Metal flotation control method, flotation equipment and medium based on expert system

Abstract

The embodiment of the present application provides a metal flotation control method, flotation equipment and medium based on an expert system, which belongs to the field of metal flotation. The method includes: obtaining current process production data, establishing a theoretical concentrate prediction grade model, and establishing a real-time concentrate prediction grade model; obtaining a dynamic concentrate grade difference according to the theoretical concentrate prediction grade model and the real-time concentrate prediction grade model; establishing a concentrate yield prediction model, and constructing a concentrate yield correction model; judging whether the current foam pump pool level is in a stable range, if not, directly adjusting the current process production data until the foam pump pool level returns to the stable range, and adjusting the current concentrate yield to the target concentrate yield according to the updated dynamic concentrate grade difference and concentrate yield correction model. The present application obtains the concentrate grade difference between theory and real time, and at the same time, through the concentrate yield correction model, accurately controls the concentrate grade and concentrate yield so that both meet the requirements.

Description

Technical Field

[0001] The present application relates to the field of metal flotation, and in particular to a metal flotation control method, flotation equipment and medium based on an expert system. Background Art

[0002] For a long time, flotation has mainly relied on experienced operators to perform single-factor or multi-factor coordinated control of flotation level, aeration, regulator, collector, frother and inhibitor variables based on on-site foam characteristics, such as foam color, size, stability, flow rate, etc. Due to the complex and changeable mineral properties of the selected raw materials, it is often the case that the concentrate grade and concentrate yield cannot meet the production requirements at the same time due to untimely control. Summary of the invention

[0003] In order to solve the above technical problems, the embodiments of the present application provide a metal flotation control method, flotation equipment and medium based on an expert system.

[0004] In a first aspect, an embodiment of the present application provides a metal flotation control method based on an expert system, the method comprising:

[0005] Acquiring current process production data under metal flotation and preprocessing the current process production data, wherein the current process production data includes current process production data based on the properties of the raw ore and current process production data based on the characteristics of the foam;

[0006] Establishing a theoretical concentrate grade prediction model based on the current process production data after pretreatment based on the properties of the raw ore, and establishing a real-time concentrate grade prediction model based on the current process production data after pretreatment based on the foam characteristics;

[0007] Obtaining a dynamic concentrate grade difference according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model;

[0008] Establishing a concentrate yield prediction model for the grade interval of the raw ore grade and the corresponding target concentrate yield interval, and constructing a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference;

[0009] Determine whether the current foam pump pool liquid level is in a stable range. If not, directly adjust the current process production data until the foam pump pool liquid level returns to the stable range, obtain the real-time concentrate prediction grade according to the adjusted process production data and the real-time concentrate prediction grade model, and update the dynamic concentrate grade difference;

[0010] According to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield.

[0011] In one embodiment, the current process production data includes ore grade, foam characteristics, aeration volume, foam pump pool level, overflow particle size, selected flow rate, selected concentration, flotation level, reagent dosage and pulp pH value, and the theoretical concentrate grade prediction model is established according to the pre-processed current process production data based on the properties of the ore, and the real-time concentrate grade prediction model is established according to the pre-processed current process production data based on the foam characteristics, including:

[0012] Establishing a theoretical concentrate grade prediction model according to formula (1) and pre-processed current process production data based on the properties of the raw ore, and outputting a theoretical concentrate grade through the theoretical concentrate grade prediction model;

[0013] Establishing a real-time concentrate grade prediction model according to formula (2) and pre-processed current process production data based on foam characteristics, and outputting the real-time concentrate grade through the real-time concentrate grade prediction model;

[0014] Formula (1): ;

[0015] Formula (2):

[0016] ;

[0017] in, For the sample, is the theoretical concentrate grade, For real-time concentrate grade, is the functional relationship between the theoretical concentrate grade and the current process production data based on the properties of the raw ore, is the functional relationship between the real-time concentrate grade and the production data based on the foam characteristics, is the original ore grade, For the foam characteristics, For the amount of gas, for the foam pump pool level, is the overflow particle size, For the selected flow, is the selected concentration, is the flotation level, For the dosage of the drug, is the pH value of the slurry.

[0018] In one embodiment, the establishment of a concentrate yield prediction model for the grade interval of the ore grade and the corresponding target concentrate yield interval, and the construction of a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference, includes:

[0019] Establishing the concentrate yield prediction model according to formula (3);

[0020] The dynamic concentrate grade difference is obtained according to formula (4), and a concentrate yield correction model is constructed according to the concentrate yield prediction model and the dynamic concentrate grade difference, wherein the concentrate yield correction model includes formula (5);

[0021] Formula (3): ;

[0022] Formula (4): ;

[0023] Formula (5): ;

[0024] in, ( ) is the grade interval of the original ore grade, ( ) is the target concentrate yield range, is the correction factor, is the dynamic concentrate grade difference, is the yield correction amount.

[0025] In one embodiment, the determining whether the current foam pump pool liquid level is in a stable range includes:

[0026] According to the mean and standard deviation of the historical foam pump pool liquid level, the upper limit of the stable interval is calculated by formula (6), and the lower limit of the stable interval is calculated by formula (7);

[0027] Determine whether the current foam pump pool liquid level is in the stable interval, the stable interval is [ ];

[0028] Formula (6): ;

[0029] Formula (7): ;

[0030] in, is the upper limit of the stability interval, is the lower limit of the stability interval, is the mean value of the historical foam pump pool liquid level, is the standard deviation of the historical foam pump pool liquid level.

[0031] In one embodiment, the adjusting the current process production data includes:

[0032] The target froth yield is determined according to the target concentrate yield, and the froth thickness standard interval is preset according to the target froth yield. The froth thickness standard interval is [ ];

[0033] According to formula (8), adjust the amount of frother and flotation liquid level in the reagent dosage;

[0034] Formula (8): ;

[0035] in, is the flotation level, is the flotation level increment, is the amount of foaming agent, is the increment of foaming agent dosage, is the lower limit of flotation foam thickness, It is the upper limit of flotation foam thickness.

[0036] In one embodiment, the adjusting the current process production data includes:

[0037] The bubble size in the foam feature is clustered by the clustering algorithm. Clustering is performed to obtain the bubble size that is too large , Foam size normal And the foam size is too small ;

[0038] According to formula (9) and the foam size, the amount of the foaming agent in the dosage of the medicine is adjusted;

[0039] Formula (9): ;

[0040] in, is the amount of foaming agent, It is the increment of foaming agent dosage.

[0041] In one embodiment, the adjusting the current process production data includes:

[0042] The pH difference between the expected pH value and the real-time pH value is obtained according to formula (10), and the pH value control model is established through formula (11);

[0043] Adjusting the amount of slurry pH adjusting agent in the dosage of the reagent according to the pH difference and the pH value control model;

[0044] Formula (10): = ;

[0045] Formula (11): + ;

[0046] in, is the expected pH value, is the real-time pH value, for The amount of slurry pH adjuster used at the time, for The amount of slurry pH adjuster used at the time, is the correction coefficient of slurry pH adjuster dosage, It is the correction amount of slurry pH adjuster.

[0047] In one embodiment, after adjusting the current concentrate yield to the target concentrate yield, the method further comprises:

[0048] A theoretical tailings grade prediction model is established based on formula (12) and the current process production data after pretreatment;

[0049] According to formula (13), the tailings grade difference between the online tailings grade and the theoretical tailings grade is obtained;

[0050] A tailings grade difference threshold is preset, and if the tailings grade difference is greater than or equal to the tailings grade difference threshold, the amount of frother in the reagent dosage and the flotation liquid level are adjusted according to formula (14) and the tailings grade difference;

[0051] Formula (12): ;

[0052] Formula (13): = ;

[0053] Formula (14): ;

[0054] in, is the theoretical tailings grade, For the sample, is the functional relationship between the theoretical tailings grade and the current process production data based on the properties of the original ore, For the online grade of tailings, The tailings grade is poor. is the target tailings concentrate yield, is the tailings grade difference threshold.

[0055] In a second aspect, an embodiment of the present application provides a flotation device, including a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is run by the processor, the expert system-based metal flotation control method provided in the first aspect is executed.

[0056] In a third aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed on a processor, executes the expert system-based metal flotation control method provided in the first aspect.

[0057] The metal flotation control method based on the expert system provided in the present application obtains the current process production data under metal flotation and preprocesses the current process production data, wherein the current process production data includes the current process production data based on the properties of the raw ore and the current process production data based on the foam characteristics; a theoretical concentrate grade prediction model is established according to the preprocessed current process production data based on the properties of the raw ore, and a real-time concentrate grade prediction model is established according to the preprocessed current process production data based on the foam characteristics; a dynamic concentrate grade difference is obtained according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model; and a raw ore grade prediction model is established. The grade interval of the ore grade and the concentrate yield prediction model of the corresponding target concentrate yield interval, and the concentrate yield correction model is constructed according to the concentrate yield prediction model and the dynamic concentrate grade difference; it is determined whether the current foam pump pool level is in a stable interval, if not, the current process production data is directly adjusted until the foam pump pool level returns to the stable interval, and the real-time concentrate prediction grade is obtained according to the adjusted process production data and the real-time concentrate prediction grade model, and the dynamic concentrate grade difference is updated; according to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield. This application obtains the concentrate grade difference between theory and real time by adjusting the process production data, and at the same time, through the concentrate yield correction model, the concentrate grade and concentrate yield are accurately controlled so that they both meet the requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and should not be regarded as limiting the scope of protection of the present application. In each of the drawings, similar components are numbered similarly.

[0059] Figure 1 A schematic diagram of a process of a metal flotation control method based on an expert system provided in an embodiment of the present application is shown;

[0060] Figure 2 Another schematic diagram of the process of the metal flotation control method based on the expert system provided in the embodiment of the present application is shown;

[0061] Figure 3 Another schematic diagram of a process of a metal flotation control method based on an expert system provided in an embodiment of the present application is shown;

[0062] Figure 4 A structural schematic diagram of a metal flotation control device based on an expert system provided in an embodiment of the present application is shown.

[0063] Icons: 400 - metal flotation control device based on expert system, 401 - first acquisition module, 402 - establishment module, 403 - second acquisition module, 404 - second establishment module, 405 - judgment module, 406 - adjustment module. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0065] The components of the embodiments of the present application generally described and shown in the drawings herein may be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0066] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present application, are intended only to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing items or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations of the foregoing items.

[0067] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0068] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meanings as those generally understood by those skilled in the art to which the various embodiments of the present application belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meanings as the contextual meanings in the relevant technical field and will not be interpreted as having idealized meanings or overly formal meanings unless clearly defined in the various embodiments of the present application.

[0069] Example 1

[0070] Metal flotation process control is a multi-coordinated, coupled, and fuzzy control. Due to the complex and changeable nature of the ore, and the fact that some mineral properties are difficult to accurately detect and quantitatively express online using existing technologies, production personnel can only qualitatively define the flotation phenomenon based on their own experience and cannot make good pre-adjustments, resulting in delayed adjustment, large production fluctuations, and affecting flotation indicators.

[0071] The embodiment of the present application provides a metal flotation control method based on an expert system.

[0072] See also Figure 1 The metal flotation control method based on the expert system includes steps S101-S106:

[0073] S101: obtaining current process production data of metal flotation and preprocessing the current process production data, wherein the current process production data includes current process production data based on the properties of the raw ore and current process production data based on the characteristics of the foam.

[0074] In this embodiment, a process investigation is carried out, and relevant production data affecting the flotation process control are collected and stored in the database, such as foam characteristics (color, size, stability, flow rate, etc.), flotation liquid level, reagent dosage (adjusting agent, collector, frother, etc.), foam pump pool level, selected flow, selected concentration, overflow particle size, ore grade, concentrate grade, tailings grade, and aeration volume. It should be noted that the current process production data based on the properties of the ore includes the ore grade, excluding foam characteristics, and the current process production data based on the foam characteristics includes the foam characteristics, excluding the ore grade, and the other common production data are obtained at the same time.

[0075] Optionally, the collected flotation process data is preprocessed by data cleaning, such as identifying isolated points, filling in missing values, and removing noise and irrelevant data from the original data. Among them, the noise and irrelevant isolated value data in the original data are removed by binning, clustering and regression methods; the missing values ​​are filled by using default values, attribute averages and similar sample averages; the data imbalance problem is solved by oversampling methods for small sample data sets and undersampling methods for large sample data sets.

[0076] Optionally, the flotation process data after cleaning can be preprocessed by data transformation. Some data need to be transformed according to the requirements of the data mining algorithm for their data types before they can be used. For example, smoothing the data can not only remove noise data, but also discretize numerical data; for example, normalizing the data, scaling the data according to a certain ratio so that it falls into a specific interval, such as [0,1] or [-1,1]. The normalization methods include minimum-maximum normalization, zero-mean normalization, and decimal calibration normalization.

[0077] Optionally, data reduction can be performed on the cleaned data, that is, reducing the amount of original data without destroying the integrity of the data, and obtaining a mining data set that is much smaller than the original data. For example, dimension reduction, that is, deleting those irrelevant, weakly related or associated attributes; for example, value reduction, that is, selecting smaller data to replace the original data to reduce the amount of data. Commonly used methods include histograms, clustering, sampling, regression and logarithmic models.

[0078] S102: Establishing a theoretical concentrate grade prediction model based on the preprocessed current process production data based on the original ore properties, and establishing a real-time concentrate grade prediction model based on the preprocessed current process production data based on the foam characteristics.

[0079] In this embodiment, a theoretical concentrate grade prediction model is defined based on the current process production data based on the properties of the raw ore after pretreatment. The model aims to predict the expected grade of the concentrate under ideal or theoretical conditions based on the current process production data of the properties of the raw ore, which can help set the target value of the concentrate grade, evaluate the process potential, and provide a reference benchmark for real-time prediction. A real-time concentrate grade prediction model is defined based on the current process production data based on the foam characteristics after pretreatment. The model predicts the current grade of the concentrate in real time based on the current process production data of the foam characteristics, so as to adjust the process parameters in time and optimize the flotation effect.

[0080] In one embodiment, the current process production data includes ore grade, foam characteristics, aeration volume, foam pump pool level, overflow particle size, selected flow rate, selected concentration, flotation level, reagent dosage and pulp pH value, and the theoretical concentrate grade prediction model is established according to the pre-processed current process production data based on the properties of the ore, and the real-time concentrate grade prediction model is established according to the pre-processed current process production data based on the foam characteristics, including:

[0081] Establishing a theoretical concentrate grade prediction model according to formula (1) and pre-processed current process production data based on the properties of the raw ore, and outputting a theoretical concentrate grade through the theoretical concentrate grade prediction model;

[0082] Establishing a real-time concentrate grade prediction model according to formula (2) and pre-processed current process production data based on foam characteristics, and outputting the real-time concentrate grade through the real-time concentrate grade prediction model;

[0083] Formula (1): ;

[0084] Formula (2):

[0085] ;

[0086] in, For the sample, is the theoretical concentrate grade, For real-time concentrate grade, is the functional relationship between the theoretical concentrate grade and the current process production data based on the properties of the raw ore, is the functional relationship between the real-time concentrate grade and the production data based on the foam characteristics, is the original ore grade, For the foam characteristics, For the amount of gas, for the foam pump pool level, is the overflow particle size, For the selected flow, is the selected concentration, is the flotation level, For the dosage of the drug, is the pH value of the slurry.

[0087] In this example, the ore grade interval is the content of useful metals in the slurry, and the corresponding range , each element in this set They all represent a specific grade range corresponding to the original ore grade; foam characteristics It includes the color, size, stability, flow rate and other aspects of the foam, and is the carrier for the mineral particles to attach during the flotation process; the amount of aeration , refers to the amount of air injected into the pulp during the flotation process. The amount of air directly affects the number and size of bubbles, which in turn affects the flotation effect; the foam pump pool level , refers to the liquid level height when the foam overflowing from the flotation machine in the flotation system flows into the foam pump pool, which is used to control the concentrate yield during the flotation process; the first overflow particle size can be the content of mineral particles with a particle size less than or equal to the 200-mesh screen aperture; the second overflow particle size can be the content of mineral particles with a particle size greater than the 80-mesh screen aperture; the selected flow rate ( ) refers to the pulp flow rate entering the flotation equipment (such as flotation machine, flotation column, etc.); the concentration of the selected pulp ( ) refers to the ratio of the mass of solid particles (i.e. useful minerals and gangue minerals) in the slurry entering the flotation equipment to the total volume of the slurry; the flotation level , refers to the liquid level of the slurry in the flotation machine or flotation column, which is used to control the bubble distribution and the suspension state of mineral particles during the flotation process; the dosage of the reagent Including dosage adjustment , Collector dosage and amount of foaming agent ; pH value of slurry , refers to the pH value of the slurry. The pH value is one of the important factors affecting the flotation effect, affecting the properties of the surface of mineral particles and the activity of the reagent.

[0088] For example, a real-time concentrate grade prediction model is constructed based on foam characteristics and other process production data. Including foam color ,size ,stability , flow rate .in, = (H, S, V), H is the foam color, S is the foam saturation, and V is the foam brightness; =(number of large bubbles, number of medium bubbles, number of small bubbles); =(large bubble stability, medium bubble stability, small bubble stability); =(X-axis flow rate, Y-axis flow rate). As the dosage of reagents is adjusted, its foam characteristics and other process production data will also change. The real-time concentrate grade prediction model predicts the real-time concentrate grade range of the concentrate that flows into the pump pool from the flotation machine through froth flotation.

[0089] S103: Obtaining a dynamic concentrate grade difference according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model.

[0090] In this embodiment, the data obtained in step S102 Theoretical concentrate grade and real-time concentrate grade , thus obtaining the concentrate grade difference: .

[0091] S104: establishing a concentrate yield prediction model for the grade range of the raw ore grade and the corresponding target concentrate yield range, and constructing a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference.

[0092] In this embodiment, by analyzing historical data, the relationship between the ore grade range and the concentrate yield is found, and a concentrate yield prediction model is established to predict the target concentrate yield at different ore grades. The concentrate yield prediction model is combined with the dynamic concentrate grade difference to construct a concentrate yield correction model.

[0093] See also Figure 2 In one embodiment, the S104 includes steps S1041 and S1042:

[0094] S1041: Establish the concentrate yield prediction model according to formula (3).

[0095] In this embodiment, by analyzing historical data, the grade interval of the original ore is matched. Corresponding concentrate yield range , such as the ore grade interval is The corresponding target concentrate yield should be ,in, , , It is the number of ore grade intervals and the number of concentrate yield intervals.

[0096] Specifically, through the ore grade interval and target concentrate yield Establish a concentrate yield prediction model, ,Right now .

[0097] S1042: Obtain the dynamic concentrate grade difference according to formula (4), and construct a concentrate yield correction model according to the concentrate yield prediction model and the dynamic concentrate grade difference, wherein the concentrate yield correction model includes formula (5).

[0098] In this embodiment, since production conditions may change at any time and the concentrate grade is constantly changing, it is necessary to update and adjust the parameters of the concentrate yield correction model in real time. By changing the amount of frother and the flotation liquid level, a new real-time concentrate grade interval and a new dynamic concentrate grade difference are obtained. , combined with the same dynamic and The concentrate yield correction model is modified to ensure that it can achieve the target concentrate yield while reaching the theoretical concentrate grade.

[0099] Specifically, the dynamic concentrate grade difference is obtained according to formula (4): ; The concentrate yield correction model is formula (5): Make adjustments, including ( ) is the grade interval of the original ore grade, ( ) is the target concentrate yield range, is the correction factor, is the dynamic concentrate grade difference, is the yield correction amount.

[0100] S105: Determine whether the current foam pump pool liquid level is in a stable range. If not, directly adjust the current process production data until the foam pump pool liquid level returns to the stable range, obtain the real-time concentrate prediction grade according to the adjusted process production data and the real-time concentrate prediction grade model, and update the dynamic concentrate grade difference.

[0101] In this embodiment, the liquid level of the foam pump pool includes a low interval, a stable interval, a high interval and a protection line. When the liquid level of the foam pump pool is in the low interval, the low interval indicates that the concentrate stock in the pump pool is small, and the liquid level is lower than the lower limit of the normal operating range. This is because the concentrate yield is low and the concentrate yield needs to be increased; the stable interval is an ideal range of the foam pump pool liquid level, indicating that the concentrate stock in the pump pool is moderate, that is, the corresponding concentrate yield is the target concentrate yield, which is conducive to the stable flotation operation; the high interval indicates that the concentrate stock in the pump pool is too much and the concentrate yield needs to be reduced; and if the pump pool liquid level is at the protection line, it is necessary to adjust the flotation liquid level in time to reduce the yield to prevent the foam pump pool from overflowing.

[0102] Optionally, when the liquid level in the foam pump pool is not in a stable range, that is, it may be in a low range, a high range or a protection line, it is necessary to adjust the current process production data until the liquid level in the foam pump pool returns to a stable range, and then obtain the real-time concentrate predicted grade based on the adjusted process production data and the real-time concentrate predicted grade model to update the dynamic concentrate grade difference.

[0103] In one embodiment, the method of determining whether the current foam pump pool liquid level is in a stable range includes: calculating the upper limit of the stable range by formula (6) and calculating the lower limit of the stable range by formula (7) based on the mean and standard deviation of the historical foam pump pool liquid level; determining whether the current foam pump pool liquid level is in the stable range, wherein the stable range is [ ]; Formula (6): ;Formula (7): ;in, is the upper limit of the stability interval, is the lower limit of the stability interval, is the mean value of the historical foam pump pool liquid level, is the standard deviation of the historical foam pump pool liquid level.

[0104] In this embodiment, by analyzing the historical data of the foam pump pool liquid level, the mean value and standard deviation as well as the foam pump pool liquid level range are obtained, and the foam pump pool liquid level is divided into pump pool control intervals, including a low interval [0, ), stable interval[ , ]、High range( , ) and protection line Among them, the foam pump pool level range Calculate the protection line, the formula is: .

[0105] See also Figure 3 In one embodiment, the adjusting the current process production data includes S301-S302:

[0106] S301: Determine a target froth yield according to the target concentrate yield, and preset a flotation froth thickness standard interval according to the target froth yield, wherein the flotation froth thickness standard interval is [ ].

[0107] In this embodiment, in the field of flotation technology, the concentrate yield and the froth yield are equivalent. According to historical data, the flotation froth thickness interval is set for the target froth yield [ ].

[0108] S302: adjusting the amount of frother and the flotation liquid level in the reagent dosage according to formula (8);

[0109] Formula (8): ;

[0110] in, is the flotation level, is the flotation level increment, is the amount of foaming agent, is the increment of foaming agent dosage, is the lower limit of flotation foam thickness, It is the upper limit of flotation foam thickness.

[0111] In this embodiment, when the flotation foam thickness is Greater than When the foam layer is too thick, it means that the amount of foaming agent needs to be reduced. and increase the flotation level , by changing the production parameters of frother dosage and flotation liquid level, a new control balance will be obtained.

[0112] For example, if the current foam layer thickness Less than When the foam layer thickness is too small, it means that the amount of foaming agent needs to be increased. and reduce the flotation level , by changing the production parameters of frother dosage and flotation liquid level, a new control balance will be obtained.

[0113] In one embodiment, the adjusting the current process production data includes: clustering the foam size in the foam feature by a clustering algorithm. Clustering is performed to obtain the bubble size that is too large , Foam size normal And the foam size is too small ; The amount of the foaming agent in the dosage of the medicine is adjusted according to formula (9) and the foam size; Formula (9): ;in, is the amount of foaming agent, It is the increment of foaming agent dosage.

[0114] In this embodiment, during the flotation process, the size of the foam is a key foam characteristic parameter. Excessive foam means excessive frother, resulting in good foam stability and not easy to break. At the same time, the foam carries too much non-target substances, reducing the real-time concentrate grade. On the contrary, too small foam indicates insufficient frother, resulting in low flotation efficiency and ineffective enrichment of target substances. The foam size is classified by data mining algorithms (such as cluster analysis). The normal foam size corresponding to the stable interval can be mined according to historical data and clustering algorithms. According to the normal foam size, the foam state can be divided into the foam size too large category, the foam size normal category, and the foam size too small category.

[0115] Exemplary, when the foam belongs to the category of foam size oversize If the amount of foaming agent is too high, it means that the amount of foaming agent is too high. , can reduce the foam over-stability and return the foam size to the normal range. If the amount of foaming agent is too small, it means that the amount of foaming agent is too small. , can improve the stability of the foam and return the foam size to the normal range. When the amount of foaming agent is moderate, no adjustment is needed. Keep the current amount of foaming agent. A stable flotation state can be maintained.

[0116] S106: According to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield.

[0117] In this embodiment, according to the new dynamic concentrate grade difference , through the formula Adjustment correction factor , yield correction , so that the current concentrate yield reaches the target concentrate yield.

[0118] In one embodiment, the adjusting the current process production data includes: obtaining a pH difference between an expected pH value and a real-time pH value according to formula (10), and establishing a pH value control model according to formula (11); and adjusting the amount of the slurry pH adjusting agent in the reagent dosage according to the pH difference and the pH value control model; formula (10): = ;Formula (11): + ;in, is the expected pH value, is the real-time pH value, for The amount of slurry pH adjuster used at the time, for The amount of slurry pH adjuster used at the time, is the correction coefficient of slurry pH adjuster dosage, It is the correction amount of slurry pH adjuster.

[0119] In this embodiment, for any mineral flotation, there is an appropriate pH value under certain flotation conditions. Only under the appropriate pulp pH conditions can better separation indicators such as concentrate grade, concentrate yield and reagent dosage be obtained. Optionally, according to the stable range of the pump pool liquid level, combined with the flotation characteristics of the mineral and historical experimental data, the expected pH value can be set. . Measure the real-time pH value of the current slurry , and Compare and calculate the difference ∆PH: , It indicates the degree of deviation between the current pH value and the expected pH value.

[0120] For example, during the flotation process, the pH value of the slurry may change due to various reasons (such as ore properties, flotation reagent consumption, water hardness, etc.). Therefore, the pH value of the slurry needs to be checked regularly. ), pH value variation range (∆PH), slurry pH adjuster dosage correction factor ( ) and the amount of slurry pH adjuster correction ( ) to calculate the current moment ( The amount of slurry pH adjuster at the time ( ),Right now , where the correction factor for the amount of slurry pH adjuster is Determined based on experimental data or experience, used to adjust the sensitivity of pH changes to the dosage of reagents, and the correction amount of the slurry pH adjuster dosage It is also determined based on experimental data or experience. Finally, the amount of the adjusting agent added at the current moment is adjusted according to the calculation results, so that the pH value of the slurry gradually approaches and stabilizes at the standard value.

[0121] In one embodiment, after adjusting the current concentrate yield to the target concentrate yield, the method further comprises: establishing a theoretical tailings grade prediction model according to formula (12) and the pre-processed current process production data; obtaining the tailings grade difference between the online tailings grade and the theoretical tailings grade according to formula (13); presetting a tailings grade difference threshold, and if the tailings grade difference is greater than or equal to the tailings grade difference threshold, adjusting the amount of frother in the reagent dosage and the flotation liquid level according to formula (14) and the tailings grade difference; Formula (12): ;Formula (13): = ;Formula (14): ;in, is the theoretical tailings grade, For the sample, is the functional relationship between the theoretical tailings grade and the current process production data based on the properties of the original ore, For the online grade of tailings, The tailings grade is poor. is the target tailings concentrate yield, is the tailings grade difference threshold.

[0122] In this embodiment, a tailings grade difference threshold is preset based on historical data. , used to determine whether the reagent dosage and flotation level need to be adjusted during tailings flotation. Theoretical tailings grade Calculated tailings grade difference , greater than or equal to the tailings grade difference threshold , it indicates that useful metals are lost and it is necessary to obtain useful metals from tailings with a tailings grade greater than the threshold. At this time, the amount of frother is increased from Increase to , the flotation level is changed from Increase to It should be noted that the online tailings grade Obtained through real-time online detection by equipment instruments.

[0123] The metal flotation control method based on the expert system provided in this embodiment obtains the current process production data under metal flotation and preprocesses the current process production data, wherein the current process production data includes the current process production data based on the properties of the raw ore and the current process production data based on the foam characteristics; a theoretical concentrate grade prediction model is established according to the preprocessed current process production data based on the properties of the raw ore, and a real-time concentrate grade prediction model is established according to the preprocessed current process production data based on the foam characteristics; a dynamic concentrate grade difference is obtained according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model; and a raw ore grade prediction model is established. The concentrate yield prediction model of the grade interval of the grade and the corresponding target concentrate yield interval is constructed, and a concentrate yield correction model is constructed according to the concentrate yield prediction model and the dynamic concentrate grade difference; it is determined whether the current foam pump pool level is in a stable interval, if not, the current process production data is directly adjusted until the foam pump pool level returns to the stable interval, and the real-time concentrate prediction grade is obtained according to the adjusted process production data and the real-time concentrate prediction grade model, and the dynamic concentrate grade difference is updated; according to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield. This application obtains the concentrate grade difference between theory and real time by adjusting the process production data, and at the same time, through the concentrate yield correction model, the concentrate grade and concentrate yield are accurately controlled so that they both meet the requirements.

[0124] Example 2

[0125] In addition, an embodiment of the present application provides a metal flotation control device based on an expert system, which is applied to flotation equipment.

[0126] like Figure 4 As shown, the metal flotation control device 400 based on the expert system includes:

[0127] The first acquisition module 401 is used to acquire current process production data under metal flotation and preprocess the current process production data, wherein the current process production data includes current process production data based on the properties of the raw ore and current process production data based on the characteristics of the foam.

[0128] Establishing module 402 is used to establish a theoretical concentrate grade prediction model based on the pre-processed current process production data based on the original ore properties, and to establish a real-time concentrate grade prediction model based on the pre-processed current process production data based on the foam characteristics.

[0129] The second acquisition module 403 is used to acquire a dynamic concentrate grade difference according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model.

[0130] The second establishing module 404 is used to establish a concentrate yield prediction model for the grade range of the raw ore grade and the corresponding target concentrate yield range, and to construct a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference.

[0131] The judgment module 405 is used to judge whether the current foam pump pool liquid level is in a stable range. If not, the current process production data is directly adjusted until the foam pump pool liquid level returns to the stable range. The real-time concentrate prediction grade is obtained according to the adjusted process production data and the real-time concentrate prediction grade model, and the dynamic concentrate grade difference is updated.

[0132] The adjustment module 406 is used to adjust the current concentrate yield to the target concentrate yield according to the updated dynamic concentrate grade difference and the concentrate yield correction model.

[0133] The metal flotation control device 400 based on the expert system provided in this embodiment can implement the metal flotation control method based on the expert system provided in Embodiment 1, and will not be described again here to avoid repetition.

[0134] The metal flotation control device based on the expert system provided in this embodiment obtains the current process production data under metal flotation and preprocesses the current process production data, wherein the current process production data includes the current process production data based on the properties of the raw ore and the current process production data based on the foam characteristics; a theoretical concentrate grade prediction model is established according to the preprocessed current process production data based on the properties of the raw ore, and a real-time concentrate grade prediction model is established according to the preprocessed current process production data based on the foam characteristics; a dynamic concentrate grade difference is obtained according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model; and a raw ore grade prediction model is established. The concentrate yield prediction model of the grade interval of the grade and the corresponding target concentrate yield interval is constructed, and a concentrate yield correction model is constructed according to the concentrate yield prediction model and the dynamic concentrate grade difference; it is determined whether the current foam pump pool level is in a stable interval, if not, the current process production data is directly adjusted until the foam pump pool level returns to the stable interval, and the real-time concentrate prediction grade is obtained according to the adjusted process production data and the real-time concentrate prediction grade model, and the dynamic concentrate grade difference is updated; according to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield. This application obtains the concentrate grade difference between theory and real time by adjusting the process production data, and at the same time, through the concentrate yield correction model, the concentrate grade and concentrate yield are accurately controlled so that they both meet the requirements.

[0135] Example 3

[0136] In addition, an embodiment of the present application provides a flotation device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program runs on the processor, the expert system-based metal flotation control method provided in Example 1 is executed.

[0137] The flotation equipment provided in the embodiment of the present invention can execute the steps of the metal flotation control method based on the expert system provided in the above method embodiment 1, which will not be described again here to avoid repetition.

[0138] Example 4

[0139] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the metal flotation control method based on the expert system provided in Example 1 is implemented.

[0140] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0141] The computer-readable storage medium provided in this embodiment can implement the metal flotation control method based on the expert system provided in Example 1, and will not be described again here to avoid repetition.

[0142] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or terminal including the element.

[0143] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0144] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A metal flotation control method based on an expert system, characterized in that: The method comprises: Acquiring current process production data under metal flotation and preprocessing the current process production data, wherein the current process production data includes current process production data based on the properties of the raw ore and current process production data based on the characteristics of the foam; Establishing a theoretical concentrate grade prediction model based on the current process production data after pretreatment based on the properties of the raw ore, and establishing a real-time concentrate grade prediction model based on the current process production data after pretreatment based on the foam characteristics; Obtaining a dynamic concentrate grade difference according to the theoretical concentrate grade prediction model and the real-time concentrate grade prediction model; Establishing a concentrate yield prediction model for the grade interval of the raw ore grade and the corresponding target concentrate yield interval, and constructing a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference; Determine whether the current foam pump pool liquid level is in a stable range. If not, directly adjust the current process production data until the foam pump pool liquid level returns to the stable range, obtain the real-time concentrate prediction grade according to the adjusted process production data and the real-time concentrate prediction grade model, and update the dynamic concentrate grade difference; According to the updated dynamic concentrate grade difference and the concentrate yield correction model, the current concentrate yield is adjusted to the target concentrate yield.

2. The method according to claim 1, characterized in that The current process production data includes raw ore grade, foam characteristics, aeration volume, foam pump pool level, overflow particle size, selected flow rate, selected concentration, flotation level, reagent dosage and pulp pH value. The theoretical concentrate grade prediction model is established based on the current process production data after pretreatment based on the raw ore properties, and the real-time concentrate grade prediction model is established based on the current process production data after pretreatment based on the foam characteristics, including: Establishing a theoretical concentrate grade prediction model according to formula (1) and pre-processed current process production data based on the properties of the raw ore, and outputting a theoretical concentrate grade through the theoretical concentrate grade prediction model; Establishing a real-time concentrate grade prediction model according to formula (2) and pre-processed current process production data based on foam characteristics, and outputting the real-time concentrate grade through the real-time concentrate grade prediction model; Formula (1): ; Formula (2): ； in, For the sample, is the theoretical concentrate grade, For real-time concentrate grade, is the functional relationship between the theoretical concentrate grade and the current process production data based on the properties of the raw ore, is the functional relationship between the real-time concentrate grade and the production data based on the foam characteristics. is the original ore grade, For the foam characteristics, For the amount of gas, for the foam pump pool level, is the overflow particle size, For the selected flow, is the selected concentration, is the flotation level, For the dosage of the drug, is the pH value of the slurry.

3. The method according to claim 1, characterized in that The method of establishing a concentrate yield prediction model for the grade interval of the raw ore grade and the corresponding target concentrate yield interval, and constructing a concentrate yield correction model based on the concentrate yield prediction model and the dynamic concentrate grade difference, includes: Establishing the concentrate yield prediction model according to formula (3); The dynamic concentrate grade difference is obtained according to formula (4), and a concentrate yield correction model is constructed according to the concentrate yield prediction model and the dynamic concentrate grade difference, wherein the concentrate yield correction model includes formula (5); Formula (3): ; Formula (4): ; Formula (5): ; in, ( ) is the grade interval of the original ore grade, ( ) is the target concentrate yield range, is the correction factor, is the dynamic concentrate grade difference, is the yield correction amount.

4. The method according to claim 1, characterized in that: The determining whether the current foam pump pool liquid level is in a stable range includes: According to the mean and standard deviation of the historical foam pump pool liquid level, the upper limit of the stable interval is calculated by formula (6), and the lower limit of the stable interval is calculated by formula (7); Determine whether the current foam pump pool liquid level is in the stable interval, the stable interval is [ ]; Formula (6): ; Formula (7): ; in, is the upper limit of the stability interval, is the lower limit of the stability interval, is the mean value of the historical foam pump pool liquid level, is the standard deviation of the historical foam pump pool liquid level.

5. The method according to claim 1, characterized in that The adjusting the current process production data includes: The target froth yield is determined according to the target concentrate yield, and the froth thickness standard interval is preset according to the target froth yield. The froth thickness standard interval is [ ]; According to formula (8), adjust the amount of frother and flotation liquid level in the reagent dosage; Formula (8): ; in, is the flotation level, is the flotation level increment, is the amount of foaming agent, is the increment of foaming agent dosage, is the lower limit of flotation foam thickness, It is the upper limit of flotation foam thickness.

6. The method according to claim 1, characterized in that The adjusting the current process production data includes: The bubble size in the foam feature is clustered by the clustering algorithm. Clustering is performed to obtain the bubble size that is too large , Foam size normal type And the foam size is too small ; According to formula (9) and the foam size, the amount of the foaming agent in the dosage of the medicine is adjusted; Formula (9): ; in, is the amount of foaming agent, It is the increment of foaming agent dosage.

7. The method according to claim 1, characterized in that The adjusting the current process production data includes: The pH difference between the expected pH value and the real-time pH value is obtained according to formula (10), and the pH value control model is established through formula (11); Adjusting the amount of slurry pH adjusting agent in the dosage of the reagent according to the pH difference and the pH value control model; Formula (10): = ; Formula (11): + ; in, is the expected pH value, is the real-time pH value, for The amount of slurry pH adjuster used at the time, for The amount of slurry pH adjuster used at the time, is the correction coefficient of slurry pH adjuster dosage, It is the correction amount of slurry pH adjuster.

8. The method according to claim 1, characterized in that After adjusting the current concentrate yield to the target concentrate yield, the method further comprises: A theoretical tailings grade prediction model is established based on formula (12) and the current process production data after pretreatment; According to formula (13), the tailings grade difference between the online tailings grade and the theoretical tailings grade is obtained; A tailings grade difference threshold is preset, and if the tailings grade difference is greater than or equal to the tailings grade difference threshold, the amount of frother in the reagent dosage and the flotation liquid level are adjusted according to formula (14) and the tailings grade difference; Formula (12): ; Formula (13): = ; Formula (14): ; in, is the theoretical tailings grade, For the sample, is the functional relationship between the theoretical tailings grade and the current process production data based on the properties of the original ore, For the online grade of tailings, The tailings grade is poor. is the target tailings concentrate yield, is the tailings grade difference threshold.

9. A flotation device, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is run on the processor, the metal flotation control method based on an expert system according to any one of claims 1 to 8 is executed.

10. A computer-readable storage medium, characterized in that: The computer program is stored therein, and when the computer program is run on a processor, the metal flotation control method based on an expert system according to any one of claims 1 to 8 is executed.

Images