A flotation dosing control method for coal preparation plants based on cascade decoupling control

Abstract

The present application provides a flotation dosing control method for a coal preparation plant based on cascade decoupling control, including the following specific steps: real-time acquisition of flotation information, obtaining the dosage of frother and recovery agent, tailings and concentrate grade information during flotation, and preprocessing the actual production data to remove abnormal points and invalid data; after obtaining the input and output sample data, design a cascade decoupling controller to control the flotation dosing object of the coal preparation plant; optimize the PID coefficient and MPC controller weight parameter according to the control scheme. The present application solves the problem of instability in the manual dosing process, adapts to the characteristics of the flotation object of the coal preparation plant, and meets the production requirements of the flotation industry.

Description

Technical Field

[0001] The present application relates to the field of information technology, and in particular to a flotation dosing control method of a coal preparation plant based on cascade decoupling control. Background Art

[0002] Flotation is a method of selectively separating useful minerals based on the hydrophobic properties of minerals. This process is a complex production process affected by many factors. The current manual dosing strategy of coal preparation plants cannot meet the production conditions of the plant, so a method is needed to meet the economic indicators of flotation and improve the efficiency and stability of flotation production.

[0003] In view of the multivariable characteristics of the flotation process, the literature (Nasseri S, Khalesi MR, Ramezani A, et al. An Adaptive Decoupling Control Design for Flotation Column: A Comparative Study Against Model Predictive Control [J]. IETE Journal of Research, 2020: 1-14.) proposed an adaptive decoupling control strategy to add the structure of the decoupler to the PID tuning to reduce the calculation time and complexity and complete the adaptive parameter adjustment. However, due to the large number of variables, large time lag, and complex working conditions in the flotation production process, the controller tuning requirements are high. In addition to the traditional PID-based method, the literature (Brooks K, Westcott M, Bauer M. A Combined MPC for Milling and Flotation–A Simulation Study [J]. IFAC-Papers On Line, 2019, 52 (14): 24-29.) uses model predictive control (MPC) to compensate for the robustness and disturbance of flotation. However, using only MPC control often leads to conservative control performance and is more dependent on an accurate system dynamics model. Therefore, it is difficult to achieve good flotation control effects using a single loop control method.

[0004] Cascade control can improve the influence of large time lag and large inertia processes, and has a simple structure and is widely used in the industrial field. The literature (Chu Yunfei, Xu Wenli, Liao Qian, Wan Weihan. Decentralized control of cascade flotation tank level [J]. Mining and Metallurgy, 2004, (2): 20-23; Wang Xu, Zhao Boxi. Design and application of intelligent control system for copper flotation process [J]. Nonferrous Metallurgy Design and Research, 2019, Vol. 40 (5): 5-9.) proposed a cascade-based control strategy, which is applied to the liquid level control of multi-stage flotation tanks, and reduces the interference effect in each link through collaborative control.

[0005] The above methods are challenging to implement in industrial sites because the coupling between control loops in the multivariable flotation process makes it impossible to control different loops individually, and the complexity, computational intensity, and low robustness of the controller make the implementation of multivariable controllers challenging. Summary of the invention

[0006] The purpose of the embodiment of the present application is to provide a flotation dosing control method for a coal preparation plant based on cascade decoupling control, which solves the problem of instability in the manual dosing process, adapts to the characteristics of the flotation objects in the coal preparation plant, and meets the production requirements of the flotation industry.

[0007] The present invention is achieved through the following technical solutions:

[0008] The embodiment of the present application provides a flotation dosing control method for a coal preparation plant based on cascade decoupling control, which is characterized by comprising the following specific steps:

[0009] Collect flotation information in real time, obtain the dosage of frother and recovery agent, tailings and concentrate grade information during flotation, and pre-process the actual production data to eliminate abnormal points and invalid data;

[0010] After obtaining the input and output sample data, the cascade decoupling controller is designed to control the flotation dosing object in the coal preparation plant;

[0011] Optimize PID coefficients and MPC controller weight parameters.

[0012] The design of the cascade decoupling controller includes using PID decoupling control in the inner loop to decouple the flotation loop and weaken the interference between variables; using MPC in the outer loop to optimize the generalized object of the inner loop and soften the set value of the production index to stabilize the control effect. Finally, the output of the MPC controller is used as the input of the densely sampled PID decoupling loop to evaluate and predict the error of the flotation process output in advance, thereby effectively improving the disturbance suppression performance and control stability of the flotation loop.

[0013] The optimization of PID coefficients and MPC controller weight parameters according to the control scheme is specifically to use a particle swarm algorithm to optimize the controller parameters, select the integral absolute error of the performance index as the optimization cost function, and provide a guarantee for the application of the cascade control strategy in the flotation dosing system.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] By studying the flotation dosing process, a decoupling controller was designed, which combines the predictive output characteristics of the MPC control strategy and the flexible control capabilities of the PID control strategy. The advantages of this method are that it solves the problem of instability in the manual dosing process, adapts to the characteristics of the flotation object in the coal preparation plant, and meets the production requirements of the flotation industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a process flow chart of flotation process;

[0018] Figure 2 It is a block diagram of decoupling structure design;

[0019] Figure 3 It is a block diagram of the cascade decoupling structure design of the present invention;

[0020] Figure 4 : It is the response comparison curve of cascade decoupling control and PID decoupling control, where (a) is the tailings grade output response, and (b) is the concentrate grade output response;

[0021] Figure 5 is the MPC output trajectory of clean coal grade. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the 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.

[0023] Due to the multivariable characteristics of the flotation process and the coupling between control loops, it is not feasible to control different loops separately. A dosing strategy based entirely on manual operation will not be able to meet the production conditions of the factory, the economic indicators of flotation, flotation production efficiency and stability. Therefore, it is necessary to use advanced control methods to study the flotation intelligent dosing control system. The present invention fully explores the flotation process and influencing factors. Figure 1 A process flow diagram showing the flotation process.

[0024] The specific steps of the present invention are as follows:

[0025] Step 1: Data Preprocessing

[0026] Obtain real-time flotation information through sensor data. The data includes the dosage of frother and recovery agent, tailings and concentrate grade information during flotation, and preprocess the actual production data, remove abnormal points through filtering, and automatically detect invalid data;

[0027] Step 2: Design a decoupled controller

[0028] For the multivariable flotation dosing process, it is necessary to decompose the high-dimensional multivariable into multiple single-input single-output (SISO) processes. In order to improve the decoupling regulation performance of the multivariable system output, the reverse decoupling method is adopted. First, the decoupling control structure design of the control object is considered. Figure 2 shown.

[0029] Among them, the output of concentrate grade and tailings grade in the economic indicators are expressed as y sp1 and sp2 ; The set values ​​of the two variables are represented by r sp1 and r sp2 ; G c1 and G c2 Represent two PID controllers respectively; u sp1 and u sp2 The two controller outputs represent the dosage of frother and collector respectively.

[0030] Assume that the transfer function expression of the flotation object model is

[0031]

[0032] in Represents the transfer function model of each circuit in the flotation object, g ij (i,j=1,2) represents the minimum phase part, τ ij (i,j=1,2) represents the delay coefficient.

[0033] Decoupler D 21 (s) and D 12 The expression of (s) is:

[0034]

[0035]

[0036] Step 3: Cascade decoupling control structure design

[0037] The use of PID decoupling to control the multivariable flotation process has excellent adaptability and robustness, however, the use of traditional PID control methods still has limitations. Although simple PID decoupling control can be applied to multivariable control loops, PID cannot meet the diversity of control scenarios and cannot handle constraint problems well.

[0038] Attached Figure 3 The overall cascade control structure shown consists of an inner loop and an outer loop. First, for the bottom-level control of the inner loop, in order to meet the PID tracking control of the concentrate and tailings grades at the set point, it is necessary to decouple the established multivariable flotation dosing object to weaken the influence of interference between variables. Secondly, according to the coupling characteristics of the variables, the feedforward decoupling method is used to solve the coupling effect of each loop, which facilitates the control adjustment of the flotation dosing object after decoupling by the PID controller. Finally, for the outer loop, the optimization of the generalized object of the inner loop is completed through MPC, where the output of the MPC controller is the input of the inner loop. The input is the outer loop softening the set values ​​of the concentrate and tailings grades in the operation process, and then the bottom-level control of the deviation value of the actual value sampled by the sensor and the softened signal in the inner loop.

[0039] In the entire cascade control structure, the inner loop uses a densely sampled multivariable decoupled PID control strategy to suppress the main disturbances of the system; the outer loop uses an MPC control strategy, which has strong robustness to model mismatch and softens the reference trajectory of the control to obtain good tracking characteristics.

[0040] Step 4: Real-time parameter optimization and dosing control

[0041] Since the flotation dosing process has multivariable characteristics, in order to apply the single-loop PID control to the multivariable flotation dosing object, the loop needs to be decoupled. Here, the feedforward decoupling method is used to weaken the interference between variables. In the control scheme, the parameters that need to be optimized mainly include PID parameters and MPC weights. The error signal between the concentrate grade and tailings grade output and the reference economic index is calculated by the performance-based objective function. Here, the objective function selects the integral absolute error (IAE) as the performance measurement indicator, and the particle swarm algorithm is used to find the global optimum through a large number of evaluations. The performance index calculation formula is:

[0042]

[0043] Attached Figure 4It shows that the response tool obtained by the cascade control method and the PID decoupling control based on multivariables have good flotation index tracking performance. Compared with the PID-based decoupling control strategy, adding MPC for main loop control can optimize the sub-loop and predict the trajectory in advance, with a more stable control effect and better tracking effect. Figure 5 It indicates that MPC has a softening effect on the flotation set value, which speeds up the tracking response of the inner loop PID.

[0044] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A flotation dosing control method for a coal preparation plant based on cascade decoupling control, characterized in that: The specific steps include: Collect flotation information in real time, obtain the dosage of frother and collector, tailings and concentrate grade information during flotation, and pre-process the actual production data to eliminate abnormal points and invalid data; After obtaining the input and output sample data, the cascade decoupling controller is designed to control the flotation dosing object in the coal preparation plant; Optimize PID coefficients and MPC controller weight parameters according to the control scheme; The design of the cascade decoupling controller includes using PID decoupling control in the inner loop to decouple the flotation loop and weaken the interference between variables; using MPC in the outer loop to optimize the generalized object of the inner loop and soften the set value of the production index to stabilize the control effect. Finally, the output of the MPC controller is used as the input of the densely sampled PID decoupling loop to evaluate and predict the error of the flotation process output in advance, thereby effectively improving the disturbance suppression performance and control stability of the flotation loop. The optimization of PID coefficients and MPC controller weight parameters according to the control scheme is specifically to use a particle swarm algorithm to optimize the controller parameters, select the integral absolute error of the performance index as the optimization cost function, and provide a guarantee for the application of the cascade control strategy in the flotation dosing system.

Images