Manufacturing energy-saving optimization method and system for smelting workshop

Abstract

The invention relates to the technical field of smelting workshop optimization scheduling, in particular to a manufacturing energy-saving optimization method and system for a smelting workshop. The method comprises the steps of full-process energy consumption data acquisition, energy consumption sensitivity quantitative analysis, dynamic optimization of a control strategy and waste heat recovery and energy scheduling. The energy consumption sensitivity quantitative analysis is used for accurately identifying key energy consumption influence factors; the dynamic optimization control strategy is used for realizing collaborative optimization of energy consumption and quality targets; waste heat recovery and energy scheduling recover waste heat in an efficient gradient mode through a temperature gradient. According to the method, through sensitivity quantitative analysis and dynamic optimization control, energy mass balance optimization in the smelting process is achieved, and smelting energy consumption is effectively reduced; and meanwhile, the comprehensive waste heat utilization rate is remarkably increased by constructing a temperature gradient efficient recovery full-chain heat energy conversion path.

Description

Technical Field

[0001] This application relates to the field of optimization scheduling technology for smelting workshops, and in particular to a manufacturing energy-saving optimization method and system for smelting workshops. Background Technology

[0002] The metallurgical and casting industries are facing the dual pressures of soaring energy costs and carbon emission limits. As a core unit of industrial energy consumption, the smelting workshop accounts for 40%-60% of a company's total energy consumption, and its energy efficiency directly impacts the company's sustainable development capabilities. However, current smelting systems still face significant technical bottlenecks in energy efficiency optimization: (1) Imbalance in energy and quality synergistic control: Traditional smelting processes rely on manual experience or single-target PID control, which makes it difficult to achieve synergistic optimization of energy consumption and metallurgical quality under dynamic working conditions; (2) Extensive utilization of waste heat resources: single-stage shell-and-tube heat exchangers are commonly used for waste heat recovery. Their heat recovery efficiency is low, and they can only utilize high-temperature flue gas above 400℃. The medium and low temperature heat energy, which accounts for 60% of the total waste heat, is directly discharged, resulting in a large amount of energy loss. Therefore, there is an urgent need to build a smelting energy-saving optimization system with multi-dimensional perception, energy quality balance, and deep utilization of waste heat. Summary of the Invention

[0003] To address the aforementioned problems in the existing technology, this invention provides a manufacturing energy-saving optimization method and system for smelting workshops. Through sensitivity quantitative analysis and dynamic optimization control, the energy and quality balance of the smelting process is optimized, effectively reducing smelting energy consumption. At the same time, by constructing a full-chain heat energy conversion path for efficient recovery of temperature gradients, the overall waste heat utilization rate is significantly improved.

[0004] In a first aspect, the present invention provides a manufacturing energy-saving optimization method for a smelting workshop, comprising the following steps: S1. Full-process energy consumption data acquisition: Data is collected through a deployed multi-dimensional sensor network, the collected data is imported into an energy-quality database, and hierarchical storage is implemented; S2. Energy consumption sensitivity quantitative analysis: A multi-dimensional factor library is constructed based on the energy-quality database. The Sobol index method is used to quantify the sensitivity of the multi-dimensional factor library to total energy consumption and quality assessment respectively. The primary key factors and secondary key factors are distinguished according to the output total energy consumption sensitivity or quality assessment sensitivity ranking. S3. Dynamic resource scheduling strategy: By optimizing the primary and secondary key factors, the smelting process parameters are dynamically adjusted and the energy consumption and quality targets are synergistically optimized. S4. Waste Heat Recovery and Energy Dispatch: A three-stage waste heat recovery system is adopted to efficiently recover smelting waste heat and convert waste heat energy into usable energy.

[0005] As a preferred embodiment of the present invention, the full-process energy consumption data acquisition is used to collect and structure the full-process energy consumption data of the smelting workshop, including: Multi-dimensional sensor network data acquisition: Planning detection points and collecting parameters for the smelting furnace body, raw material pretreatment area, cooling and forming area; constructing an energy-quality database; and importing the collected data into the energy-quality database. The energy-quality database is stored in a hierarchical manner: the energy-quality database includes a real-time layer, an aggregation layer, and an analysis layer. The real-time layer is used to store the data collected in real time; the aggregation layer is used to store the minute average / peak values ​​of the collected data; and the analysis layer is used to store statistical reports and correlation model data of the collected data.

[0006] As a preferred embodiment of the present invention, the energy consumption sensitivity quantitative analysis is used to identify key energy consumption influencing factors, including: Construct a multidimensional factor library based on the energy-quality database; The Sobol index method is used to quantify the sensitivity of the multidimensional factor library to total energy consumption and quality assessment, and thresholds for total energy consumption sensitivity and quality assessment sensitivity are set. Factors with the highest total energy consumption sensitivity or quality assessment sensitivity in the output are listed as primary key factors, and those with the highest sensitivity in the range of a%-b% are listed as secondary key factors; a < b; and both primary and secondary key factors need to be greater than the thresholds for total energy consumption sensitivity or quality assessment sensitivity.

[0007] As a preferred embodiment of the present invention, the Sobol exponent method includes: The Sobol index method is used to calculate the sensitivity using the following formula: ; in, This is a parameter affecting total energy consumption or a quality assessment parameter. For the first A multidimensional factor, For the first The expected values ​​of the influence parameters or quality assessment parameters of the multidimensional factors on total energy consumption. The total variance is the sum of all the factors at work. for Change caused variance for right Sensitivity.

[0008] As a preferred embodiment of the present invention, the dynamic resource scheduling strategy, which guides the priority of resource allocation and optimization strategies based on the results of energy consumption sensitivity quantification analysis, includes: The primary key factors include primary total energy consumption key factors and primary quality key factors, and the secondary key factors include secondary total energy consumption key factors and secondary quality key factors. For the key factors of total energy consumption at level 1, including smelting temperature and waste heat recovery, the control of smelting temperature is determined by electricity consumption or gas consumption; the waste heat recovery is detected by a distributed fiber optic thermometer. A correlation model is constructed for electricity consumption, gas consumption and waste heat recovery values, and a total energy consumption threshold is set. For the primary quality critical factors, including alloy composition content, deploy LIBS equipment to monitor composition deviations in real time and set alloy composition thresholds under qualified conditions. By deeply integrating the NSGA-III algorithm with the smelting process, the synergistic optimization of the primary total energy consumption key factor and the primary quality key factor is achieved. For key factors of secondary total energy consumption, including cooling water volume, an electromagnetic flowmeter is used to monitor the cooling water volume in real time, and combined with historical cooling water volume, the true value of alloy composition is dynamically corrected through a correction algorithm formula. For the secondary quality key factors, including feeding rate and feeding moisture content, a nonlinear dynamic optimization formula is used to construct a quantitative relationship between quality score and secondary quality key factors, and a quality score threshold is set so that the quality score is less than the quality score threshold.

[0009] As a preferred embodiment of the present invention, the NSGA-III algorithm includes the following steps: Quantifying the core objective function: Defining the expression for minimum total energy consumption: β is the calorific value coefficient. This represents the energy consumption value. This represents the gas consumption value. To recover waste heat, and to ensure the value is less than the total energy consumption threshold; alloy composition content expression: , For the first The composition of the alloy, The threshold value for alloy composition under acceptable conditions; Initialize population A: Initialize the population based on historical energy-quality database data, set the population size, and define individuals as two-dimensional process parameters including total energy consumption and alloy composition content; Digital twin simulation evaluation B: Input two-dimensional process parameters into the digital twin model to simulate the smelting process and output two-dimensional objective function values ​​of total energy consumption and alloy composition content; Non-dominated sorting C: If solution A is not inferior to solution B on all objectives and is better on at least one objective, then solution A dominates solution B. Individuals not dominated by any solution are classified into the first level, and the remaining solutions are recursively sorted. Reference point association and niche preservation D: Map the solution to a preset reference point. Solutions in the same reference point region are filtered by crowding distance. Densely distributed inferior solution reference points are removed, and sparsely distributed reference points are retained. Crossover mutation to generate offspring E: Randomly select two parent individuals, simulate binary crossover to generate offspring, and apply random perturbations to the offspring parameters; Merge parent and child generations F: Merge N parent individuals and N child individuals into 2N individuals, providing a richer candidate set for subsequent non-dominated sorting; Termination condition G: Set a maximum number of iterations to prevent infinite loops, and ensure that the Pareto front improvement is less than a threshold; Output Pareto front H: Outputs a set of final optimized nondominated solutions, reflecting the optimal trade-off between total energy consumption and quality.

[0010] As a preferred embodiment of the present invention, the correction algorithm formula includes: ; in, This represents the actual value of the cooling water volume. The amount of cooling water detected by the electromagnetic flowmeter. For correction factors, This represents the historical cooling water volume.

[0011] As a preferred embodiment of the present invention, the nonlinear dynamic optimization formula includes: ; in, To rate the quality, , and These are the weighting coefficients, For feeding rate, For the moisture content of the feed, The time decay factor, This refers to the melting temperature.

[0012] As a preferred embodiment of the present invention, the waste heat recovery and energy dispatch adopts a three-stage cascade efficient recovery of smelting waste heat, converting waste heat energy into usable energy, including: The three-stage waste heat stage includes a high-temperature section of a primary heat exchanger, a medium-temperature section of a secondary heat exchanger, and a low-temperature section of a tertiary heat storage tank. The high-temperature section of the primary heat exchanger provides recovered waste heat for preheating raw materials in the smelting furnace and for system energy compensation, thereby reducing energy consumption in the smelting furnace. The secondary heat exchanger uses titanium alloy corrugated plates for heating water in the medium-temperature section. The temperature is set to a comfortable temperature for the human body through an intelligent temperature control system, which meets the heating needs of the workshop in winter and saves coal consumption. The three-stage thermal storage tank uses MXene-based composite phase change material to heat the waste heat of the exhaust gas to amorphous salt to liquid state and store the thermal power. The three-stage waste heat cascade also includes a high-temperature section fault bypass mechanism for the first-stage heat exchanger and a second-stage heat exchanger protection mechanism, which are used to protect the three-stage waste heat cascade so that it can continuously and efficiently recover waste heat. The high-temperature section fault bypass mechanism of the primary heat exchanger includes a non-stop guarantee mechanism. When the primary heat exchanger is blocked, leaked, or the control system fails, the waste heat recovery is maintained through the bypass system to avoid the smelting furnace shutdown. Secondary heat exchanger protection mechanism: This includes an emergency cooling tower. When the heating water temperature of the secondary heat exchanger rises abnormally, the emergency cooling tower is activated to protect the heat exchange equipment and pipeline network.

[0013] Secondly, the present invention also provides a manufacturing energy-saving optimization system for a smelting workshop, which implements the manufacturing energy-saving optimization method for a smelting workshop as described above, including a data acquisition module, a sensitivity analysis module, a dynamic resource scheduling module, and a waste heat recovery module. The data acquisition module achieves data acquisition of the entire process in the smelting workshop by deploying a multi-dimensional sensor network; The sensitivity analysis module uses the Sobol index method to quantify the sensitivity of the multidimensional factor library to total energy consumption and quality assessment, and distinguishes between primary key factors and secondary key factors based on the output total energy consumption sensitivity or quality assessment sensitivity ranking. The dynamic resource scheduling module dynamically adjusts the primary and secondary key factors and coordinates the optimization of energy consumption and quality targets. The waste heat recovery module adopts a three-stage waste heat cascade utilization method to efficiently recover waste heat from the smelting process and convert waste heat energy into usable energy.

[0014] The beneficial effects of this invention include: This invention achieves accurate collection and structured storage of energy consumption data throughout the entire smelting workshop process through the deployment of a multi-dimensional sensor network and the construction of an energy database. A multi-dimensional factor library is constructed based on the energy-quality database, and the Sobol index method is used to quantify the sensitivity of the multi-dimensional factor library to total energy consumption and quality assessment. Primary and secondary key factors are distinguished based on the output ranking of total energy consumption sensitivity or quality assessment sensitivity. Furthermore, a dynamic resource scheduling strategy is implemented for the primary and secondary key factors to achieve dynamic adjustment of smelting process parameters and synergistic optimization of energy consumption and quality objectives.

[0015] Meanwhile, this invention innovatively designs a three-stage gradient waste heat recovery system. By constructing a full-chain thermal energy conversion path of "high-temperature waste heat recovery - medium-temperature heating water supply - low-temperature phase change heat storage", the overall waste heat utilization rate is significantly improved. In addition, the system has a built-in fault tolerance mechanism to effectively ensure the continuity of waste heat recovery and avoid secondary energy waste caused by local failures in traditional systems. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic flowchart of an energy-saving optimization method for smelting workshop manufacturing provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the NSGA-III algorithm in this invention; Figure 3 This is a schematic diagram of the structure of the energy-saving optimization system for manufacturing in a smelting workshop provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are also within the scope of protection of this application.

[0019] The preferred embodiments of the present invention will be further described below with reference to the accompanying drawings; Please see Figure 1-3 This embodiment provides a manufacturing energy-saving optimization method for a smelting workshop, including the following steps: S1. Full-process energy consumption data acquisition: By deploying a multi-dimensional sensor network to build an energy database, accurate acquisition and structured storage of full-process energy consumption data in the smelting workshop are achieved, including: Multi-dimensional sensor network deployment: Monitoring points are planned and parameters are collected for key areas such as the smelting furnace body, raw material pretreatment zone, and cooling and forming zone. An energy database is then constructed, and the monitored data is imported into the energy database. For example, an infrared array thermometer is used to detect the temperature of the smelting furnace; a distributed fiber optic thermometer is used to detect the recovered waste heat; a microwave moisture sensor is used to detect the moisture content of the raw materials in the pretreatment zone; an electromagnetic flowmeter is used to detect the cooling water volume in the cooling and forming zone; and a spectrometer is used to detect the alloy composition content of the castings. Energy database framework division: The energy database is divided into a real-time layer, an aggregation layer, and an analysis layer. The real-time layer is used to store the collected parameters in real time; the aggregation layer is used to store the minute average / peak values ​​of the collected parameters; and the analysis layer is used to store statistical reports and related model data of the collected parameters.

[0020] S2. Energy consumption sensitivity quantitative analysis: A multi-dimensional factor library is constructed based on the energy-quality database. The Sobol index method is used to quantify the sensitivity of the multi-dimensional factor library to total energy consumption and quality assessment respectively. The primary key factors and secondary key factors are distinguished according to the output total energy consumption sensitivity or quality assessment sensitivity ranking. The sensitivity quantification analysis, used to accurately identify key energy consumption influencing factors, includes: A multidimensional factor library is constructed based on parameter information from the energy database; Construct total energy consumption parameters and quality assessment parameters to output sensitivity quantification analysis results; The Sobol index method is used to quantify the sensitivity of the multidimensional factor library to total energy consumption parameters and quality assessment parameters, respectively. Thresholds for total energy consumption sensitivity and quality assessment sensitivity are set. Factors with the highest total energy consumption sensitivity or quality assessment sensitivity in the output are listed as primary key factors, and those with the highest sensitivity in the range of a%-b% are listed as secondary key factors; a < b; and both primary and secondary key factors need to be greater than the threshold for total energy consumption sensitivity or quality assessment sensitivity. The core formula of the Sobol exponent method is: ; in, This is a parameter affecting total energy consumption or a quality assessment parameter. For the first A multidimensional factor, For the first The expected values ​​of the influence parameters or quality assessment parameters of the multidimensional factors on total energy consumption. The total variance is the sum of all the factors at work. for Change caused variance for right Sensitivity.

[0021] S3. Dynamic resource scheduling strategy: By optimizing the primary and secondary key factors, the smelting process parameters are dynamically adjusted and the energy consumption and quality targets are synergistically optimized. The dynamic resource scheduling strategy, based on the sensitivity threshold results, guides the priority of resource allocation and optimization strategies, including: The primary key factors include primary total energy consumption key factors and primary quality key factors, and the secondary key factors include secondary total energy consumption key factors and secondary quality key factors. For the key factors of total energy consumption at level 1, including smelting temperature and waste heat recovery, the control of smelting temperature is determined by electricity consumption or gas consumption; the waste heat recovery is detected by a distributed fiber optic thermometer. A correlation model is constructed for electricity consumption, gas consumption and waste heat recovery values, and a total energy consumption threshold is set. For the primary quality critical factors, including alloy composition content, deploy LIBS equipment to monitor composition deviations in real time and set alloy composition thresholds under qualified conditions. By deeply integrating the NSGA-III algorithm with the smelting process, the synergistic optimization of the primary total energy consumption key factor and the primary quality key factor is achieved. For key factors of secondary total energy consumption, including cooling water volume, an electromagnetic flowmeter is used to monitor the cooling water volume in real time, and combined with historical cooling water volume, the true value of alloy composition is dynamically corrected through a correction algorithm formula. The correction algorithm formula is as follows: ; Among them, the This represents the actual value of the cooling water volume. The amount of cooling water detected by the electromagnetic flowmeter. For correction factors, Historical cooling water volume; For secondary quality key factors, including feeding rate and feeding moisture content, a nonlinear dynamic optimization formula is used to construct a quantitative relationship between quality score and secondary quality key factors, and a quality score threshold is set so that the quality score is less than the quality score threshold. The nonlinear dynamic optimization formula is: ; in, To rate the quality, , and These are the weighting coefficients, For feeding rate, For the moisture content of the feed, The time decay factor, This refers to the melting temperature. The NSGA-III algorithm includes the following steps: Quantization of the core objective function: Expression for minimizing total energy consumption: β is the calorific value coefficient. This represents the energy consumption value. This represents the gas consumption value. To recover waste heat; alloy composition content expression: , for The composition of the alloy The alloy threshold under qualified conditions; Initialize population A: Initialize the population based on historical data, set the population size, and set the individual representatives as two-dimensional process parameters of total energy consumption and alloy composition content; Digital twin simulation evaluation B: Input two-dimensional process parameters into the digital twin model to simulate the smelting process and output two-dimensional objective function values ​​of total energy consumption and alloy composition content; Non-dominated sorting C: If the total energy consumption and alloy composition of solution A are both no worse than those of solution B, and at least one of the solutions A (total energy consumption or alloy composition) is better than that of solution B, then solution A dominates solution B; among them, the first layer of individuals that are not dominated by any solution is called the Pareto front, and the second layer of solutions are dominated by the first layer of solutions, but they do not dominate each other. Reference point association and niche retention D: The solution is mapped to a preset reference point. Solutions in the same reference point region are filtered according to the crowding distance. Densely distributed inferior solution reference points are removed, and sparsely distributed reference points are retained. If two solutions are close to each other, and the total energy consumption of solutions X and Y differs by 0.1 degrees and the alloy composition content differs by 0.01%, the denser solution can be deleted. If two solutions are sparsely distributed, they are retained as key exploration points. The association of the reference point with niche preservation D can avoid the occurrence of local optima; Crossover mutation to generate offspring E: Randomly select two parent individuals, simulate binary crossover to generate offspring, and apply random perturbations to the offspring parameters; Merge parent and child generations F: Merge N parent individuals and N child individuals into 2N individuals, providing a richer candidate set for subsequent non-dominated sorting; Termination condition G: Set a maximum number of iterations to prevent infinite loops, and ensure that the Pareto front improvement is less than a threshold; Output Pareto front H: Outputs a set of final optimized nondominated solutions, reflecting the optimal trade-off between energy consumption and quality.

[0022] S4. Waste Heat Recovery and Energy Dispatch: A three-stage waste heat recovery system is adopted to efficiently recover waste heat from the smelting process, converting waste heat energy into usable energy and significantly improving overall energy efficiency, including: The three-stage waste heat treatment system includes a high-temperature section of a primary heat exchanger, a medium-temperature section of a secondary heat exchanger, and a low-temperature section of a tertiary heat storage tank; the temperature range of the high-temperature section of the primary heat exchanger is 400~1100°C; the temperature range of the medium-temperature section of the secondary heat exchanger is 200~400°C; and the temperature range of the low-temperature section of the tertiary heat storage tank is 100~200°C. The high-temperature section of the primary heat exchanger provides recovered waste heat for preheating raw materials in the smelting furnace and for system energy compensation, thereby reducing energy consumption in the smelting furnace. The secondary heat exchanger uses titanium alloy corrugated plates for heating water in the medium-temperature section. The temperature is set to a comfortable temperature for the human body through an intelligent temperature control system, which meets the heating needs of the workshop in winter and saves coal consumption. The three-stage thermal storage tank uses MXene-based composite phase change materials to heat amorphous salt to liquid state using waste heat from exhaust gas, thus storing thermal power.

[0023] The three-stage waste heat cascade also includes a high-temperature section fault bypass mechanism for the first-stage heat exchanger and a second-stage heat exchanger protection mechanism, which are used to protect the three-stage waste heat cascade so that it can continuously and efficiently recover waste heat. The high-temperature section fault bypass mechanism of the primary heat exchanger includes a non-stop guarantee mechanism. When the primary heat exchanger is blocked, leaked, or the control system fails, the waste heat recovery is maintained through the bypass system to avoid the smelting furnace shutdown. Secondary heat exchanger protection mechanism: This includes an emergency cooling tower. When the heating water temperature of the secondary heat exchanger rises abnormally, the emergency cooling tower is activated to protect the heat exchange equipment and pipeline network.

[0024] This invention also provides a manufacturing energy-saving optimization system for a smelting workshop, which implements the aforementioned manufacturing energy-saving optimization method for a smelting workshop, including a data acquisition module, a sensitivity analysis module, a dynamic resource scheduling module, and a waste heat recovery module. The data acquisition module achieves data acquisition of the entire process in the smelting workshop by deploying a multi-dimensional sensor network; The sensitivity analysis module uses the Sobol index method to quantify the sensitivity of the multidimensional factor library to total energy consumption and quality assessment, and distinguishes between primary key factors and secondary key factors based on the output total energy consumption sensitivity or quality assessment sensitivity ranking. The dynamic resource scheduling module dynamically adjusts the primary and secondary key factors and coordinates the optimization of energy consumption and quality targets. The waste heat recovery module adopts a three-stage waste heat cascade utilization method to efficiently recover waste heat from the smelting process and convert waste heat energy into usable energy.

[0025] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing energy conservation in a smelting workshop, characterized in that: Includes the following steps: S1. Full-process energy consumption data acquisition: Data is collected through a deployed multi-dimensional sensor network, the collected data is imported into an energy-quality database, and hierarchical storage is implemented; S2. Energy consumption sensitivity quantitative analysis: A multi-dimensional factor library is constructed based on the energy-quality database. The Sobol index method is used to quantify the sensitivity of the multi-dimensional factor library to total energy consumption and quality assessment respectively. The primary key factors and secondary key factors are distinguished according to the output total energy consumption sensitivity or quality assessment sensitivity ranking. S3. Dynamic resource scheduling strategy: By optimizing the primary and secondary key factors, the smelting process parameters are dynamically adjusted and the energy consumption and quality targets are synergistically optimized. S4. Waste Heat Recovery and Energy Dispatch: A three-stage waste heat recovery system is adopted to efficiently recover smelting waste heat and convert waste heat energy into usable energy.

2. The energy-saving optimization method for smelting workshops according to claim 1, characterized in that, The full-process energy consumption data collection is used to collect energy consumption data for the entire smelting workshop process and store it in a structured manner, including: Multi-dimensional sensor network data acquisition: Planning detection points and collecting parameters for the smelting furnace body, raw material pretreatment area, cooling and forming area; constructing an energy-quality database; and importing the collected data into the energy-quality database. The energy-quality database is stored in a hierarchical manner: the energy-quality database includes a real-time layer, an aggregation layer, and an analysis layer. The real-time layer is used to store the data collected in real time; the aggregation layer is used to store the minute average / peak values ​​of the collected data; and the analysis layer is used to store statistical reports and correlation model data of the collected data.

3. The energy-saving optimization method for smelting workshops according to claim 1, characterized in that, The energy consumption sensitivity quantitative analysis is used to identify key energy consumption influencing factors, including: Construct a multidimensional factor library based on the energy-quality database; The Sobol index method is used to quantify the sensitivity of a multidimensional factor library to total energy consumption and quality assessment, respectively. Thresholds for total energy consumption sensitivity and quality assessment sensitivity are set, and the output total energy consumption sensitivity or quality assessment sensitivity is ranked among the top. The factor is listed as a first-level key factor, ranking first in sensitivity. Listed as secondary key factors; among which Furthermore, both the primary and secondary key factors need to be greater than the total energy consumption sensitivity threshold or the quality assessment sensitivity threshold.

4. The energy-saving optimization method for smelting workshops according to claim 3, characterized in that, The Sobol exponent method includes: The Sobol index method is used to calculate the sensitivity using the following formula: ； in, This refers to parameters affecting total energy consumption or parameters for quality assessment. For the first A multidimensional factor, For the first The expected values ​​of the influence parameters or quality assessment parameters of the multidimensional factors on total energy consumption. The total variance is the sum of all the factors at work. for Change caused variance for right Sensitivity.

5. The energy-saving optimization method for smelting workshops according to claim 1, characterized in that, The dynamic resource scheduling strategy, based on the energy consumption sensitivity quantitative analysis results, guides the priority of resource allocation and optimization strategies, including: The primary key factors include primary total energy consumption key factors and primary quality key factors, and the secondary key factors include secondary total energy consumption key factors and secondary quality key factors. For the key factors of total energy consumption at level 1, including smelting temperature and waste heat recovery, the control of smelting temperature is determined by electricity consumption or gas consumption; the waste heat recovery is detected by a distributed fiber optic thermometer. A correlation model is constructed for electricity consumption, gas consumption and waste heat recovery values, and a total energy consumption threshold is set. For the primary quality critical factors, including alloy composition content, deploy LIBS equipment to monitor composition deviations in real time and set alloy composition thresholds under qualified conditions. By deeply integrating the NSGA-III algorithm with the smelting process, the synergistic optimization of the primary total energy consumption key factor and the primary quality key factor is achieved. For key factors of secondary total energy consumption, including cooling water volume, an electromagnetic flowmeter is used to monitor the cooling water volume in real time, and combined with historical cooling water volume, the true value of alloy composition is dynamically corrected through a correction algorithm formula. For the secondary quality key factors, including feeding rate and feeding moisture content, a nonlinear dynamic optimization formula is used to construct a quantitative relationship between quality score and secondary quality key factors, and a quality score threshold is set so that the quality score is less than the quality score threshold.

6. The energy-saving optimization method for smelting workshops according to claim 5, characterized in that, The NSGA-III algorithm includes the following steps: Quantifying the core objective function: Defining the expression for minimum total energy consumption: β is the calorific value coefficient. This represents the energy consumption value. This represents the gas consumption value. To recover waste heat, and to ensure the value is less than the total energy consumption threshold; alloy composition content expression: , For the first The composition of the alloy The threshold value for alloy composition under acceptable conditions; Initialize population A: Initialize the population based on historical energy-quality database data, set the population size, and define individuals as two-dimensional process parameters including total energy consumption and alloy composition content; Digital twin simulation evaluation B: Input two-dimensional process parameters into the digital twin model to simulate the smelting process and output two-dimensional objective function values ​​of total energy consumption and alloy composition content; Non-dominated sorting C: If solution A is not inferior to solution B on all objectives and is better on at least one objective, then solution A dominates solution B. Individuals not dominated by any solution are classified into the first level, and the remaining solutions are recursively sorted. Reference point association and niche preservation D: Map the solution to a preset reference point. Solutions in the same reference point region are filtered by crowding distance. Densely distributed inferior solution reference points are removed, and sparsely distributed reference points are retained. Crossover mutation to generate offspring E: Randomly select two parent individuals, simulate binary crossover to generate offspring, and apply random perturbations to the offspring parameters; Merge parent and child generations F: Merge N parent individuals and N child individuals into 2N individuals, providing a richer candidate set for subsequent non-dominated sorting; Termination condition G: Set a maximum number of iterations to prevent infinite loops, and ensure that the Pareto front improvement is less than a threshold; Output Pareto front H: Outputs a set of final optimized nondominated solutions, reflecting the optimal trade-off between total energy consumption and quality.

7. The energy-saving optimization method for smelting workshops according to claim 6, characterized in that, The correction algorithm formula includes: ； in, This represents the actual value of the cooling water volume. The amount of cooling water detected by the electromagnetic flowmeter. For correction factors, This represents the historical cooling water volume.

8. The energy-saving optimization method for smelting workshops according to claim 7, characterized in that, The nonlinear dynamic optimization formula includes: ； in, To rate the quality, , and These are the weighting coefficients, For feeding rate, For the moisture content of the feed, The time decay factor, This refers to the melting temperature.

9. The energy-saving optimization method for smelting workshops according to claim 8, characterized in that, The waste heat recovery and energy dispatch adopts a three-stage cascade efficient recovery of smelting waste heat, converting waste heat energy into usable energy, including: The three-stage waste heat stage includes a high-temperature section of a primary heat exchanger, a medium-temperature section of a secondary heat exchanger, and a low-temperature section of a tertiary heat storage tank. The high-temperature section of the primary heat exchanger provides recovered waste heat for preheating raw materials in the smelting furnace and for system energy compensation, thereby reducing energy consumption in the smelting furnace. The secondary heat exchanger uses titanium alloy corrugated plates for heating water in the medium-temperature section. The temperature is set to a comfortable temperature for the human body through an intelligent temperature control system, which meets the heating needs of the workshop in winter and saves coal consumption. The three-stage thermal storage tank uses MXene-based composite phase change material to heat the waste heat of the exhaust gas to amorphous salt to liquid state and store the thermal power. The three-stage waste heat cascade also includes a high-temperature section fault bypass mechanism for the first-stage heat exchanger and a second-stage heat exchanger protection mechanism, which are used to protect the three-stage waste heat cascade so that it can continuously and efficiently recover waste heat. The high-temperature section fault bypass mechanism of the primary heat exchanger includes a non-stop guarantee mechanism. When the primary heat exchanger is blocked, leaked, or the control system fails, the waste heat recovery is maintained through the bypass system to avoid the smelting furnace shutdown. Secondary heat exchanger protection mechanism: This includes an emergency cooling tower. When the heating water temperature of the secondary heat exchanger rises abnormally, the emergency cooling tower is activated to protect the heat exchange equipment and pipeline network.

10. A manufacturing energy-saving optimization system for a smelting workshop, comprising executing a manufacturing energy-saving optimization method for a smelting workshop as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, a sensitivity analysis module, a dynamic resource scheduling module, and a waste heat recovery module. The data acquisition module achieves data acquisition of the entire process in the smelting workshop by deploying a multi-dimensional sensor network; The sensitivity analysis module uses the Sobol index method to quantify the sensitivity of the multidimensional factor library to total energy consumption and quality assessment, and distinguishes between primary key factors and secondary key factors based on the output total energy consumption sensitivity or quality assessment sensitivity ranking. The dynamic resource scheduling module dynamically adjusts the primary and secondary key factors and coordinates the optimization of energy consumption and quality targets. The waste heat recovery module adopts a three-stage waste heat cascade utilization method to efficiently recover waste heat from the smelting process and convert waste heat energy into usable energy.

Images