A method and system for heat feedback regulation of a copper smelting process

Abstract

The application discloses a copper smelting process heat feedback regulation method and system, aiming at solving the problems of low heat control precision, poor hysteresis and difficulty in considering multiple element process constraints. The application constructs a double-layer closed-loop control architecture composed of a dynamic control layer and an optimization decision layer: in the dynamic control layer, the required heat adjustment amount is calculated according to the real-time deviation between the target heat and the actual heat of the smelting process; in the optimization decision layer, the ratio of each ore is adjusted under the premise of meeting the preset multiple element process constraints and other conditions with the heat adjustment amount as the guide. The application deeply cooperates dynamic adjustment and multivariable constraint optimization, can realize accurate, rapid and non-overshoot control of the smelting process heat, ensure that the key parameters strictly meet the process requirements, significantly improve the self-heating level and energy efficiency of copper smelting, reduce harmful element emissions, and has high industrial application value.

Description

Technical Field

[0001] This invention relates to the field of copper smelting process control technology, specifically to a method and system for heat feedback regulation in copper smelting processes, applicable to precise heat control and ore ratio optimization in various copper smelting processes such as oxygen-enriched bottom blowing smelting. Background Technology

[0002] Copper smelting is a core high-temperature metallurgical process in copper smelting production, aiming to convert copper concentrate into copper matte and complete gangue separation. The precision of heat control in this process directly affects energy efficiency, product quality stability, equipment lifespan, and environmental emission standards. Ideally, copper smelting operations should strive to achieve "self-heating smelting," that is, maintaining heat balance with the ore's own heat of reaction, minimizing dependence on external fuels. Simultaneously, it is necessary to ensure that the content of key elements and process parameters strictly comply with established process constraints to prevent problems such as abnormal slag formation, substandard copper matte grade, or excessive emissions of harmful elements.

[0003] However, the copper smelting process is characterized by strong nonlinearity, large time delay, and frequent fluctuations in raw material composition, posing a severe challenge to heat control. Specifically, changes in the phase composition and elemental content of the ore fed into the furnace significantly affect the heat released by the reaction; these heat changes, in turn, affect the reaction rate and elemental distribution. Furthermore, the smelting process must also consider the upper and lower limits of the content of multiple elements, further increasing the difficulty of control.

[0004] Conventional control strategies typically separate heat balance regulation from element content constraints. On the one hand, they rely on operational experience to dynamically compensate for operational variables such as oxygen concentration and feed rate to track the heat setpoint. On the other hand, they indirectly meet constraints such as copper matte grade, slag copper content, and sulfide emissions through offline testing and manual adjustments. However, in actual operation, there is a strong mutual constraint and linkage between heat dynamics and multi-element distribution. Temperature deviations not only change the reaction equilibrium but also drive the redistribution of impurities such as arsenic, antimony, and bismuth between the copper matte and slag phases. Local thermal fluctuations can easily trigger the breach of composition constraint boundaries. When raw materials fluctuate frequently and operating conditions shift over a wide range, this separate control structure, which adjusts heat first and then tests composition, reveals inherent defects such as lag in response and insufficient global coordination, making it difficult to achieve consistency between heat regulation targets and multi-element constraint satisfaction on the same time scale.

[0005] Existing methods for controlling heat in copper smelting suffer from a disconnect between dynamic adjustment and global optimization, making it difficult to simultaneously achieve precise heat tracking and coordinated control of multiple element contents. This hinders the further development of copper smelting processes towards higher efficiency, stability, and environmental friendliness. Therefore, there is an urgent need for a control method that can deeply integrate real-time dynamic adjustment with multivariate constraint optimization. Summary of the Invention

[0006] To address the aforementioned problems, this invention provides a method, namely, an intelligent control method and system capable of achieving precise heat control and synergistic optimization of multi-element content in the copper smelting process.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for heat feedback regulation in a copper smelting process, specifically comprising the following steps: S1. Based on the phase composition of copper concentrate, calculate the heat release per unit mass and the current heat of each mineral type, and establish a two-layer closed-loop mechanism for heat feedback regulation that includes a dynamic control layer and an optimization decision layer. S2. In the dynamic control layer, a parameter adaptive PID controller is used to calculate the required heat adjustment amount based on the real-time deviation between the target heat and the actual heat in the smelting process. Furthermore, S2 specifically includes the following steps: S2.1 Calculate the real-time heat deviation; Let the target heat of the smelting process be... Q target The actual heat is Q actual The real-time heat deviation is defined as: ; S2.2 Based on the absolute value of the real-time heat deviation and the preset error threshold, the PID control parameters are optimized using a segmented parameter adjustment strategy, and the PID control parameters are output. The PID control parameters include: proportional coefficient. Differential coefficients and integral coefficient ; The specific optimization strategy for segmented parameter adjustment is as follows: Based on real-time heat deviation The absolute value of the error is optimized using a piecewise parameter adjustment strategy based on the magnitude of the error. (1) When At that time, increase the proportionality coefficient Reduce differential coefficients Integral coefficient Set to zero or keep the small value; (2) When When, reduce the scaling factor Gradually introduce integral coefficients Increase the differential coefficient ; (3) When At the same time, maintain or slightly increase the proportionality coefficient. Increase the integral coefficient Maintain appropriate differential coefficients ; Among them, the first threshold Set as Second threshold Set as .

[0008] S2.3. Based on the real-time heat deviation and PID control parameters, the control output is calculated using a discrete PID control law to obtain the heat adjustment amount; The heat adjustment amount serves as the input for the desired heat change in the optimization decision layer.

[0009] In each sampling period, the controller adjusts the real-time thermal deviation. The dynamic adjustment ratio coefficient of its changing trend Differential coefficients and integral coefficient This allows the controller characteristics to adapt to the operating state in real time. Its discrete PID control law can be expressed as: in, u(k) For the first k The control output at each sampling time, where T is the sampling period. e(k) This represents the real-time heat deviation of the system at the current moment. k For sampling time index, This is the cumulative sum of deviations from the initial time to the current time.

[0010] S3. In the optimization decision layer, guided by the heat adjustment amount, a cost function is constructed consisting of a weighted sum of heat balance penalty, total mass fixed penalty, element content exceeding limit penalty and mass non-negative penalty. With the mass adjustment amount of each mineral type as the variable, under the condition of satisfying the preset multi-element process constraints, the adaptive differential evolution algorithm is used to solve the problem and obtain the adjusted proportion of each mineral type. The heat adjustment amount calculated by the dynamic control layer in step S2 is used as the expected heat change input in the cost function of the optimization decision layer; the optimization decision layer adjusts the mineral ratio so that the actual achievable heat change caused by the ratio change is as close as possible to the heat adjustment amount.

[0011] The heat adjustment amount is specifically used as the guide: the heat adjustment amount calculated by the dynamic control layer in S2 is used as the input of the expected heat change in the cost function of the optimization decision layer; the optimization decision layer adjusts the mineral ratio so that the actual achievable heat change caused by the ratio change is as close as possible to the heat adjustment amount.

[0012] Furthermore, S3 specifically includes the following steps: S3.1 Construct a cost function based on the heat adjustment amount; The cost function J (ΔM The following penalty terms constitute a weighted sum to unify heat balance, fixed mass, element content constraints, and non-negativity mass constraints into a single optimization framework for solving the problem: (1) Heat balance penalty term: The actual achievable heat change and the expected heat adjustment amount Δ Q The square of the difference; (2) Total mass fixed penalty: the square of the difference between the optimized total material mass and the preset fixed total mass; (3) Penalty for exceeding the limit of element content: the weighted sum of the squares of the portion of the mass percentage of each constrained element that exceeds its preset upper or lower limit; (4) Non-negative quality penalty: Penalty for each mineral type whose quality is negative after optimization.

[0013] The aforementioned penalty terms are uniformly quantified in the objective function, and can be flexibly adapted to different process requirements by adjusting the weight coefficients.

[0014] The cost function J( M The calculation expression for ) is as follows: In the formula, To preset the weighting coefficient of the heat balance penalty term, As a heat balance penalty term, The weighting coefficient for the fixed penalty term of total mass. A fixed penalty item for total mass. The weighting coefficient for the penalty term for exceeding the element content limit. This is a penalty item for exceeding the limit for element content. The weighting coefficient for the non-negative quality penalty term. This is a non-negative penalty for quality. S3.2. Adaptive Differential Evolution Algorithm (ADE) is used to adjust the quality of each mineral type by Δ. M Cost function with variables J (Δ M The optimal solution is obtained by performing a search and finding the quality adjustment amount of each mineral type corresponding to the optimal solution, thus obtaining the adjusted mineral type ratio. The adaptive differential evolution algorithm balances global search capability and local convergence speed by dynamically adjusting the scaling factor and crossover probability. The specific steps are as follows: S3.2.1 ADE Algorithm Parameter Initialization and Population Initialization; Within the feasible solution space determined by the multi-element process constraints and the fixed total material mass condition, a population of size NP is randomly generated, with each individual... x i,g =( x i,1,g, x i,2,g ,..., x i,D,g ) represents a D-dimensional solution vector, where g For evolutionary generations, g ∈{0,1,2,...,G max}, where D is the dimension of the decision variable. The maximum number of generations; i To assign an individual number to a member of the population. i ∈{1,2,...,NP}; j For the dimension index of the solution vector, j ∈{1,2,...,D}; x i,j,g Indicates the first g The first generation of the population i The first individual j Each dimension component value.

[0015] S3.2.2, Parameter adaptive adjustment; During the g-th generation of evolution, the scaling factor F and crossover probability CR are updated for each individual: (a) The scaling factor F decays exponentially from its initial maximum value F max Gradually decrease to the minimum value F min Specific scaling factor update formula for: Among them, F max F min Let F be the upper and lower bounds, respectively, where F is the scaling factor. λ is a decay coefficient between 0 and 1 (0 < λ < 1), and g is the current iteration number.

[0016] (b) The crossover probability (CR) is adaptively adjusted based on the historical improvement of an individual's fitness, so that individuals with significant fitness improvement receive a larger CR to enhance the retention of superior genes, while individuals with stagnant improvement receive a smaller CR to increase variation diversity and escape local optima.

[0017] S3.2.3, Mutation and Crossover Operations; Mutation: For each target vector x i,g Generate a mutation vector v i,g+1 The most commonly used DE / rand / 1 mutation strategy is: v i , g+1 =x r1 , g + F ( x r2 , g x r3,g ) in, r 1 ,r 2 ,r 3 represents a unique index randomly selected from {1,2,...,NP}, and none of these indices are equal to the target vector. i Therefore, the population size NP must be at least 4.

[0018] Crossover: To increase the diversity of the population, crossover is performed on the target vector. x i ,g and mutation vector v i,g+1 Perform cross operations between them to generate test vectors. u i , g+1 .

[0019] u i , g+1 =( u i,1,g+1 , u i , 2 , g+1 ,..., u i,D,g+1 ) The core formula for crossover is: in, rand(j) This represents a random number that follows a uniform distribution on the interval [0,1]. rnbr(i) ∈{1,2,...,D} represents a randomly selected dimension index, and this constraint ensures that the experimental vector... u i,g+1 At least from the mutation vector v i,g+1 Inherit parameters in one dimension to avoid the experimental vector and the target vector being completely identical; CR ∈[0,1] represents the crossover probability.

[0020] S3.2.4, Select Operation; Selection: After mutation and crossover operations, the differential evolution algorithm selects the trial vector and the target vector in the current population according to a greedy criterion. x i,g The comparisons are made, and the best individuals are retained for the next generation of the population. For the minimization optimization problem, the mathematical expression of the selection criterion is: in, f (·) represents the fitness function. That is, if the fitness of the experimental vector is better, the original target vector is replaced in the next generation; otherwise, the original target vector is retained.

[0021] S3.2.5 Termination conditions and optimal solution output yield the adjusted proportions of each mineral type; Repeat steps S3.2.2 to S3.2.4 until the number of generations reaches [number missing]. The individual with the smallest cost function value is selected from the final population. The corresponding mineral quality adjustment amount is the optimal solution that satisfies the heat balance and multi-element process constraints, thus obtaining the adjusted mineral ratio scheme.

[0022] S4. The copper smelting system is regulated using the adjusted ratio of each mineral, and the actual heat and element content after regulation are collected. The actual heat is fed back to the dynamic control layer in real time to update the deviation and maintain the heat closed-loop regulation. The actual element content is fed back to the optimization decision layer, which is used to verify the satisfaction of multi-element process constraints and as a reference for the next round of ratio optimization iteration, thereby realizing the two-layer closed-loop regulation of the whole process. Furthermore, the actual achievable heat change is adjusted according to the mass adjustment amount Δ for each mineral type. m Its heat release per unit mass Δ q Calculate the sum of the products.

[0023] Furthermore, a positive heat adjustment indicates that heat needs to be increased, while a negative heat adjustment indicates that heat needs to be decreased.

[0024] Furthermore, the multi-element process constraints include mass percentage constraint ranges for copper (Cu), iron (Fe), sulfur (S), silicon (Si), lead (Pb), and arsenic (As).

[0025] Furthermore, the dynamic control layer uses a parameter self-tuning adaptive PID controller to calculate the heat adjustment amount; the optimization decision layer uses an adaptive differential evolution algorithm to solve for the adjusted mineral ratio.

[0026] The present invention also provides a heat feedback regulation system for a copper smelting process, used to implement the above method, characterized in that the system comprises: The data acquisition module is used to obtain data on the actual heat, actual element content, and current ore ratio during the copper smelting process. The dynamic control module is used to calculate and output the heat adjustment amount based on the real-time deviation between the preset target heat and the actual heat. The optimization decision module is used to solve and output the adjusted mineral ratio scheme based on the heat adjustment amount, under the premise of meeting the preset multi-element process constraints and the total material mass being fixed. The execution and feedback module is used to send the adjusted ore ratio to the copper smelting system for regulation, and to collect the actual heat and element content data after regulation and feed them back to the dynamic control module and the optimization decision module respectively; wherein, the actual heat is used to update the deviation input of the dynamic control module, and the actual element content is used to verify the satisfaction of multi-element process constraints and as a reference for the next round of solution by the optimization decision module.

[0027] Furthermore, the present invention also provides a heat feedback regulation and control system for a copper smelting process, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0028] Furthermore, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0029] Compared with the prior art, the present invention provides a heat feedback regulation and control method and system for copper smelting process, which has the following beneficial effects: (1) Based on the phase composition of copper concentrate, the present invention accurately calculates the heat release per unit mass and actual furnace temperature of each mineral, breaking through the limitations of the existing rough estimation; through a double-layer closed-loop mechanism, the dynamic control layer calculates the heat deviation in real time and outputs the adjustment amount, optimizes the decision layer to quickly adjust the mineral ratio, and the data closed-loop iteration significantly improves the adjustment accuracy and response speed, avoids furnace temperature fluctuations, and ensures continuous and stable production.

[0030] (2) The optimization decision layer of the present invention is guided by the amount of heat adjustment, strictly meets the process constraints of multiple elements such as Cu, Fe, and S, and constrains the element content to exceed the limit through the cost function penalty term, thereby avoiding the problem of heat adjustment and process disconnection, improving the purity of copper products, reducing the energy consumption of impurity removal, reducing the emission of harmful impurities, and meeting environmental protection requirements.

[0031] (3) This invention achieves data acquisition, automatic heat control, ratio optimization and closed-loop feedback through the collaboration of various modules. No manual intervention is required, avoiding operational errors, reducing labor intensity and risk. It improves the reliability and intelligence level of regulation through iterative optimization, and adapts to the automation needs of modern copper smelting.

[0032] (4) The system of the present invention can be adapted to smelting production lines of different scales and different raw material types. It can achieve precise heat control by adjusting parameters, without the need for large-scale modification of existing equipment. The modification cost is low, which facilitates industry promotion and promotes the technological upgrading and sustainable development of the copper smelting industry. Attached Figure Description

[0033] Figure 1 This is a block diagram illustrating the heat feedback regulation principle of the present invention; Figure 2 This is a flowchart of the two-layer closed-loop system of the APID-ADE fusion strategy of the present invention; Figure 3 This is a comparison of the heat response curves of the present invention with those of other control methods; Figure 4 This is a comparison chart of the quality before and after the optimization of the mineral ratio in this invention; Figure 5 This is a comparison chart of elements before and after the optimization of the mineral composition in this invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail and completely below with reference to the accompanying drawings and specific embodiments. It should be noted that the described embodiments are merely some embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0035] This embodiment details the specific application and verification process of the method of the present invention in heat feedback regulation during copper smelting. In the dual-layer closed-loop mechanism of heat feedback regulation in this embodiment, the dynamic control layer adopts an adaptive PID controller, and the optimization decision layer adopts an adaptive differential evolution algorithm (ADE). The principle diagram of heat feedback regulation is shown below. Figure 1 .

[0036] (1) Experimental parameter settings; (1.1) Basic parameters of the mineral type; Five typical copper concentrates (ore A to ore E) were selected as research objects. The elemental contents of Cu, Fe, S, Si, Pb and As and the heat release per unit mass of each ore are shown in Table 1. The total feed amount was fixed at 911,771 kg, and the initial ore ratio was ore A 13.71%, ore B 27.75%, ore C 15.52%, ore D 22.68%, and ore E 20.35%.

[0037] Table 1: Elemental Content and Heat Release per Unit Mass of Minerals Table 2: Phase Composition of Minerals (1.2) Adaptive PID parameters; Initial scaling factor =1.5, initial integration coefficient =0.08, initial differential coefficient =0.3; Error threshold , The sampling period is T = 1 min.

[0038] (1.3) ADE algorithm parameters: Population size NP=50, maximum number of iterations =500; Scaling factor Fmin=0.4, Fmax=0.9, Attenuation coefficient λ=0.98; Crossover probability =0.7, =0.95.

[0039] (1.4) Cost Function J( M It consists of four weighted penalty terms, each corresponding to a different optimization objective: Where γ, λ1, λ2, and λ3 are weighting coefficients, with γ = 10.3, λ1 = 10.5, λ2 = 10.2, and λ3 = 10.6. The specific meanings of each component are as follows: Heat balance : in, This refers to the actual achievable heat change; m i Indicates the first i The quality of the ore; q i For the first i The heat release per unit mass of the mineral (kJ / kg).

[0040] Total mass fixed : in M target This is a fixed total feed rate. This penalty ensures that the optimized total quality is strictly equal to the target value required by the process, avoiding production instability caused by quality fluctuations.

[0041] Element content : in: For the set of elements that need to be constrained; C j For the first j The mass percentage of each element in the mixed ore; w j This represents the weight coefficient of the corresponding element; f j (C j ) =max(0, C j min - C j, C j -C j max ) 2 This is a penalty function for exceeding the limit for element content.

[0042] Non-negative mass : This penalty ensures that the quality of all minerals remains non-negative. m i +Δ m i ≥ 0, i =1,…,5.

[0043] (1.5) Elemental process constraints: Cu (19±0.5)%, Fe (25.5±0.5)%, S (29±0.5)%, Si (6±0.5)%, Pb (0.45±0.1)%, As (0.4±0.1)%.

[0044] (2) Implementation steps; Step S1: Calculate the heat release per unit mass and current heat of each mineral type based on the phase composition of copper concentrate, and establish a two-layer closed-loop mechanism for heat feedback regulation that includes a dynamic control layer and an optimization decision layer.

[0045] Based on the chemical composition of each ore type in Tables 1 and 2 and the principle of copper smelting heat balance calculation, the total heat generated by the material fed into the furnace at the current feed rate is calculated to be 2.239 × 10^8 kJ, which is 2.482 × 10^6 kJ short of the target heat required to achieve self-heating smelting. The flow chart of the dual-layer closed-loop control system in this embodiment is as follows: Figure 2 As shown, it includes a dynamic control layer (adaptive PID) and an optimization decision layer (ADE algorithm).

[0046] Step S2: In the dynamic control layer, the required heat adjustment amount is calculated based on the real-time deviation between the target heat and the actual heat in the melting process.

[0047] Initial thermal deviation detected = 2.482 × 10^6 kJ, its absolute value is greater than E1. Based on the segmented adjustment strategy optimization of the adaptive PID controller: ... Upgraded to 2.2 to improve response speed. Set to 0 to prevent integral saturation. The value is lowered to 0.15 to avoid derivative sensitivity. The desired heat adjustment Δ is then output after PID calculation. Q (0).

[0048] Step S3: In the optimization decision layer, guided by the heat adjustment amount, a cost function with the quality adjustment amount of each mineral as a variable is constructed and solved. Under the premise of satisfying the preset multi-element process constraints and other conditions, the adjusted proportion of each mineral is obtained.

[0049] The optimization decision module receives the heat adjustment instruction and uses it as the input of the expected heat change in the cost function. The ADE algorithm initializes the mineral quality adjustment population and adaptively adjusts the scaling factor F according to the number of generations g and the crossover probability CR according to the individual fitness. The algorithm iteratively optimizes through mutation, crossover, and selection operations to make the actual achievable heat change caused by the change in ratio as close as possible to the heat adjustment amount. After 500 iterations, the algorithm converges to the optimal solution in 287 generations, obtaining the mineral quality adjustment amounts: ore A increases to 142575.50 kg, ore D decreases to 189195.50 kg, and the masses of ores B, C, and E remain unchanged.

[0050] Step S4: The copper smelting system is regulated using the adjusted ore ratio, and the actual heat and element content after regulation are collected. The actual heat is fed back to the dynamic control layer in real time to update the deviation and maintain the heat closed-loop regulation. The actual element content is fed back to the optimization decision layer, which is used to verify the satisfaction of multi-element process constraints and as a reference for the next round of ratio optimization iteration, thereby realizing the two-layer closed-loop regulation of the whole process.

[0051] The optimized proportioning correction plan described above will be issued and implemented. For example... Figure 4 As shown, the optimized ore mix ratio is: Ore A 15.64%, Ore B 27.75%, Ore C 15.52%, Ore D 20.75%, and Ore E 20.35%, with a total feed amount remaining at 911,771 kg. The actual total calorific value after implementation was measured to be 2.264 × 10⁻⁶ kg. 8The accuracy is measured in kJ with an error of only 0.10 kJ. Simultaneously, the content of key elements is detected: Cu 19.32%, Fe 25.55%, S 29.5%, Si 5.55%, Pb 0.45%, and As 0.36%. All element contents strictly fall within the preset process constraint range. Actual heat data is fed back to the dynamic control layer to update the deviation, while actual element content data is fed back to the optimization decision layer as a reference for the next round of optimization, forming a complete two-layer closed-loop iteration.

[0052] (3) Comparative experiment; The APID-ADE method of this invention is compared with traditional PID, fuzzy PID, and particle swarm optimization PID (PSO-PID) methods. The core performance indicators of each method are shown in Table 3. As can be seen from Table 3, the method of this invention is far superior to other methods in terms of steady-state error, overshoot, settling time, and IAE value, achieving precise heat control and overshoot-free smooth response.

[0053] Table 3: Quantitative Comparison Table of Key Performance Indicators From Table 3 and Figure 3 It can be seen that this invention (APID-ADE) achieves smooth adjustment with almost no overshoot (0.03%) while maintaining a fast response (Tr=4 min), with the shortest adjustment time (15 min) and the best overall control efficiency (IAE=200.1%, a reduction of 44.4% compared to PID). Comparison of element content before and after optimization is shown below. Figure 5 As shown, Pb decreased from 1.16% to 0.50%, and As decreased from 0.58% to 0.36%, both falling within the constraint range.

[0054] In summary, this invention effectively solves the problems of poor heat control accuracy, response lag, and difficulty in balancing elemental constraints in the copper smelting process by constructing a dual-layer heat feedback regulation strategy that integrates APID and ADE. It achieves coordinated and precise control of heat and element content, providing an advanced technical path for efficient, stable, and low-carbon smelting of complex copper raw materials.

[0055] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for heat feedback regulation in a copper smelting process, characterized in that, Includes the following steps: S1. Based on the phase composition of copper concentrate, calculate the heat release per unit mass and the current heat of each mineral type, and establish a two-layer closed-loop mechanism for heat feedback regulation that includes a dynamic control layer and an optimization decision layer. S2. In the dynamic control layer, a parameter adaptive PID controller is used to calculate the required heat adjustment amount based on the real-time deviation between the target heat and the actual heat in the smelting process. S3. In the optimization decision layer, guided by the heat adjustment amount, a cost function is constructed consisting of a weighted sum of heat balance penalty, total mass fixed penalty, element content exceeding limit penalty and mass non-negative penalty. With the mass adjustment amount of each mineral type as the variable, under the condition of satisfying the preset multi-element process constraints, the adaptive differential evolution algorithm is used to solve the problem and obtain the adjusted proportion of each mineral type. S4. The copper smelting system is regulated using the adjusted proportions of various minerals, and the actual heat and element content after regulation are collected. The actual heat is fed back to the dynamic control layer in real time to update the deviation and maintain the closed-loop heat regulation. The actual element content is fed back to the optimization decision layer, which is used to verify the satisfaction of multi-element process constraints and as a reference for the next round of proportion optimization iteration. This achieves a two-layer closed-loop regulation of the entire process and completes the heat feedback regulation method for the copper smelting process.

2. The method for heat feedback regulation in a copper smelting process according to claim 1, characterized in that, S2 specifically includes the following steps: S2.1 Calculate the real-time heat deviation; Let the target heat of the smelting process be... Q target The actual heat is Q actual Define real-time heat deviation for: ； S2.2 Based on the absolute value of the real-time heat deviation and the preset error threshold, a segmented parameter adjustment strategy is adopted to input PID control parameters; The PID control parameters include: proportional coefficient. Differential coefficients and integral coefficient ; S2.

3. Based on the real-time heat deviation and PID control parameters, the control output is calculated using a discrete PID control law to obtain the heat adjustment amount.

3. The method for heat feedback regulation in a copper smelting process according to claim 1, characterized in that, Specifically, in S3, the heat adjustment amount is used as the guide: the heat adjustment amount calculated by the dynamic control layer in S2 is used as the input of the expected heat change in the cost function of the optimization decision layer; the optimization decision layer adjusts the mineral ratio so that the actual achievable heat change caused by the ratio change is as close as possible to the heat adjustment amount.

4. The method for heat feedback regulation in a copper smelting process according to claim 1, characterized in that, S3 specifically includes the following steps: S3.1 Construct a cost function based on the heat adjustment amount; When the heat adjustment amount is positive, it means that heat needs to be increased; when it is negative, it means that heat needs to be decreased. The cost function J (Δ M The following penalty terms constitute a weighted sum to unify heat balance, fixed mass, element content constraints, and non-negativity mass constraints into a single optimization framework for solving the problem: (1) Heat balance penalty term: The actual achievable heat change and the expected heat adjustment amount Δ Q The square of the difference; (2) Total mass fixed penalty: the square of the difference between the optimized total material mass and the preset fixed total mass; (3) Penalty for exceeding the limit of element content: the weighted sum of the squares of the portion of the mass percentage of each constrained element that exceeds its preset upper or lower limit; (4) Non-negative quality penalty: Penalty for each mineral type whose quality is negative after optimization; S3.

2. Adaptive Differential Evolution Algorithm (ADE) is used to adjust the quality of each mineral type by Δ. M Cost function with variables J (Δ M The optimal solution is obtained by performing optimization and outputting the quality adjustment amount of each mineral type corresponding to the optimal solution, thus obtaining the adjusted mineral type ratio.

5. The method for heat feedback regulation in a copper smelting process according to claim 4, characterized in that, The actual achievable heat change is based on the mass adjustment Δ of each mineral type. m Its heat release per unit mass Δ q Calculate the sum of the products.

6. The method for heat feedback regulation in a copper smelting process according to claim 1, characterized in that, In S4, the multi-element process constraints include mass percentage constraint ranges for copper (Cu), iron (Fe), sulfur (S), silicon (Si), lead (Pb), and arsenic (As).

7. The method for heat feedback regulation in a copper smelting process according to claim 4, characterized in that, The specific steps for solving S3.2 using the adaptive differential evolution algorithm are as follows: S3.2.1 Initialization of ADE algorithm parameters and population: Within the feasible solution space determined by the multi-element process constraints and the fixed total material mass, an initial population of NP size is randomly generated. S3.2.2, Parameter adaptive adjustment; During the g-th generation of evolution, the scaling factor F and crossover probability CR are updated for each individual. Scaling factor update formula for: Among them, F max F min Let F be the upper and lower bounds, respectively, where F is the scaling factor. The decay coefficient is a value between 0 and 1, and g is the current iteration number. S3.2.

3. Use the updated parameters to perform mutation, crossover, and selection operations, iterate to the maximum number of generations, and output the adjusted ratio of each mineral type.

8. A heat feedback regulation system for a copper smelting process, used to implement the method according to any one of claims 1 to 6, characterized in that, The system includes: The data acquisition module is used to obtain data on the actual heat, actual element content, and current ore ratio during the copper smelting process. The dynamic control module is used to calculate and output the heat adjustment amount based on the real-time deviation between the preset target heat and the actual heat. The optimization decision module is used to solve and output the adjusted mineral ratio scheme based on the heat adjustment amount, under the premise of meeting the preset multi-element process constraints and the total material mass being fixed. The execution and feedback module is used to send the adjusted ore ratio to the copper smelting system for regulation, and to collect the actual heat and element content data after regulation and feed them back to the dynamic control module and the optimization decision module respectively; wherein, the actual heat is used to update the deviation input of the dynamic control module, and the actual element content is used to verify the satisfaction of multi-element process constraints and as a reference for the next round of solution by the optimization decision module.

9. A heat feedback regulation and control system for a copper smelting process, characterized in that, include: Memory, which stores computer programs; A processor for executing the computer program to implement the method of any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

Images