Optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process

Abstract

This invention discloses an optimization control method for deep purification and high-value copper and arsenic recovery in a heavy metal wastewater treatment process. The heavy metal wastewater treatment process utilizes a cascade of multiple unit-level equipment, with process state variables of adjacent unit-level equipment coupled to each other. The optimization control method at each time step includes: first, substituting current data into a global optimization problem model and solving the optimization problem to obtain the current setpoint concentration of each heavy metal ion for each unit-level equipment; the global optimization objective is to achieve the heavy metal ion removal rate and economic benefit of heavy metal recovery for the entire heavy metal wastewater treatment process; then, a decentralized control scheme is proposed, whereby each unit-level equipment uses a model predictive control strategy based on its current setpoint concentration of heavy metal ions to obtain the current control input for each unit-level equipment. This invention can improve the purity of sulfide slag and the economic benefits of the process.

Description

Technical Field

[0001] The present invention belongs to the field of industrial control technology, and in particular relates to an optimization control method for deep purification of heavy metal wastewater treatment and high-value recovery of copper and arsenic. Background Art

[0002] The non-ferrous metallurgical production process is the process of refining the metals contained in minerals into high-purity metals. This production process will produce a large amount of heavy metal wastewater, which usually contains a large amount of high-concentration impurity ions (copper, arsenic, antimony, bismuth, nickel, etc.). Its direct discharge will cause irreversible harm to the environment. Therefore, as the back-end of the non-ferrous metallurgical production process, the heavy metal wastewater treatment process is crucial to its green operation. Most elements in heavy metal wastewater can react with sulfur ions to form insoluble sulfide precipitates. Sulfidation technology is widely used in heavy metal wastewater treatment processes based on this principle. Currently, there are two main ways to introduce sulfur ions into heavy metal wastewater: using a solid sulfiding agent and directly introducing H2S. The solid sulfiding agent refers to Na2S or NaHS. When it is added to heavy metal wastewater, it will react with sulfuric acid to produce H2S as follows:

[0003] Na2S+H2SO4=Na2SO4+H2S↑ (1)

[0004] 2NaHS+H2SO4=Na2SO4+2H2S↑ (2)

[0005] This method is not only prone to H2S leakage but also introduces new cations into the wastewater. In comparison, the method of directly introducing H2S is safer and more reliable.

[0006] The heavy metal wastewater treatment is carried out by introducing hydrogen sulfide. Figure 1 As shown, the whole structure consists of multiple sulfidation reactors and thickeners connected in series. The wastewater flows through the sulfidation reactors in sequence, where it undergoes a sulfidation reaction with the introduced H2S to form sulfide slag precipitation. Sulfide slag of each level is obtained at the thickener corresponding to the sulfidation reactor. In wastewater containing copper and arsenic, H2S reacts with copper ions and arsenous acid (the form of arsenic in wastewater) to form copper sulfide and arsenic sulfide, respectively. If there are residual copper ions in the wastewater, the following reaction will occur:

[0007] 3Cu 2+ +As2S3+4H2O→3CuS+2HAsO2+6H + (3)

[0008] The generated arsenic sulfide is converted into copper sulfide. Specifically, when H2S is introduced into wastewater containing both copper and arsenic, copper sulfide is generated before arsenic sulfide. This provides the theoretical basis for obtaining high-purity sulfide slag at each thickener and achieving high-value copper and arsenic recovery. However, in actual production, the amount of H2S introduced into each sulfide reactor directly affects the purity of the sulfide slag. Specifically, when excessive H2S is introduced into the primary sulfidation reactor, a mixed slag of copper sulfide and arsenic sulfide is produced in the primary thickener, while high-purity arsenic sulfide slag is produced in the secondary thickener. When insufficient H2S is introduced into the primary sulfidation reactor, a mixed slag of copper sulfide and arsenic sulfide is produced in the primary thickener, while high-purity arsenic sulfide slag is produced in the secondary thickener. When an appropriate amount of H2S is introduced into the primary sulfidation reactor or an excessive amount of H2S is introduced into the secondary sulfidation reactor, pure copper sulfide slag is produced in the primary and secondary thickeners, respectively. However, this wastes H2S resources and increases the burden on subsequent tail gas treatment processes. When an appropriate amount of H2S is introduced into the primary sulfidation reactor or an insufficient amount of H2S is introduced into the secondary sulfidation reactor, pure copper sulfide slag is produced in the primary and secondary thickeners, respectively, but some arsenic still remains in the wastewater, failing to meet wastewater treatment requirements. Balancing the deep purification of heavy metal wastewater with high-value resource recovery is a key issue. Summary of the Invention

[0009] The present invention provides an optimized control method for deep purification and high-value recovery of copper and arsenic in a heavy metal wastewater treatment process, thereby improving the purity of sulfide slag and the economic benefits of the process.

[0010] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0011] An optimization control method for deep purification and high-value copper and arsenic recovery in a heavy metal wastewater treatment process. The heavy metal wastewater treatment process is constructed using a cascade of multiple unit-level equipment. The process state variables of adjacent unit-level equipment are coupled to each other through physical connectivity. The optimization control method at each time step includes:

[0012] First, the input variable data and state variable data of each unit-level equipment are substituted into the global optimization problem model of the heavy metal wastewater treatment process, and the optimization problem is solved to obtain the concentration of copper and arsenic ions in each unit-level equipment;

[0013] The global optimization problem model is as follows: the copper-arsenic ion removal rate of the entire heavy metal wastewater treatment process and the economic benefits of copper-arsenic sulfide slag recovery are taken as optimization targets, the relationship between the optimization target and the input variables, state variables, and output variables of each unit-level equipment is used as the objective function, and the concentration of copper-arsenic ions in the output variables of each unit-level equipment is set as the optimization variable;

[0014] Then, each unit-level equipment uses the concentration of copper arsenic ions obtained by optimization as the set value, adopts the model predictive control strategy, performs predictive control on the input variables of each unit-level equipment, and obtains the control input of each unit-level equipment in the next time step, that is, the input variable of the next time step.

[0015] Furthermore, the input variable of each unit-level equipment is the amount of H2S introduced.

[0016] Furthermore, the state variables of each unit-level equipment include: volume of contaminated acid, copper ion concentration, arsenic ion concentration, copper sulfide content, and arsenic sulfide content.

[0017] Furthermore, the output variables of each unit-level equipment include: copper ion concentration, arsenic ion concentration, copper sulfide content, and arsenic sulfide content.

[0018] Furthermore, the heavy metal wastewater treatment process is globally optimized based on the following process steady-state model:

[0019] x i =f s (x i,0 ,u i )

[0020] y i =g s (x i )

[0021] x i+1,0 =x i ,i=1,…,N

[0022] Among them, x i，0 is the initial value of the state variable of the i-th unit-level equipment; u i ,x i ,y i represents the input variables, state variables, and output variables of the i-th unit-level equipment; x i+1，0 is the initial value of the state variable of the i+1th unit-level equipment, which is determined by the state variable x of the i-th unit-level equipment. i Determine; N is the number of unit levels in the cascade of heavy metal wastewater treatment process.

[0023] Furthermore, the objective function in the global optimization problem is:

[0024]

[0025] Among them, P represents the optimization target of the heavy metal wastewater treatment process, including the copper-arsenic ion removal rate of the entire heavy metal wastewater treatment process and the economic benefit of copper-arsenic sulfide slag recovery; is the objective function.

[0026] Furthermore, the constraints of the global optimization problem are:

[0027]

[0028] in, and The input variable u for the i-th unit level equipment is i lower and upper limits.

[0029] Furthermore, a two-stage sulfidation reactor cascade is used to purify heavy metal wastewater in which copper and arsenic coexist; the first-stage sulfidation reactor is used to purify copper ions, and its global optimization problem only includes one optimization variable, namely the copper ion concentration; the second-stage sulfidation reactor is used to purify arsenic ions, and its global optimization problem only includes one optimization variable, namely the arsenic ion concentration.

[0030] Furthermore, each unit-level equipment uses a model predictive control strategy based on its own heavy metal ion concentration set value to predictively control its input variables. This is expressed as solving the following optimization problem, obtaining the optimal control sequence, and taking the first value of the sequence as the input variable for the next time step:

[0031]

[0032] Among them, m c Indicates the number of steps in the control time domain, m p Indicates the number of steps in the prediction time domain, j represents the current moment; Indicates that the i-th unit-level equipment uses the input variable x at the current j moment i (j) and the initial value of the state variable x i,0 (j) The optimal sequence value of the input variables obtained by solving, that is, the optimal control sequence; y i (j+l) represents the output variable of the i-th unit-level equipment at time j+l, u i (j+l) represents the input variable of the i-th unit-level equipment at time j+l; is the optimization variable of the i-th unit-level equipment, that is, the concentration setting value of the heavy metal ions to be purified by the i-th unit-level equipment; and The input variable u for the i-th unit level equipment is i The lower and upper bounds of the objective function are: Composition, the |||| internal uniform use of x to represent, then where x T is the transpose of x, and P and R are the weights of each item.

[0033] Beneficial effects

[0034] This paper proposes an optimized control method for deep purification and high-value copper and arsenic recovery in heavy metal wastewater treatment. First, a global optimization of the cascade process is performed to improve process efficiency. The overall economic benefit of the process is studied and optimized to ensure global optimality of the setpoints for each unit-level device. Next, a decentralized control scheme is proposed, with controllers designed based on the unit-level devices, effectively improving control efficiency and ensuring effective control.

[0035] Applying this method to industrial wastewater treatment processes significantly improves the purity of sulfide slag and the economic benefits of the process. Specifically, in the treatment of wastewater containing copper and arsenic, the purity of copper sulfide slag increased by over 15% and the economic benefits by over 6% compared to traditional methods. Therefore, the method of the present invention is highly effective in purifying heavy metal wastewater and recovering high-value heavy metal resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a schematic diagram of hierarchical sulfidation for treating heavy metal wastewater by introducing hydrogen sulfide;

[0037] Figure 2 It is the global optimization-decentralized control architecture of the embodiment of the present application;

[0038] Figure 3 is a schematic diagram of the cascade vulcanization process of an embodiment of the present application;

[0039] Figure 4 It is a schematic diagram of global optimization and local optimization;

[0040] Figure 5 This is a model predictive control framework diagram of an embodiment of the present application;

[0041] Figure 6 is the liquid phase material control result of the experiment described in the embodiment of the present application;

[0042] Figure 7 is the solid phase material control result of the experiment described in the examples of this application;

[0043] Figure 8 It is a process economic indicator of the experiment described in the examples of this application. DETAILED DESCRIPTION

[0044] The following is a detailed description of an embodiment of the present invention. This embodiment is based on the technical solution of the present invention, provides a detailed implementation method and a specific operation process, and further explains the technical solution of the present invention.

[0045] This embodiment provides an optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process. The overall framework is as follows: Figure 1As shown in the figure, a global optimization-decentralized control architecture is adopted. In global optimization, the heavy metal ion removal rate of the entire heavy metal wastewater treatment process and the economic benefits of heavy metal sulfide slag recovery are used as optimization targets. By establishing a relationship between the overall benefits of the entire process and the key states of the system, the optimization problem is solved to obtain the control set values ​​of each unit-level equipment, thereby ensuring the global optimality of the set values ​​of each unit-level equipment. In decentralized control, after obtaining the control set values ​​of each unit-level equipment, the controller is designed with the unit-level equipment as the control object to achieve the optimal control target, effectively improving control efficiency and ensuring control effectiveness.

[0046] 1. Global process optimization

[0047] The whole process of heavy metal wastewater treatment of the present invention adopts a cascade of multiple unit-level equipment, and the process state variables of adjacent unit-level equipment are cascaded through pipes and other structures. Figure 3 As shown in Figure 2, the state variables of adjacent unit-level devices are coupled to each other due to physical connections.

[0048] like Figure 3 The cascade vulcanization process shown can be expressed as:

[0049] x i (k+1)=f i (x i,0 (k),u i (k))

[0050] y i (k+1)=g i (x i (k+1)) (4)

[0051] x i+1,0 (k) = x i (k-1),i=1,…,N

[0052] Among them, i represents the unit-level equipment index, k represents time, and u i (k), x i (k) and y i (k) represents the input variable, state variable and output variable of the i-th unit-level equipment at time k, x i,0 (k), x i+1,0 (k) represents the initial state of the i-th unit level and the i+1-th unit level at time k; f i and g i is the mapping relationship between the input variables and state variables of the i-th unit-level equipment, g i is the mapping relationship between the input variables and output variables of the i-th unit-level equipment; for the cascade vulcanization system, its biggest feature is reflected in x i+1,0 (k) = xi (k-1), that is, the state variables of each unit-level equipment are mutually correlated and coupled, and the output of each unit-level equipment is not only affected by the input of the current equipment, but also by the upstream equipment.

[0053] This embodiment of the present invention treats wastewater containing both copper and arsenic by introducing hydrogen sulfide, using two separate unit stages to treat the copper and arsenic ions in the wastewater. In each stage of the sulfidation reactor, the input variable is the amount of hydrogen sulfide introduced, and the state variables include: waste acid volume, copper ion concentration, arsenite concentration, copper sulfide content, and arsenic sulfide content. The output variables include: copper ion concentration, arsenite concentration, copper sulfide content, and arsenic sulfide content.

[0054] The optimization control objectives of the cascade vulcanization process are to ensure safe operation, namely:

[0055]

[0056] in, and are the lower and upper limits of the input variables of the i-th unit-level equipment, respectively.

[0057] If dismantling is done according to unit level equipment, it is dismantled into Perform local optimization separately and optimize the overall benefit by optimizing the benefit of each equipment. Although each unit level can obtain the local optimum, it ignores the coupling relationship between equipment. The existence of the cascade structure makes the independent sum of the local optimum of the unit level equipment not equal to the global optimum, such as Figure 4 At the same time, the control effect of each unit-level equipment is difficult to guarantee, so the present invention proposes a process global optimization method.

[0058] In order to ensure the overall benefits of the cascade process, the process decision variables are globally optimized. First, a steady-state model of the process needs to be obtained. Optimization based on the steady-state model can ensure the feasibility of the decision target. The steady-state model corresponding to Equation (4) is obtained by the following formula:

[0059]

[0060] The global optimization goal of the cascade vulcanization process is:

[0061]

[0062] Where P represents the optimization target of the heavy metal wastewater treatment process, including the copper-arsenic ion removal rate of the entire heavy metal wastewater treatment process and the economic benefit of recovering copper-arsenic sulfide slag; is the objective function, i.e., the mapping relationship between the optimization objective and the key variables of each unit-level equipment, u i ,x i ,y irepresents the key variables of the i-th unit-level equipment, including input variables, state variables and output variables. N is the number of unit-level cascades in the heavy metal wastewater treatment process. p1, p2, p3, p4, q1, q2 are the weights of each item, and x 1,1 、x 1,2 are the copper ion concentration and arsenite concentration in the primary sulfidation reactor, respectively, and x 2,1 、x 2,2 where u1 and u2 are the copper ion concentration and arsenite concentration in the secondary sulfidation reactor, respectively. The amounts of hydrogen sulfide input to the primary and secondary sulfidation reactors, respectively, are used. Equation (7) improves global efficiency by directly optimizing the overall process efficiency, effectively avoiding the drawbacks of optimizing unit-level equipment efficiency.

[0063] Global optimization is based on the following process steady-state model:

[0064] x i =f s (x i,0 ,u i )

[0065] y i =g s (x i ) (8)

[0066] x i+1,0 =x i ,i=1,…,N

[0067] Formulas (5), (7), and (8) together constitute the global optimization model of the cascade process. Optimizing this model yields the set value of the cascade sulfidation process, i.e., the copper ion concentration set value of the first-stage sulfidation reactor. and the set value of arsenic ion concentration in the secondary sulfidation reactor Finally, in order to ensure the control effect, each unit-level equipment is controlled separately.

[0068] 2. Decentralized control

[0069] The decentralized control strategy is based on Model Predictive Control (MPC). Compared with model-free control methods such as PID, MPC can consider more system states and handle constraints. It predicts the future state of the system in multiple steps within the current prediction step, while considering the system constraints to determine the next control input, such as Figure 5 shown.

[0070] Rolling optimization is an important mechanism of MPC, which is achieved by p =m p By minimizing the objective function within Δt, the optimization problem is solved to obtain a control time domain Tc =m c Optimal control sequence within Δt Δt is the step size of the discrete model. Usually the control time domain is less than or equal to the prediction time domain. When the control time domain is less than the prediction time domain, [m c+1 ,…,m l ] range and the last control quantity u(j+m c ), and finally the first value u(j+1) in the optimal control sequence is applied to the process. Repeating the above process at each moment can obtain the control rate:

[0071]

[0072] in, It means that the first control variable u(j+1) in the optimal control sequence is obtained by taking x(j) as the initial state.

[0073] The first-stage sulfurization reactor controller is responsible for tracking the copper ion concentration set value obtained by global optimization The control optimization problem is:

[0074]

[0075] Where m c Indicates the number of steps in the control time domain, m p represents the number of steps in the prediction time domain, Δt is the step size of the discrete model, and j represents the current moment; Indicates that the first stage sulfidation reactor is in the initial state x at the current time j 1,0 (j) The optimal sequence value of the input variables obtained by solving, that is, the optimal control sequence; y1(j+l) represents the output variable of the first-stage vulcanization reactor at time j+l, and u1(j+l) represents the input variable of the first-stage vulcanization reactor at time j+l.

[0076] The second-stage sulfidation reactor controller is responsible for tracking the arsenic ion concentration set value obtained by global optimization The control optimization problem is:

[0077]

[0078] Indicates that the second stage sulfidation reactor is in the initial state x at the current time j 2,0 (j) The optimal sequence value of the input variables obtained by solving, that is, the optimal control sequence; y2(j+l) represents the output variable of the second-stage vulcanization reactor at time j+l, and u2(j+l) represents the input variable of the second-stage vulcanization reactor at time j+l.

[0079] In the objective function of the above-mentioned optimization problem for the control of each stage of the sulfidation reactor, each item is calculated in the form of a weighted vector. x T is the transpose of x. Represents the deviation of the process state from the reference trajectory, g * In (10) and (11), they represent and

[0080] By solving (10) and (11) respectively, the optimal control sequence of the first-stage sulfidation reactor at the current time j is obtained and the optimal control sequence of the second stage sulfidation reactor The first value in the optimal control sequence is applied to the corresponding sulfidation reactor, that is, the corresponding amount of hydrogen sulfide is introduced into the sulfidation reactor to achieve decentralized control of the cascade sulfidation treatment of copper and arsenic coexisting wastewater.

[0081] In order to verify the effectiveness of the proposed method, the proposed method is compared with the global optimization-global control (GlobalOptimization-Global Control, GO-GC), local optimization-decentralized control (LocalOptimization-Decentralized Control, LO-DC) and rule control methods. The experimental results are as follows: Figure 6-8 and as shown in Table 1.

[0082] Table 1: Control effect

[0083]

[0084] From Table 1, it can be seen that the method of the present invention can achieve a higher purity of the first-level copper sulfide slag. In terms of the purity of the second-level arsenic sulfide slag, there is no significant difference between the various methods, all reaching more than 98%. At the same time, the distributed control scheme can also ensure the timeliness of the control. Figure 8 It can be seen that the proposed method requires lower processing costs and can achieve higher process benefits.

[0085] The above embodiments are preferred embodiments of the present application. Ordinary technicians in this field can also make various changes or improvements on this basis. Without departing from the overall concept of the present application, these changes or improvements should fall within the scope of protection required by the present application.

Claims

1. An optimization control method for deep purification of heavy metal wastewater treatment and high-value recovery of copper and arsenic, characterized in that: The heavy metal wastewater treatment process is composed of a plurality of unit-level equipment in cascade, and the process state variables of adjacent unit-level equipment are coupled to each other through physical connection. The optimization control method at each time step includes: First, the input variable data and state variable data of each unit-level equipment are substituted into the global optimization problem model of the heavy metal wastewater treatment process, and the optimization problem is solved to obtain the concentration of copper and arsenic ions in each unit-level equipment; The global optimization problem model is as follows: the copper-arsenic ion removal rate of the entire heavy metal wastewater treatment process and the economic benefits of copper-arsenic sulfide slag recovery are taken as optimization targets, the relationship between the optimization target and the input variables, state variables, and output variables of each unit-level equipment is used as the objective function, and the concentration of copper-arsenic ions in the output variables of each unit-level equipment is set as the optimization variable; Then, each unit-level equipment uses the concentration of copper arsenic ions obtained by optimization as the set value, adopts the model predictive control strategy, performs predictive control on the input variables of each unit-level equipment, and obtains the control input of each unit-level equipment in the next time step, that is, the input variable of the next time step.

2. The optimization control method for deep purification of heavy metal wastewater treatment and high-value recovery of copper and arsenic according to claim 1 is characterized in that: The input variable of each unit-level equipment is the amount of H2S input.

3. The optimization control method for deep purification of heavy metal wastewater treatment and high-value recovery of copper and arsenic according to claim 1 is characterized in that: The state variables of each unit-level equipment include: contaminated acid volume, copper ion concentration, arsenic ion concentration, copper sulfide content, and arsenic sulfide content.

4. The optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process according to claim 1 is characterized in that: The output variables of each unit-level equipment include: copper ion concentration, arsenic ion concentration, copper sulfide content, and arsenic sulfide content.

5. The optimization control method for deep purification of heavy metal wastewater treatment and high-value recovery of copper and arsenic according to claim 1 is characterized in that: The heavy metal wastewater treatment process is globally optimized based on the following process steady-state model: x i =f s (x i,0 ,u i ) y i =g s (x i ) x i+1,0 =x i ,i=1,…,N Among them, x i，0 is the initial value of the state variable of the i-th unit-level equipment; u i ,x i ,y i represents the input variables, state variables, and output variables of the i-th unit-level equipment; x i+1，0 is the initial value of the state variable of the i+1th unit-level equipment, which is determined by the state variable x of the i-th unit-level equipment. i Determine; N is the number of unit levels in the cascade of heavy metal wastewater treatment process.

6. The optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process according to claim 5 is characterized in that: The objective function in the global optimization problem is: Among them, P represents the optimization target of the heavy metal wastewater treatment process, including the copper-arsenic ion removal rate of the entire heavy metal wastewater treatment process and the economic benefit of copper-arsenic sulfide slag recovery; is the objective function.

7. The optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process according to claim 1 is characterized in that: The constraints of the global optimization problem are: in, and The input variable u for the i-th unit level equipment is i lower and upper limits.

8. The optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process according to claim 1 is characterized in that: A two-stage sulfidation reactor cascade is used to purify heavy metal wastewater containing coexisting copper and arsenic. The first-stage sulfidation reactor is used to purify copper ions, and the global optimization problem only includes one optimization variable, copper ion concentration. The second-stage sulfidation reactor is used to purify arsenic ions, and the global optimization problem only includes one optimization variable, arsenic ion concentration.

9. The optimization control method for deep purification and high-value recovery of copper and arsenic in heavy metal wastewater treatment process according to claim 1, characterized in that: Each unit-level equipment adopts the model predictive control strategy to predict and control the input variables of its own unit-level equipment. This is expressed as solving the following optimization problem, obtaining the optimal control sequence and taking the first value of it as the input variable of the next time step: Among them, m c Indicates the number of steps in the control time domain, m p Indicates the number of steps in the prediction time domain, j represents the current moment; Indicates that the i-th unit-level equipment uses the input variable x at the current j moment i (j) and the initial value of the state variable x i,0 (j) The optimal sequence value of the input variables obtained by solving, that is, the optimal control sequence; y i (j+l) represents the output variable of the i-th unit-level equipment at time j+l, u i (j+l) represents the input variable of the i-th unit-level equipment at time j+l; is the optimization variable of the i-th unit-level equipment, that is, the concentration setting value of the heavy metal ions to be purified by the i-th unit-level equipment; and The input variable u for the i-th unit level equipment is i The lower and upper bounds of the objective function are: Composition, the || || internal uniform use of x to represent, then where x T is the transpose of x, and P and R are the weights of each item.

Images