Intelligent identification and optimization system and method for thallium enrichment and cadmium recovery in zinc hydrometallurgy process

Abstract

The invention provides an intelligent identification and optimization system and method for thallium enrichment and cadmium recovery in the zinc hydrometallurgy process, and belongs to the field of non-ferrous metal metallurgy and intelligent optimization control. The system comprises a thallium enrichment process intelligent optimization module based on incoming material identification and a cadmium recovery process target optimization module. According to the method, thallium content data of mixed zinc concentrate and thallium-containing and cadmium-containing slag in MES are grabbed, 8 g / t, 15 g / t and 200 g / t are taken as threshold values respectively, thallium removal and cadmium recovery processes are intelligently matched, after process rationality verification, a multi-objective function is established through a deep learning model, and an optimal operation scheme is solved by adopting an NSGA-II algorithm. The method adapts to the raw material thallium content fluctuation scene, no toxic reagent is introduced, collaborative optimization of thallium enrichment and cadmium recovery is achieved, the process stability and economical efficiency are improved, and cost reduction and efficiency improvement are achieved for enterprises.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of non-ferrous metal smelting and intelligent optimization control, and specifically relates to an intelligent identification and optimization system and method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process. Background Technology

[0002] Hydrometallurgical zinc smelting, as one of the mainstream zinc smelting processes, requires not only efficient zinc recovery during production but also the simultaneous control and treatment of associated elements such as thallium and cadmium in the raw materials. Thallium accumulation within the system can severely impact the normal operation of the zinc electrowinning process and even lead to substandard product quality. Cadmium, as a high-value by-product metal, has its recovery efficiency directly affecting the overall profitability of smelting enterprises. Therefore, the targeted enrichment of thallium and the efficient and economical recovery of cadmium have become key technical issues urgently needing to be addressed in the hydrometallurgical zinc smelting industry.

[0003] Existing technologies have proposed some solutions for the treatment of thallium and cadmium in the hydrometallurgical zinc smelting process, but they still have significant shortcomings and cannot meet the industry's requirements for process adaptability, safety, and overall efficiency: First, some technologies (such as thallium recovery processes for acidic leaching solutions in hydrometallurgical zinc smelting) do not clearly define the suitable thallium concentration range. Relying solely on zinc powder to replace low-concentration thallium (a few mg / L) has limited effectiveness, and it requires the use of excessive potassium ferrocyanide to precipitate thallium, introducing toxic cyanide and posing safety and environmental risks. Second, some thallium recovery technologies (such as methods for treating thallium and cadmium flue dust from lead smelting) have strict requirements for the thallium content of the raw materials (requiring 0.4-1.0%, i.e., 4000-10000 mg / L). The first limitation is that the concentration of thallium in the raw materials for hydrometallurgical zinc refining varies greatly (5-20 g / t), which is not fully suitable for the actual scenario where the thallium content in the flue dust products containing thallium flue dust is low, thus limiting its applicability. Secondly, for multi-metal co-recovery schemes (such as the comprehensive recovery process of lead system flue dust), the impact of key process parameters (such as the thallium concentration in the liquid after the first replacement) on the recovery effect is not clearly defined. When the thallium concentration in the liquid after the first replacement is low, the effect of zinc powder replacing thallium will be greatly reduced, and thallium will be difficult to completely enter the first replacement slag, resulting in insufficient thallium removal. Fourth, existing optimization technologies for wet zinc smelting processes (such as potential control in the cobalt removal process) only focus on the removal of a single, high-content (tens of g / t) element (cobalt), without considering the synergistic treatment of harmful and low-content elements such as thallium and valuable elements such as cadmium. Furthermore, due to the strong coupling of multiple factors, large time delays, and nonlinear mapping between various influencing factors and process objectives in the process, it is difficult to coordinate and optimize the design of process objectives, and it is impossible to maximize the comprehensive benefits of "thallium enrichment-cadmium recovery-cost control".

[0004] In summary, existing technologies have significant shortcomings in terms of process adaptability, element synergistic processing capability, safety, and economy. There is an urgent need for an intelligent optimization scheme that can dynamically adapt the process according to the characteristics of raw materials and take into account both efficient enrichment of thallium and economical recovery of cadmium, in order to solve the actual technical pain points of the hydrometallurgical zinc smelting industry. Summary of the Invention

[0005] To overcome the problems of existing technologies, this invention proposes an intelligent identification and optimization system and method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process. This system adapts to the actual scenario of fluctuating thallium content in hydrometallurgical zinc smelting raw materials, avoids the introduction of toxic reagents, achieves directional enrichment of thallium and efficient and economical recovery of cadmium, and simultaneously controls energy consumption and reagent costs. This addresses the shortcomings of existing technologies in terms of process adaptability, collaborative processing capabilities, safety, and economy, thereby reducing costs and increasing efficiency for enterprises.

[0006] The objective of this invention is achieved as follows:

[0007] The first aspect of the present invention provides an intelligent identification and optimization system for thallium enrichment and cadmium recovery in a hydrometallurgical zinc smelting process, including an intelligent optimization module for thallium enrichment process based on incoming material identification and a target optimization module for cadmium recovery process.

[0008] The intelligent optimization module for thallium enrichment process based on incoming material identification includes: a multi-source mixed zinc concentrate incoming material identification unit, a thallium removal process determination unit, an automatic verification unit for process rationality, a thallium- and cadmium-containing slag identification unit, and a cadmium recovery process determination unit.

[0009] The multi-source mixed zinc concentrate incoming material identification unit is used to automatically capture the thallium content data of mixed zinc concentrate in the associated enterprise manufacturing execution system (MES) and identify thallium content <8g / t, 8g / t≤thallium content≤15g / t or >15g / t;

[0010] The thallium removal process determination unit is used to prompt the user to enter a specific thallium removal process mode based on the thallium content identification result of the incoming material: "thallium removal in purification process", "thallium removal in leaching control + thallium removal in purification process" or "thallium removal in leaching and purification processes".

[0011] The automatic verification unit for process rationality is used to further verify the accuracy and rationality of the prompts from the thallium removal process determination unit based on the actual thallium concentration data of the zinc leaching solution in the MES, according to whether the thallium concentration is <0.1mg / L or ≥0.1mg / L.

[0012] The thallium-cadmium slag identification unit is used to automatically capture the thallium content data of the associated thallium-cadmium slag in the MES and identify whether the thallium content is <200g / t or ≥200g / t.

[0013] The cadmium recovery process determination unit is used to indicate whether to enter a specific cadmium recovery process mode based on the thallium content identification result of thallium-containing cadmium slag: "direct refining" or "multi-stage thallium removal in cadmium recovery process". After entering "multi-stage thallium removal in cadmium recovery process", it is necessary to enter the "direct refining" process again.

[0014] The cadmium recovery process target optimization module includes a cadmium recovery optimization path determination unit, which is used to determine the incoming material "direct refining target optimization" or "'cadmium recovery process multi-stage thallium removal process target optimization' → 'direct refining target optimization'" based on the cadmium recovery process determined by the cadmium recovery process determination unit.

[0015] Furthermore, thallium content of 8 g / t and 15 g / t are the critical start-up thresholds for different thallium removal processes, and 200 g / t is the critical start-up threshold for different cadmium recovery processes.

[0016] The second aspect of this invention provides an intelligent identification and optimization method for thallium enrichment and cadmium recovery in a hydrometallurgical zinc smelting process. The evaluation method is based on the system described in the first part and includes the following steps:

[0017] Step 1: Intelligent matching and execution of thallium removal processes based on the thallium content of the incoming material.

[0018] The thallium content of multi-source zinc concentrate is identified by an intelligent optimization module based on incoming material recognition. The next thallium removal process is selected based on the obtained thallium content of the mixed zinc concentrate in the MES.

[0019] When the thallium content of the mixed zinc concentrate is identified as <8g / t, the system prompts that the zinc leaching solution directly enters the "purification process for thallium removal" and verifies the thallium concentration of the zinc leaching solution from the roasting dust and roasting sand: if the thallium concentration of the zinc leaching solution is <1.0mg / L, it is considered reasonable and the zinc leaching solution enters the "purification process for thallium removal". In this process, thallium removal is enhanced by "lead salt + zinc powder", producing thallium-cadmium-containing slag; if the thallium concentration of the zinc leaching solution is ≥1.0mg / L, it enters the "leaching and purification process for tiered thallium removal". In the leaching process, the zinc leaching solution first undergoes permanganate or ozone oxidation precipitation to remove thallium, producing thallium-containing leaching slag and thallium-removed leaching solution. The thallium-removed leaching solution enters the purification process, where thallium removal is further enhanced by "lead salt + zinc powder", producing thallium-cadmium-containing slag.

[0020] When the system identifies a thallium content of 8g / t ≤ 15g / t in the mixed zinc concentrate, it prompts the user to enter the "leaching control + purification process for thallium removal". The control method involves reducing the amount of roasting dust from the primary dust collector by 1 / 2 to 1 / 3, and mixing all the roasting dust from the secondary dust collector with the roasted sand for leaching. This staged leaching of roasting dust reduces the concentration of the zinc leachate. Then, the thallium concentration in the zinc leachate is verified: a thallium concentration < 1.0mg / L is considered reasonable, and the zinc leachate enters the next stage. The "Thallium Removal in Purification Process" involves using "lead salt + zinc powder" to enhance thallium removal during cadmium removal, producing thallium- and cadmium-containing slag. When the thallium concentration in the zinc leaching solution is ≥1.0 mg / L, it enters the "Step-by-Step Thallium Removal in Leaching and Purification Process." In the leaching process, the zinc leaching solution first undergoes permanganate or ozone oxidation precipitation to remove thallium, producing thallium-containing leaching slag and thallium-removed leaching solution. The thallium-removed leaching solution then enters the purification process, where "lead salt + zinc powder" is used to further enhance thallium removal, producing thallium- and cadmium-containing slag.

[0021] When the thallium content of the mixed zinc concentrate is identified as ≥8g / t, the system prompts to enter the "thallium removal in the leaching and purification process" and verifies the thallium concentration of the zinc leaching solution from the roasting dust and roasting sand: if the thallium concentration of the leaching solution is <1.0mg / L, return to the "thallium removal in the purification process"; if the thallium concentration of the leaching solution is ≥1.0mg / L, operate according to the "thallium removal in the leaching and purification process" process, and remove thallium from the zinc leaching solution in the leaching process and the purification process respectively, producing thallium-cadmium-containing slag.

[0022] Step 2: Screening and implementation of cadmium recovery processes based on the characteristics of thallium- and cadmium-containing slag.

[0023] Based on the obtained thallium content in the thallium-cadmium slag within the MES, the next cadmium recovery process will be selected:

[0024] When the thallium content of the thallium-cadmium slag is identified as <200g / t, the system prompts that the thallium-cadmium slag should enter "direct refining" to produce refined cadmium through refining and vacuum distillation;

[0025] When the thallium content of the thallium-cadmium slag is identified as ≥200g / t, the system prompts to enter the "multi-stage thallium removal process in cadmium recovery". Through multi-stage thallium removal, including thallium-cadmium slag stockpiling oxidation, acid leaching to make liquor, and zinc powder replacement, thallium-rich liquor-making slag and sponge cadmium are produced. The sponge cadmium then enters the "direct refining" process.

[0026] Step 3: Multi-objective intelligent optimization solution for cadmium recycling process

[0027] The target optimization module for the cadmium recovery process selects the target optimization path for the next step of cadmium recovery based on the cadmium recovery process:

[0028] When thallium-cadmium-containing slag enters the "direct refining" process, the target optimization process for this step is implemented:

[0029] Establish a set of decision variables: ;

[0030] Among them, setting The iron content of thallium-cadmium slag; The cadmium content of thallium-cadmium slag; This refers to the refining temperature; This refers to the distillation temperature. This is the condensation temperature; The melting and casting temperature; n≥6;

[0031] Establish the objective function through deep learning training:

[0032] Cadmium recovery rate ;

[0033] System energy consumption ;

[0034] Sodium hydroxide addition amount ;

[0035] In the formula, CNN represents a deep learning model trained on a production dataset;

[0036] For the objective function Solve the problem and obtain the optimal solution set P. The system will then display the message "Optimization complete".

[0037] When thallium-cadmium-containing slag enters the "multi-stage thallium removal process in cadmium recovery → direct refining," a "two-step optimization" process is implemented:

[0038] The first step in optimization is to optimize the multi-stage thallium removal process in the cadmium recovery process:

[0039] Establish a set of decision variables: ;

[0040] in, The iron content of thallium-cadmium slag; The lead content of thallium-cadmium slag; The cadmium content of thallium-cadmium slag; The thallium content of thallium-cadmium slag; The zinc content of thallium-cadmium slag; This refers to the amount of oxidant added; This refers to the amount of lead salt added; m ≥ 7;

[0041] Establish the objective function using a deep learning model:

[0042] Cadmium recovery rate in sponge cadmium ;

[0043] System energy consumption ;

[0044] Zinc powder addition amount ;

[0045] Thallium enrichment ratio in liquid residue ;

[0046] In the formula, DNN represents a deep learning model trained on a production dataset;

[0047] For the objective function Solve to obtain the set of optimal solutions. ;

[0048] The second optimization step involves optimizing the cadmium sponge refining target:

[0049] The cadmium sponge product obtained in the first optimization step is then compressed and refined, following the direct refining target optimization process to obtain the optimal solution set. The system then displayed a message: "Optimization complete."

[0050] Furthermore, the deep learning model includes a convolutional neural network or a long short-term memory neural network.

[0051] In step 3, the decision variables and their number n for "target optimization of direct refining" and the decision variables and their number m for "target optimization of multi-stage thallium removal in cadmium recovery process" are determined based on grey relational analysis and can be selected and replaced according to actual target requirements.

[0052] Furthermore, in step 3, the constraints for directly refining the target optimization process are:

[0053] st ∈ [ , ], i =1,2,...,n;

[0054] in, , These represent the lower and upper limits of the decision variables obtained by the MES and the distributed control system (DCS), respectively, which are determined by statistical analysis of more than 1,000 sets of historical datasets.

[0055] Furthermore, in step 3, the constraints for optimizing the multi-stage thallium removal objective in the cadmium recovery process are as follows:

[0056] st ∈ [ , ], i =1,2,...,m;

[0057] in, , These represent the lower and upper limits of the decision variables obtained from MES and DCS, respectively, which are determined by statistical analysis of more than 1,000 historical datasets.

[0058] Furthermore, the method for solving the objective function includes the fast non-dominated sorting genetic algorithm.

[0059] The advantages and beneficial effects of this invention are:

[0060] 1. High process adaptability: Based on the thallium content of mixed zinc concentrate (with 8g / t and 15g / t as the boundary), the thallium concentration of leachate (with 1.0mg / L as the boundary), and the thallium content of thallium-cadmium slag (with 200g / t as the boundary), it can automatically match processes such as "thallium removal in purification process", "thallium removal in leaching control + purification process", "thallium removal in tiered leaching and purification processes", "direct cadmium refining", and "multi-stage thallium removal in cadmium recovery", adapting to the actual scenario of fluctuating thallium content (5-20g / t) in hydrometallurgical zinc smelting raw materials, breaking through the limitations of existing technologies in adapting to thallium concentration;

[0061] 2. Excellent safety and environmental protection: The entire process adopts a combination of permanganate / ozone, lead salt, and zinc powder processes, without the introduction of toxic reagents such as cyanide; after thallium is enriched in the liquid-making slag, it can be further processed and enriched in high-fluorine chlorine fume (which can be used to extract thallium), realizing the reduction of thallium dispersion pollution and the recycling of valuable metals such as cadmium, which meets the requirements of green metallurgy.

[0062] 3. Significant synergistic benefits: By coordinating thallium enrichment and cadmium recovery, a multi-objective optimization model is used to achieve the maximum thallium enrichment ratio, the maximum cadmium recovery rate, and the lowest energy consumption and reagent costs. At the same time, it ensures that the thallium concentration in the new electrolytic solution is ≤0.1mg / L, avoids zinc electrowinning failure, and solves the shortcomings of existing technologies that only process one element.

[0063] 4. Intelligent and economical: The intelligent optimization module for thallium enrichment process based on incoming material recognition described in this application automatically captures data and matches processes, reducing human error and over-processing; through the coupling of big data-driven deep learning models and optimization algorithms, it quickly provides multiple sets of selectable optimal operation schemes for production, taking into account process stability, economy and practicality, and reducing costs and increasing efficiency for enterprises. Attached Figure Description

[0064] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0065] Figure 1 This is a hierarchical structure diagram of the intelligent identification and optimization system for thallium enrichment and cadmium recovery in the wet zinc smelting process described in this invention.

[0066] Figure 2 This is a schematic diagram of the target optimization process for the cadmium recovery procedure described in this invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments disclosed herein will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0068] Example:

[0069] This embodiment provides an intelligent identification and optimization system and method for thallium enrichment and cadmium recovery in a hydrometallurgical zinc smelting process, such as... Figure 1 As shown, the system includes a thallium enrichment process intelligent optimization module based on incoming material identification and a cadmium recovery process target optimization module. This embodiment takes the production process of a hydrometallurgical zinc smelting enterprise as an example. Based on the system's identification of its mixed zinc concentrate incoming material and determination of the thallium enrichment and cadmium recovery optimization schemes, the specific steps include:

[0070] Step 1: Using the "Intelligent Optimization Module for Thallium Enrichment Process Based on Incoming Material Recognition," the thallium content of the mixed zinc concentrate from the enterprise is identified. The system automatically captures the thallium content data of the mixed zinc concentrate in the MES, which is 15.68 g / t. The system automatically identifies and prompts the zinc leaching solution to enter the "step-by-step thallium removal in leaching and purification process" setting. After the roasted sand is leached, the system captures the thallium concentration of the zinc leaching solution in the MES for verification. The thallium concentration is identified as 1.83 mg / L. The zinc leaching solution enters the "step-by-step thallium removal in leaching and purification process" as prompted. In the leaching process, the zinc leaching solution first undergoes oxidation precipitation with permanganate to remove thallium, producing thallium-containing leaching residue and thallium-removed leaching solution. The thallium-removed leaching solution enters the purification process, where it is further enhanced by "lead acetate + zinc powder" to remove thallium again, producing thallium-cadmium-containing residue.

[0071] Step 2: The system automatically captures and identifies the thallium content data in the thallium-cadmium slag from Step 1. The thallium content is 2190 g / t, which is higher than the starting threshold for "direct refining". Following the system prompts, the system enters the "multi-stage thallium removal process in cadmium recovery": After the cadmium slag is oxidized in the stockpile, it undergoes thallium removal through electrolytic zinc anode mud and zinc powder two-stage replacement processes, producing thallium-rich liquid slag and sponge cadmium. According to the thallium content data of 147 g / t in the sponge cadmium in the MES, the sponge cadmium enters the "pyrometallurgical refining" setting. The thallium-rich liquid slag can be treated by fumigation and defluorination / chlorination to obtain high-fluorine / chlorine dust with a higher degree of thallium enrichment.

[0072] Step 3: Using the "Cadmium Recovery Process Target Optimization Module", the target optimization path for the cadmium recovery process in Step 2, "'Multi-stage thallium removal in cadmium recovery process' → 'Direct refining'", is determined to be "two-step optimization".

[0073] like Figure 2 As shown, the first step of optimization is the optimization of the "multi-stage thallium removal in cadmium recovery process" process objective;

[0074] Establish a set of decision variables: ;

[0075] in, The content of thallium-cadmium slag iron is % The lead content of thallium-cadmium slag is % The cadmium content of the thallium-cadmium slag is % The thallium content of thallium-cadmium slag; The zinc content of thallium-cadmium slag is % This refers to the amount of oxidant added, in tons of cadmium slag. This refers to the amount of lead salt added, in tons of zinc powder;

[0076] Furthermore, the MES and DCS acquire a dataset of 1000 decision variables and determine the lower and upper limits of 7 decision variables to form constraints.

[0077] Furthermore, an objective function is established using a convolutional neural network model (DNN) trained on a production dataset:

[0078] Cadmium recovery rate in sponge cadmium ;

[0079] System energy consumption ;

[0080] Zinc powder addition amount ;

[0081] Thallium enrichment ratio in liquid residue ;

[0082] Optimize the objective function: .

[0083] The optimal solution set is obtained by using the Fast Non-Dominated Sorting Genetic Algorithm (NSGA-II). The first 5 solutions are:

[0084]

[0085] This solution set corresponds to the optimal operating scheme set that includes efficient and economical processes for recovering cadmium from sponges and enriching for thallium removal.

[0086] Step Two Optimization: Optimization of the "Cadmium Sponge Refining" Target

[0087] Establish a set of decision variables: ,

[0088] Among them, setting The cadmium and iron content of the sponge is % The cadmium content of the sponge is % (%). The refining temperature is ℃; The distillation temperature is ℃. The condensation temperature is in °C. , where is the melting and casting temperature, in °C.

[0089] Furthermore, the MES and DCS acquire a dataset of 1000 decision variables and determine the lower and upper limits of 6 decision variables to form constraints.

[0090] Furthermore, an objective function is established using a convolutional neural network model (DNN) trained on a production dataset:

[0091] Cadmium recovery rate ;

[0092] System energy consumption ;

[0093] Sodium hydroxide addition amount ;

[0094] Optimize the objective function: .

[0095] The NSGA-II algorithm is continued to be used to solve the problem, and the optimal solution set is obtained. The first 5 solutions are:

[0096]

[0097] This solution set corresponds to the set of optimal operating schemes for the efficient and economical direct refining process of sponge cadmium.

[0098] Finally, it should be noted that the above is only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention (such as the selection of decision variables, training of deep learning models, establishment of objective functions, selection of optimization algorithms, application of various formulas, and the order of steps) without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent identification and optimization system for thallium enrichment and cadmium recovery in a hydrometallurgical zinc smelting process, characterized in that, The optimization system includes a thallium enrichment process intelligent optimization module based on incoming material identification and a cadmium recovery process target optimization module. The intelligent optimization module for thallium enrichment process based on incoming material identification includes: a multi-source mixed zinc concentrate incoming material identification unit, a thallium removal process determination unit, an automatic verification unit for process rationality, a thallium- and cadmium-containing slag identification unit, and a cadmium recovery process determination unit. The multi-source mixed zinc concentrate incoming material identification unit is used to automatically capture the thallium content data of mixed zinc concentrate in the associated enterprise manufacturing execution system (MES) and identify thallium content < 8 g / t, 8 g / t ≤ thallium content ≤ 15 g / t, or thallium content > 15 g / t; The thallium removal process determination unit is used to prompt the user to enter a specific thallium removal process mode based on the thallium content identification result of the incoming material: "thallium removal in purification process", "thallium removal in leaching control + thallium removal in purification process" or "thallium removal in leaching and purification processes in stages". The automatic verification unit for process rationality is used to further verify the accuracy and rationality of the prompts from the thallium removal process determination unit based on the actual thallium concentration data of the zinc leaching solution in the MES, according to whether the thallium concentration is <0.1mg / L or ≥0.1mg / L. The thallium-cadmium slag identification unit is used to automatically capture the thallium content data of the associated thallium-cadmium slag in the MES and identify whether the thallium content is <200g / t or ≥200g / t. The cadmium recovery process determination unit is used to identify the thallium content in thallium-containing cadmium slag and prompt the user to enter a specific cadmium recovery process mode: "direct refining" or "multi-stage thallium removal in cadmium recovery process". After entering "multi-stage thallium removal in cadmium recovery process", the user must then enter the "direct refining" process. The cadmium recovery process target optimization module includes a cadmium recovery optimization path determination unit, which is used to determine the two-step optimization path of incoming material entering either "direct refining target optimization" or "cadmium recovery process multi-stage thallium removal process target optimization → direct refining target optimization" based on the cadmium recovery process determined by the cadmium recovery process determination unit.

2. The intelligent identification and optimization system for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process according to claim 1, characterized in that, The critical start-up thresholds for thallium content of 8 g / t and 15 g / t are used to distinguish different thallium removal processes, and the critical start-up threshold for 200 g / t is used to distinguish different cadmium recovery processes.

3. A smart identification and optimization method for thallium enrichment and cadmium recovery in a hydrometallurgical zinc smelting process, characterized in that, The method is implemented based on the system described in claim 1 or 2, and includes the following steps: Step 1: Intelligent matching and execution of thallium removal processes based on the thallium content of the incoming material. The thallium content of multi-source zinc concentrate is identified by an intelligent optimization module based on incoming material recognition. The next thallium removal process is selected based on the obtained thallium content of the mixed zinc concentrate in the MES. When the thallium content of the mixed zinc concentrate is identified as <8g / t, the system prompts to directly enter the "purification process for thallium removal" and verifies the thallium concentration of the zinc leachate from the roasting dust and roasting sand: if the thallium concentration of the zinc leachate is <1.0mg / L, it is considered reasonable and the zinc leachate enters the "purification process for thallium removal". In this process, thallium removal is enhanced by "lead salt + zinc powder", producing thallium-cadmium-containing slag; if the thallium concentration of the zinc leachate is ≥1.0mg / L, it enters the "leaching and purification process for tiered thallium removal". In the leaching process, the zinc leachate is first subjected to permanganate or ozone oxidation precipitation to remove thallium, producing thallium-containing leaching slag and thallium-removed leachate. The thallium-removed leachate enters the purification process, where thallium removal is further enhanced by "lead salt + zinc powder", producing thallium-cadmium-containing slag. When the system identifies a thallium content of 8g / t ≤ 15g / t in the mixed zinc concentrate, it prompts the user to enter the "leaching control + purification process for thallium removal". The control method involves reducing the amount of roasting dust from the primary dust collector by 1 / 2 to 1 / 3, and mixing all the roasting dust from the secondary dust collector with the roasted sand for leaching. This staged leaching of roasting dust reduces the concentration of the zinc leachate. Then, the thallium concentration in the zinc leachate is verified: a thallium concentration < 1.0mg / L is considered reasonable, and the zinc leachate enters the next stage. The "Thallium Removal in Purification Process" involves using "lead salt + zinc powder" to enhance thallium removal during cadmium removal, producing thallium- and cadmium-containing slag. When the thallium concentration in the zinc leaching solution is ≥1.0 mg / L, it enters the "Step-by-Step Thallium Removal in Leaching and Purification Process." In the leaching process, the zinc leaching solution first undergoes permanganate or ozone oxidation precipitation to remove thallium, producing thallium-containing leaching slag and thallium-removed leaching solution. The thallium-removed leaching solution then enters the purification process, where "lead salt + zinc powder" is used to further enhance thallium removal, producing thallium- and cadmium-containing slag. When the thallium content in the mixed zinc concentrate is identified as >15g / t, the system prompts to enter the "thallium removal in the leaching and purification process" and verifies the thallium concentration of the zinc leaching solution from the roasting dust and roasting sand: if the thallium concentration in the leaching solution is <1.0mg / L, return to the "thallium removal in the purification process"; if the thallium concentration in the leaching solution is ≥1.0mg / L, operate according to the "thallium removal in the leaching and purification process" procedure, and remove thallium from the zinc leaching solution in the leaching and purification processes respectively, producing thallium-cadmium-containing slag; Step 2: Screening and Implementation of Cadmium Recovery Processes Based on the Characteristics of Thallium- and Cadmium-Containing Slag The next cadmium recovery process is selected based on the thallium content of the thallium-cadmium slag obtained from the MES: When the thallium content of the thallium-cadmium slag is identified as <200g / t, the system prompts that the thallium-cadmium slag should enter "direct refining" to produce refined cadmium through refining and vacuum distillation; When the thallium content of the thallium-cadmium slag is identified as ≥200g / t, the system prompts to enter the "multi-stage thallium removal process in cadmium recovery". Through multi-stage thallium removal, including thallium-cadmium slag stockpiling oxidation, acid leaching to make liquor, and zinc powder replacement, thallium-rich liquor-making slag and sponge cadmium are produced. The sponge cadmium enters the "direct refining" process. Step 3: Multi-objective intelligent optimization solution for cadmium recycling process The target optimization module for the cadmium recovery process selects the target optimization path for the next step of cadmium recovery based on the cadmium recovery process: When thallium-cadmium-containing slag enters the "direct refining" process, the target optimization process for this step is implemented: Establish a set of decision variables: ; Among them, setting The iron content of thallium-cadmium slag; The cadmium content of thallium-cadmium slag; This refers to the refining temperature; This refers to the distillation temperature; This is the condensation temperature; The melting and casting temperature; n≥6; Establish the objective function through deep learning training: Cadmium recovery rate ; System energy consumption ; Sodium hydroxide addition amount ; In the formula, CNN represents a deep learning model trained on a production dataset; For the objective function Solve the problem and obtain the optimal solution set P. The system will then display the message "Optimization complete". When thallium-cadmium-containing slag enters the "multi-stage thallium removal process in cadmium recovery → direct refining," a "two-step optimization" process is implemented: The first step in optimization is to optimize the multi-stage thallium removal process in the cadmium recovery process: Establish a set of decision variables: ; in, The iron content of thallium-cadmium slag; The lead content of thallium-cadmium slag; The cadmium content of thallium-cadmium slag; The thallium content of thallium-cadmium slag; The zinc content of thallium-cadmium slag; This refers to the amount of oxidant added; This refers to the amount of lead salt added; m≥7; Establish the objective function using a deep learning model: Cadmium recovery rate in sponge cadmium ; System energy consumption ; Zinc powder addition amount ; Thallium enrichment ratio in liquid residue ; In the formula, DNN represents a deep learning model trained on a production dataset; For the objective function Solve to obtain the set of optimal solutions. ; The second optimization step involves optimizing the cadmium sponge refining target: The cadmium sponge product obtained in the first optimization step is then compressed and refined, following the direct refining target optimization process to obtain the optimal solution set. The system then displayed a message: "Optimization complete." 4. The intelligent identification and optimization method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process according to claim 3, characterized in that, The deep learning models include convolutional neural networks and long short-term memory neural networks.

5. The intelligent identification and optimization method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process according to claim 3, characterized in that, In step 3, the constraints for directly refining the target optimization process are: s.t. ∈ [ , ], i =1,2,...,n; in, , These represent the lower and upper limits of the decision variables obtained by the MES and the distributed control system (DCS), respectively, which are determined by statistical analysis of more than 1,000 sets of historical datasets.

6. The intelligent identification and optimization method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process according to claim 3, characterized in that, In step 3, the constraints for optimizing the multi-stage thallium removal objective in the cadmium recovery process are: s.t. ∈ [ , ], i =1,2,...,m; in, , These represent the lower and upper limits of the decision variables obtained from MES and DCS, respectively, which are determined by statistical analysis of more than 1,000 historical datasets.

7. The intelligent identification and optimization method for thallium enrichment and cadmium recovery in the hydrometallurgical zinc smelting process according to claim 3, characterized in that, The objective function is solved using a fast non-dominated sorting genetic algorithm.

Images