Method and system for optimizing wet ball mill grinding medium supplement and industrial controller readable storage medium

Abstract

The invention relates to a method and system for optimizing wet ball mill grinding medium supplement and an industrial controller readable storage medium. The invention provides a method for optimizing the supplement of a grinding medium of a mill. The invention further provides an industrial controller readable storage medium. The invention also provides a system for optimizing the supplement of the grinding medium of the mill.

Description

TECHNICAL FIELD

[0001] The present invention relates to the field of control systems specifically for comminution processes. In particular, the present invention relates to a method and system for optimizing wet ball mill grinding media replenishment. BACKGROUND

[0002] Comminution plays a vital role in mineral processing, aiming to achieve three main objectives: adjusting the material size to meet the requirements of subsequent processes, increasing the surface area of particles, and meeting the size requirements for direct sales.

[0003] The energy consumed in comminution and the mechanical load involved in the process have a significant impact on the operating and capital costs of crushing and grinding circuits. Therefore, improving the energy efficiency of comminution operations helps to improve economic efficiency and reduce global energy demand.

[0004] Given the high cost and high energy consumption of the grinding process, the industry urgently needs to find effective ways to reduce energy consumption per unit of feed. Autogenous grinding (AG), semi-autogenous grinding (SAG), and ball mills have been widely used for this purpose.

[0005] The operating performance of such mills is influenced by a number of operating parameters, of which the filling rate is one of the most critical parameters. The parameter represents the ratio of the volume of material load in the mill to the effective volume of the mill. The parameter not only affects the operating performance and energy consumption of the mill, but also determines the particle size of the final product.

[0006] Therefore, controlling the filling parameter of the rotary mill within the ideal range is crucial for optimizing its energy consumption, as the parameter determines the maximum production capacity of the equipment.

[0007] A number of methods and systems for optimizing mill operating parameters have been disclosed in the prior art. Patent document WO2023242752(A1) discloses a system and method for real-time optimization of beneficiation processes, aiming to reduce equipment downtime, improve operating accuracy, optimize efficiency, and minimize energy consumption and grinding media loss. The method is based on historical data, real-time operating data, and laboratory data, combined with a special model established for the grinding system, using a feed prediction model, a grinding media load model, a drive system model, and a model of the movement of material in the mill, aiming to reduce the variability of the final product particle size. For the replenishment of grinding media, a wear prediction model based on data such as the movement trajectory of steel balls, mill power, equipment size, physical properties of grinding media, and discharge systems is used to estimate the filling rate and make replenishment recommendations. However, the method does not use a prediction model to plan the replenishment of grinding media in the mill within a shift.

[0008] Patent document CN112588424(B) discloses a control method for a ball mill grinding system based on a cloud-based intelligent model. This document aims to provide a control method that utilizes artificial intelligence and fuzzy control technology to achieve automated operation control of the equipment. The method includes a local controller and an intelligent model configured with an output conversion algorithm. The controller analyzes the system's output signals and adjusts the input parameters accordingly. The system employs multiple output signals and conversion algorithms to achieve the target output, including powder output. The adjustment amount of the local controller serves as the system input, while the status monitoring signal serves as the output; the input vector includes hot air, circulating air, and coal feed rate, while the output vector includes outlet temperature, inlet negative pressure, pressure difference, and mill load. The powder output is determined based on the above parameters using the output conversion algorithm. However, the document does not employ a fuzzy controller based on laboratory data and specific energy consumption to generate the optimal power setpoint for the mill within a planned time window.

[0009] Patent document WO2007110466 (A1) discloses a method, apparatus, and computer program for estimating the volume percentage of steel balls in the ore material within a mill. Preferably, the invention relates to a semi-autogenous grinding mill (SAG). The document employs an extended Kalman filter to estimate the ball charge mass based on measured process data and a process model. The document combines precise measurements of the mill filling rate with another type of measurement signal that depends on both the filling rate and the ball charge mass. By inputting this information into the extended Kalman filter algorithm, a precise estimate of the ball charge mass can be calculated. However, the document neither explicitly employs a fuzzy controller nor discloses a model for planning grinding media replenishment within a single work shift.

[0010] Although there is existing literature on mill operation optimization, none of them disclose a method for determining the required power of a mill using a model, nor do they disclose a complete technical solution for determining the timing and appropriate replenishment of grinding media for each mill during a single work shift.

[0011] Therefore, there is an urgent need for a method for optimizing mill operation, which estimates the ideal power and instantaneous power of the mill to coordinate the replenishment sequence of grinding media for each mill within a work shift. Summary of the Invention

[0012] The purpose of this invention is to provide a system that implements a method for optimizing the replenishment of grinding media in a wet ball mill. The method can determine the required power of the mill using a fuzzy controller; construct a model based on operating parameters such as feed rate, slurry solids content, and discharge flow rate to estimate the mill power; and determine the timing and appropriate amount of grinding media replenishment for each mill using a metaheuristic algorithm.

[0013] Another object of the present invention is to provide a storage medium storing computer program instructions executable by an industrial controller for implementing an optimized method for replenishing grinding media in a ball mill.

[0014] This invention discloses a method for optimizing the replenishment of grinding media in a wet ball mill, comprising the following steps: determining the required power value of at least one mill using a fuzzy controller; estimating the power of the at least one mill using a mill power estimation model based on operating data; determining the grinding media replenishment parameters of the at least one mill using a grinding media replenishment decision module; and performing an automatic replenishment operation on the at least one mill according to the grinding media replenishment parameters using an automatic grinding media replenishment device.

[0015] The present invention also discloses an industrial controller-readable storage medium including instructions that, when executed by at least one processor, cause the at least one processor to perform various steps of a method for optimizing the replenishment of grinding media in a ball mill.

[0016] The present invention also provides a system for optimizing the replenishment of grinding media in a mill, the system comprising: a grinding media replenishment device; at least one mill connectable to the grinding media replenishment device; and an industrial controller readable storage medium. Attached Figure Description

[0017] Brief descriptions of each attached figure are as follows:

[0018] Figure 1 —A schematic diagram of a system for optimizing the replenishment of grinding media in a mill;

[0019] Figure 2 —The graph shows a curve showing the correlation between specific energy and the percentage of material passed through a +0.15mm grinding screen;

[0020] Figure 3 —A schematic diagram of the structural configuration of the fuzzy controller;

[0021] Figure 4 —The charts show (a) the trends of actual and estimated power over time during the grinding stage and (b) the correlation between actual and estimated power models; the charts also show (c) the trends of actual and estimated power over time during the regrinding stage and (d) the correlation between actual and estimated power models.

[0022] Figure 5 —The charts show the correlation between (a) mill filling rate and motor power, and (b) filling rate and load mass;

[0023] Figure 6—The chart shows the power variation trend of the mill under the conditions of feeding and no feeding. Detailed Implementation

[0024] The following description is based on a preferred embodiment of the present invention, applied to a method and system for optimizing the replenishment of grinding media in a ball mill, and an industrial controller-readable storage medium.

[0025] This invention provides a method for optimizing the replenishment of grinding media in a mill, comprising the following steps: determining the required power value of at least one mill using a fuzzy controller; estimating the power of the at least one mill using a mill power estimation device based on operating data; determining the grinding media replenishment parameters of the at least one mill using a grinding media replenishment decision module; and performing an automatic replenishment operation on the at least one mill according to the grinding media replenishment parameters using an automatic grinding media replenishment device.

[0026] The present invention also discloses an industrial controller-readable storage medium, including instructions that, when executed by at least one processor, cause the at least one processor to perform steps of a method for optimizing the replenishment of grinding media in a ball mill, including: determining the required power value of at least one mill by a fuzzy controller; estimating the power of the at least one mill based on operating data of the at least one mill by a mill power estimation device; determining the grinding media replenishment parameters of the at least one mill by a grinding media replenishment decision device; and performing an automatic replenishment operation on the at least one mill according to the grinding media replenishment parameters by an automatic grinding media replenishment device.

[0027] The present invention also provides a system for optimizing the replenishment of grinding media in a mill, the system comprising: a grinding media replenishment device; at least one mill connectable to the grinding media replenishment device; and an industrial controller readable storage medium including instructions, which, when executed by at least one processor, cause the at least one processor to perform steps of a method for optimizing the replenishment of grinding media in a mill, including: determining a required power value for at least one mill using a fuzzy controller; estimating the power of the at least one mill based on operating data of the at least one mill using a mill power estimation device; determining grinding media replenishment parameters for the at least one mill using a grinding media replenishment decision device; and performing an automatic replenishment operation on the at least one mill based on the grinding media replenishment parameters using an automatic grinding media replenishment device; wherein the automatic grinding media replenishment device is further configured to control the grinding media replenishment device.

[0028] Figure 1 A system for optimizing the replenishment of grinding media in mills is shown. The system includes a grinding media replenishment device 50 and at least one mill M1, M2, M3 that can be connected to the grinding media replenishment device 50.n The system also includes an industrial controller-readable storage medium comprising instructions that, when executed by at least one processor, cause the at least one processor to perform steps for optimizing a method of replenishing grinding media in mills, including: determining at least one mill M1, M2, M... n The required power value; based on at least one mill M1, M2, M... using a mill power estimation device. n The operating data estimates the at least one mill M1, M2, M n The power of the at least one mill M1, M2, and M3 is determined by the grinding media replenishment decision device 30. n The grinding media replenishment parameters; and the grinding media replenishment device 40 replenishing the at least one mill M1, M2, M3 based on the grinding media replenishment parameters. n Perform automatic replenishment.

[0029] In the embodiment described above, the automatic grinding media replenishment device 40 is further configured to control the grinding media replenishment device 50.

[0030] Preferably, at least one mill M1, M2, M is selected. n The steps to determine the required power values ​​include constructing at least one mill M1, M2, M... n Energy consumption prediction model.

[0031] More preferably, the energy consumption prediction model is based on the method developed by Donda J., Galinari C., and Rabelo P. in "Energy Efficiency Control of Pre-grinding and Primary Grinding Loops at Samarco Mining Company," published in "The Second Brazilian Iron Ore Symposium - ABM," Orupreto, 1999, pp. 144-150, with the aim of estimating unit energy consumption in primary grinding and regrinding processes.

[0032] This method involves conducting grinding tests at different time periods, which are then converted into unit energy consumption values, such as... Figure 2 As shown. For this transformation, the model considers parameters including mill diameter, filling rate, and critical speed fraction.

[0033] In the embodiment described, at least one mill M1, M2, M n The energy consumption prediction model includes formula (1):

[0034]

[0035] in:

[0036] E represents the energy consumed per unit mass of feed in the loop on the pinion shaft (kWh / t);

[0037] R f To control the percentage of material on the screen within the target particle size range;

[0038] R p To control the percentage of material on the screen within the target particle size range on the screen for product control;

[0039] K is a characteristic parameter of the ore (t / kWh).

[0040] according to Figure 2 The data shown indicates that reducing the percentage of particles passing through a 0.15mm sieve from 39% to 5% requires a unit energy consumption E of 5.44 kWh / t. This calculation considers parameters K = 0.377 and R... f =39% of the values, and the correlation index R f =0.9945.

[0041] In the embodiment described, the fuzzy controller 10 is connected to at least one mill M1, M2, M... n The energy consumption prediction model works in conjunction with the existing grinding energy consumption value and the percentage of the target particle size in the grinding product to predict whether at least one mill M1, M2, and M3 needs to be adjusted. n The power. Preferably, the at least one mill M1, M2, M... n The energy consumption prediction model is built based on historical operating data.

[0042] Figure 3 A configuration scheme for the fuzzy controller 10 is shown. More preferably, the method includes E and R p The model is constructed as fuzzy variables, which are then used by the fuzzy controller 10 to determine mill parameters M1, M2, and M. n Ideal power configuration.

[0043] In this embodiment, at least one mill M1, M2, M is estimated. n The power steps include constructing a method for estimating the power of at least one mill M1, M2, M... n A power model that varies over time.

[0044] Preferably, to study the relationship between power decay and power increase in the mill, average values ​​from the operating database can be used, including feed rate, and values ​​of each mill M1, M2, and M... n The running time, grinding media replenishment amount, motor power, solid phase percentage, mass flow rate of the mill discharge slurry tank outlet pump, and mass flow rate of the hydrocyclone feed pump are all considered.

[0045] More preferably, it is used to estimate at least one mill M1, M2, M nThe model for power decay over time during the grinding stage includes formula (2), which can be obtained through regression analysis between variables:

[0046]

[0047] in:

[0048] H = Discretized planning time range, in hours, H = {1, 2, ..., |H|};

[0049] P d ij For at least one mill j (M1, M2, M n The power attenuation value at time i∈H, in kW;

[0050] Q p ij The outlet mass flow rate of mill j at time i is expressed in t / h.

[0051] α ij For at time i, at least one mill j (M1, M2, M...) n The feed rate is expressed in t / h.

[0052] φ ij For at least one mill j (M1, M2, M n The percentage of solid phase at time i, in %;

[0053] Used to estimate the values ​​of M1, M2, and M for at least one mill. n The model for power increase over time during the grinding stage also includes formula (3), which is obtained through regression analysis and describes the correlation between power increase and other variables of the mill:

[0054]

[0055] in:

[0056] p r ij To ensure that at time i, at least one mill j (M1, M2, M...) is connected to the mill j. n The power increase caused by adding grinding media, in kW;

[0057] x ij For at least one mill j (M1, M2, M n The mass of the ball at time i, in kg;

[0058] More preferably, it is used to estimate at least one mill M1, M2, M nThe model for power decay over time during the regrinding stage includes formula (4), which can be obtained through regression analysis between variables:

[0059]

[0060] in:

[0061] Q h ij =mill j(M1, M2, M n The feed mass flow rate of the hydrocyclone at time i∈H is expressed in t / h.

[0062] Used to estimate the values ​​of M1, M2, and M for at least one mill. n The model for power increase over time during the regrinding stage also includes formula (5), which is obtained through regression analysis and describes the correlation between power increase and other mill variables:

[0063]

[0064] Models M1, M2, and M used to estimate the power of at least one mill. n The model for power variation over time during the grinding and regrinding stages also includes formula (6):

[0065]

[0066] in:

[0067] For mill j (M1, M2, M) n The estimated power at time i, in kW;

[0068] For mill j (M1, M2, M) n Average power over the past 15 minutes, in kW.

[0069] Formula (4) estimates the mill parameters M1, M2, and M. n The power varying over time, based on analysis of relevant variables, provides an estimate and serves as a basis for understanding the mill's M1, M2, and M... n A reference point for power changes at different times.

[0070] To evaluate the model, estimate the time-dependent parameters of at least one mill, M1, M2, and M3, during the grinding stage. n The effectiveness of the power was assessed by comparing the measured operating data of a mill M1 with the model predictions, such as... Figure 4 As shown in (a) and 4(b).

[0071] The model exhibits an accuracy rate of 79%, indicating that its power estimates are in high agreement with actual measurements. This fully validates the model's ability to continuously predict mill power and provides crucial data support for process control.

[0072] To evaluate the model's estimation of at least one mill's M1, M2, and M during the regrinding phase over time. n The effectiveness of the power was assessed by comparing the measured operating data of a mill M1 with the model predictions, such as... Figure 4 As shown in (c) and 4(d).

[0073] The model exhibits an accuracy rate of 76%, indicating that its power estimates are in high agreement with actual measurements. This fully validates the model's ability to continuously predict mill power and provides crucial data support for process control.

[0074] In the described embodiment, at least one mill M1, M2, M is estimated. n The power generation steps include constructing a model to establish at least one mill M1, M2, M... n The correlation between power and filling rate.

[0075] Preferably, it is used to establish at least one mill M1, M2, M n The model for the correlation between power and filling rate includes formula (7):

[0076]

[0077] in:

[0078] x ij For mill j (M1, M2, M) n The grinding media load at time i, in kg;

[0079] D j For at least one mill j (M1, M2, M n The inner diameter of the lining plate, in meters;

[0080] L j For at least one mill j (M1, M2, M n The length of () is in meters;

[0081] ρ j For mill j (M1, M2, M) n The apparent density of the material load within the container, in kg / m³. 3 ;

[0082] e ij For at least one mill j (M1, M2, M n The filling rate at time i.

[0083] Used to establish at least one mill M1, M2, M n The model relating power to filling rate also includes formula (8) for estimating the power per ton of steel balls in mills with a diameter greater than 2.44 meters:

[0084]

[0085] in:

[0086] p b ij For mill j (M1, M2, M) n The power of the pinion at time i per ton of steel balls is expressed in kW / t.

[0087] D j For at least one mill j (M1, M2, M n The inner diameter of the lining plate, in meters;

[0088] e ij For at least one mill j (M1, M2, M n The filling rate at time i, in %;

[0089] V j For mill j (M1, M2, M) n The critical speed fraction, expressed as a percentage;

[0090] S j For mill j (M1, M2, M) n The steel ball size correction factor.

[0091] Used to establish at least one mill M1, M2, M n The power correction factor model also includes formula (9), which is applicable to mills with a diameter greater than 3.3 meters:

[0092]

[0093] in:

[0094] B j For mill j (M1, M2, M) n The diameter (apex diameter) of the largest steel ball used in the design, in millimeters;

[0095] S j For only when the diameter of the steel ball is smaller than that of at least one mill j (M1, M2, M n This parameter only needs to be considered when the diameter is 1 / 80.

[0096] Used to establish at least one mill M1, M2, M nThe model relating the power and filling rate also includes formula (10) for estimating the power of at least one mill M1, M2, M... n Power at the pinion:

[0097]

[0098] in:

[0099] p b ij For mill j (M1, M2, M) n The power of the pinion at time i is expressed in kW.

[0100] Used to establish at least one mill M1, M2, M n The model relating power to filling rate also includes formula (11) for estimating the mechanical power on the motor shaft, which is the ratio of power at the pinion to gear transmission efficiency:

[0101]

[0102] in:

[0103] p m ij For at least one mill j (M1, M2, M n The mechanical power on the motor shaft at time i, expressed in kW;

[0104] η is the gear transmission efficiency.

[0105] Formulas (7) to (11) can be used to estimate the filling rate e. ij Mechanical power p of the motor shaft m ij Relationships, such as Figure 5 As shown in (a) and 5(b), this provides information about mills M1, M2, and M n Information on the quantity of internal grinding media.

[0106] according to Figure 5 The regression analysis results shown in (a) can establish at least one mill M1, M2, M n A model of the correlation between power and filling rate during the grinding stage, which also includes formula (12):

[0107]

[0108] Furthermore, based on the results obtained from regression analysis, at least one mill M1, M2, and M3 were used to establish... n The model for the correlation between power and fill rate during the regrinding stage includes formula (13):

[0109]

[0110] Used to establish at least one mill M1, M2, M n The model for the correlation between power and filling rate also includes formula (14):

[0111]

[0112] in:

[0113] a = -1.7227;

[0114] b = 183.76 (applicable to mills in the grinding stage) and 180.38 (applicable to mills in the regrinding stage);

[0115] c = -p m ij.

[0116] Preferably, it is used to establish at least one mill M1, M2, M n The model for the correlation between power and filling rate also includes formula (15), used to estimate the power of at least one mill M1, M2, M... n Internal load mass:

[0117] x ij =9,0139e ij (15)

[0118] In the described embodiment, when considering a mechanical power of 4.188 kW, based on Figure 5 The data in (a) suggests that the mill filling rate was 33% during the grinding stage and 34.8% during the regrinding stage. Furthermore, as... Figure 5 As shown in (b), the estimated internal load mass of the mill during the grinding stage is 297.46t, and the internal load mass of the mill during the regrinding stage is 313.69t.

[0119] Formula (11) is used to describe the power on the pinion shaft when the mill is only loaded with grinding media. Figure 6 Mills M1, M2, and M are shown. n The power difference is observed when materials (ore and water) are fed together with the grinding media, versus when no materials are fed. When material feeding is interrupted, the increased free volume inside the mill facilitates the movement of the grinding media load, leading to an increase in mill speed and consequently, increased energy consumption.

[0120] Furthermore, the imbalance of the grinding media load can also lead to an increase in kinetic forces and kinetic energy, thereby affecting the mill's M1, M2, and M... n Energy consumption.

[0121] Preferably, it is used to establish at least one mill M1, M2, Mn The model relating the power to the filling rate also includes formula (16) for determining the relationship between at least one mill M1, M2, and M3 containing grinding media. n Net power:

[0122]

[0123] in:

[0124] p l ij For mill j (M1, M2, M) n The net power at time i; and

[0125] ΔP represents at least one mill M1, M2, M n The difference between the net power and the mechanical power of the motor shaft.

[0126] In the embodiment described, there are at least one mill M1, M2, and M n The steps for determining the grinding media replenishment parameters include: constructing a metaheuristic optimization model containing an objective function, wherein the objective function is used to minimize at least one mill M1, M2, M... n The required power is related to at least one mill M1, M2, M n Estimate the difference between power values. Preferably, the optimization model uses a mixed integer linear programming (MILP) method to express the problem; it includes an optimization algorithm based on local search class heuristics and integrates an enhanced ILS algorithm, the objective function of which is described by the following formula (17):

[0127]

[0128] In addition, for at least one mill M1, M2, M n The steps for determining the grinding media replenishment parameters include: defining at least one mill M1, M2, M... n The grinding media replenishment sequence, and the definition of at least one mill M1, M2, M n The grinding media accelerates the rate.

[0129] The advantage of this invention is that the proposed method for optimizing the replenishment of grinding media in mills ensures higher energy efficiency and improved particle size control, because the method can maintain the efficiency of mills M1, M2, and M... n The filling rate is at the optimal operating value, thereby avoiding under-grinding and over-grinding of the ore.

[0130] Another advantage of this invention is that it maintains the operation of mills M1, M2, and M n Finding the optimal filling rate not only improves ore crushing efficiency but also has a positive effect on subsequent processes such as flotation.

[0131] This invention provides a heuristic method that dynamically infers the optimal power value of a mill by measuring the percentage of material ground, and estimates the instantaneous power of the mill based on operating data such as feed rate and solids percentage, thereby assisting in decisions on when to supply power to each mill M1, M2, and M3. n Add grinding media and the amount to add.

[0132] Therefore, the present invention can achieve reasonable planning and higher efficiency in replenishing grinding media during work shifts.

Claims

1. A method for optimizing the replenishment of grinding media for a mill, characterized in that comprising the steps of: determining a required power value for at least one mill (M1, M2, M n ) by means of a fuzzy controller (10); estimating the power of the at least one mill (M1, M2, M n ) by means of a mill power estimation device (20) on the basis of operating data of the at least one mill (M1, M2, M n ) determining grinding media replenishment parameters of said at least one mill (M1, M2, M n ) by means of a grinding media replenishment decision device (30); and The at least one mill (M1, M2, M3) is automatically replenished by means of an automatic grinding media replenishment device (40) based on the grinding media replenishment parameters. n ) 2. The method of claim 1, characterized in that the step of automatically replenishing the grinding media further comprises controlling a grinding media replenishment device (50) by the grinding media automatic replenishment device (40).

3. The method according to claim 1 or 2, characterized in that , the step of determining the required power value of at least one mill (M1, M2, M n ) comprises building an energy consumption prediction model of at least one mill (M1, M2, M n ).

4. The method according to claim 3, characterized in that , the energy consumption prediction model of the at least one mill (M1, M2, M n ) comprises equation (1): wherein: E is the energy consumed on the pinion shaft per unit mass of circuit feed (kWh / t); R f To control the percentage of material on the screen within the target size range for the feed. R p To control the percentage of material on the screen within the target size range for the product. K is an ore characteristic parameter (t / kWh).

5. The method according to claim 1 or 2, characterized in that , said step of estimating the power of at least one mill (M1, M2, M n ) comprises constructing a model for estimating the power of said at least one mill (M1, M2, M n ) as a function of time.

6. The method according to claim 5, characterized in that , the model for estimating the power of at least one mill (M1, M2, M n ) in the ore grinding phase as a function of time comprises the formula (2): wherein: P d ij Pj(i) is the power decay value for at least one mill j (M1, M2, M n ) at time i in kW; Q p ij Qi,j is the outlet mass flow of at least one mill j (M1, M2, M n ) at time i, in t / h; a ij is the feed rate of at least one mill j (M1, M2, M n ) at time i in t / h; φ ij is the percentage of solid phase of at least one mill j (M1, M2, M n ) at time i; The model for estimating the power of at least one mill (M1, M2, M n ) over time in the grinding phase also comprises equation (3): wherein: p r ij Pj(i) = Pj(i-1) + Pj(i-1) * (1 - e^(-a * (Mj(i) - Mj(i-1)) / Mj(i-1))) (1) for the power increase at time i due to the addition of grinding media to at least one mill j (M1, M2, M n ) in kW; x ij is the mass of the ball load of at least one mill (M1, M2, M n ) at time i in kg.

7. The method according to claim 5, characterized in that , the model for estimating the power of at least one mill (M1, M2, M n ) over time in the regrinding phase comprises the formula (4): wherein: Q h ij = mass flow rate of feed material to mill j (M1, M2, M n ) at time i, in t / h The method for estimating the power of at least one mill (M1, M2, M n ) over time in the regrinding phase also comprises equation (5): The model for estimating the power of at least one mill (M1, M2, M n ) over time also comprises equation (6): wherein: The estimated power of the mill j (M1, M2, M n ) at time i in kW; The average power in the first 15 minutes of the mill j (M1, M2, M n ) is given in kW.

8. The method according to claim 1 or 2, characterized in that said step of estimating the power of at least one mill (M1, M2, M n ) comprises establishing a model of the correlation between the power and the filling rate of at least one mill (M1, M2, M n ).

9. The method according to claim 7, characterized in that , said model establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate comprises the formula (7): wherein: x ij Mj(i) is the mill j (M1, M2, M n ) grinding media load at time i in kg; D j Ri is the inner radius of the liner of at least one mill j (M1, M2, M n ) in meters; L j is the length of at least one mill j (M1, M2, M n ) in meters; ρ j is the apparent density of the material load in the mill j (M1, M2, M n ) in kg / m 3 ; e ij the filling rate of at least one mill j (M1, M2, M n ) at time i; said model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate further comprises equation (8) for estimating the power per ton of steel balls for a mill having a diameter greater than 2.44 meters: wherein: p b ij The power per ton of steel balls at the pinion of the mill j (M1, M2, M n ) at time i, in kW / t. D j Ri is the inner radius of the liner of at least one mill j (M1, M2, M n ) in meters; e ij Filling rate of at least one mill j (M1, M2, M n ) at time i; V j is the fraction of the critical speed of the mill j (M1, M2, M n ) in percent; S j is a size correction factor for the steel balls of the mill j (M1, M2, M n ). said model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate further comprises formula (9) which is applicable to mills having a diameter greater than 3.3 meters: wherein: B j Diameter of the largest steel ball used in the mill j (M1, M2, M n ) in millimeters. S j is a parameter to be considered only when the diameter of the steel ball is less than 1 / 80 of the diameter of at least one of the mills j (M1, M2, M n ) The model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate further comprises equation (10) for estimating the power at the pinion of at least one mill (M1, M2, M n ). wherein: p p ij Pmillj(M1, M2, M n ) is the power at the pinion of the mill j (M1, M2, M n ) at time i in kW. said model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate further comprises equation (16) for estimating the mechanical power on the motor shaft, said mechanical power being the ratio between the power at the pinion and the gear transmission efficiency: η is the gear transmission efficiency; p m ij Pj(i) is the mechanical power on the motor shaft of at least one mill j (M1, M2, M n ) at time i, in kW; wherein: The model for establishing a correlation between the power of at least one mill (M1, M2, M n ) in the ore grinding phase and the filling rate also comprises equation (12): said model for establishing a correlation between the power of at least one mill (M1, M2, M n ) in the regrinding phase and the filling rate also comprises equation (13): The model for establishing the correlation between the power of at least one mill (M1, M2, M n ) and the filling rate also comprises equation (14): b = 183.76 for mills in the grinding stage and 180.38 for mills in the regrinding stage; a＝-1.7227； wherein: c = -p m ij The model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate further comprises equation (15) for estimating the load mass in at least one mill (M1, M2, M n ). x ij = 9,0139e ij (15) The model for establishing a correlation between the power of at least one mill (M1, M2, M n ) and the filling rate also comprises formula (16) for determining the net power of at least one mill (M1, M2, M n ) equipped with grinding media: characterized in that p l ij Pj(i) is the net power of the mill j (M1, M2, M n ) at time i; and ΔP is the difference between the net power of at least one mill (M1, M2, M n ) and the mechanical power of the motor shaft.

10. The method according to claim 1 or 2, characterized in that determining a grinding media replenishment parameter for at least one mill (M1, M2, M n ) comprises: A metaheuristic optimization model is constructed comprising an objective function for minimizing the difference between the required power of the at least one mill (M1, M2, M n ) and the estimated power of the at least one mill (M1, M2, M n ).

11. The method of claim 9, characterized in that The meta-heuristic optimization model is of local search type and comprises an enhanced ILS algorithm.

12. The method of claim 9, grinding media make-up device (50); determining a grinding media replenishment parameter for the at least one mill (M1, M2, M n ) comprises: defining a sequence of replenishment of grinding media of said at least one mill (M1, M2, M n ) and defining a rate of replenishment of grinding media of said at least one mill (M1, M2, M n ).

13. A controller readable storage medium, characterized in that comprising instructions that, when executed by at least one processor, cause the at least one processor to perform the steps of the method for optimizing the make-up of grinding media of a mill as defined in any one of claims 1 to 11, comprising: determining a required power value for at least one mill (M1, M2, M n ) by means of a fuzzy controller (10); estimating the power of the at least one mill (M1, M2, M n ) by means of a mill power estimation device (20) on the basis of operating data of the at least one mill (M1, M2, M n ) determining grinding media replenishment parameters of said at least one mill (M1, M2, M n ) by a grinding media replenishment decision device (30); and The at least one mill (M1, M2, M3) is automatically replenished by means of an automatic grinding media replenishment device (40) based on the grinding media replenishment parameters. n ) 14. A system for optimizing the replenishment of grinding media for a mill, characterized in that said system comprising: The industrial controller readable storage medium as defined in claim 12, said storage medium comprising instructions which, when executed by at least one processor, cause the at least one processor to perform the individual steps of the method for optimizing grinding media make-up of a mill, comprising: at least one mill (M1, M2, M3) connectable to said grinding medium replenishment device (50); and n ) The grinding media automatic make-up device (40) is further configured for controlling the grinding media make-up device (50). determining a required power value for at least one mill (M1, M2, M n ) by means of a fuzzy controller (10); estimating the power of the at least one mill (M1, M2, M n ) by means of a mill power estimation device (20) on the basis of operating data of the at least one mill (M1, M2, M n ) determining grinding media replenishment parameters of said at least one mill (M1, M2, M n ) by a grinding media replenishment decision device (30); and The at least one mill (M1, M2, M3) is automatically replenished by means of an automatic grinding medium replenishment device (40) based on the grinding medium replenishment parameters. n ) ​

Images