Steel industry green short process operation management and control system based on digital twinning

Abstract

The present application relates to the production technology field of steel industry, more specifically, it relates to a kind of steel industry green short process operation control system based on digital twinning, for solving the problems that existing technology cannot accurately optimize key production parameters, increase energy waste and material loss, reduce overall resource utilization, cannot provide scientific production scheduling scheme, reduce the stability and reliability of production;The present application adjusts key production parameters using optimization algorithm through production optimization scheduling module, to minimize energy consumption and maximize output, adapt to different needs by flexibly adjusting weight coefficient, ensure the continuity and stability of production, reduce energy waste and material loss by accurately optimizing production parameters, improve overall resource utilization, the best production scheduling scheme generated provides scientific basis for management, enhances the stability and reliability of production.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of steel industry production, more particularly, to a green short process operation control system for steel industry based on digital twinning. BACKGROUND

[0002] As an important pillar of the national economy, the steel industry has the characteristics of long process, high energy consumption and large pollution in the production process. With the enhancement of environmental awareness and the deepening of the concept of sustainable development, the steel industry is facing tremendous transformation pressure. Green short process steel production technology has become an important direction for the transformation and upgrading of the steel industry due to its short process, low energy consumption and small pollution.

[0003] The patent application with publication number CN117473770A discloses a steel equipment intelligent management system based on digital twinning information, which includes a digital twinning module, a data acquisition module, a data processing module and an analysis control module. The digital twinning module is set up to create a digital twinning model based on steel equipment data and corresponding production process data. The digital twinning model is synchronized with the space and time of the steel equipment and the corresponding production process. The analysis control module can compare, analyze and output control instructions based on real-time data and the digital twinning model in the digital twinning module, and update the digital twinning model. This achieves real-time monitoring and predictive maintenance of steel equipment, improves equipment operation efficiency and reduces operating costs.

[0004] However, the above-mentioned reference patent creates a digital twinning model synchronized with the steel equipment and the process through a digital twinning module, and sets up an analysis control module for real-time monitoring, analysis and instruction output to achieve efficient operation of the equipment and cost reduction. However, it cannot accurately determine the production state and cannot distinguish the actual operating conditions, which may cause unplanned downtime and reduce overall production efficiency. Precise simulation of operating scenarios and prediction of production and energy consumption cannot achieve early identification of problems and optimization of production plans, reducing production efficiency. In addition, it cannot accurately optimize key production parameters, increasing energy waste and material loss, reducing overall resource utilization, and cannot provide a scientific production scheduling scheme, reducing production stability and reliability.

[0005] Therefore, we propose a green short process operation control system for steel industry based on digital twinning to solve the above problems. SUMMARY

[0006] The present application aims to provide a steel industry green short process operation management and control system based on digital twinning, which solves the problem that the prior art cannot accurately judge the production state, cannot distinguish the actual running condition, is easy to cause unplanned shutdown, reduces the overall production efficiency, accurately simulates the running scene and predicts the yield and energy consumption, cannot realize early identification of problems and optimization of production plan, reduces the production efficiency, at the same time cannot accurately optimize the key production parameters, increases energy waste and material loss, reduces the overall resource utilization, cannot provide a scientific production scheduling scheme, and reduces the stability and reliability of production.

[0007] The purpose of the present application is achieved by the following technical solutions:

[0008] A steel industry green short process operation management and control system based on digital twinning applied to an operation management and control platform, comprising:

[0009] A data acquisition and preprocessing module for real-time acquisition of full-link data in the steel production process and preprocessing operation of the acquired full-link data;

[0010] An abnormal early warning module for real-time monitoring of key operating parameters in the steel production process and triggering an abnormal early warning immediately when an abnormal condition is detected;

[0011] A digital twinning prediction module for constructing an electric furnace steelmaking digital twinning model, simulating the running state under different scenes through the model, and predicting the molten steel yield and power consumption based on the full-link data;

[0012] A production optimization and scheduling module for optimizing the production parameters by using an optimization algorithm based on the prediction results of the electric furnace steelmaking digital twinning model and the feedback information provided by the abnormal early warning module, and generating an optimal production scheduling scheme.

[0013] As a preferred embodiment of the present application, the process of the abnormal early warning module for real-time monitoring of the key operating parameters in the steel production process comprises:

[0014] Obtaining the key operating parameters in the scrap steel smelting production process, the key operating parameters including the internal temperature of the electric furnace, the smelting process pressure, the power input, and the key equipment vibration acceleration, generating a monitoring period, and equally dividing the monitoring period into multiple monitoring time periods;

[0015] Obtaining the internal temperature imbalance value of the electric furnace in the monitoring period, and marking the internal temperature imbalance value of the electric furnace as DWS, the internal temperature imbalance value of the electric furnace representing the ratio between the part where the internal temperature variation difference of the electric furnace in each monitoring time period is greater than a preset internal temperature variation difference threshold value and the internal temperature variation difference of the electric furnace, and the internal temperature variation difference of the electric furnace representing the difference between the maximum value and the minimum value of the internal temperature of the electric furnace.

[0016] As a preferred embodiment of the present application, the process of the abnormal early warning module for early warning of abnormal conditions in the scrap steel smelting production process comprises:

[0017] The method for calculating the internal temperature imbalance value DWS of the electric furnace can also be used to obtain the smelting process pressure imbalance value DYS, the power input imbalance value DSS, and the key equipment vibration acceleration imbalance value SZS. The production operation state evaluation coefficient SYP is calculated by the following formula: ;

[0018] wherein w1, w2, w3, and w4 are all preset proportional factor coefficients, w1> w2> w3> w4> 0. The production operation state evaluation coefficient SYP is compared with a preset first production operation state evaluation coefficient threshold and a preset second production operation state evaluation coefficient threshold, wherein the preset first production operation state evaluation coefficient threshold is less than the preset second production operation state evaluation coefficient threshold:

[0019] If the production operation state evaluation coefficient SYP is less than the preset first production operation state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is excellent;

[0020] If the production operation state evaluation coefficient SYP is greater than or equal to the preset first production operation state evaluation coefficient threshold, and the production operation state evaluation coefficient SYP is less than the preset second production operation state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is normal;

[0021] If the production operation state evaluation coefficient SYP is greater than or equal to the preset second production operation state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is abnormal, an alarm signal is generated and sent to the operation control platform;

[0022] After receiving the alarm signal, the operation control platform immediately notifies the management personnel and takes corresponding measures for processing.

[0023] As a preferred embodiment of the present application, the process of the digital twin prediction module for constructing the electric furnace steelmaking digital twin model comprises:

[0024] The full-link data in the scrap steel smelting process is obtained, including the key material component ratio, the smelting process temperature, the pressure, the power input, the oxygen supply, the molten steel yield, and the environmental temperature.

[0025] The specific steps of constructing the electric furnace steelmaking twin model by digital twin technology are as follows:

[0026] A physical model is constructed based on the furnace charge ratio, process parameters and equipment operation principle in the electric furnace steelmaking process, the thermodynamics, heat transfer and chemical reaction kinetics equations are used to simulate the temperature change, composition change and energy transfer in the smelting process, the electric furnace smelting model is used to simulate the electrode current, voltage, power and the melting, oxidation and decarburization process of the furnace charge, and the three-dimensional modeling technology and sensor data are used to construct a three-dimensional visualization model of the electric furnace steelmaking process.

[0027] As a preferred embodiment of the present application, the process of the digital twin prediction module simulating the running state under different scenarios includes:

[0028] Real-time full-link data is obtained, real-time data synchronization technology is used to synchronize the real-time full-link data to the electric furnace steelmaking twin model, and the running state of the electric furnace steelmaking process is displayed in real time through the electric furnace steelmaking twin model;

[0029] The running state under different scenarios is simulated through the electric furnace steelmaking twin model:

[0030] Define the simulation scenario, and divide the simulation scenario into normal smelting scenario, abnormal smelting scenario, different furnace charge ratio scenario and different production target scenario;

[0031] The thermodynamics, heat transfer and chemical reaction kinetics model is used to simulate the temperature change, composition change and energy transfer in the electric furnace steelmaking process, the smelting process under different scenarios is simulated by adjusting the model parameters, and the internal temperature distribution, composition distribution and energy transfer of the electric furnace under different scenarios are displayed in real time through the three-dimensional visualization model.

[0032] As a preferred embodiment of the present application, the process of the digital twin prediction model processing full-link data to facilitate the prediction of molten steel yield and power consumption includes:

[0033] Obtain historical full-link data in the scrap steel smelting production process, generate a collection period, and divide the collection period into multiple collection time periods;

[0034] Obtain the key material composition proportion change rate in the multiple collection time periods, the key material composition proportion change rate represents the ratio between the key material composition proportion change amount and the corresponding time period length, and the set A of the key material composition proportion change rate is constructed in this way, and the mean value of the difference between the maximum subset and the minimum subset in set A is recorded as the key material composition proportion change rate difference CBC;

[0035] The method for calculating the key material composition proportion change rate difference CBC can also be used to obtain the smelting process temperature change rate difference DWC, the power input amount change rate difference DSC and the oxygen supply amount change rate difference YGC;

[0036] obtaining the proportion of the key material component in a plurality of collection periods, and calculating the arithmetic mean of the plurality of obtained proportions of the key material component, and denoting the arithmetic mean of the plurality of proportions of the key material component as the average proportion of the key material component PCB;

[0037] The average smelting process temperature PDW, the average power input amount PDS and the average oxygen supply amount PYG can be obtained in the same way as the average proportion of the key material component PCB.

[0038] As a preferred embodiment of the present application, the process in which the digital twin prediction model predicts the molten steel yield and the power consumption based on the whole-link data comprises:

[0039] The yield-energy consumption prediction matrix CNY is constructed by combining the key material component proportion change rate difference CBC, the smelting process temperature change rate difference DWC, the power input amount change rate difference DSC, the oxygen supply amount change rate difference YGC, the average proportion of the key material component PCB, the average smelting process temperature PDW, the average power input amount PDS and the average oxygen supply amount PYG, the yield-energy consumption prediction matrix CNY is used as the input of the machine learning model, the molten steel yield GCC and the power consumption DXL in a future period of time corresponding to each group of yield-energy consumption prediction matrix CNY are used as the output of the machine learning model, the molten steel yield GCC and the power consumption DXL in the future period of time are used as the prediction target, the sum of the prediction errors of all training data is minimized as the training target, the machine learning model is trained until the sum of the prediction errors converges, and the yield-energy consumption prediction model is obtained;

[0040] Real-time whole-link data in the scrap steel smelting production process is obtained, which is converted into the corresponding yield-energy consumption prediction matrix CNY and input into the yield-energy consumption prediction model, and the real-time molten steel yield GCC and the real-time power consumption DXL in a future period of time are obtained through the yield-energy consumption prediction model.

[0041] As a preferred embodiment of the present application, the process in which the production optimization scheduling module optimizes the production parameters by using the optimization algorithm comprises:

[0042] The prediction results of the electric furnace steelmaking digital twin model and the feedback information provided by the abnormal early warning module are obtained, the prediction results are the real-time molten steel yield GCC and the real-time power consumption DXL in a future period of time, and the feedback information is that the operation of the scrap steel smelting production is excellent, normal or abnormal.

[0043] The production parameters are defined as a vector z = [T, O, P, M] T , wherein T is the smelting process temperature, O is the oxygen supply amount, P is the power input amount, and M is the proportion of the key material component;

[0044] The production parameters are optimized using an optimization algorithm. The optimization objective is to minimize power consumption and maximize steel output while satisfying production constraints. The objective function is:

[0045] f(z) = v1·DXL(z) + v2·(GCC) target -GCC(z)), where DXL(z) is the input production parameter z, the power consumption output by the electric arc furnace steelmaking digital twin model, and GCC(z) is the input production parameter z, the steel output by the electric arc furnace steelmaking digital twin model. target For the target steel production output, v1 and v2 are both weighting coefficients;

[0046] Constraints: .

[0047] In a preferred embodiment of the present invention, the process by which the production optimization scheduling module solves the objective function and generates the optimal production scheduling scheme includes:

[0048] The steps to solve the objective function are as follows:

[0049] S1: Initialization: Randomly generate a set of production parameters z;

[0050] S2: Calculate the objective function: Calculate f(z) based on the mathematical model;

[0051] S3: Evaluate constraints: Check whether the production parameters meet the constraints;

[0052] S4: Update parameters: Update process parameters using an optimization algorithm;

[0053] S5: Iteration: Repeat steps S2-S4 until the convergence condition or the maximum number of iterations is reached;

[0054] The optimal production parameter vector z is obtained by solving the objective function. * =[T * O * P * M * ] T According to the optimal production parameter vector z * Generate the optimal production scheduling plan, which includes:

[0055] Adjust the current smelting process temperature T to the optimal smelting process temperature T. * ;

[0056] Adjust the current oxygen supply level O to the optimal oxygen supply level O. * ;

[0057] Adjust the current power input P to the optimal power input P. * ;

[0058] adjusting the current key material component ratio M to the optimal key material component ratio M * .

[0059] Compared with the prior art, the present application has the advantages that:

[0060] In the present application, the key operating parameters are monitored in real time by the abnormality early warning module, the abnormal conditions are quickly detected and responded, the production operation state evaluation coefficient SYP is calculated using the imbalance values of various parameters, the production state is accurately judged, the actual operating conditions are distinguished through hierarchical thresholds, the accuracy of management decisions is improved, the unplanned downtime is reduced, and the overall production efficiency is improved;

[0061] In the present application, the digital twin prediction module is used to construct a digital twin model of the electric furnace steelmaking, accurately simulate the operating scenario and predict the yield and energy consumption, realize early identification of problems and optimization of production plan, generate a prediction matrix through historical data analysis and machine learning, optimize production parameters to reduce waste, improve efficiency and reduce cost, support preventive maintenance, reduce fault risk and prolong equipment life;

[0062] In the present application, the production optimization scheduling module uses an optimization algorithm to adjust key production parameters to minimize power consumption and maximize yield, adjusts the weight coefficient flexibly to adapt to different needs, ensures the continuity and stability of production, reduces energy waste and material loss through accurate optimization of production parameters, improves overall resource utilization, and generates the best production scheduling scheme to provide a scientific basis for management and enhance the stability and reliability of production. BRIEF DESCRIPTION OF DRAWINGS

[0063] Figure 1 is a system block diagram of example one in the present application;

[0064] Figure 2 is a system block diagram of example two in the present application;

[0065] Figure 3 is a logic flow diagram of example one in the present application;

[0066] Figure 4 is a flowchart of the solving steps of the objective function in the present application. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application; obviously, the described embodiments are only part of the embodiments of the present application, but not all the embodiments; based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0068] Embodiment one: as Figure 1 and Figure 3 The present application proposes a steel industry green short process operation control system based on digital twinning, applied to an operation control platform, comprising:

[0069] A data acquisition and preprocessing module is configured to acquire real-time full-link data in the steel production process and perform preprocessing operations on the acquired full-link data, including but not limited to data cleaning, denoising, outlier processing, and data conversion.

[0070] The data acquisition and preprocessing module effectively improves the accuracy and reliability of the data through data cleaning, denoising, outlier processing, and data conversion, providing solid decision support for management. High-quality data not only helps to optimize production processes, identify and improve production bottlenecks, but also reduces the burden on subsequent data analysis and storage systems, improving overall operational efficiency.

[0071] An abnormality early warning module is configured to monitor key operating parameters in the steel production process in real time and trigger an abnormality early warning when an abnormality is detected.

[0072] The process of the abnormality early warning module monitoring key operating parameters in the steel production process in real time includes:

[0073] Obtain the key operating parameters in the scrap steel smelting production process, including the temperature, pressure, power input, and key equipment vibration acceleration of the electric furnace smelting process, generate a monitoring period, and divide the monitoring period into multiple monitoring time intervals.

[0074] Obtain the electric furnace internal temperature imbalance value within the monitoring period and mark the electric furnace internal temperature imbalance value as DWS. The electric furnace internal temperature imbalance value represents the ratio between the part where the electric furnace internal temperature variation difference is greater than the preset electric furnace internal temperature variation difference threshold and the electric furnace internal temperature variation difference in each monitoring time interval. The electric furnace internal temperature variation difference represents the difference between the maximum and minimum values of the electric furnace internal temperature.

[0075] The process of the abnormality early warning module warning about abnormal conditions in the scrap steel smelting production process includes:

[0076] The method used to calculate the electric furnace internal temperature imbalance value DWS can also be used to obtain the smelting process pressure imbalance value DYS, power input imbalance value DSS, and key equipment vibration acceleration imbalance value SZS. The production operation state assessment coefficient SYP is calculated using the following formula:

[0077] ;

[0078] wherein w1, w2, w3 and w4 are preset proportion factor coefficients, w1 > w2 > w3 > w4 > 0, the production operation state evaluation coefficient SYP is compared with a preset first production operation state evaluation coefficient threshold and a preset second production operation state evaluation coefficient threshold, the preset first production operation state evaluation coefficient threshold is less than the preset second production operation state evaluation coefficient threshold:

[0079] If the production operation state evaluation coefficient SYP is less than the preset first production operation state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is excellent;

[0080] If the production operation state evaluation coefficient SYP is greater than or equal to the preset first running state evaluation coefficient threshold, and the production operation state evaluation coefficient SYP is less than the preset second running state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is normal;

[0081] If the production operation state evaluation coefficient SYP is greater than or equal to the preset second running state evaluation coefficient threshold, it indicates that the running condition of the scrap steel smelting production is abnormal, an alarm signal is generated and sent to the operation control platform;

[0082] After receiving the alarm signal, the operation control platform immediately notifies the management personnel and takes corresponding measures for processing;

[0083] Through the abnormal early warning module, the key operation parameters (such as the internal temperature of the electric furnace, the operating pressure, the power input amount and the vibration acceleration of the key equipment) are monitored in real time, the abnormal conditions in the production process can be found immediately, and the fast response and processing are ensured. By calculating the imbalance values (DWS, DYS, DSS, SZS) of each parameter and comprehensively evaluating the production operation state evaluation coefficient SYP, a quantitative method is provided to accurately judge the production state, the state classification is carried out by using the layered threshold (the first and second production operation state evaluation coefficient thresholds), the excellent, normal and abnormal conditions of the running condition can be effectively distinguished, the accuracy and flexibility of the management decision are improved, the module supports the preventive maintenance strategy, the risk of sudden failure is reduced by continuously monitoring the equipment state, and the equipment life is prolonged. These measures help to improve the safety and stability of the production line, reduce the unplanned downtime, optimize the resource utilization, and improve the overall production efficiency and product quality.

[0084] The specific content of the treatment measures is:

[0085] Immediately check and adjust the key parameters (such as temperature, pressure, etc.) in the production process to ensure that they are in the best range;

[0086] Check the key equipment that may cause abnormality, and timely repair or replace the faulty parts;

[0087] Adjust the production process according to abnormal situations, such as modifying the raw material ratio or changing the cooling speed, to quickly restore normal production;

[0088] Increase the frequency of testing product intermediates and final products to ensure product quality;

[0089] Strengthen the monitoring of emissions such as waste gas and waste water, and take necessary measures to reduce pollution and ensure environmental compliance;

[0090] Evaluate the safety conditions of the work site, suspend work with safety hazards for inspection and rectification, and ensure personnel safety;

[0091] Post-analysis of abnormal causes, development of long-term prevention measures, and incorporation of lessons learned into daily management and training to prevent similar incidents from occurring again.

[0092] Digital twin prediction module for building an electric furnace steelmaking digital twin model, simulating the operating state under different scenarios through the model, and predicting the molten steel yield and power consumption based on full-link data;

[0093] The process of building an electric furnace steelmaking digital twin model by the digital twin prediction module includes:

[0094] Obtain full-link data during scrap steel smelting production, including key material composition ratio, smelting process temperature, smelting process pressure, power input, oxygen supply, molten steel yield, and environmental temperature;

[0095] The specific steps of building an electric furnace steelmaking twin model through digital twin technology are as follows:

[0096] Based on the charge ratio, process parameters and equipment operation principle of electric furnace steelmaking, a physical model is built to simulate the temperature change, composition change and energy transfer during the smelting process using thermodynamics, heat transfer and chemical reaction kinetics equations. The electrode current, voltage, power, and the melting, oxidation, and decarburization processes of the charge are simulated using the electric furnace smelting model. A three-dimensional visualization model of the electric furnace steelmaking process is built using three-dimensional modeling technology and sensor data. The models, parameters, reaction equations, sensor data and three-dimensional modeling technology used in building the electric furnace steelmaking twin model are all prior art and will not be described in detail here.

[0097] The process of simulating the operating state under different scenarios by the digital twin prediction module includes:

[0098] Obtain real-time full-link data, use real-time data synchronization technology to synchronize real-time full-link data to the electric furnace steelmaking twin model, and display the operating state of the electric furnace steelmaking process in real time through the electric furnace steelmaking twin model. The real-time data synchronization technology is prior art, such as MQTT protocol or OPCUA, and will not be described in detail here.

[0099] Simulate the running state of different scenarios through the twin model of electric furnace steelmaking:

[0100] Define the simulation scenarios, divide the simulation scenarios into normal smelting scenarios, abnormal smelting scenarios (for example: power failure, abnormal raw material composition, equipment failure), different furnace charge ratio scenarios and different production targets (for example: producing different steel grades) scenarios;

[0101] Use thermodynamics, heat transfer and chemical reaction kinetics model to simulate the temperature change, composition change and energy transfer in the process of electric furnace steelmaking, adjust the model parameters (such as electrode current, voltage, power, furnace charge ratio, oxygen flow, etc.) to simulate the smelting process under different scenarios, and display the internal temperature distribution, composition distribution and energy transfer of the electric furnace under different scenarios through three-dimensional visualization model in real time;

[0102] The process of predicting the molten steel yield and power consumption by the digital twin prediction model based on the whole link data includes:

[0103] Obtain the historical whole link data in the steel production process, generate a collection period, and divide the collection period into multiple collection time periods;

[0104] Obtain the key material composition ratio change rate in multiple collection time periods, the key material composition ratio change rate represents the ratio between the key material composition ratio change amount and the corresponding time period length, and the set A of key material composition ratio change rates is constructed in this way, and the average value of the difference between the maximum subset and the minimum subset in set A is recorded as the key material composition ratio change rate difference CBC;

[0105] The method for calculating the key material composition ratio change rate difference CBC can be used to obtain the smelting process temperature change rate difference DWC, the power input change rate difference DSC and the oxygen supply change rate difference YGC;

[0106] Obtain the key material composition ratio in multiple collection time periods, and calculate the arithmetic mean of the obtained multiple key material composition ratios, and record the arithmetic mean of the multiple key material composition ratios as the average key material composition ratio PCB;

[0107] The method for calculating the average key material composition ratio PCB can be used to obtain the average smelting process temperature PDW, the average power input PDS and the average oxygen supply PYG;

[0108] The process of predicting the molten steel yield and power consumption by the digital twin prediction model based on the whole link data includes:

[0109] The critical material component proportion change rate difference CBC, the smelting process temperature change rate difference DWC, the power input quantity change rate difference DSC, the oxygen supply quantity change rate difference YGC, the average critical material component proportion PCB, the average smelting process temperature PDW, the average power input quantity PDS, and the average oxygen supply quantity PYG are combined to construct a yield-energy consumption prediction matrix CNY, the yield-energy consumption prediction matrix CNY is taken as the input of the machine learning model, and the molten steel yield GCC and the power consumption DXL in a future period of time corresponding to each group of yield-energy consumption prediction matrix CNY are taken as the output of the machine learning model, the molten steel yield GCC and the power consumption DXL in a future period of time are taken as the prediction target, the sum of prediction errors of all training data is minimized as the training target, the machine learning model is trained until the sum of prediction errors converges, and a yield-energy consumption prediction model is obtained;

[0110] Real-time full-link data in the scrap smelting production process is obtained, which is converted into a corresponding yield-energy consumption prediction matrix CNY and input into the yield-energy consumption prediction model, and the real-time molten steel yield GCC and the real-time power consumption DXL in a future period of time are obtained through the yield-energy consumption prediction model;

[0111] The digital twin prediction module accurately simulates different operating scenarios and predicts molten steel yield and power consumption by building an electric furnace steelmaking digital twin model, realizes early identification of potential problems and production plan optimization, uses real-time data synchronization technology (such as MQTT or OPCUA) for real-time monitoring and dynamic adjustment of the production process, enhances transparency and control, supports simulation of multiple smelting scenarios, helps enterprises develop effective strategies, generates prediction matrices through historical data analysis and machine learning, provides accurate yield and energy consumption prediction, assists decision-making, optimizes production parameters to reduce energy waste, improves efficiency and reduces costs, supports preventive maintenance, reduces failure risk and prolongs equipment life.

[0112] Embodiment two: The technical solution of the embodiment of the application is different from that of embodiment one in that:

[0113] As shown in Figure 2 and Figure 4 , the production optimization scheduling module optimizes production parameters using an optimization algorithm based on the prediction results of the electric furnace steelmaking digital twin model and the feedback information provided by the abnormality early warning module to generate the best production scheduling scheme.

[0114] The process in which the production optimization scheduling module optimizes production parameters using an optimization algorithm includes:

[0115] The prediction result of the digital twin model of the electric furnace steelmaking and the feedback information provided by the abnormal early warning module are obtained, the prediction result is the real-time molten steel yield GCC and the real-time power consumption DXL in a future period of time, and the feedback information is that the operation condition of the scrap steel smelting production is excellent, normal or abnormal.

[0116] The production parameters are defined as a vector z = [T, O, P, M] T , wherein T is a smelting process temperature, O is an oxygen supply amount, P is an electric power input amount, and M is a key material component ratio;

[0117] The production parameters are optimized using an optimization algorithm, the optimization target is to minimize the electric power consumption and maximize the molten steel yield while satisfying the production constraint conditions, and the objective function is:

[0118] f(z) = v1·DXL(z) + v2·(GCC target -GCC(z)), wherein DXL(z) is the electric power consumption output by the electric furnace steelmaking digital twin model by inputting the production parameter z, GCC(z) is the molten steel yield output by the electric furnace steelmaking digital twin model by inputting the production parameter z, GCC target is the target molten steel yield, and v1 and v2 are weight coefficients for balancing the electric power consumption and yield targets (v1 > v2 > 0), which need to be adjusted according to actual production requirements;

[0119] The constraint conditions are: ;

[0120] The process of the production optimization scheduling module for solving the objective function and generating the best production scheduling scheme includes:

[0121] The solving steps of the objective function are as follows:

[0122] S1: initialization: a set of production parameters z is randomly generated;

[0123] S2: calculating the objective function: f(z) is calculated according to the mathematical model;

[0124] S3: evaluating the constraint conditions: whether the production parameters satisfy the constraint conditions is checked;

[0125] S4: updating the parameters: the process parameters are updated using the optimization algorithm;

[0126] S5: iteration: steps S2-S4 are repeated until a convergence condition is reached or a maximum number of iterations is reached;

[0127] The optimal production parameter vector z * = [T * , O * , P * , M *] T , according to the optimal production parameter vector z * generate the optimal production scheduling scheme, the scheme content including:

[0128] adjust the current smelting process temperature T to the optimal smelting process temperature T * ;

[0129] adjust the current oxygen supply O to the optimal oxygen supply O * ;

[0130] adjust the current power input P to the optimal power input P * ;

[0131] adjust the current key material component ratio M to the optimal key material component ratio M * ;

[0132] The production optimization scheduling module adjusts the key production parameters accurately based on the prediction results and abnormal early warning feedback of the electric furnace steelmaking digital twin model, uses the optimization algorithm to minimize the power consumption and maximize the molten steel yield, adjusts the weight coefficient flexibly, and can adapt to different production demands to ensure that the optimal solution is found. Using real-time data synchronization technology, quickly respond to production changes, ensure the continuity and stability of production, optimize production parameters to reduce energy waste and material loss, improve efficiency and reduce cost, and the generated optimal production scheduling scheme provides a scientific basis for the management layer. In combination with abnormal early warning information, problems are identified in advance to reduce the occurrence of failures and prolong the service life of equipment.

[0133] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can make equivalent replacement or change within the technical range disclosed by the present application according to the technical solution and improvement concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. A green short-process operation and management system for the steel industry based on digital twins, applied to an operation and management platform, characterized in that: include: The data acquisition and preprocessing module is used to collect data from all stages of the scrap steel smelting production process in real time and to perform preprocessing operations on the collected data. The anomaly warning module is used to monitor key operating parameters in the steel production process in real time and immediately trigger an anomaly warning when an anomaly is detected. The digital twin prediction module is used to build a digital twin model of electric arc furnace steelmaking. The model simulates the operating status under different scenarios and predicts steel production and electricity consumption based on data from all stages. The digital twin prediction model processes data from all stages to facilitate the prediction of steel production and electricity consumption, including the following steps: Acquire historical data from all stages of the scrap steel smelting production process, generate a collection cycle, and divide the collection cycle into multiple collection periods. The rate of change of the proportion of key material components is obtained in multiple collection periods. The rate of change of the proportion of key material components represents the ratio between the change in the proportion of key material components and the duration of the corresponding time period. A set A of the rate of change of the proportion of key material components is constructed in this way, and the mean of the difference between the largest and smallest subsets in set A is denoted as the difference of the rate of change of the proportion of key material components CBC. Similarly, the difference in the rate of change of the proportion of key material components (CBC) can be obtained by using the same method as the difference in the rate of change of the temperature during the smelting process (DWC), the difference in the rate of change of the power input (DSC), and the difference in the rate of change of the oxygen supply (YGC). The proportions of key material components within multiple acquisition periods are obtained, and the arithmetic mean of the obtained proportions of key material components is calculated. The arithmetic mean of the proportions of key material components is recorded as the average key material component proportion (PCB). Similarly, the average smelting process temperature PDW, average power input PDS, and average oxygen supply PYG can be obtained by using the method of calculating the average critical material composition ratio of PCB. The process by which the digital twin prediction model predicts steel production and electricity consumption based on data from all stages includes: The production energy consumption prediction matrix CNY is constructed by combining the difference in the rate of change of the proportion of key material components (CBC), the difference in the rate of change of smelting process temperature (DWC), the difference in the rate of change of power input (DSC), the difference in the rate of change of oxygen supply (YGC), the average proportion of key material components (PCB), the average smelting process temperature (PDW), the average power input (PDS), and the average oxygen supply (PYG). The production energy consumption prediction matrix CNY is used as the input of the machine learning model, and the steel production (GCC) and power consumption (DXL) for a future period corresponding to each production energy consumption prediction matrix CNY are used as the output of the machine learning model. The prediction targets are steel production (GCC) and power consumption (DXL) for a future period, and the training objective is to minimize the sum of prediction errors of all training data. The machine learning model is trained until the sum of prediction errors converges, and then training stops, thus obtaining the production energy consumption prediction model. Real-time data from all stages of the scrap steel smelting process is acquired, converted into a corresponding production and energy consumption prediction matrix CNY, and input into the production and energy consumption prediction model. The production and energy consumption prediction model is used to obtain the real-time steel production GCC and real-time electricity consumption DXL for a future period of time. The production optimization and scheduling module, based on the prediction results of the digital twin model of electric arc furnace steelmaking and the feedback information provided by the anomaly early warning module, uses optimization algorithms to optimize production parameters and generate the best production scheduling plan.

2. The green short-process operation and management system for the steel industry based on digital twins according to claim 1, characterized in that, The process by which the anomaly early warning module monitors key operating parameters in the steel production process in real time includes: Key operating parameters in the scrap steel smelting process are obtained, including the internal temperature of the electric furnace, the pressure during the smelting process, the power input, and the vibration acceleration of key equipment. A monitoring cycle is generated and the monitoring cycle is divided into multiple monitoring periods. The internal temperature imbalance value of the electric furnace is obtained within the monitoring period and marked as DWS. The internal temperature imbalance value of the electric furnace represents the ratio between the portion of the internal temperature variation difference of the electric furnace that is greater than the preset threshold for the internal temperature variation difference of the electric furnace and the internal temperature variation difference of the electric furnace. The internal temperature variation difference of the electric furnace represents the difference between the maximum and minimum internal temperature values ​​of the electric furnace.

3. The green short-process operation and management system for the steel industry based on digital twins according to claim 2, characterized in that, The process by which the anomaly early warning module provides early warnings for abnormal situations in the steel production process includes: Similarly, using the method for calculating the internal temperature imbalance value DWS of the electric furnace, the pressure imbalance value DYS, the power input imbalance value DSS, and the vibration acceleration imbalance value SZS of key equipment in the smelting process can be obtained. The production operation status evaluation coefficient SYP is then calculated using the following formula: ； Where w1, w2, w3, and w4 are all preset scaling factor coefficients, w1 > w2 > w3 > w4 > 0. The production operation status evaluation coefficient SYP is compared with the preset first production operation status evaluation coefficient threshold and the preset second production operation status evaluation coefficient threshold. The preset first production operation status evaluation coefficient threshold is less than the preset second production operation status evaluation coefficient threshold. If the production operation status evaluation coefficient SYP is less than the preset first production operation status evaluation coefficient threshold, it indicates that the scrap steel smelting production is in excellent condition. If the production operation status evaluation coefficient SYP is greater than or equal to the preset first operation status evaluation coefficient threshold, and the production operation status evaluation coefficient SYP is less than the preset second operation status evaluation coefficient threshold, it indicates that the scrap steel smelting production is operating normally. If the production operation status assessment coefficient SYP is greater than or equal to the preset second operation status assessment coefficient threshold, it indicates that the operation status of scrap steel smelting production is abnormal, an alarm signal is generated and sent to the operation control platform; Upon receiving an alarm signal, the operation and management platform immediately notifies management personnel and takes appropriate measures to handle the situation.

4. The green short-process operation and management system for the steel industry based on digital twins according to claim 1, characterized in that, The process of constructing a digital twin model of electric arc furnace steelmaking by the digital twin prediction module includes: Acquire data from all stages of the scrap steel smelting production process, including the proportion of key material components, smelting process temperature, smelting process pressure, power input, oxygen supply, molten steel output, and ambient temperature. The specific steps for constructing a digital twin model of electric arc furnace steelmaking are as follows: A physical model is constructed based on the furnace charge ratio, process parameters, and equipment operation principles in the electric arc furnace steelmaking process. Thermodynamic, heat transfer, and chemical reaction kinetic equations are used to simulate temperature changes, composition changes, and energy transfer during the smelting process. The electric arc furnace smelting model is used to simulate electrode current, voltage, power, and the melting, oxidation, and decarburization processes of the furnace charge. A three-dimensional visualization model of the electric arc furnace steelmaking process is constructed using three-dimensional modeling technology and sensor data.

5. A green short-process operation and control system for the steel industry based on digital twins according to claim 4, characterized in that, The process by which the digital twin prediction module simulates the operating state under different scenarios through a model includes: Acquire real-time data from all stages and use real-time data synchronization technology to synchronize the real-time data from all stages to the electric arc furnace steelmaking twin model. The electric arc furnace steelmaking twin model can then display the real-time operating status of the electric arc furnace steelmaking process. Simulate the operating status under different scenarios using an electric arc furnace steelmaking twin model: Define simulation scenarios, which are divided into normal smelting scenarios, abnormal smelting scenarios, scenarios with different furnace charge ratios, and scenarios with different production targets. Thermodynamic, heat transfer, and chemical reaction kinetic models are used to simulate temperature changes, composition changes, and energy transfer during electric arc furnace steelmaking. By adjusting the model parameters, the smelting process under different scenarios is simulated, and a three-dimensional visualization model is used to display the temperature distribution, composition distribution, and energy transfer inside the electric arc furnace under different scenarios in real time.

6. The green short-process operation and management system for the steel industry based on digital twins according to claim 1, characterized in that, The process by which the production optimization scheduling module optimizes production parameters using optimization algorithms includes: The prediction results of the digital twin model of electric arc furnace steelmaking and the feedback information provided by the anomaly early warning module are obtained. The prediction results are the real-time steel production GCC and real-time power consumption DXL in the future period. The feedback information is whether the scrap steel smelting production is in excellent, normal or abnormal. The production parameters are defined as a vector z = [T, O, P, M]. T Where T is the smelting process temperature, O is the oxygen supply, P is the power input, and M is the proportion of key material components. The production parameters are optimized using an optimization algorithm. The optimization objective is to minimize power consumption and maximize steel output while satisfying production constraints. The objective function is: f(z) = v1·DXL(z) + v2·(GCC) target -GCC(z)), where DXL(z) is the input production parameter z, the power consumption output by the electric arc furnace steelmaking digital twin model, and GCC(z) is the input production parameter z, the steel output by the electric arc furnace steelmaking digital twin model. target For the target steel production output, v1 and v2 are both weighting coefficients; Constraints: .

7. A green short-process operation and control system for the steel industry based on digital twins as described in claim 6, characterized in that, The process by which the production optimization scheduling module solves the objective function and generates the optimal production scheduling scheme includes: The steps to solve the objective function are as follows: S1: Initialization: Randomly generate a set of production parameters z; S2: Calculate the objective function: Calculate f(z) based on the mathematical model; S3: Evaluate constraints: Check whether the production parameters meet the constraints; S4: Update parameters: Update process parameters using an optimization algorithm; S5: Iteration: Repeat steps S2-S4 until the convergence condition or the maximum number of iterations is reached; The optimal production parameter vector z is obtained by solving the objective function. * =[T * O * P * M * ] T According to the optimal production parameter vector z * Generate the optimal production scheduling plan, which includes: Adjust the current smelting process temperature T to the optimal smelting process temperature T. * ; Adjust the current oxygen supply level O to the optimal oxygen supply level O. * ; Adjust the current power input P to the optimal power input P. * ; Adjust the current key material composition ratio M to the optimal key material composition ratio M. * .

Images