that the modules in the apparatus of the implementation scenario can be distributed within the apparatus of the implementation scenario as described, or they can be located in one or more apparatuses different from this implementation scenario, with corresponding changes. The modules of the above-described implementation scenario can be combined into one module, or they can be further divided into multiple sub-modules.

[0182] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenario.

[0183] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A flotation process prediction method based on a hybrid mechanism and data-driven approach, characterized in that, include: Obtain the flotation pulp foam diameter model, flotation pulp foam rise velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell; The bubble diameter and bubble rise velocity of the current flotation cell are calculated based on the flotation pulp foam diameter model and flotation pulp foam rise velocity model corresponding to the current flotation cell. Read the tailings flow rate of the previous flotation cell, and input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate and concentrate flow rate of the current flotation cell. Based on the flotation pulp particle adsorption model corresponding to the current flotation cell, the useful mineral flotation constant and gangue flotation constant of the current flotation cell are calculated using the bubble diameter of the current flotation cell. Based on the flotation model corresponding to the current flotation cell, the useful mineral flotation constant and gangue flotation constant of the current flotation cell are used to calculate the useful mineral flotation mass and gangue flotation mass of the current flotation cell. The tailings concentration and tailings grade of the previous flotation cell are read, and the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of useful minerals floated out and the mass of gangue floated out of the current flotation cell are used to calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell.

2. The method according to claim 1, characterized in that, The calculation of the bubble diameter and bubble rise velocity in the current flotation cell based on the flotation pulp foam diameter model and the flotation pulp foam rise velocity model corresponding to the current flotation cell includes: The bubble diameter of the current flotation cell is calculated based on the flotation pulp foam diameter model corresponding to the current flotation cell. c = c 水 [1-aln(1+c / b)], Among them, D b ρ represents the diameter of the bubble in the current flotation cell. f γ represents the apparent density of the pulp in the current flotation cell, γ represents the gas-liquid interfacial tension in the current flotation cell, and ε represents the average energy dissipation rate of the current flotation cell. 水 The gas-water interfacial tension is represented by 'a', the first model coefficient is represented by 'b', the second model coefficient is represented by 'c', the frother concentration in the current flotation cell is represented by 'n', and the impeller speed in the current flotation cell is represented by 'n'. min n represents the minimum impeller speed of the current flotation cell. max The maximum rotational speed of the impeller in the current flotation cell is represented by α, which represents the third model coefficient. Here, a and b are determined based on the type of frother in the current flotation cell, and α is determined based on the impeller structure of the current flotation cell. The flotation pulp foam diameter model is constructed based on the operating data of the current flotation cell and the flotation pulp foam generation mechanism. The bubble rise velocity of the current flotation cell is calculated based on the flotation pulp foam rise velocity model corresponding to the current flotation cell. Among them, U b ρ represents the rising velocity of the bubbles in the current flotation cell. f ρ represents the apparent density of the slurry in the current flotation cell. g denoted by air density, g represents gravitational acceleration, and μ represents the viscosity of the slurry in the current flotation cell. The flotation slurry foam rise velocity model is constructed based on the operating data of the current flotation cell and the flotation slurry flow mechanism.

3. The method according to claim 1, characterized in that, The step of inputting the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rise velocity of the current flotation cell into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate and concentrate flow rate of the current flotation cell includes: The tailings flow rate of the previous flotation cell, the bubble diameter and bubble rise velocity of the current flotation cell are input into the single-cell slurry flow rate model corresponding to the current flotation cell to obtain the tailings flow rate of the current flotation cell. Among them, Q i Q represents the tailings flow rate of the current flotation cell. i-1 q represents the tailings flow rate of the previous flotation cell. i The aeration rate of the current flotation cell is represented by t, the bubble rise time of the current flotation cell is represented by r, and the bubble radius of the current flotation cell is represented by D. b The bubble diameter of the current flotation cell is represented by d, the bubble thickness of the current flotation cell is represented by H, and the height of the current flotation cell is represented by U. b This represents the bubble rising speed in the current flotation cell, and the single-cell slurry flow rate model is constructed based on the operating data and overall balance mechanism of the current flotation cell. The difference between the tailings flow rate of the previous flotation cell and the tailings flow rate of the current flotation cell is taken as the concentrate flow rate of the current flotation cell.

4. The method according to claim 1, characterized in that, The calculation of the useful mineral flotation constant and gangue flotation constant of the current flotation cell based on the flotation pulp particle adsorption model corresponding to the current flotation cell and using the bubble diameter of the current flotation cell includes: Obtain the flotation pulp particle adsorption model corresponding to the current flotation cell. Where K represents the flotation constant of the current flotation cell, and P a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, where Re represents the bubble Reynolds number, and D represents the apparent gas velocity in the current flotation cell. b This indicates the bubble diameter in the current flotation cell. The flotation slurry particle adsorption model is constructed based on the operating data of the current flotation cell and the collision and adhesion mechanism between the particles and the bubbles. For the flotation process of valuable minerals, the useful mineral correction hyperparameter is obtained. This useful mineral correction hyperparameter is then used to adjust the flotation pulp particle adsorption model corresponding to the current flotation cell, resulting in the useful mineral flotation pulp particle adsorption model for the current flotation cell. The bubble diameter of the current flotation cell is then input into the useful mineral flotation pulp particle adsorption model for the current flotation cell to obtain the useful mineral flotation constant for the current flotation cell. Among them, K n P represents the useful mineral flotation constant of the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, and the bubble Reynolds number is represented by D. b The bubble diameter of the current flotation cell is represented by α, and the useful mineral correction hyperparameter is represented by α. α is calculated by inputting the collector dosage, frother dosage, lime dosage, and the aeration volume of the current flotation cell into a BP neural network. For the gangue flotation process, gangue correction hyperparameters are obtained. These hyperparameters are then used to adjust the particle adsorption model of the flotation pulp corresponding to the current flotation cell, resulting in the gangue flotation pulp particle adsorption model for the current flotation cell. The bubble diameter of the current flotation cell is then input into the gangue flotation pulp particle adsorption model for the current flotation cell to obtain the gangue flotation constant for the current flotation cell. Among them, K m P represents the gangue flotation constant of the current flotation cell. a D represents the probability of successful adhesion in the current flotation cell. p V represents the diameter of the mineral particles in the current flotation cell. g The apparent gas velocity in the current flotation cell is represented by Re, and the bubble Reynolds number is represented by D. b β represents the bubble diameter of the current flotation cell, and β represents the gangue correction hyperparameter. β is calculated by inputting the collector dosage, the frother dosage, the lime dosage, and the current aeration volume of the flotation cell into the BP neural network.

5. The method according to claim 4, characterized in that, The method further includes: Obtain the induction time model corresponding to the current flotation cell, and input the bubble diameter and bubble rising speed of the current flotation cell into the induction time model corresponding to the current flotation cell to obtain the induction time of the current flotation cell. N=-0.437Re 1.0562 , Where T represents the sensing time of the current flotation cell, and D p D represents the diameter of the mineral particles in the current flotation cell. b U represents the diameter of the bubble in the current flotation cell. b The value represents the rising velocity of the bubbles in the current flotation cell, A represents the first induction time model coefficient, B represents the second induction time model coefficient, and φ represents the rising velocity of the bubbles in the current flotation cell. c φ represents the collision angle. cr Indicates the capture angle, V p The following values ​​represent the settling velocity of the mineral particles in the current flotation cell: X represents the first hydrodynamic coefficient, Y represents the second hydrodynamic coefficient, Re represents the bubble Reynolds number, M represents the third hydrodynamic coefficient, N represents the fourth hydrodynamic coefficient, C represents the comprehensive coefficient, and ρ represents the total settling velocity. f ρ represents the apparent density of the slurry in the current flotation cell. g The value represents the air density, r represents the bubble radius of the current flotation cell, g represents the gravitational acceleration, and θ represents the hydrodynamic viscosity coefficient. Obtain the probability model corresponding to the current flotation cell, and input the induction time, bubble diameter, and bubble rising speed of the current flotation cell into the probability model corresponding to the current flotation cell to obtain the adhesion success probability of the current flotation cell. Among them, P a The value represents the probability of successful adhesion in the current flotation cell, δ represents the probability model coefficient, Re represents the bubble Reynolds number, and U represents the bubble number. b D represents the rising speed of the bubbles in the current flotation cell, T represents the induction time of the current flotation cell, and D represents the rising speed of the bubbles in the current flotation cell. b D represents the diameter of the bubble in the current flotation cell. p This indicates the diameter of the mineral particles in the current flotation cell; Based on the collision and adhesion mechanism between the mineral particles and bubbles, an adsorption model for the flotation pulp particles corresponding to the current flotation cell is constructed using the adhesion success probability of the current flotation cell, the diameter of the mineral particles in the current flotation cell, the apparent gas velocity of the current flotation cell, the bubble diameter of the current flotation cell, and the bubble Reynolds number.

6. The method according to claim 1, characterized in that, The calculation of the useful mineral floatation mass and gangue floatation mass in the current flotation cell, based on the flotation model corresponding to the current flotation cell and using the useful mineral flotation constant and gangue flotation constant of the current flotation cell, includes: Obtain the useful mineral flotation model from the flotation model corresponding to the current flotation cell, input the useful mineral flotation constant of the current flotation cell into the useful mineral flotation model, and obtain the useful mineral float-out mass of the current flotation cell. M n =K n V i C ri , Among them, M n K represents the mass of useful minerals floated out of the current flotation cell. n V represents the useful mineral flotation constant of the current flotation cell. i C represents the volume of the current flotation cell. ri This indicates the average concentration of the pulp in the current flotation cell; Obtain the gangue flotation model from the flotation model corresponding to the current flotation cell, input the gangue flotation constant of the current flotation cell into the gangue flotation model, and obtain the gangue floatation mass of the current flotation cell. M m =K m V i C mi , Among them, M m K represents the mass of gangue floated out of the current flotation cell. m V represents the gangue flotation constant of the current flotation cell. i C represents the volume of the current flotation cell. ri This indicates the average concentration of the pulp in the current flotation cell.

7. The method according to claim 1, characterized in that, The calculation of the tailings concentration, concentrate grade, and tailings grade of the current flotation cell using the tailings flow rate, tailings concentration, and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, and the mass of valuable minerals and gangue floated out of the current flotation cell includes: The concentrate grade of the current flotation cell is calculated using the mass of useful minerals and gangue floated out of the current flotation cell. Among them, G j M represents the concentrate grade of the current flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m This indicates the mass of gangue that floats to the surface in the current flotation cell; The tailings grade of the current flotation cell is calculated using the tailings flow rate, tailings concentration, and tailings grade of the previous flotation cell, as well as the mass of valuable minerals and gangue floated out of the current flotation cell. Among them, G i M represents the tailings grade of the current flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m Q represents the mass of gangue floated out of the current flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1 G represents the tailings concentration of the previous flotation cell. i-1 This indicates the tailings grade of the previous flotation cell; The tailings concentration of the current flotation cell is calculated using the tailings flow rate of the previous flotation cell, the tailings concentration of the previous flotation cell, the tailings flow rate of the current flotation cell, the mass of valuable minerals floated out of the current flotation cell, and the mass of gangue floated out of the current flotation cell. Among them, C i Q represents the tailings concentration in the current flotation cell. i Q represents the tailings flow rate of the current flotation cell. i-1 C represents the tailings flow rate of the previous flotation cell. i-1 M represents the tailings concentration of the previous flotation cell. j M represents the mass of useful minerals floated out of the current flotation cell. m This indicates the quality of the gangue flotation in the current flotation cell.

8. The method according to claim 1, characterized in that, The method further includes: For a flotation process with return material, obtain the concentrate grade and concentrate flow rate of the previous flotation cell, calculate the flow rate difference between the concentrate flow rate of the current flotation cell and the concentrate flow rate of the previous flotation cell, and the grade difference between the concentrate grade of the current flotation cell and the concentrate grade of the previous flotation cell. Obtain a preset flow rate difference threshold and a preset grade difference threshold. If the absolute value of the flow rate difference is greater than or equal to the preset flow rate difference threshold and the absolute value of the grade difference is greater than or equal to the preset grade difference threshold, then use the tailings of the current flotation cell and the tailings of the previous flotation cell as the input slurry of the current flotation cell to recalculate the tailings flow rate, concentrate flow rate, useful mineral float-out mass, gangue float-out mass, tailings concentration, concentrate grade, and tailings grade of the current flotation cell.

9. The method according to claim 1, characterized in that, The method further includes: For a flotation subprocess consisting of multiple flotation cells connected in series, the output tailings of the previous flotation cell are used as the input slurry of the current flotation cell. The tailings flow rate, concentrate flow rate, useful mineral float-out mass, gangue float-out mass, tailings concentration, concentrate grade, and tailings grade of each flotation cell are calculated sequentially. The output tailings of the last flotation cell in the series are used as the output tailings of the flotation subprocess. Multiple flotation sub-processes are connected in series to obtain the complete flotation process. The output tailings of the previous flotation sub-process are used as the input slurry of the current flotation sub-process. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of each flotation sub-process are calculated sequentially. The tailings flow rate, concentrate flow rate, useful mineral float mass, gangue float mass, tailings concentration, concentrate grade, and tailings grade of the last flotation sub-process in the series are used as the flotation process prediction result of the complete flotation process.

10. A flotation process prediction device based on a hybrid mechanism and data-driven approach, characterized in that, include: The model acquisition module is used to acquire the flotation pulp foam diameter model, flotation pulp foam rising velocity model, single-cell pulp flow rate model, flotation pulp particle adsorption model, and flotation model corresponding to the current flotation cell. The bubble parameter calculation module is used to calculate the bubble diameter and bubble rising speed of the current flotation cell based on the flotation pulp foam diameter model and flotation pulp foam rising speed model corresponding to the current flotation cell. The flow distribution calculation module is used to read the tailings flow rate of the previous flotation cell, input the tailings flow rate of the previous flotation cell, the bubble diameter and bubble rising speed of the current flotation cell into the single cell slurry flow rate model corresponding to the current flotation cell, and obtain the tailings flow rate and concentrate flow rate of the current flotation cell. The flotation constant calculation module is used to calculate the useful mineral flotation constant and gangue flotation constant of the current flotation cell based on the flotation pulp particle adsorption model corresponding to the current flotation cell and using the bubble diameter of the current flotation cell. The mineral floatation calculation module is used to calculate the useful mineral floatation mass and gangue floatation mass of the current flotation cell based on the flotation model corresponding to the current flotation cell and using the useful mineral flotation constant and gangue flotation constant of the current flotation cell. The grade and concentration calculation module is used to read the tailings concentration and tailings grade of the previous flotation cell, and calculate the tailings concentration, concentrate grade and tailings grade of the current flotation cell using the tailings flow rate, tailings concentration and tailings grade of the previous flotation cell, the tailings flow rate of the current flotation cell, and the floated mass of useful minerals and gangue in the current flotation cell.

Images