A predictive control method for intelligent process of dry heavy medium coal preparation

Abstract

The invention discloses a method for predicting and controlling an intelligent process of dry heavy medium coal preparation, comprising the following steps: firstly collecting data of raw coal entering the washing process, and collecting statistics on a priori selectivity information of the collected data; then establishing a bed density dynamic model based on evolution time, and forming a stable bed density; then calculating and obtaining a sorting density model, establishing an empirical allocation function model based on four parameters of the sorting density model, possible deviations of the sorting system, and boundary control parameters, and calculating the product allocation rate corresponding to each density level interval; finally, predicting and calculating the product yield and ash content of the coal preparation system through the a priori selectivity information and the product allocation rate corresponding to each density level interval, and making corrections; the invention integrates different dry heavy medium sorting mathematical models, enhances the model's ability to describe the steady-state and dynamic characteristics of the sorting process, improves the prediction accuracy of the ash content and yield of clean coal, and solves the problem of lack of theoretical models in the field of dry heavy medium and difficulty in implementing dynamic control.

Description

Technical Field

[0001] The invention relates to a dry heavy medium coal separation technology, belongs to the field of coal separation, and specifically relates to a dry heavy medium coal separation intelligent process prediction and control method. Background Art

[0002] In the traditional coal sorting industry, the production process parameters need to be adjusted through manual decision-making. This method is highly subjective, resulting in large lag and poor reliability in the adjustment of the process parameters of the sorting process, making the quality of the sorted clean coal products unstable. Implementing effective intelligent control of the sorting system can effectively improve the stability and economy of the coal preparation plant operation.

[0003] Dry heavy medium separation can strictly sort coal according to density during the coal sorting process, so that it can efficiently separate coal particles of different densities. It has the advantages of high sorting accuracy and quantity efficiency, strong adaptability, wide particle size range, and easy automation.

[0004] When the dynamic model of dry heavy medium separation is applied to the control algorithm, the related mathematical models are mainly divided into four categories, namely, the first category is the distribution model, which is often statistically analyzed through the distribution curve to ensure the quality of the clean coal products, such as different distribution curve models such as empirical function distribution, normal distribution, hyperbolic tangent distribution, etc.; the second category is the process parameter model, some key process parameters such as sorting density and possible deviation are often closely related to the operating conditions, the properties of raw coal and the medium, etc., and need to be modeled and regressed through particle kinematics, fluid dynamics, multiphase flow theory, etc.; the third category is the computational simulation model, such as computational fluid dynamics, discrete element analysis, finite element analysis and other methods, which capture the dynamic parameters that are difficult to obtain through experimental measurement through simulation; the fourth category is the state space model, which is mathematically expressed as a differential equation that evolves over time;

[0005] The first type of distribution model and the second type of process parameter model are often static algebraic equations, which do not conform to the equation form required for intelligent control. Although the third type is a differential equation, it is too complex to solve and cannot achieve online prediction and feedback control. The fourth type of state space model is generally a comprehensive model that combines the characteristics of the first three types of models, which can reduce the problems of the first to third types. However, the fourth type of state space model is currently difficult to establish and lacks an accurate and effective state space model, which cannot meet the precise prediction and intelligent control of the dry coal preparation process. Summary of the invention

[0006] The purpose of the present invention is to provide a dry heavy medium coal preparation intelligent process prediction and control method, integrate different dry heavy medium separation mathematical models, enhance the model's ability to describe the steady-state and dynamic characteristics of the separation process, improve the prediction accuracy of clean coal ash content and yield, achieve stable control of the separation process, and solve the problem of lack of theoretical models in the field of dry heavy medium and difficulty in implementing dynamic control.

[0007] To achieve the above purpose, the present invention provides a dry heavy medium coal preparation intelligent process prediction and control method, which specifically includes the following steps:

[0008] S1: Collect the raw coal data for washing, and calculate the prior selectivity information {ρ i ,λ raw,i ,γ raw,i} i＝1:n ;

[0009] Where n is the number of density level intervals of raw coal, ρ i is the average value of the ith density level interval, γ raw,i is the yield of the ith density level interval, λ raw,i is the ash content of the ith density level interval;

[0010] S2, establish a bed density dynamic model based on evolution time The bed density dynamic model is controlled in real time through the model control algorithm to form a stable bed density ρ bed ′;

[0011] S3, considering the influence of the interaction between gas and particles in the sorting system on the sorting density, and combining the sorting test to correct the empirical parameters, the sorting density model ρ is calculated 50 ;

[0012] S4, establish a sorting density model ρ 50 , possible deviation of the sorting system E p , boundary control parameters α and β based on the empirical allocation function model, calculate the product allocation rate Y corresponding to each density level interval i ;

[0013] S5, using the prior optional information {ρ i ,λ raw,i ,γ raw,i} i＝1:n , and the product distribution rate Y corresponding to each density level interval in step S4 i , predict and calculate the clean coal yield and ash content, gangue yield and ash content of the coal preparation system;

[0014] S6, correcting the clean coal yield and ash content, and the gangue yield and ash content in step S5 to obtain corrected clean coal yield and ash content, and gangue yield and ash content.

[0015] In some examples of the present invention, in step S2, the bed density dynamic model based on evolution time The calculation formula is:

[0016]

[0017] Among them, γ is the coal powder content of the bed, and t represents the evolution time;

[0018] is the evolution equation of coal powder content over time, and the calculation formula is:

[0019]

[0020] Among them, W q is the mass flux of qualified medium, k q is the mass proportion of pulverized coal in the qualified medium, W h is the mass flux of the high-density medium, k h is the mass proportion of pulverized coal in high-density medium, W m is the mass flux discharged from the mixed medium, W gen is the mass flux of the secondary medium in the sorting system, H bed is the bed height, A is the effective surface area of ​​the separator;

[0021] ρ bed is the empirical model of bed density with coal powder content as the single independent variable, and its calculation formula is:

[0022]

[0023] is the evolution equation of bed density with respect to coal powder content, and the calculation formula is:

[0024]

[0025] Among them, b 1 and b 2 is an empirical parameter.

[0026] In some examples of the present invention, in step S3, the sorting density model p 50 The calculation formula is:

[0027] ρ 50 =K 1 ρ bed ′+K 2 ρ drag

[0028]

[0029] Among them, K 1 and K 2 Correct the empirical parameters for the sorting test, u g is the superficial gas velocity, μ g is the air viscosity, d med is the average particle size of the mixed medium in the system, ρ g is the air density and g is the acceleration due to gravity.

[0030] In some examples of the present invention, in step S4, the product allocation rate Y corresponding to each density level interval is i The calculation formula is:

[0031]

[0032] Among them, x i =ρ i / ρ 50 is the normalized density.

[0033] In some examples of the present invention, the calculation formula of the gangue yield is:

[0034]

[0035] Among them, γ d,i is the gangue yield in each density level interval;

[0036] Gangue lime fraction d The calculation formula is:

[0037]

[0038] Clean coal yieldγ c The calculation formula is:

[0039] γ c =1-γ d

[0040] Clean coal ash content c The calculation formula is:

[0041]

[0042] Among them, the raw coal ash content raw The calculation formula is:

[0043]

[0044] In some examples of the present invention, in step S6, the corrected clean coal yield The calculation formula is:

[0045]

[0046] Among them, θ is the mass efficiency parameter;

[0047] Corrected gangue yield The calculation formula is:

[0048]

[0049] Corrected gangue lime content The calculation formula is:

[0050]

[0051] Corrected clean coal ash The calculation formula is:

[0052]

[0053] In some examples of the present invention, in step S1, the prior selectivity information {ρ i ,λ raw,i ,γ raw,i} i＝1:n The process includes two situations:

[0054] In the first case, when the detection means are relatively insufficient and the source of raw coal is relatively stable, the selected raw coal can be sampled and subjected to natural screening and floating and sinking tests to obtain a priori selectivity information of the raw coal;

[0055] In the second case, when advanced detection methods are used, online coal quality analysis instruments are used to conduct real-time statistics on the properties of the selected raw coal, and update the prior selectivity information of the raw coal at certain intervals to obtain the prior selectivity information of the raw coal.

[0056] Compared with the existing technology, this intelligent process predictive control method for dry heavy medium coal preparation establishes a dynamic model of bed density based on evolution time, avoids the inability to simulate the real coal preparation system by using a static model, and solves the problem of difficulty in implementing dynamic control due to the lack of theoretical models in the field of dry heavy medium. An empirical distribution function model based on four parameters, namely, a sorting density model, possible deviations of the sorting system, and boundary control parameters, is established to construct a process predictive intelligent control framework to fully capture the steady-state and dynamic characteristics of the dry heavy medium coal preparation system. The clean coal yield and ash content, and the gangue yield and ash content are corrected through quality efficiency parameters to improve the prediction accuracy of the yield and ash content of clean coal and gangue. This predictive control method integrates different dry heavy medium sorting mathematical models to enhance the model's ability to describe the steady-state and dynamic characteristics of the sorting process, improve the prediction accuracy of the clean coal ash content and yield, and achieve stable control of the sorting process through a model predictive control algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of the overall steps of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the technical solution of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described in conjunction with the drawings of specific embodiments of the present invention. The same figure marks in the drawings represent the same parts. It should be noted that the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Unless otherwise defined, the technical terms or scientific terms used herein shall have the usual meanings understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar words used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not necessarily indicate a quantity limitation. Words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0060] like Figure 1As shown, the present invention is a dry heavy medium coal preparation intelligent process prediction and control method, which specifically includes the following steps:

[0061] S1: Collect the raw coal data for washing, and calculate the prior selectivity information {ρ i ,λ raw,i ,γ raw,i} i＝1:n ;

[0062] Where n is the number of density level intervals of raw coal, ρ i is the average value of the ith density level interval, γ raw,i is the yield of the ith density level interval, λ raw,i is the ash content of the ith density level interval;

[0063] S2, establish a bed density dynamic model based on evolution time The bed density dynamic model is controlled in real time through the model control algorithm to form a stable bed density ρ bed ′;

[0064] S3, considering the influence of the interaction between gas and particles in the sorting system on the sorting density, and combining the sorting test to correct the empirical parameters, the sorting density model ρ is calculated 50 ;

[0065] S4, establish a sorting density model ρ 50 , possible deviation of the sorting system E p , boundary control parameters α and β based on the empirical allocation function model, calculate the product allocation rate Y corresponding to each density level interval i ;

[0066] S5, using the prior optional information {ρ i ,λ raw,i ,γ raw,i} i＝1:n , and the product distribution rate Y corresponding to each density level interval in step S4 i , predict and calculate the clean coal yield and ash content, gangue yield and ash content of the coal preparation system;

[0067] S6, correcting the clean coal yield and ash content, and the gangue yield and ash content in step S5 to obtain corrected clean coal yield and ash content, and gangue yield and ash content;

[0068] Specifically, in step S1, by collecting the raw coal data for washing, the prior selectivity information {ρ i ,λ raw,i ,γ raw,i} i＝1:n The process mainly includes two situations:

[0069] In the first case, when the detection means are relatively insufficient and the source of raw coal is relatively stable, the selected raw coal can be sampled and subjected to natural screening and floating and sinking tests to obtain a priori selectivity information of the raw coal;

[0070] In the second case, when the detection means are advanced, online coal quality analysis instruments are used to conduct real-time statistics on the properties of the raw coal, and the prior selectivity information of the raw coal is updated at certain intervals to obtain the prior selectivity information of the raw coal;

[0071] In step S2, the bed density dynamic model based on evolution time It can be decomposed into two processes, namely the evolution of coal powder content over time and the evolution of bed density with respect to coal powder content;

[0072] The model predictive control algorithm MPC is used to control the bed density dynamic model in real time. This model predictive control algorithm uses the mathematical model of the system to predict the future state and optimize the control based on these predictions. That is, by calculating the bed density dynamic model ρ with evolution time bed Predict the behavior in the future, find a series of control inputs in the future prediction period, and re-enter the next control cycle to convert the bed density dynamic model ρ bed Stable control to the target value, that is, stable bed density ρ bed ′; Model predictive control algorithm MPC is a classic control algorithm, which is widely used in the field of industrial process control. It can achieve efficient and accurate control of complex dynamic systems. It will not be further elaborated here.

[0073] In step S3, based on the two-phase flow theoretical model, the effect of the interaction between gas and particles in the sorting system on the sorting density is considered. The interaction can be bubbling behavior, particle agglomeration, and medium backmixing. The empirical parameters are corrected in combination with the sorting test, and finally the sorting density model ρ is calculated. 50 ;

[0074] In the steps of this predictive control method, a dynamic model of bed density based on evolution time is established to avoid the inability to simulate the real coal preparation system by using a static model, and to solve the problem that the lack of theoretical models in the field of dry heavy medium makes it difficult to implement dynamic control; an empirical distribution function model based on four parameters, namely, a sorting density model, possible deviations of the sorting system, and boundary control parameters, is established to construct a process predictive intelligent control framework to fully capture the steady-state and dynamic characteristics of the dry heavy medium coal preparation system; and the clean coal yield and ash content, and the gangue yield and ash content are corrected through quality efficiency parameters to improve the prediction accuracy of the yield and ash content of clean coal and gangue; this predictive control method integrates different dry heavy medium sorting mathematical models to enhance the model's ability to describe the steady-state and dynamic characteristics of the sorting process, improve the prediction accuracy of the clean coal ash content and yield, and achieve stable control of the sorting process through the model predictive control algorithm, accelerate the intelligent construction of coal preparation plants, and promote high-quality coal development.

[0075] In some examples of the present invention, in step S2, the bed density dynamic model based on evolution time ρ bed The calculation formula is:

[0076]

[0077] Among them, γ is the coal powder content of the bed, and t represents the evolution time;

[0078] Furthermore, is the evolution equation of coal powder content over time, and the calculation formula is:

[0079]

[0080] Among them, W q is the mass flux of qualified medium, k q is the mass proportion of pulverized coal in the qualified medium, W h is the mass flux of the high-density medium, k h is the mass proportion of pulverized coal in high-density medium, W m is the mass flux discharged from the mixed medium, W gen is the mass flux of the secondary medium in the sorting system, H bed is the bed height, A is the effective surface area of ​​the separator;

[0081] Based on the law of conservation of mass, an empirical model of bed density with coal powder content as a single independent variable is introduced, and its calculation formula is:

[0082]

[0083] Among them, b 1 and b 2 is an empirical parameter;

[0084] Based on the above empirical model of bed density with coal powder content as a single independent variable, the evolution equation of bed density with respect to coal powder content is obtained: The calculation formula is:

[0085]

[0086] In some examples of the present invention, in step S3, the sorting density model p 50 The calculation formula is:

[0087] ρ 50 =K 1 ρ bed ′+K 2 ρ drag

[0088]

[0089] Among them, K 1 and K 2 Correct the empirical parameters for the sorting test, u g is the superficial gas velocity, d med is the average particle size of the mixed medium in the system, μ g is the air viscosity, ρ g is the air density, g is the acceleration due to gravity;

[0090] In some examples of the present invention, in step S4, the product allocation rate Y corresponding to each density level interval is i The calculation formula is:

[0091]

[0092] Among them, x i =ρ i / ρ 50 is the normalized density, so the product distribution rate Y corresponding to each density level interval 1-n is obtained i ;

[0093] Explain the possible deviation of the sorting system E p The value is generally selected as 0.03-0.07, and the boundary control parameters α and β are generally selected as 0.5-1;

[0094] In some examples of the present invention, based on the ash consistency assumption, the raw coal ash content λ before sorting in each density level interval is raw,i and ash content of clean coal after sorting c,i , Gangue lime separation d,i Equal, that is, λ raw,i =λ c,i =λ d,i ;

[0095] The calculation formula of the gangue yield is:

[0096]

[0097] Among them, γ d,i is the gangue yield in each density level interval;

[0098] Gangue lime fraction d The calculation formula is:

[0099]

[0100] Clean coal yieldγ c The calculation formula is:

[0101] γ c =1-γ d

[0102] Clean coal ash content c The calculation formula is:

[0103]

[0104] In some examples of the present invention, in step S6, the corrected clean coal yield The calculation formula is:

[0105]

[0106] Among them, θ is the mass efficiency parameter;

[0107] Corrected gangue yield The calculation formula is:

[0108]

[0109] Corrected gangue lime content The calculation formula is:

[0110]

[0111] Corrected clean coal ash The calculation formula is:

[0112]

[0113] Where the a priori optional information {ρ i ,λ raw,i ,γ raw,i} i＝1:n , raw coal ash content λ raw The calculation formula is:

[0114]

[0115] Example

[0116] The present invention provides a dry heavy medium coal preparation intelligent process prediction and control method, which specifically comprises the following steps:

[0117] S1, collect the raw coal data of the coal preparation plant, divide the raw coal into n = 8 density level intervals, and the average value of the corresponding density level interval is ρ i , the yield is γ raw,i , ash content is λ raw,i As shown in Table 1; the group of prior optional information {ρ i ,λ raw,i ,γ raw,i} i＝1:n The input information known to the overall method;

[0118] Table 1 shows the statistical prior optional information of the collected data

[0119]

[0120] S2, based on the evolution equation of coal powder content over time and the evolution equation of bed density with respect to coal powder content, a bed density dynamic model ρ is established bed The calculation formula is used to set the parameters in the bed density dynamic model as shown in Table 2;

[0121] Table 2: Parameters in the bed density dynamic model

[0122]

[0123]

[0124] S3, set the sorting test correction empirical parameter K 1 and K 2 , apparent gas velocity u g , the average particle size d of the mixed medium in the system med , air viscosity μ g , air density ρ g , the value of gravitational acceleration g, as shown in Figure 3;

[0125] Table 3: Parameter settings in sorting density

[0126]

[0127] Calculate the sorting density ρ under this parameter 50 =1.4958;

[0128] S4, possible deviations in setting the sorting system E p =0.055, boundary control parameters α=0.9550 and β=0.9555, and combined with the sorting density ρ calculated in step S350 =1.4958, these four parameters {ρ 50 ,E p ,α,β} into the calculation of the product allocation rate Y corresponding to each density level interval i ;

[0129] S5, the data collected in Table 1 are statistically analyzed for the prior optional information {ρ i ,λ raw,i ,γ raw,i} i＝1:n , and combined with the product distribution rate Y calculated in step S4 i , predict and calculate the clean coal yield and ash content, gangue yield and ash content of the coal preparation system;

[0130] That is, by calculating the clean coal yield and ash content, gangue yield and ash content of the coal preparation system, they are predicted. It is explained that the calculated value is the predicted value of this method. The true value and the predicted value are shown in Table 4. The calculation expression of the error is:

[0131]

[0132] Table 4: Actual and predicted values ​​of products

[0133]

[0134] S6, establish a correction algorithm, that is, set the quality efficiency parameter θ = 0.98, and correct the prediction to The true value and the corrected value are shown in Table 5. The error calculation expression is:

[0135]

[0136] The key indicator λ can be obtained c The error is reduced by one order of magnitude, and other indicators are at the same order of magnitude:

[0137] Table 5: Corrected product actual value and predicted value

[0138]

[0139] In summary, it can be seen from Table 4 that the prediction accuracy of the four indicators of clean coal yield and ash content, gangue yield and ash content of the coal preparation system is relatively high, and the error of the three indicators of clean coal yield, gangue yield and ash content is within 10 -2 -10 -3 The prediction accuracy of clean coal ash content is slightly lower than that of the other three indicators, with an error of 10 -1Order of magnitude; In the coal preparation process, the clean coal ash content index is the most important index, because clean coal is the real product needed, while the index for gangue can be relatively relaxed. Therefore, the prediction accuracy of clean coal ash content should be improved as much as possible;

[0140] As shown in Table 5, the prediction accuracy of clean coal ash content ranges from 10 -1 The order of magnitude was increased to 10 -2 At the same time, the errors of the other three indicators remain at 10 -2 The order of magnitude is not sacrificed,without sacrificing the accuracy of other indicators;

[0141] The advantage of the present invention is that it can greatly improve the prediction accuracy of clean coal ash content. The high-precision prediction of clean coal ash content can significantly optimize production processes and decisions, and more accurately understand the output and quality of clean coal, thereby reasonably arranging production plans, avoiding overproduction or underproduction, and improving production efficiency.

[0142] An exemplary implementation method of a dry heavy medium coal preparation intelligent process predictive control method proposed in the present invention is described in detail above with reference to a preferred embodiment. However, those skilled in the art will appreciate that, without departing from the concept of the present invention, various modifications and variations may be made to the above-mentioned specific embodiments, and various technical features and structures proposed in the present invention may be combined in various ways without exceeding the protection scope of the present invention, which is determined by the appended claims.

Claims

1. A method for predicting and controlling an intelligent process of dry heavy medium coal preparation, comprising the following steps: S1: Collect the raw coal data for washing, and calculate the prior selectivity information {ρ i ,λ raw,i ,γ raw,i } i＝1:n ; Where n is the number of density level intervals of raw coal, ρ i is the average value of the ith density level interval, γ raw,i is the yield of the ith density level interval, λ raw,i is the ash content of the ith density level interval; S2, establish a bed density dynamic model based on evolution time The bed density dynamic model is controlled in real time through the model control algorithm to form a stable bed density ρ bed ′; S3, considering the influence of the interaction between gas and particles in the sorting system on the sorting density, and combining the sorting test to correct the empirical parameters, the sorting density model ρ is calculated 50 ; S4, establish a sorting density model ρ 50 , possible deviation of the sorting system E p , boundary control parameters α and β based on the empirical allocation function model, calculate the product allocation rate Y corresponding to each density level interval i ; S5, using the prior optional information {ρ i ,λ raw,i ,γ raw,i } i＝1:n , and the product distribution rate Y corresponding to each density level interval in step S4 i , predict and calculate the clean coal yield and ash content, gangue yield and ash content of the coal preparation system; S6, correcting the clean coal yield and ash content, and the gangue yield and ash content in step S5 to obtain corrected clean coal yield and ash content, and gangue yield and ash content; In step S2, the bed density dynamic model based on evolution time The calculation formula is: Among them, γ is the coal powder content of the bed, and t represents the evolution time; is the evolution equation of coal powder content over time, and the calculation formula is: Among them, W q is the mass flux of qualified medium, k q is the mass proportion of coal powder in the qualified medium, W h is the mass flux of the high-density medium, k h is the mass proportion of pulverized coal in high-density medium, W m is the mass flux discharged from the mixed medium, W gen is the mass flux of the secondary medium in the sorting system, H bed is the bed height, A is the effective surface area of ​​the separator; ρ bed is the empirical model of bed density with coal powder content as the single independent variable, and its calculation formula is: is the evolution equation of bed density with respect to coal powder content, and the calculation formula is: Among them, b1 and b2 are empirical parameters.

2. The method for predicting and controlling the intelligent process of dry heavy medium coal preparation according to claim 1, characterized in that: In step S3, the sorting density model ρ 50 The calculation formula is: r 50 =K1ρ bed ′+K2ρ drag Among them, K1 and K2 are the empirical parameters for sorting test correction, u g is the superficial gas velocity, μ g is the air viscosity, d med is the average particle size of the mixed medium in the system, ρ g is the air density and g is the acceleration due to gravity.

3. The method for predicting and controlling the intelligent process of dry heavy medium coal preparation according to claim 2, characterized in that: In step S4, the product allocation rate Y corresponding to each density level interval i The calculation formula is: Among them, x i =ρ i / ρ 50 is the normalized density.

4. The method for predicting and controlling the intelligent process of dry heavy medium coal preparation according to claim 3 is characterized in that: The calculation formula of the gangue yield is: Among them, γ d,i is the gangue yield in each density level interval; Gangue lime fraction d The calculation formula is: Clean coal yieldγ c The calculation formula is: c c =1-c d Clean coal ash content c The calculation formula is: Among them, the raw coal ash content raw The calculation formula is:

5. The method for predicting and controlling the intelligent process of dry heavy medium coal preparation according to claim 4, characterized in that: In step S6, the corrected clean coal yield The calculation formula is: Among them, θ is the mass efficiency parameter; Corrected gangue yield The calculation formula is: Corrected gangue lime content The calculation formula is: Corrected clean coal ash The calculation formula is:

6. A method for predicting and controlling an intelligent process of dry heavy medium coal preparation according to any one of claims 1 to 5, characterized in that: In step S1, the prior selectivity information {ρ i ,λ raw,i ,γ raw,i } i＝1:n The process includes two situations: In the first case, when the detection means are relatively insufficient and the source of raw coal is relatively stable, the selected raw coal can be sampled and subjected to natural screening and floating and sinking tests to obtain a priori selectivity information of the raw coal; In the second case, when advanced detection methods are used, online coal quality analysis instruments are used to conduct real-time statistics on the properties of the selected raw coal, and update the prior selectivity information of the raw coal at certain intervals to obtain the prior selectivity information of the raw coal.

Images