A heavy medium coal separation full-process intelligent control method

Abstract

The application discloses a heavy medium coal separation full-process intelligent control method, which comprises the following steps: data acquisition and preprocessing of a coal separation plant; a heavy medium coal separation model is constructed to describe the dynamic characteristics of a controlled variable in a clean coal ash content control process; control loops are established respectively for heavy medium suspension density and clean coal ash content; an edge controller is developed to obtain corresponding intelligent control software; and field debugging is performed to verify the effect. The method designed in the application is based on coal separation professional knowledge and artificial intelligence means, is suitable for coal quality change and working condition fluctuation in an industrial field, and realizes the goal of heavy medium coal separation full-process intelligent control, guides actual production control, and effectively improves the intelligent level of the coal separation plant.

Description

Technical Field

[0001] This invention relates to the field of heavy medium coal preparation technology, specifically to an intelligent control method for the entire heavy medium coal preparation process. Background Technology

[0002] In recent years, the coal preparation industry has entered a period of rapid development towards intelligent manufacturing. Heavy media coal preparation, as the most widely used separation method, boasts a wide feed range and high separation accuracy, making it the most crucial link in determining the coal separation effect. However, due to my country's poor coal resource endowment, variable coal quality, and large feed volume, there is a lack of control equipment that can guide on-site production for the complex industrial separation processes in coal preparation plants, which involve multiple levels, multiple time scales, and unknown operational models. Relying solely on traditional control methods makes it difficult to achieve intelligent control and decision-making in the heavy media separation process. Furthermore, existing process control theories and models cannot form a closed loop with actual product indicators, severely restricting the requirements for intelligent construction of coal preparation plants.

[0003] Therefore, there is an urgent need to rely on coal preparation expertise and artificial intelligence to study advanced ash content optimization control technologies that take into account practical factors such as changes in raw coal ash content and the load capacity of production equipment, so as to guide coal preparation plants to carry out production under optimized control strategies that conform to actual conditions. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control method for the entire process of heavy media coal preparation. Relying on coal preparation expertise and artificial intelligence, this method adapts to changes in coal quality and fluctuations in operating conditions at the industrial site, achieves the goal of intelligent control of the entire heavy media coal preparation process, guides actual production control, and effectively improves the level of intelligence in coal preparation plants.

[0005] To achieve the above functions, this invention designs an intelligent control method for the entire heavy media coal preparation process, executing the following steps S1-S5 to realize dynamic control of heavy media coal preparation in the target coal preparation plant:

[0006] Step S1: Collect production data from the target coal preparation plant, and preprocess the collected data to remove noise and invalid data;

[0007] Step S2: Based on the controlled autoregressive integral moving average model, construct a heavy media coal preparation model to describe the dynamic characteristics of the controlled variable in the process of controlling the ash content of clean coal, as well as the effects of the control variables and disturbance variables on the controlled variable.

[0008] Step S3: Based on the heavy medium coal preparation model, establish a control loop with the density of the heavy medium suspension and the ash content of the clean coal as the controlled variables, and construct the objective function and the control rate that makes the controlled variables reach the optimal level.

[0009] In one control method, the density of the heavy medium suspension is used as the controlled variable, the frequency of the water supply valve, the diversion valve, and the medium supply pump is used as the control variable, and the liquid level of the heavy medium suspension tank is used as the disturbance variable. The generalized predictive control algorithm is used to control the density of the heavy medium suspension. In another control method, the ash content of the clean coal is used as the controlled variable, the ash content of the raw coal, the coal flow rate, and the magnetic content are used as the disturbance variables, and the set value of the density of the heavy medium suspension is used as the control variable. The generalized predictive control algorithm is used to control the ash content of the clean coal.

[0010] Step S4: Using a control protocol and PLC control system, develop an edge controller and develop heavy media coal preparation control software based on GPC algorithm to realize the data acquisition, processing and dynamic control process of heavy media suspension density and clean coal ash content in steps S1-S3.

[0011] Step S5: Based on the environment of the target coal preparation plant, implement the edge controller architecture, conduct on-site debugging and verification of the heavy medium coal preparation control software developed in Step S4, and run the debugged edge controller and heavy medium coal preparation control software to achieve dynamic control of heavy medium coal preparation in the target coal preparation plant.

[0012] Beneficial effects: Compared with the prior art, the advantages of the present invention include:

[0013] This invention designs an intelligent control method for the entire process of heavy media coal preparation, which solves the problem that existing process control theories and models cannot form a closed loop with actual product indicators, severely restricting the intelligent construction of coal preparation plants. This method relies on coal preparation expertise and artificial intelligence, adapts to changes in coal quality and fluctuations in operating conditions at the industrial site, achieves the goal of intelligent control of the entire process of heavy media coal preparation, guides actual production control, and effectively improves the level of intelligence in coal preparation plants. Attached Figure Description

[0014] Figure 1 This is a flowchart of an intelligent control method for the entire process of heavy media coal preparation according to an embodiment of the present invention;

[0015] Figure 2 This is an execution flowchart of the heavy media coal preparation control software provided according to an embodiment of the present invention;

[0016] Figure 3 This is an edge controller architecture diagram provided according to an embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0018] This invention provides an intelligent control method for the entire process of heavy media coal preparation, referring to... Figure 1Perform the following steps S1-S5 to achieve dynamic control of heavy media coal preparation in the target coal preparation plant:

[0019] Step S1: Collect production data from the target coal preparation plant, and preprocess the collected data to remove noise and invalid data;

[0020] The preprocessing of the collected data includes: for noise (random error), digital filtering is used, including high-pass filtering, low-pass filtering, or data smoothing; for invalid data (significant error), statistical hypothesis testing is used, including residual analysis, correction analysis, generalized likelihood ratio, Bayesian method, incremental method, or principal component analysis.

[0021] Step S2: Based on the controlled autoregressive integral moving average model, construct a heavy media coal preparation model to describe the dynamic characteristics of the controlled variable in the clean coal ash content control process, as well as the effects of the control variables and disturbance variables on the controlled variable; thereby realizing the understanding, control, optimization and diagnosis of industrial processes.

[0022] For nonparametric models, system identification methods are used, including but not limited to step response method, frequency method, correlation analysis method and spectral analysis method; for parametric models, identification methods need to first assume a model structure and determine the model parameters by minimizing the error criterion function between the model and the process, including but not limited to least squares method, gradient method, Gauss-Newton method and simplex method.

[0023] Because operating conditions frequently change at coal preparation plants, precise model parameters are unavailable. Therefore, generalized predictive control uses various performance index optimizations as control conditions to achieve optimized regulation. A controlled autoregressive integral moving average (CARIMA) model is used to describe the controlled object subjected to random disturbances, such as various industrial processes (e.g., steady-state, integral, and unstable processes). The effects of disturbances and noise are also considered, and feedback correction is optimized. In step S2, a heavy media coal preparation model is constructed based on the CARIMA model, and its expression is as follows:

[0024] ;

[0025] In the formula, , Represents the shift operator A polynomial of the form has the following form:

[0026] ;

[0027] ;

[0028] In the formula, The output vector for the object at time k. Let k be the control input vector at time k. Given a zero-mean noise sequence at time k that is uncorrelated, consider the case of white noise. , It is a difference operator.

[0029] Step S3: Based on the heavy medium coal preparation model, establish a control loop with the density of the heavy medium suspension and the ash content of the clean coal as the controlled variables, and construct the objective function and the control rate that makes the controlled variables reach the optimal level.

[0030] In one control method, the density of the heavy medium suspension is used as the controlled variable (CV), the frequency of the water supply valve, the diversion valve, and the medium supply pump is used as the control variable (MV), and the liquid level of the heavy medium suspension tank is used as the disturbance variable (DV). A generalized predictive control algorithm is used to control the density of the heavy medium suspension. In another control method, the ash content of the clean coal is used as the controlled variable (CV), the ash content of the raw coal, the coal flow rate, and the magnetic content are used as disturbance variables, and the setpoint of the density of the heavy medium suspension is used as the control variable (MV). A generalized predictive control algorithm is used to control the ash content of the clean coal.

[0031] The control principle of the density control loop for heavy medium suspension is as follows: density is detected using a densitometer. The density of the suspension in the heavy medium suspension tank is adjusted by connecting the clean water valve on the tank to the underflow box of the clean coal screen, making its value slightly higher than the theoretical separation density. Then, clean water is added through the water inlet valve on the pump root of the heavy medium suspension tank, causing the density to decrease and reach the separation density required by the hydrocyclone. When the density of the heavy medium suspension is low, the flow rate is increased, or the water inlet is reduced, or the medium inlet pump is turned on; when the density is high, the flow rate is decreased, or the water inlet is increased. Furthermore, fluctuations in the liquid level in the heavy medium suspension tank also significantly affect the density control effect and need to be considered as a disturbance variable.

[0032] In step S3, the object output vector of the heavy medium coal preparation model at time k-1 in the control loop for the density of the heavy medium suspension. For the density of the heavy medium suspension, control the input vector. The vector combination is as follows:

[0033] ;

[0034] In the formula, Let k-1 be the opening degree of the water supply valve. Let k-1 be the opening degree of the diverter valve. The frequency of the medium pump is applied at time k-1. Let be the liquid level of the heavy medium suspension tank at time k-1; where the opening degree of the water supply valve, the opening degree of the diversion valve, and the frequency of the medium supply pump are control variables, and the liquid level of the heavy medium suspension tank is a disturbance variable.

[0035] Polynomials in heavy media coal preparation models , Organize it into the following format:

[0036] ;

[0037] ;

[0038] If each controlled object is described using a first-order inertial time-delay system, then its transfer function has the following form:

[0039] ;

[0040] Where K is the proportional coefficient of the system, T is the time constant, τ is the time delay, and Ts is the sampling time of the system. Discretizing the above first-order inertial time-delay system yields the following discrete equation:

[0041] ;

[0042] In the formula, This is the output at time k;

[0043] therefore and It has the following form:

[0044] ;

[0045] ;

[0046] ;

[0047] in , , These represent the proportionality coefficient, time constant, and time delay of the effect of the water supply valve opening on the suspension density control. , , These represent the proportionality coefficient, time constant, and time delay, respectively, of the effect of the diverter valve opening on the suspension density control. , , These represent the proportionality coefficient, time constant, and time delay, respectively, of the effect of the pump frequency on the density control of the suspension. , , These represent the proportionality coefficient, time constant, and time delay of the effect of the liquid level in the suspension tank on the density control of the suspension.

[0048] In the control loop for clean coal ash content, the object output vector at time k-1 of the heavy medium coal preparation model For clean coal ash content, control input vector The vector combination is as follows:

[0049] ;

[0050] In the formula, This is the setpoint for the density of the heavy medium suspension at time k-1. Let k-1 be the ash content of raw coal. Let K be the coal flow rate at time k-1. The magnetic content at time k-1; where the density of the heavy medium suspension is the setpoint as the control variable, and the ash content of raw coal, coal flow rate, and magnetic content are the disturbance variables;

[0051] Therefore, the polynomial in the heavy medium coal preparation model and It has the following form:

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] ;

[0057] in , , These represent the proportionality coefficient, time constant, and time delay of the effect of the water supply valve opening on the suspension density control. , , These are the proportional coefficients representing the effect of raw coal ash content on the density control of the suspension. , , These are the time constants for the effect of coal flow rate on the density control of the suspension. , , These represent the time lag of the effect of magnetic material content on the density control of the suspension.

[0058] Step S4: Using a control protocol and PLC control system, develop an edge controller, and develop heavy media coal preparation control software based on GPC algorithm and computer language programming to realize the data acquisition, processing and dynamic control process of heavy media suspension density and clean coal ash content in steps S1-S3.

[0059] Reference Figure 2 The execution steps of the heavy media coal preparation control software are as follows:

[0060] Step S4.1: Read the production data of the target coal preparation plant from the PLC control system through a dedicated control protocol;

[0061] Step S4.2: Preprocess the collected data;

[0062] Step S4.3: Input the preprocessed data into the heavy medium coal preparation model, and obtain the control rate that makes the control variable reach the predetermined value based on the GPC algorithm;

[0063] Step S4.4: The obtained control frequency is transmitted to the PLC system through a dedicated control protocol, thereby controlling the movement of the actuator.

[0064] In step S4, a control loop is established based on the GPC algorithm, and an objective function is constructed. The optimal control rate is determined by finding the control rate that makes the control variable reach a predetermined value. The objective function is as follows:

[0065] ;

[0066] In the formula, J represents the objective function. This represents the output at time k+j. The output tracking sequence at time k+j is expressed as follows:

[0067] ;

[0068] In the formula, It outputs the set value. For the change in input, It is a softening factor, where N is the prediction time domain and M is the control time domain. For error-weighted sequences, It is a control weighted sequence;

[0069] Based on the known input and output at the current moment and the input values ​​at future moments, the predicted output of the object at future moments is as follows:

[0070] ;

[0071] The optimal control law is obtained as follows:

[0072] ;

[0073] in:

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] Using the Diophantine equation, the intermediate variable vector is defined as follows:

[0079] ;

[0080] ;

[0081] in:

[0082] ;

[0083] ;

[0084] in, , They are respectively , The coefficient;

[0085] ;

[0086] ;

[0087] ;

[0088] initial value when hour, , , ;

[0089] Define the step response polynomial:

[0090] ;

[0091] ;

[0092] The elements in matrix G represent the step response coefficients of the system;

[0093] Define an intermediate variable vector:

[0094] ;

[0095] in:

[0096] ;

[0097] in, As an intermediate variable;

[0098] ;

[0099] ;

[0100] when hour, , ; For coefficients;

[0101] The optimal input at the current moment is as follows:

[0102] ;

[0103] In the formula, d is The first line.

[0104] Step S5: Refer to Figure 3 For the target coal preparation plant environment, an edge controller architecture is constructed, and the heavy medium coal preparation control software developed in step S4 is debugged and verified on-site. The debugged edge controller and heavy medium coal preparation control software are run, and finally the heavy medium coal preparation target that meets the production control requirements is obtained, realizing the dynamic control of heavy medium coal preparation in the target coal preparation plant.

[0105] An electronic device includes a storage device and one or more processors. The storage device is used to store one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the intelligent control method for the entire process of heavy media coal preparation.

[0106] A computer-readable storage medium storing a computer program therein, which, when executed by a processor, implements the aforementioned intelligent control method for the entire process of heavy medium coal preparation.

[0107] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for intelligent control of the entire heavy media coal preparation process, characterized in that, Perform the following steps S1-S5 to achieve dynamic control of heavy media coal preparation in the target coal preparation plant: Step S1: Collect production data from the target coal preparation plant, and preprocess the collected data to remove noise and invalid data; Step S2: Based on the controlled autoregressive integral moving average model, construct a heavy media coal preparation model to describe the dynamic characteristics of the controlled variable in the process of controlling the ash content of clean coal, as well as the effects of the control variables and disturbance variables on the controlled variable. In step S2, a heavy media coal preparation model is constructed based on the controlled autoregressive integral moving average model, and its expression is as follows: ； In the formula, , Represents the shift operator A polynomial of the form has the following form: ； ； In the formula, The output vector for the object at time k. Let k be the control input vector at time k. Given a zero-mean noise sequence at time k that is uncorrelated, consider the case of white noise. , It is a difference operator; Step S3: Based on the heavy medium coal preparation model, establish a control loop with the density of the heavy medium suspension and the ash content of the clean coal as the controlled variables, and construct the objective function and the control rate that makes the controlled variables reach the optimal level. In one control method, the density of the heavy medium suspension is used as the controlled variable, the frequency of the water supply valve, the diversion valve, and the medium supply pump is used as the control variable, and the liquid level of the heavy medium suspension tank is used as the disturbance variable. The generalized predictive control algorithm is used to control the density of the heavy medium suspension. In another control method, the ash content of the clean coal is used as the controlled variable, the ash content of the raw coal, the coal flow rate, and the magnetic content are used as the disturbance variables, and the set value of the density of the heavy medium suspension is used as the control variable. The generalized predictive control algorithm is used to control the ash content of the clean coal. In step S3, the object output vector of the heavy medium coal preparation model at time k-1 in the control loop for the density of the heavy medium suspension. For the density of the heavy medium suspension, control the input vector. The vector combination is as follows: ； In the formula, Let k-1 be the opening degree of the water supply valve. Let k-1 be the opening degree of the diversion valve. The frequency of the medium pump is applied at time k-1. Let be the liquid level of the heavy medium suspension tank at time k-1; where the opening degree of the water supply valve, the opening degree of the diversion valve, and the frequency of the medium supply pump are control variables, and the liquid level of the heavy medium suspension tank is a disturbance variable. Polynomials in heavy media coal preparation models , Organize it into the following format: ； ； If each controlled object is described using a first-order inertial time-delay system, then its transfer function has the following form: ； Where K is the proportional coefficient of the system, T is the time constant, τ is the time delay, and Ts is the sampling time of the system. Discretizing the above first-order inertial time-delay system yields the following discrete equation: ； In the formula, This is the output at time k; therefore and It has the following form: ； ； ； in , , These represent the proportionality coefficient, time constant, and time delay of the effect of the water supply valve opening on the suspension density control. , , These represent the proportionality coefficient, time constant, and time delay, respectively, of the effect of the diverter valve opening on the suspension density control. , , These represent the proportionality coefficient, time constant, and time delay, respectively, of the effect of the pump frequency on the density control of the suspension. , , These are the proportionality coefficient, time constant, and time delay, respectively, representing the effect of the liquid level in the suspension tank on the density control of the suspension. In step S3, the object output vector of the heavy medium coal preparation model at time k-1 in the control loop for clean coal ash content is... For clean coal ash content, control input vector The vector combination is as follows: ； In the formula, This is the setpoint for the density of the heavy medium suspension at time k-1. Let k-1 be the ash content of raw coal. Let K be the coal flow rate at time k-1. The magnetic content at time k-1; where the density of the heavy medium suspension is the setpoint as the control variable, and the ash content of raw coal, coal flow rate, and magnetic content are the disturbance variables; Therefore, the polynomial in the heavy medium coal preparation model and It has the following form: ； ； ； ； ； in , , These represent the proportionality coefficient, time constant, and time delay of the effect of the water supply valve opening on the suspension density control. , , These are the proportional coefficients representing the effect of raw coal ash content on the density control of the suspension. , , These are the time constants for the effect of coal flow rate on the density control of the suspension. , , These represent the time lag of the effect of magnetic material content on the density control of the suspension; Step S4: Using a control protocol and PLC control system, develop an edge controller and develop heavy media coal preparation control software based on GPC algorithm to realize the data acquisition, processing and dynamic control process of heavy media suspension density and clean coal ash content in steps S1-S3. Step S5: Based on the environment of the target coal preparation plant, implement the edge controller architecture, conduct on-site debugging and verification of the heavy medium coal preparation control software developed in Step S4, run the debugged edge controller and heavy medium coal preparation control software, and realize the dynamic control of heavy medium coal preparation in the target coal preparation plant.

2. The intelligent control method for the entire process of heavy media coal preparation according to claim 1, characterized in that, The preprocessing of the collected data in step S1 includes: for noise, digital filtering is used, including high-pass filtering, low-pass filtering, or data smoothing; for invalid data, statistical hypothesis testing is used, including residual analysis, correction analysis, generalized likelihood ratio, Bayesian method, incremental method, or principal component analysis.

3. The intelligent control method for the entire process of heavy media coal preparation according to claim 1, characterized in that, In step S4, a control loop is established based on the GPC algorithm, and an objective function is constructed. The optimal control rate is obtained by solving for the control rate that makes the control variable reach the predetermined value.

4. The intelligent control method for the entire process of heavy media coal preparation according to claim 1, characterized in that, The heavy media coal preparation control software executes the following steps in step S4: Step S4.1: Read the production data of the target coal preparation plant from the PLC control system through a dedicated control protocol; Step S4.2: Preprocess the collected data; Step S4.3: Input the preprocessed data into the heavy medium coal preparation model, and obtain the control rate that makes the control variable reach the predetermined value based on the GPC algorithm; Step S4.4: The obtained control frequency is transmitted to the PLC system through a dedicated control protocol, thereby controlling the movement of the actuator.

5. An electronic device, characterized in that, It includes a storage device, one or more processors, and the storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement a method for intelligent control of the entire process of heavy media coal preparation as described in any one of claims 1-4.

6. A computer-readable storage medium internally storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent control method for the entire process of heavy medium coal preparation as described in any one of claims 1-4.

Images