Fine grinding control method for mineral separation and control device thereof

Abstract

The invention relates to the technical field of fine grinding of ore grinding machines, in particular to a fine grinding control method for ore dressing and a control device of the fine grinding control method for ore dressing. The weight ratio, the average particle size and the classification efficiency of the minerals in different storage areas at each collection moment after the different minerals are fed into the corresponding storage areas by the classifier are collected in real time; a processing step: reflecting the overall ore grinding efficiency of the ore grinding machine by analyzing data characteristics of all minerals collected at all collection moments; and a control step: self-adaptively adjusting the rotating speed of the current ore grinding machine by utilizing the overall ore grinding efficiency of the ore grinding machine. According to the method, the rotating speed of the ore grinding machine is adaptively adjusted through the overall ore grinding efficiency obtained through analysis, and a more accurate rotating speed control strategy is achieved.

Description

Technical Field

[0001] This application relates to the technical field of fine grinding of grinding mills, and specifically relates to a fine grinding control method for ore dressing and its control device. Background Art

[0002] Fine grinding in ore dressing is a key step in mineral processing. With the reduction of easily selected mineral resources, the proportion of poor, fine, and complex refractory ores increases, and more refined grinding technologies are needed to achieve effective separation of minerals. The application of fine grinding technologies such as agitation mills has significantly improved the grinding efficiency and the monomer dissociation degree of target minerals, reduced production costs, and enhanced economic benefits. Fine grinding in ore dressing can separate the useful components in the ore into monomers, creating conditions for subsequent operations and directly affecting the technical and economic indicators of the ore dressing plant. Appropriate grinding fineness can maximize the throughput, improve production efficiency, and at the same time ensure the full dissociation of minerals and gangue, which is beneficial to improving the metal recovery rate. The control of grinding fineness is particularly crucial for subsequent flotation operations, directly affecting the flotation behavior of minerals, including the surface properties and pulp properties of minerals. Therefore, fine grinding control is not only related to ore dressing efficiency and cost, but also directly affects the quality and economic benefits of the final product, and is an indispensable part of the ore dressing process.

[0003] During the ore dressing process, due to the different hardness, particle size, and dissociation characteristics of different minerals, the suitable rotational speeds of the grinding mills are also different. However, most of the existing ore dressing technologies rely on the initially set fixed rotational speeds, and a single rotational speed is difficult to adapt to the characteristics of all minerals, thus affecting the grinding efficiency and the dissociation degree of minerals. It may not only lead to over-grinding or under-grinding of useful minerals, but also affect the subsequent flotation effect, reduce the metal recovery rate, and increase energy consumption and production costs. Summary of the Invention

[0004] To solve the above technical problems, this application provides a fine grinding control method for ore dressing and its control device. The specific technical solutions adopted are as follows: In a first aspect, an embodiment of this application provides a fine grinding control method for ore dressing. The method includes the following steps: Collection step: After the grinding mill starts to operate stably, the weight ratio, average particle size, and classification efficiency of minerals in different storage areas at each collection moment are collected in real time after the classifier sends different kinds of minerals into the corresponding storage areas respectively. Processing step: By analyzing the data characteristics collected for all kinds of minerals at all collection moments, the overall grinding efficiency of the grinding mill is reflected. Specifically: S1. After performing curve fitting on the weight percentage and average particle size of each mineral at any collection time and all previous collection times respectively, analyze the curvature characteristics of the corner points on the fitted curve and the range characteristics of the data around the corner points to determine the grinding compliance at the any collection time for each mineral. S2. For each mineral, arrange the grinding compliance calculated at the current and all previous collection times in ascending order of time to obtain the compliance sequence of each mineral at present. S3. Calculate the similarity between the compliance sequence of each mineral at present and the classification efficiency sequence composed of the classification efficiencies at all collection times at each collection time; and use the similarities of each mineral at all collection times as each row, and the similarities of different minerals at the same collection time as each column to construct a trend change matrix. S4. Perform clustering processing on the elements in the trend change matrix; based on the proportion of mineral types in each clustering cluster and the average distance between the center of each clustering cluster and the centers of all other clustering clusters in the clustering result, determine the overall grinding efficiency of the grinding mill. Control step: Use the overall grinding efficiency of the grinding mill to adaptively adjust the rotation speed of the current grinding mill.

[0005] Preferably, in the collection step, the time when the grinding mill starts to operate stably is the moment one minute after the grinding mill starts working.

[0006] Preferably, in step S1, the data around the corner point is all the data within a preset time window on the fitted curve where the corner point is located, with the corner point as the center.

[0007] Preferably, in step S1, denote the grinding compliance at the any collection time for each mineral as A, and the calculation formula is: ; where exp represents the exponential function with the natural constant e as the base; represents the number of corner points on the i-th fitted curve; represents the curvature of the j-th corner point on the i-th fitted curve; represents the range of the data within the preset time window centered on the j-th corner point on the i-th fitted curve.

[0008] Preferably, in step S3, the similarity at each collection time is obtained by calculating the similarity between the two sequences at each collection time using the dynamic kernel correlation algorithm.

[0009] Preferably, in step S4, the clustering distance in the clustering processing is the numerical difference and position difference between elements.

[0010] Preferably, in step S4, the calculation formula for the overall grinding efficiency characteristic value of the grinding mill is as follows: Calculate the ratio result of the proportion of the mineral types in each clustering cluster and the average distance between the center of this clustering cluster and the centers of all other clustering clusters; Take the average level of the ratio results of all clustering clusters as the overall grinding efficiency characteristic value of the grinding mill.

[0011] Preferably, in the control step, the relationship formula for adaptively adjusting the rotation speed of the current grinding mill using the overall grinding efficiency of the grinding mill is as follows: ; where is the rotation speed of the current grinding mill after adjustment, is the preset adjustment parameter, e is the natural constant, B is the overall grinding efficiency of the grinding mill, and R is the rotation speed of the current grinding mill.

[0012] Preferably, the initial rotation speed when the grinding mill starts operating is a preset percentage of the critical rotation speed.

[0013] In a second aspect, another embodiment of the present application provides a fine grinding control device for ore dressing, which includes a grinding mill and a classifier; the grinding mill is used to control its rotation speed to grind the material and send it into the classifier from the mineral outlet of the grinding mill; the classifier is used to divide the ground product into different types of minerals and send it into the corresponding storage area from the overflow outlet of the classifier, so that the selected minerals can be temporarily stored in specific areas respectively for further processing or separation, and send the unqualified minerals back to the grinding mill from the classifier return sand outlet for re-grinding to ensure that the particle size of the grinding product meets the requirements of subsequent ore dressing operations.

[0014] The present application has at least the following beneficial effects: The present application proposes a fine grinding control method and its control device for ore dressing. Aiming at the problem of the grinding fit degree of different minerals by the grinding mill, the corner point distribution characteristics on the curve after curve fitting are used to reflect the complexity and severity of the mineral distribution and particle size change during the grinding process, and solve the problem of the fit degree between the preset parameters of the grinding mill and the mineral grinding effect; aiming at the overall grinding efficiency problem of the grinding mill, the clustering method is used for the trend change matrix to reflect the distribution, change trend of different mineral types and the influence on the classification efficiency, and exclude the influence of the characteristics of a single mineral on the evaluation of the overall grinding efficiency; by using the overall grinding efficiency to adaptively adjust the rotation speed of the grinding mill, a more accurate rotation speed control strategy is realized to adapt to the characteristics of different minerals, improve the efficiency and economic benefits of the entire ore dressing process, and the improved rotation speed control strategy can dynamically adjust the operating parameters of the grinding mill, optimize the grinding process, enable the grinding mill to better adapt to the physical characteristics of different minerals, and achieve more effective grinding and separation. Description of the Drawings

[0015] Figure 1 The flowchart of steps of a fine grinding control method for ore dressing provided by an embodiment of the present application; Figure 2 The flowchart of the determination process of the overall grinding efficiency of the grinding mill provided by an embodiment of the present application. Detailed implementation manners

[0016] Embodiment 1 During the ore dressing process, due to the different hardness, particle size and dissociation characteristics of different minerals, the suitable rotational speeds of the grinding mills are also different. Most of the existing ore dressing technologies rely on the initially set fixed rotational speeds, and a single rotational speed is difficult to adapt to the characteristics of all minerals, thus affecting the grinding efficiency and the dissociation degree of minerals. It may not only cause over-grinding or under-grinding of valuable minerals, but also affect the subsequent flotation effect, reduce the metal recovery rate, and increase the energy consumption and production cost.

[0017] Accordingly, this embodiment provides a flowchart of steps of a fine grinding control method for ore dressing, as shown in the appendix Figure 1 The method includes the following steps: Collection step: After the grinding mill starts to operate stably, the weight ratio, average particle size and classification efficiency of the minerals in different storage areas at each collection moment are collected in real time after the classifier sends different kinds of minerals into the corresponding storage areas respectively.

[0018] Since when the grinding mill just starts to operate, the physical states such as the temperature, humidity and friction coefficient inside the machine have not reached the stable state, and these factors will affect the grinding effect and the accuracy of data, relevant data is collected in real time after the grinding mill starts to operate stably.

[0019] Among them, the time when the grinding mill starts to operate stably is set to the moment one minute after the grinding mill starts to work in this embodiment, and it can be specifically set by the implementer himself.

[0020] In addition, in this embodiment, the time interval for collecting the above relevant data is set to 1 s, and the collected data is subjected to overall normalization processing. The normalization processing method adopts the maximum-minimum normalization method, and the maximum-minimum normalization method is a well-known technology and will not be elaborated here.

[0021] Processing step: By analyzing the data characteristics of all kinds of minerals collected at all collection moments, the overall grinding efficiency of the grinding mill is reflected.

[0022] The weight percentage reflects the weight percentage of minerals in each storage area, that is, the ratio of the weight of minerals in each storage area to the total weight of minerals in all storage areas. By analyzing the changes in the percentage of different minerals, the distribution of minerals during the grinding process can be understood. If the percentage of a certain mineral increases, it may indicate that the grinding efficiency of the grinding mill for this mineral is relatively high because more of this mineral is effectively separated. By monitoring the average particle size of different minerals, the grinding efficiency of the grinding mill for different minerals can be evaluated.

[0023] If the average particle size of the target mineral after grinding is low, it indicates that the grinding effect of the grinding mill on the target mineral is good. Conversely, if the average particle size of the target mineral after grinding is high, it indicates that the grinding effect of the grinding mill on the target mineral is poor.

[0024] Compared with the weight percentage and average particle size, the classification efficiency reflects the ability of the classifier to separate ore particles according to particle size. A high classification efficiency means that more fine-grained minerals are effectively separated, which is positively correlated with the grinding efficiency. If the grinding efficiency is low, the mineral particles may not be fully dissociated, resulting in a decrease in the classification efficiency. It is directly related to the performance and operation effect of the classifier, indicating the separation ability of the classifier for materials of the target particle size, showing the separation effect of the classifier on materials of a given particle size, that is, the enrichment degree of fine-grained materials in the overflow, and is directly related to the control ability of the particle size distribution of the grinding product, which is crucial for the overall grinding efficiency of subsequent beneficiation operations.

[0025] Based on the above analysis, it can be seen that the weight percentage, average particle size, and classification efficiency of each mineral in the storage area directly reflect the overall grinding efficiency of the grinding mill.

[0026] Accordingly, in this application, the flowchart for determining the overall grinding efficiency of the grinding mill is as shown in the appendix Figure 2 as follows: S1. After performing curve fitting on the weight percentage and average particle size of each mineral at any collection moment and all collection moments before it respectively, analyze the curvature characteristics of the corner points on the fitted curve and the range characteristics of the data around the corner points to determine the grinding compliance of each mineral at the said any collection moment.

[0027] For each mineral, taking a random mineral as an example here, in order to reflect the change situation of this mineral during grinding, all the weight percentages and all the average particle sizes of this mineral at any collection moment and all collection moments before it are sorted in the order of collection time and then curve fitting is performed.

[0028] Subsequently, taking the two fitted curves as inputs respectively, use the CSS (Curvature Scale Space) algorithm to output the corner points of each fitted curve. The detected corner points represent the turning points of the change trend of the weight percentage or average particle size, that is, at these points, the distribution or grinding efficiency of the mineral has changed significantly.

[0029] Among them, the least squares method and the CSS algorithm are well-known technologies and will not be described in detail. In other embodiments of the present application, the principal component analysis method can be used for curve fitting, and the corner point analysis algorithm can be used to obtain the corner points on the curve.

[0030] Furthermore, suppose that the i-th fitting curve has detected corner points, calculate the curvature of the j-th corner point on the i-th fitting curve And with each corner point as the center, a time window of preset length N is constructed. In this embodiment, N is 5. It should be noted that if the length of the time window exceeds the length of the fitting curve, mean padding is performed and the data range in each time window is calculated. Among them, mean padding is a well-known technology and will not be described in detail.

[0031] Based on the above analysis, the grinding suitability A of each mineral at any collection time is calculated. The specific calculation formula is: ; Wherein, exp represents the exponential function with the natural constant e as the base; Indicates the number of corner points on the i-th fitting curve; represents the curvature of the jth corner point on the i-th fitting curve; Represents the data range within the preset time window centered at the j-th corner point on the i-th fitting curve.

[0032] It should be understood that the number of corner points It reflects the number of turning points in the weight percentage or average particle size change trend of the mineral. The larger the value of , the more turning points there are in the changing trend, which means the greater the fluctuation in the grinding process, thus having a greater negative impact on the value of A. Describes the degree of curvature of the curve at the jth corner point on the i-th fitting curve of the mineral. The larger the curvature, the more severe the curve bends at that point. That is, during the grinding process, the greater the abnormal change in the weight proportion or average particle size of the mineral, the worse the grinding level of the mill for the mineral, and the smaller the A value. The size of represents the range of weight proportion or average particle size variation in the time period near the jth corner point on the i-th fitting curve of the mineral. This means that the change near the corner is more significant, which is consistent with Similarly, the smaller the A value.

[0033] A describes the complexity and severity of the mineral distribution and particle size changes during the grinding process, thereby characterizing the degree to which the mill meets the grinding requirements of the mineral. The larger the A value, the more stable the grinding process, that is, the more the grinding parameters of the mill meet the grinding conditions of the mineral.

[0034] S2. For each mineral, arrange the grinding fitting degrees calculated at the current and all previous collection times in ascending order of time to obtain the fitting degree sequence of each mineral at present.

[0035] Furthermore, in order to analyze the influence of the grinding effect of the grinding mill on each mineral on the classification efficiency, for each mineral, after the grinding mill starts to operate stably, arrange the grinding fitting degrees calculated at the current and all previous collection times in ascending order of time to obtain the fitting degree sequence of each mineral.

[0036] S3. Calculate the similarity between the fitting degree sequence of each mineral at present and the classification efficiency sequence composed of the classification efficiencies at all collection times at each collection time; and use the similarities of each mineral at all collection times as each row, and the similarities of different minerals at the same collection time as each column to construct a trend change matrix.

[0037] For each mineral, form a classification efficiency sequence by arranging the classification efficiencies at the current and all previous collection times in the order of collection time.

[0038] Furthermore, take the fitting degree sequence and the classification efficiency sequence of each mineral at present as inputs, and use the dynamic kernel correlation algorithm to calculate the similarity between the two sequences at each collection time. Among them, the dynamic kernel correlation algorithm is a well-known technology and will not be elaborated here.

[0039] That is, at each collection time, there is a similarity representing the historical trend similarity between the fitting degree sequence and the classification efficiency sequence at the selected time. Use the similarities of each mineral at all collection times as each row, and the similarities of different minerals at the same collection time as each column to construct a trend change matrix.

[0040] S4. Perform clustering processing on the elements in the trend change matrix; determine the overall grinding efficiency of the grinding mill based on the proportion of the mineral types in each clustering cluster in the clustering result and the average distance between the center of each clustering cluster and the centers of all other clustering clusters.

[0041] In order to reveal the stability of the grinding mill during the grinding process of different minerals, in this embodiment, take the trend change matrix as the input of the DBSCAN density clustering algorithm, set the minimum number of points to 2, the neighborhood radius to 0.5, and the clustering distance to the numerical difference and position difference between elements, and use the DBSCAN density clustering algorithm to output each clustering cluster of the trend change matrix after clustering. Among them, the DBSCAN density clustering algorithm is a well-known technology and will not be elaborated here.

[0042] Among them, take the clustering distance d between any two elements as an example: , where S represents the absolute value of the difference in values between any two elements, and W represents the positional distance between any two elements.

[0043] For each output clustering cluster, which represents the similar trend change of different minerals on the classification efficiency. Suppose a total of P clustering clusters are obtained. Calculate the ratio between the number of mineral types in the p-th clustering cluster and the total number of mineral types, which is the proportion of mineral types in the p-th clustering cluster. , and calculate the average distance between the center of the p-th clustering cluster and the centers of all other clustering clusters. .

[0044] Based on the above analysis, calculate the overall grinding efficiency B of the grinding mill. The specific calculation formula is , where P represents the number of clustering clusters, represents the proportion of mineral types in the p-th clustering cluster, represents the average distance between the center of the p-th clustering cluster and the centers of all other clustering clusters.

[0045] It should be understood that reflects the proportion of mineral types in the p-th clustering cluster relative to the total number of mineral types, which can be understood as the richness of mineral types in the p-th clustering cluster. The larger its value, the more mineral types are included in this clustering cluster, which may indicate that the minerals in this clustering cluster have a more overall impact on the classification efficiency, that is, a greater contribution to the eigenvalue of the overall grinding efficiency, and B increases accordingly; is used to measure the separation degree between clustering clusters, that is, the distribution distance of clustering clusters in the feature space. The larger its value, the more obvious the difference in position characteristics between this clustering cluster and other clustering clusters, that is, it indicates that the impact of mineral types in this clustering cluster on the classification efficiency is more different from that of other clustering clusters, and the value of B is smaller.

[0046] As the eigenvalue of the overall grinding efficiency, B comprehensively considers the distribution, change trend of different mineral types and their influence on the classification efficiency to quantitatively evaluate the performance of the grinding mill. The larger the value of B, the better the grinding mill can adapt to the characteristics of different minerals and achieve more effective grinding and separation.

[0047] Control step: Use the overall grinding efficiency of the grinding mill to adaptively adjust the rotation speed of the current grinding mill.

[0048] In this embodiment, the initial rotation speed of the grinding mill is set to 80% of the critical rotation speed. The critical rotation speed refers to the rotation speed when the outermost grinding medium just rotates with the cylinder without falling when the cylinder of the grinding mill rotates. In other embodiments, the initial rotation speed of the grinding mill is set by the implementer himself.

[0049] During the operation of the grinding mill, when the rotational speed of the grinding mill exceeds the critical speed, the grinding media will rotate with the cylinder body without falling, entering the so-called "centrifugal operation state". In this state, the grinding media have no impact effect and the grinding effect is also very small, resulting in almost no grinding effect; if the rotational speed of the grinding mill is too low, the height to which the grinding media are lifted is small, and they will move in a cascading state. At this time, the impact force of the grinding media is very small, and the ore is mainly crushed by the abrasion effect, resulting in poor grinding efficiency.

[0050] Therefore, the rotational speed of the current grinding mill is adaptively adjusted using the overall grinding efficiency of the grinding mill. The specific adjustment relationship is as follows: ; where is the rotational speed of the current grinding mill after adjustment, is the preset adjustment parameter, which takes a value of 0.25 in this embodiment. Its function is to prevent the rotational speed of the grinding mill after adjustment from exceeding the critical speed, resulting in a reduction in the grinding effect; e is the natural constant, B is the overall grinding efficiency of the grinding mill, and R is the rotational speed of the current grinding mill.

[0051] By adaptively adjusting the rotational speed of the current grinding mill, the operating parameters of the grinding mill can be dynamically adjusted to adapt to different working conditions and mineral characteristics, thereby optimizing the grinding process.

[0052] It should be understood that when the value of B is larger, it indicates that the grinding mill can better adapt to the characteristics of different minerals and achieve more effective grinding and separation. At this time, the rotational speed and operating parameters of the current grinding mill can match the physical characteristics of the minerals, and the adjustment degree of the rotational speed R of the current grinding mill is relatively small, infinitely close to the current rotational speed R; the smaller the value of B means that the rotational speed and operating parameters of the grinding mill do not match the physical characteristics of the minerals, resulting in low grinding efficiency. At this time, the rotational speed is adjusted more intensively, and the adjusted rotational speed is limited by the adjustment parameter not to exceed the critical speed, and on this basis, the rotational speed is increased to improve the grinding efficiency.

[0053] Based on the above adaptive modification of the rotational speed, a more precise rotational speed control strategy can be achieved to adapt to the characteristics of different minerals, improve the efficiency and economic benefits of the entire beneficiation process. The grinding and mineral classification are completed through the set rotational speed, thereby differentiating different minerals and realizing the beneficiation operation.

[0054] Embodiment 2 A fine grinding control device for ore dressing, which device comprises a grinding mill and a classifier; the grinding mill is used to control its rotation speed to grind materials and send them into the classifier from the mineral outlet of the grinding mill; the classifier is used to divide the ground products into different kinds of minerals and send them into the corresponding storage areas from the overflow outlet of the classifier, so that the selected minerals can be temporarily stored in specific areas respectively for further processing or separation, and the unqualified minerals are sent back to the grinding mill from the classifier's return sand outlet for re-grinding to ensure that the particle size of the ground products meets the requirements of subsequent ore dressing operations. Among them, the classifier uses the gravity classification method to classify by the balance relationship between the buoyancy and gravity that the ground products receive in the liquid medium, and by using the difference in the density of ore particles, by adjusting the medium concentration and flow rate, the settling speeds of ore particles with different densities are made different, so as to achieve the classification between different minerals.

[0055] Among them, each storage area is provided with a weighing sensor for detecting the weight of the minerals stored in the storage area, and by using the weights of the minerals stored in all storage areas at the same moment, the weight proportion of each storage area at each moment is determined; screening analysis equipment is provided at both the overflow outlet and the return sand outlet of the classifier for calculating the classification efficiency of the storage area at each moment by the classification efficiency calculation method; a laser particle size analyzer is provided at the overflow outlet of the classifier for collecting the average particle size of the storage area.

[0056] It should be noted that both the gravity classification method and the classification efficiency calculation method are well-known technologies and will not be elaborated here.

[0057] The above technical features constitute the best embodiment of this application, which has strong adaptability and the best implementation effect. Non-essential technical features can be added or reduced according to actual needs to meet the requirements of different situations.

Claims

1. A fine grinding control method for ore dressing, characterized in that The method includes: Collection step: After the grinding mill operation starts and runs stably, in real time, collect the weight ratio, average particle size, and classification efficiency of the minerals in different storage areas at each collection moment after the classifier sends different types of minerals into the corresponding storage areas. Processing step: Reflect the overall grinding efficiency of the grinding mill by analyzing the data characteristics of all types of minerals collected at all collection moments. Specifically: S1. After performing curve fitting on the weight ratio and average particle size of each type of mineral at any collection moment and all previous collection moments respectively, analyze the curvature characteristics of the corner points on the fitting curve and the range characteristics of the data around the corner points to determine the grinding fitting degree of each type of mineral at the said any collection moment. S2. For each type of mineral, arrange the grinding fitting degrees calculated at the current and all previous collection moments in ascending order of time to obtain the fitting degree sequence of the current type of mineral. S3. Calculate the similarity between the fitting degree sequence of each current type of mineral and the classification efficiency sequence composed of the classification efficiencies at all collection moments at each collection moment; and use the similarities of each type of mineral at all collection moments as each row, and the similarities of different types of minerals at the same collection moment as each column to construct a trend change matrix. S4. Perform clustering processing on the elements in the trend change matrix; based on the proportion of the types of minerals in each clustering cluster in the clustering result and the average distance between the center of each clustering cluster and the centers of all other clustering clusters, determine the overall grinding efficiency of the grinding mill. Control step: Use the overall grinding efficiency of the grinding mill to adaptively adjust the rotation speed of the current grinding mill.

2. The fine grinding control method for ore dressing according to claim 1, wherein In the collection step, the time when the grinding mill operation starts and runs stably is the moment one minute after the grinding mill starts working.

3. A fine grinding control method for ore dressing according to claim 1, characterized in that, In step S1, the data around the corner point is all the data within a preset time window on the fitting curve where the corner point is located, with the corner point as the center.

4. A fine grinding control method for ore dressing according to claim 3, characterized in that, In step S1, record the grinding fitting degree of each type of mineral at the said any collection moment as A, and the calculation formula is: ; where exp represents the exponential function with the natural constant e as the base; represents the number of corner points on the i-th fitting curve; represents the curvature of the j-th corner point on the i-th fitting curve; represents the data range within a preset time window centered on the j-th corner point on the i-th fitting curve.

5. A fine grinding control method for ore dressing according to claim 1, characterized in that, In step S3, the similarity at each collection moment is obtained by calculating the similarity between the two sequences at each collection moment using the dynamic kernel correlation algorithm.

6. A fine grinding control method for ore dressing according to claim 1, characterized in that, In step S4, the clustering distance in the clustering processing is the numerical difference and position difference between elements.

7. A fine grinding control method for ore dressing according to claim 6, characterized in that, In step S4, the calculation formula for the characteristic value of the overall grinding efficiency of the grinding mill is: Calculate the ratio result of the proportion of the types of minerals in each clustering cluster and the average distance between the center of this clustering cluster and the centers of all other clustering clusters; Take the average level of the ratio results of all clustering clusters as the characteristic value of the overall grinding efficiency of the grinding mill.

8. A fine grinding control method for ore dressing according to claim 1, characterized in that, In the control step, the relational formula for adaptively adjusting the rotation speed of the current grinding mill using the overall grinding efficiency of the grinding mill is: ; wherein, is the adjusted rotational speed of the current grinding mill, is the preset adjustment parameter, e is the natural constant, B is the overall grinding efficiency of the grinding mill, and R is the rotational speed of the current grinding mill.

9. A fine grinding control method for ore dressing according to claim 8, characterized in that, The initial rotation speed when the grinding mill starts operating is a preset percentage of the critical rotation speed.

10. A fine grinding control device for ore dressing, which realizes a fine grinding control method for ore dressing as described in any one of claims 1-9, characterized in that, The device includes a grinding mill and a classifier; the grinding mill is used to control its rotation speed to grind materials and send them into the classifier from the mineral outlet of the grinding mill; the classifier is used to divide the ground products into different minerals and send them into the corresponding storage areas from the overflow outlet of the classifier, so that the selected minerals can be temporarily stored in specific areas respectively for further processing or separation, and send the unqualified minerals back to the grinding mill from the classifier's return sand outlet for re-grinding to ensure that the particle size of the ground products meets the requirements of subsequent ore dressing operations.

Images