Aluminum electrolysis green electricity regulation and control method and system based on electric power fluctuation regulation and control model

Abstract

The invention relates to the technical field of aluminum electrolysis, in particular to an aluminum electrolysis green electricity regulation and control method and system based on an electric power fluctuation regulation and control model.The method comprises the steps that multi-source data in the aluminum electrolysis process of an electrolytic cell is collected, and an initial data set is formed; performing data fusion processing and blind area filling processing on the initial data set, and constructing a multi-dimensional energy regulation and control model based on the processed initial data set; based on the multi-dimensional energy regulation and control model, generating a dynamic parameter adjustment strategy by using a machine learning algorithm; predicting the running trend of the electrolytic cell through the target deep learning model, diagnosing the abnormal state of the electrolytic cell, and generating an adjustment suggestion of a dynamic parameter adjustment strategy; and on the basis of the dynamic parameter adjustment strategy and the adjustment suggestion, generating and executing a regulation and control instruction of the aluminum electrolysis green electricity regulation and control system. Efficient, stable and low-carbon operation of electrolytic aluminum production under green electricity fluctuation is achieved, and meanwhile the operation cost of an enterprise is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of aluminum electrolysis, and particularly to a method and system for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model. Background Art

[0002] As a key link in the aluminum industry chain, the production process of aluminum electrolysis has extremely high requirements for the stability and efficiency of energy supply. In recent years, with the increasing global attention to green and sustainable energy, the application of green electricity (such as renewable energy sources like solar and wind energy) in the aluminum electrolysis industry has gradually increased.

[0003] However, the output of green electricity energy is unstable and somewhat unpredictable, which poses significant challenges to both the grid side and the load side. As an important user on the load side of electricity consumption, the aluminum electrolysis industry has extremely high requirements for the stability of power supply, and any power supply fluctuation may have an adverse impact on production. Although traditional energy storage technologies and regulation systems can alleviate the problems caused by power supply fluctuations to a certain extent, they often have disadvantages such as high cost, low efficiency, and slow response speed. Especially in the case of a high ratio of green electricity energy access, the adjustment and optimization of traditional strategies seem inadequate. Moreover, existing aluminum electrolysis production control systems also have deficiencies in data processing and algorithm application. For example, the ability to collect, process, and analyze real-time production data is limited, and it is difficult to fully explore and utilize the valuable information in the data. In addition, existing systems lack customized algorithms and models for the characteristics of green electricity energy, resulting in limitations in the accuracy and practicality of algorithms in actual applications. These problems not only affect the efficiency and stability of aluminum electrolysis production but also limit the further promotion and application of green electricity energy in the industry.

[0004] Therefore, there is an urgent need for a green electricity regulation technology for aluminum electrolysis that integrates multi-source data and has dynamic response capabilities to support the realization of the industry's green transformation goals. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model to solve the problems of slow response, high cost, insufficient data utilization, and poor algorithm adaptability in existing aluminum electrolysis production control technologies when dealing with green electricity fluctuations, resulting in insufficient stability of aluminum electrolysis production.

[0006] In a first aspect, the present invention provides a method for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model, and the method includes:

[0007] Collect multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set;

[0008] Perform data fusion processing and blind spot filling processing on the initial data set, and construct a multi-dimensional energy regulation model based on the processed initial data set;

[0009] Based on the multi-dimensional energy regulation model, use machine learning algorithms to generate dynamic parameter adjustment strategies;

[0010] Predict the operation trend of the electrolytic cell through the target deep learning model, diagnose the abnormal state of the electrolytic cell, and generate adjustment suggestions for the dynamic parameter adjustment strategy;

[0011] Based on the dynamic parameter adjustment strategy and the adjustment suggestions, generate and execute the regulation instructions of the green power regulation system for aluminum electrolysis.

[0012] In an optional implementation manner, the multi-source data includes electrolytic cell operation parameters, rectifier station system production data, chemical analysis data, and intelligent device real-time operation data.

[0013] In an optional implementation manner, the performing data fusion processing and blind spot filling processing on the initial data set, and constructing a multi-dimensional energy regulation model based on the processed initial data set includes:

[0014] Perform data cleaning, association analysis, and interpolation algorithm processing on the initial data set in sequence to complete the data blind spots required for electrolytic cell analysis and diagnosis;

[0015] Based on the processed initial data set and combined with simulation technology, construct a multi-dimensional energy regulation model; the multi-dimensional energy regulation model is used to dynamically optimize the production parameters in the aluminum electrolysis process of the electrolytic cell.

[0016] In an optional implementation manner, the generating dynamic parameter adjustment strategies based on the multi-dimensional energy regulation model and using machine learning algorithms includes:

[0017] Based on the multi-dimensional energy regulation model, use the target deep learning model to analyze the processed initial data set to generate dynamic parameter adjustment strategies; the dynamic parameter adjustment strategies include adjusting the NB interval, the fluoride salt feeding amount, and the set voltage.

[0018] In an optional implementation manner, the predicting the operation trend of the electrolytic cell through the target deep learning model, diagnosing the abnormal state of the electrolytic cell, and generating adjustment suggestions for the dynamic parameter adjustment strategy includes:

[0019] Predict the operation trend of the electrolytic cell through the target deep learning model and evaluate whether there is an abnormal cell condition; the abnormal cell condition includes voltage abnormality and cell temperature abnormality;

[0020] The target deep learning model is used to perform real-time diagnosis on energy supply fluctuations and process parameter deviations, and determine the impacts of the energy supply fluctuations and the process parameter deviations on the operation of the electrolytic cell;

[0021] According to the prediction and diagnosis results of the target deep learning model, adjustment suggestions for the dynamic parameter adjustment strategy are generated; the adjustment suggestions for the dynamic parameter adjustment strategy include a set voltage correction value, an optimized value of the fluoride salt feeding amount, and an adjustment value of the NB interval.

[0022] In an optional implementation manner, generating and executing a control instruction for the green power regulation system of aluminum electrolysis based on the dynamic parameter adjustment strategy and the adjustment suggestions includes:

[0023] According to the dynamic parameter adjustment strategy and the adjustment suggestions, dynamically match the changes in green power supply on the grid side, and output a control instruction to the electrolytic cell control system through the green power regulation system of aluminum electrolysis to meet the production regulation requirements under different green power access ratios;

[0024] Real-time monitor the cell condition data after execution and feedback it to the multi-dimensional energy regulation model and the target deep learning model to form a closed-loop optimization.

[0025] In a second aspect, the present invention provides a green power regulation system for aluminum electrolysis based on a power fluctuation regulation model. The system is used to execute the above-mentioned method for regulating green power of aluminum electrolysis based on a power fluctuation regulation model. The system includes:

[0026] A data acquisition unit, which is used to acquire multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set;

[0027] A data processing unit, connected to the data acquisition unit, which is used to perform data fusion processing and blind area filling processing on the initial data set, and construct a multi-dimensional energy regulation model based on the processed initial data set;

[0028] An electrolysis big data platform, connected to the data processing unit, which is used to store, manage and analyze the data processed by the data processing unit, and provide it to the intelligent flexible regulation system to generate a dynamic parameter adjustment strategy;

[0029] An intelligent flexible regulation system, connected to the electrolysis big data platform, which is used to generate a dynamic parameter adjustment strategy based on the multi-dimensional energy regulation model and using machine learning algorithms; predict the operation trend of the electrolytic cell through a target deep learning model, diagnose the abnormal state of the electrolytic cell, and generate adjustment suggestions for the dynamic parameter adjustment strategy; generate and execute a control instruction for the green power regulation system of aluminum electrolysis based on the dynamic parameter adjustment strategy and the adjustment suggestions.

[0030] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the above-mentioned method for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored. The computer instructions are used to cause a computer to execute the method for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model according to the first aspect or any corresponding embodiment thereof.

[0032] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, which are used to cause a computer to execute the above-mentioned method for regulating green electricity in aluminum electrolysis based on a power fluctuation regulation model.

[0033] The technical solution provided by the present invention may include the following beneficial effects:

[0034] By establishing a multi-dimensional energy regulation model to dynamically match the changes in green electricity supply on the grid side, and combining machine learning algorithms to optimize production parameters in real time (such as adjusting current efficiency and setting voltage), the present invention maximizes the proportion of green electricity consumption, reduces the dependence on traditional fossil energy, reduces the direct current power consumption per unit of aluminum product, and significantly reduces the energy cost.

[0035] By using a target deep learning model to predict the operation trend of the electrolytic cell in real time (such as voltage swing and abnormal cell temperature), and dynamically adjusting parameters through a closed-loop optimization mechanism, the present invention responds to green electricity fluctuations in seconds, reduces the number of unplanned cell stops, avoids abnormal effects and out-of-control cell conditions caused by power supply fluctuations, and ensures the continuity of product quality and production capacity.

[0036] The present invention integrates multi-source data and provides real-time monitoring and multi-dimensional analysis functions through a visual interface. This enables enterprises to quickly locate production bottlenecks based on data, optimize the process path, and improve decision-making efficiency. The present invention combines digital twin simulation to verify the effectiveness of the dynamic parameter adjustment strategy, and iteratively updates the model parameters through real-time feedback data.

[0037] Through intelligent algorithm drive, multi-model collaboration, and a closed-loop optimization mechanism, the present invention realizes the efficient, stable, and low-carbon operation of aluminum electrolysis production under green electricity fluctuations, while reducing the enterprise operation cost, and provides a feasible technical path for the green transformation of the industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0039] Figure 1 It is a schematic structural diagram of an aluminum electrolysis green power regulation system based on a power fluctuation regulation model according to an embodiment of the present invention;

[0040] Figure 2 It is a schematic functional diagram of an aluminum electrolysis green power regulation system based on a power fluctuation regulation model according to an embodiment of the present invention;

[0041] Figure 3 It is a flowchart of a method for regulating aluminum electrolysis green power based on a power fluctuation regulation model according to an embodiment of the present invention;

[0042] Figure 4 It is a flowchart of another method for regulating aluminum electrolysis green power based on a power fluctuation regulation model according to an embodiment of the present invention;

[0043] Figure 5 It is a flowchart of yet another method for regulating aluminum electrolysis green power based on a power fluctuation regulation model according to an embodiment of the present invention;

[0044] Figure 6 It is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Specific Embodiments

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0046] According to an embodiment of the present invention, an embodiment of an aluminum electrolysis green power regulation system based on a power fluctuation regulation model is provided. Figure 1 It is a schematic structural diagram of an aluminum electrolysis green power regulation system based on a power fluctuation regulation model according to an embodiment of the present invention. As Figure 1 shown, the system includes:

[0047] A data acquisition unit for collecting multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set;

[0048] A data processing unit, connected to the data acquisition unit, for performing data fusion processing and blind area filling processing on the initial data set, and constructing a multi-dimensional energy regulation model based on the processed initial data set;

[0049] An electrolysis big data platform, connected to the data processing unit, for storing, managing and analyzing the data processed by the data processing unit, and providing it to the intelligent flexible regulation system to generate a dynamic parameter adjustment strategy;

[0050] An intelligent flexible regulation system, connected to the electrolysis big data platform, for generating a dynamic parameter adjustment strategy based on the multi-dimensional energy regulation model and using machine learning algorithms; predicting the operation trend of the electrolytic cell through the target deep learning model, diagnosing the abnormal state of the electrolytic cell, and generating adjustment suggestions for the dynamic parameter adjustment strategy; generating and executing the regulation instructions of the green power regulation system for aluminum electrolysis based on the dynamic parameter adjustment strategy and the adjustment suggestions.

[0051] Furthermore, the data acquisition unit in this embodiment is responsible for collecting multi-source data generated during the aluminum electrolysis process, and these data are integrated to form an initial data set. The multi-source data includes the operating parameters of the electrolytic cell, the production data of the rectifier station, the chemical analysis data, and the real-time operating data of intelligent devices, etc. The data processing unit is connected to the data acquisition unit and processes the collected initial data set. The processing process includes data fusion and blind area filling to ensure the integrity and accuracy of the data. The processed data is used to construct a multi-dimensional energy regulation model, which can dynamically optimize the production parameters of the electrolytic cell. The electrolysis big data platform is connected to the data processing unit and is used to store, manage and analyze the processed data. This platform not only stores the data, but also is responsible for providing the data to the intelligent flexible regulation system to generate a dynamic parameter adjustment strategy. The intelligent flexible regulation system is connected to the electrolysis big data platform and is the core part of the system. It uses the multi-dimensional energy regulation model and machine learning algorithms (target deep learning model) to predict the operation trend of the electrolytic cell, diagnose the abnormal state, and generate the dynamic parameter adjustment strategy and its adjustment suggestions. Finally, based on these strategies and suggestions, the system will generate and execute specific regulation instructions to adapt to the changes in the green power supply on the grid side and ensure the stable operation of the electrolytic cell under different green power access ratios.

[0052] Furthermore, as Figure 1As shown in the figure, the green power regulation system for aluminum electrolysis is divided into four main levels. The data acquisition unit obtains the basic data of the electrolytic cell and combines the original data of the existing system to obtain multi-source data, so as to form an initial data set. For example, in the existing system's cell control system, the basic data included in the cell control system are core data such as the working voltage, series current, and molecular ratio of the electrolytic cell. The data processing unit performs key operations such as classifying and screening all the collected multi-source data, and uploads the processed data to the electrolysis big data platform. The electrolysis big data platform visualizes, digitalizes, and intelligentizes the data. Its operation is to classify the processed data, such as process data, basic data, operation data, energy consumption data, and alarm data. Process data includes molecular ratio, alumina concentration, etc.; among them, basic data is voltage (working voltage, set voltage, average voltage), cell temperature, electrolyte level, etc.; operation data is tapping amount, anode replacement, etc., and energy consumption data is DC power consumption, AC power consumption, total energy consumption of the electrolytic cell, etc.; the alarm data system will monitor the key parameters of each electrolytic cell. When the parameters exceed the preset threshold range, the system will give a warning / alarm prompt in the way of different text colors. The electrolysis big data platform transmits all the summarized data to the intelligent flexible regulation system, and the intelligent flexible regulation system is displayed according to functions. The main displayed functions include key concerns, adjustment strategies, evaluation reports, warning cells, parameter configuration, etc.

[0053] Furthermore, please refer to Figure 2 the functional schematic diagram of a green power regulation system for aluminum electrolysis based on a power fluctuation regulation model shown in the figure. The intelligent flexible regulation system processes the historical data of the existing electrolysis aluminum big data platform, efficiently accumulates data sets, empowers production application scenarios, and realizes flexible regulation of production strategies.

[0054] In this embodiment, based on four major supporting hardware (basic IT resources, high-frequency voltage acquisition, single-cell task display, intelligent broadcast system), a set of green power regulation system for aluminum electrolysis is constructed, aiming to achieve efficient and stable operation of aluminum electrolysis production under green power fluctuations through data-driven and multi-model collaboration. On the basis of the construction of the four major supporting hardware such as "basic IT resources, high-frequency voltage acquisition, single-cell task display, intelligent broadcast system", the green power regulation system for aluminum electrolysis is constructed, including:

[0055] 1. Various data reports such as the total number of concerned cells, cells with abnormal effects, voltage swing cells, voltage abnormal cells, feeding abnormalities, upper and lower level fluctuation cells, and cell temperature abnormal cells.

[0056] 2. Provide real-time data of cells with abnormal effects, voltage swing cells, voltage abnormal cells, feeding abnormalities, upper and lower level fluctuation cells, and cell temperature abnormal cells, and draw a curve graph to provide a basis for decision-making.

[0057] 3. Provide an overview of the cells, real-time data of abnormal cells, online single-cell monitoring, and list data such as cell numbers, voltages, high-frequency / low-frequency noises of the electrolytic cells. The real-time data, current abnormal voltage change trends, and current abnormal fluctuation change trends corresponding to each single cell can be viewed.

[0058] 4. For the decision-making level, the centralized control center, and the section chiefs: Provide series control current, series stop pole-changing operation current, fluorine-bearing alumina stop feeding current, butterfly valve opening, adjustment of the purification alumina feeding time, etc. Alumina NB interval, set voltage, aluminum fluoride feeding amount (or interval), aluminum fluoride stop feeding current, alumina stop feeding current, aluminum tapping adjustment amount, etc. Pole-changing additional voltage, pole-changing additional voltage time, aluminum tapping additional voltage, pole-changing additional voltage time.

[0059] 5. Conduct loss statistical analysis on the primary aluminum loss, effect loss, material loss, fluoride salt volatilization loss, increase in labor volume, and environmental pollution during the process of current anomaly control. Accurately calculate the losses from aspects such as product loss, energy consumption loss, manpower loss, and material loss. The analysis of the current anomaly process includes parameter adjustment situations, effect month-on-month comparison, process diagnosis (data analysis of each section), key concerns, etc.

[0060] Furthermore, the development test method and technical route of the aluminum electrolysis green power regulation system in this embodiment include:

[0061] 1. Upgrade the intelligent acquisition equipment to assist the aluminum electrolysis green power regulation system:

[0062] According to the actual situation of the production site data acquisition platform, analyze the data acquisition requirements for the aluminum electrolysis green power regulation system, upgrade the existing intelligent acquisition equipment, effectively improve the data acquisition range and accuracy of electrolytic cell parameters in the data acquisition unit, and then help the electrolytic cell to quickly give feedback to the intelligent system when the energy changes. The aluminum electrolysis green power regulation system makes corresponding control adjustments according to the data feedback analysis results. That is to say, in this embodiment, high-precision sensors (such as infrared thermometers, current transformers) are deployed at key points of the electrolytic cell, and multi-source data (voltage, temperature, chemical concentration) are aligned and preprocessed in real time through edge computing devices to reduce the cloud transmission pressure.

[0063] 2. Real-time monitoring of electrolysis data, visual analysis, and artificial intelligence to assist in cell condition warning:

[0064] The data processing unit converges, processes, and analyzes the data from the production site and the inventory business system, deeply mines the data, and monitors the production data of the electrolytic cell process in real time. Through the visualization of data monitoring, it lists the data such as the cell number, voltage, high-frequency / low-frequency noise, etc. of the electrolytic cell. According to the real-time data of a single cell, it analyzes the abnormal change trend of voltage, the abnormal fluctuation change of current, etc. The artificial intelligence technology is used to realize the early warning of special situations in the cell condition, and cooperate to improve the management efficiency. That is to say, this embodiment can store data in a time series database and dynamically display the voltage fluctuation heat map and abnormal cell distribution of the electrolytic cell cluster through Grafana. The LSTM neural network is used to predict the cell temperature trend in the next 30 minutes. If the predicted value exceeds the threshold, an early warning is triggered and a cooling strategy is generated (such as reducing the current efficiency by 3%).

[0065] 3. Mechanism modeling to achieve precise control:

[0066] Based on the process control center and the technical support center, the data processing unit builds an intelligent flexible regulation center around production operation. The data is updated in real time, and the parameter adaptability is improved through intelligent means such as self-adaptation. A special model for energy change is established, and through big data simulation iteration, precise control of various parameters under special working conditions is achieved. The reaction speed is focused on in real time, and a quick reminder function is provided to achieve technical means such as precise control, task-driven, and standardized operation management, so as to ensure the production stability of the electrolytic cell when the external environment such as energy changes, and then improve the efficiency, reduce unnecessary energy consumption, and reduce the abnormal handling time of employees. That is to say, this embodiment calculates the theoretical optimal parameters based on the thermodynamic equation and material balance formula of the electrolytic cell. The historical data is analyzed by the random forest algorithm to correct the deviation of the mechanism model and improve the parameter adaptability. The influence of different green power access ratios on the electrolytic cell is simulated, and a regulation plan is generated in advance (such as starting the low-energy consumption mode when the green power is insufficient).

[0067] 4. Artificial intelligence for rapid evaluation to provide intelligent loss analysis under different working conditions:

[0068] Through the construction of the data system on the electrolysis big data platform, the electrolysis process is systematically managed to create standardized operations for electrolysis management, control, and operators. The system conducts process management on aluminum electrolysis production, energy, operations, etc., and conducts online diagnosis, automatic focusing, and task push on aspects such as electrolysis process, on-site control, cell stability, and energy loss. It provides various types of visualization applications for decision-making managers and workshop managers in a simple and clear manner. That is to say, this embodiment can conduct root cause analysis on current abnormal events. For example, if the voltage swing is caused by too low electrolyte level, the system recommends adjusting the aluminum liquid height and supplementing fluoride salts.

[0069] 5. Flexible solutions suitable for different users:

[0070] The intelligent flexible control system emphasizes the representability, displayability, and tasking of production process control data. Customize on-site and scenario configurations, such as parameters like molten aluminum level, standard current value, and ambient temperature. In the intelligent flexible control system, process, energy, and other information can be directly viewed. According to different needs, it provides classified applicable service pages for different roles, with rich content and simple operation. The background provides data services based on physical and chemical and mathematical models, which form a knowledge graph through deep learning. That is to say, this embodiment focuses on macro indicators (green electricity utilization rate, carbon emissions per ton of aluminum), and supports one-click generation of emission reduction reports. Users can customize alarm thresholds (such as the upper limit of cell temperature is 950 °C) or strategy priorities (energy consumption optimization priority / stability priority).

[0071] This aluminum electrolysis green electricity control system supports tasking the flexible control process. The system automatically learns the manual optimization plan through machine learning algorithms and has the function of cyclic optimization of the plan. The system integrates core functions such as data management, task scheduling, multi-dimensional analysis, real-time warning, and alarm, realizing centralized control of the whole process. All operations are completed on the same platform, supporting instant message push and dynamic reminder to ensure the coordination and efficiency of the production process.

[0072] To sum up, this embodiment dynamically matches the changes in green electricity supply on the grid side by establishing a multi-dimensional energy control model, and combines machine learning algorithms to optimize production parameters in real time (such as adjusting current efficiency and setting voltage), maximizing the proportion of green electricity consumption, reducing the dependence on traditional fossil energy, reducing the direct current power consumption per unit of aluminum product, and significantly reducing energy costs.

[0073] This embodiment uses a target deep learning model to predict the operation trend of the electrolytic cell in real time (such as voltage swing and abnormal cell temperature), and dynamically adjusts parameters through a closed-loop optimization mechanism, responding to green electricity fluctuations in seconds, reducing the number of unplanned cell stops, and avoiding abnormal effects and out-of-control cell conditions caused by power supply fluctuations, ensuring product quality and production capacity continuity.

[0074] This embodiment integrates multi-source data and provides real-time monitoring and multi-dimensional analysis functions through a visual interface. It enables enterprises to quickly locate production bottlenecks based on data, optimize the process path, and improve decision-making efficiency. This embodiment combines digital twin simulation to verify the effectiveness of the dynamic parameter adjustment strategy and iteratively updates model parameters through real-time feedback data.

[0075] This embodiment realizes the efficient, stable, and low-carbon operation of electrolytic aluminum production under green electricity fluctuations through intelligent algorithm drive, multi-model collaboration, and closed-loop optimization mechanism. At the same time, it reduces the enterprise operation cost and provides a feasible technical path for the green transformation of the industry.

[0076] According to an embodiment of the present invention, an embodiment of an aluminum electrolysis green power regulation method based on a power fluctuation regulation model is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0077] In this embodiment, an aluminum electrolysis green power regulation method based on a power fluctuation regulation model is provided, which can be used in Figure 1 an aluminum electrolysis green power regulation system based on a power fluctuation regulation model. Figure 3 It is a flowchart of an aluminum electrolysis green power regulation method based on a power fluctuation regulation model according to an embodiment of the present invention. As Figure 3 shown, this process includes the following steps:

[0078] Step S301, collect multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set.

[0079] Furthermore, in this embodiment, data can be collected in real time through high-precision sensors (such as infrared thermometers, current transformers), and the stock data of the enterprise's existing data platform (such as the MES system) can be integrated to ensure the comprehensiveness of the data. Output the initial data set, which contains multi-source data with time series alignment, for subsequent modeling and analysis.

[0080] Step S302, perform data fusion processing and blind area filling processing on the initial data set, and build a multi-dimensional energy regulation model based on the processed initial data set.

[0081] Furthermore, the data fusion processing and blind area filling processing can include data cleaning, data fusion, interpolation and filling, etc.; and combined with physical simulation (such as a heat balance model) and machine learning algorithms, a multi-dimensional energy regulation model covering four dimensions of process, energy, economy, and safety is established. This multi-dimensional energy regulation model can optimize the production parameters of the electrolytic cell in real time. For example, if the input of the multi-dimensional energy regulation model is the cleaned voltage and current data, the output is the adjustment suggestion for the fluoride salt feeding amount.

[0082] Step S303, based on the multi-dimensional energy regulation model, use machine learning algorithms to generate a dynamic parameter adjustment strategy.

[0083] Furthermore, based on the multi-dimensional energy regulation model, this embodiment uses machine learning algorithms (such as LSTM) to analyze historical data and real-time operating conditions, and generates dynamic parameter adjustment strategies, including NB intervals, fluoride salt feeding amounts, and set voltages, etc. This dynamic nature can update the strategy every target number of seconds to adapt to green power fluctuations (such as reducing the set voltage when wind power drops suddenly); the dynamic parameter adjustment strategy of this embodiment can generate differentiated strategies for different abnormal types (such as voltage swings) to achieve real-time matching of production parameters and power fluctuations.

[0084] Step S304, predict the operation trend of the electrolytic cell through the target deep learning model, diagnose the abnormal state of the electrolytic cell, and generate adjustment suggestions for this dynamic parameter adjustment strategy.

[0085] Furthermore, the target deep learning model of this embodiment can be a trained LSTM neural network and a convolutional neural network (CNN). The LSTM neural network can input the time series data (voltage, cell temperature, noise signal) of the electrolytic cell to predict the operation trend within a future target time period; the convolutional neural network (CNN) can analyze the high-frequency noise spectrum to diagnose the risks of voltage swings and effect abnormalities. This embodiment uses the target deep learning model to predict the operation trend of the electrolytic cell, evaluate whether there are abnormal cell conditions (such as voltage and cell temperature abnormalities); perform real-time diagnosis on energy supply fluctuations and process parameter deviations to determine their impacts on the operation of the electrolytic cell; according to the prediction and diagnosis results, generate adjustment suggestions for the dynamic parameter adjustment strategy, discover potential problems in advance, provide adjustment directions, and ensure the stability of the production process.

[0086] Step S305, generate and execute the regulation instructions of the green power regulation system for aluminum electrolysis based on this dynamic parameter adjustment strategy and this adjustment suggestion.

[0087] Furthermore, this embodiment combines the dynamic parameter adjustment strategy and the adjustment suggestion to generate the regulation instructions of the green power regulation system for aluminum electrolysis. Send the regulation instructions to the electrolytic cell control system to dynamically adapt to the changes in green power supply on the grid side and meet the production requirements under different green power access ratios. Achieve precise regulation of the electrolytic cell to ensure the stability and efficiency of the production process.

[0088] In summary, this embodiment dynamically matches the changes in green power supply on the grid side by establishing a multi-dimensional energy regulation model, combines machine learning algorithms to optimize production parameters in real time (such as adjusting current efficiency and set voltage), maximizes the green power consumption ratio, reduces the dependence on traditional fossil energy, reduces the direct current power consumption per unit of aluminum product, and significantly reduces the energy cost.

[0089] In this embodiment, the target deep learning model is used to predict the operation trend of the electrolytic cell in real time (such as voltage swing and abnormal cell temperature), and the parameters are dynamically adjusted through a closed-loop optimization mechanism to respond to green power fluctuations in seconds, reduce the number of unplanned cell stops, avoid abnormal effects and out-of-control cell conditions caused by power supply fluctuations, and ensure product quality and production capacity continuity.

[0090] This embodiment integrates multi-source data and provides real-time monitoring and multi-dimensional analysis functions through a visualization interface. It enables enterprises to quickly locate production bottlenecks based on data, optimize the process path, and improve decision-making efficiency. This embodiment combines digital twin simulation to verify the effectiveness of the dynamic parameter adjustment strategy and iteratively updates the model parameters through real-time feedback data.

[0091] This embodiment realizes the efficient, stable, and low-carbon operation of electrolytic aluminum production under green power fluctuations through intelligent algorithm drive, multi-model collaboration, and a closed-loop optimization mechanism. At the same time, it reduces the enterprise operation cost and provides an implementable technical path for the green transformation of the industry.

[0092] In this embodiment, another method for regulating green power in electrolytic aluminum based on a power fluctuation regulation model is provided, which can be used for Figure 1 in a system for regulating green power in electrolytic aluminum based on a power fluctuation regulation model, Figure 4 is a flowchart of another method for regulating green power in electrolytic aluminum based on a power fluctuation regulation model according to an embodiment of the present invention, as Figure 4 shown, and this process includes the following steps:

[0093] Step S401, collect multi-source data during the electrolytic aluminum process of the electrolytic cell to form an initial data set. The multi-source data includes electrolytic cell operation parameters, production data of the rectifier station system, chemical analysis data, and real-time operation data of intelligent devices.

[0094] Furthermore, in this embodiment, multi-source data during the electrolytic aluminum process is collected in real time to form an initial data set, providing a comprehensive and real-time data basis for subsequent data processing and analysis. The data types of the multi-source data include electrolytic cell operation parameters, production data of the rectifier station system, chemical analysis data, and real-time operation data of intelligent devices. The electrolytic cell operation parameters include cell temperature, working voltage, average voltage, set voltage, aluminum liquid level, electrolyte level, etc. The production data of the rectifier station system includes current, voltage, rectification efficiency, etc. The chemical analysis data includes alumina concentration, molecular ratio, etc. The real-time operation data of intelligent devices includes the operation status of devices such as high-frequency voltage acquisition modules and cell controllers.

[0095] Furthermore, please refer to Figure 5Flowchart of another aluminum electrolysis green power regulation method based on a power fluctuation regulation model. In this embodiment, the basic data of the electrolytic cell (tank number, working voltage, average voltage, set voltage, cell temperature, electrolyte level, aluminum liquid level, etc.) is collected in real time through hardware devices, and the original data of the customer's existing data platform is integrated, including: production data such as current, voltage, and rectification efficiency of the rectifier station system; chemical analysis data such as alumina concentration and molecular ratio in the laboratory; real-time operation data of intelligent devices such as high-frequency voltage acquisition modules and cell controllers.

[0096] Step S402: Perform data cleaning, correlation analysis, and interpolation algorithm processing on the initial data set in sequence to complement the data blind spots required for electrolytic cell analysis and diagnosis.

[0097] Furthermore, in this embodiment, the initial data set is sequentially subjected to data cleaning, correlation analysis, and interpolation algorithm processing to complement the data blind spots, improve the accuracy and integrity of the data, and provide high-quality data for subsequent model construction. Data cleaning includes removing noise, outliers, and incorrect data to ensure data quality. Correlation analysis is used to identify the correlation between data to enhance data availability. The interpolation algorithm is used to fill in missing data points to ensure data integrity.

[0098] Step S403: Based on the processed initial data set and combined with simulation technology, construct a multi-dimensional energy regulation model; the multi-dimensional energy regulation model is used to dynamically optimize the production parameters in the aluminum electrolysis process of the electrolytic cell.

[0099] Furthermore, in this embodiment, based on the processed initial data set, combined with simulation technology, a multi-dimensional energy regulation model is constructed to provide a scientific basis for subsequent strategy generation and ensure the efficiency and stability of the production process. The simulation technology is used to simulate and optimize the operating state of the electrolytic cell and predict the energy demand and supply under different working conditions. The multi-dimensional energy regulation model is used to dynamically optimize the production parameters in the aluminum electrolysis process of the electrolytic cell, such as cell temperature, voltage, aluminum liquid level, etc.

[0100] As Figure 5 shown, in this embodiment, data cleaning, correlation analysis, and interpolation algorithm are used to complement the data blind spots required for electrolytic cell analysis and diagnosis. Combined with simulation technology, a multi-dimensional energy regulation model of the electrolytic cell is constructed to dynamically optimize production parameters.

[0101] Step S404: Based on the multi-dimensional energy regulation model, use the target deep learning model to analyze the processed initial data set and generate a dynamic parameter adjustment strategy; the dynamic parameter adjustment strategy includes adjusting the NB interval, fluoride salt feeding amount, and set voltage.

[0102] Furthermore, the multi-dimensional energy regulation model is constructed based on the in-depth understanding and analysis of the processed initial data set in the aluminum electrolysis production process. It integrates multi-faceted data and can comprehensively and accurately reflect the real-time operating status of the electrolytic cell and the energy supply and consumption situation, providing basic information for parameter adjustment. With the help of simulation technology, the multi-dimensional energy regulation model can simulate and predict the operating trend of the electrolytic cell, reflecting the current operating status and predicted operating trend of the electrolytic cell. The target deep learning model, namely machine learning algorithms (such as LSTM or Transformer), analyzes the processed initial data set based on the current operating status and predicted operating trend of the electrolytic cell reflected by the multi-dimensional energy regulation model, and generates corresponding dynamic parameter adjustment strategies. At the same time, the target deep learning model can automatically adjust and optimize its own model parameters and structure to better adapt to the complex situations during the operation of the electrolytic cell. With the support of the rich data and accurate operating trend prediction provided by the multi-dimensional energy regulation model, the deep learning model can be trained and learned from a large number of processed data samples (i.e., the processed initial data set), continuously improving its knowledge system and decision-making ability, and then generating more reasonable and effective dynamic parameter adjustment strategies to achieve intelligent and refined control of the aluminum electrolysis production process. In this embodiment, the target deep learning model is used to analyze the processed initial data set to generate dynamic parameter adjustment strategies to optimize production parameters, improve production efficiency and stability, and adapt to the volatility of green electricity energy. The adjustment of the NB interval in the dynamic parameter adjustment strategy is to dynamically adjust the feeding frequency according to the alumina concentration; the fluoride salt feeding amount is to optimize the addition amount based on the predicted result of the molecular ratio; the set voltage is to match the fluctuation of green electricity supply and balance the power consumption and thermal stability. The characteristics of the dynamic parameter adjustment strategy can include real-time and personalized. The real-time can update the strategy every target number of seconds to respond to the green electricity fluctuation; the personalized can generate differentiated strategies for different cell conditions (such as giving priority to adjusting the fluoride salt for electrolytic cells with high cell temperature).

[0103] Furthermore, this embodiment supports dynamically configuring data standards according to the number of data collection points. When the number of collection points is small, it operates according to the minimum standard; after the number of collection points increases, the data standard can be seamlessly extended to ensure that reliable services can be obtained in data environments of different scales. Data encryption processing and private deployment are provided. By encrypting the original plaintext data, it is ensured that the data cannot be deciphered even if it is illegally intercepted, improving the data security of the system. At the same time, this encryption service can be provided to other enterprise applications, enabling the enterprise to improve operation efficiency, reduce management risks, and standardize the application of relevant regulations while ensuring privacy.

[0104] Step S405, use the target deep learning model to predict the operating trend of the electrolytic cell and evaluate whether there is an abnormal cell condition; the abnormal cell condition includes voltage abnormality and cell temperature abnormality.

[0105] Furthermore, in this embodiment, the target deep learning model is used to predict the operation trend of the electrolytic cell to discover potential problems in advance, ensure the stability of the production process, and take preventive measures in a timely manner. The prediction content includes evaluating whether there are abnormal cell conditions, such as abnormal voltage, abnormal cell temperature, etc. Exemplarily, the target deep learning model predicts the changes in cell temperature and voltage within a future time period; an abnormal cell temperature can be that the cell temperature deviates from the reference value, and an abnormal voltage can be that the voltage swing amplitude exceeds the target value; the output of this step can be the abnormal probability (such as the probability of voltage swing occurring is 80%) and the abnormal impact level (severe / medium / slight).

[0106] Furthermore, in this embodiment, the target deep learning model is applied to analyze data, predict the operation trend of the electrolytic cell, combine artificial process experience and the deep learning model to intelligently evaluate whether the cell condition of the electrolytic cell is normal, and generate parameter adjustment strategies, such as adjusting the NB interval, adjusting the fluoride salt feeding amount, etc. Real-time diagnosis is carried out on problems such as abnormal energy supply and process parameter deviation, adjustment suggestions are provided, and potential consequences are predicted, such as abnormal cell temperature and abnormal voltage. Parameter adjustments are made for different abnormalities, and multi-dimensional analysis (energy consumption analysis, electrolytic cell stability analysis, etc.) is carried out according to the parameter adjustment strategy in combination with the operations of on-site operators.

[0107] Step S406, through the target deep learning model, real-time diagnosis is carried out on the energy supply fluctuation and process parameter deviation to determine the influence of the energy supply fluctuation and the process parameter deviation on the operation of the electrolytic cell.

[0108] Furthermore, in this embodiment, the target deep learning model is used to carry out real-time diagnosis on the energy supply fluctuation and process parameter deviation, determine the influence of the energy supply fluctuation and process parameter deviation on the operation of the electrolytic cell, and timely adjust the production strategy to reduce the adverse effects and ensure the smooth progress of the production process. For example, analyze the influence of the change in green power supply on the current stability, identify process problems such as too long NB interval and insufficient fluoride salt addition, and conduct attribution analysis and sensitivity analysis.

[0109] Step S407, according to the prediction and diagnosis results of the target deep learning model, generate adjustment suggestions for the dynamic parameter adjustment strategy; the adjustment suggestions for the dynamic parameter adjustment strategy include setting voltage correction values, optimized fluoride salt feeding amounts, and NB interval adjustment values.

[0110] Furthermore, in this embodiment, according to the prediction and diagnosis results of the deep learning model, generate adjustment suggestions for the dynamic parameter adjustment strategy, including setting voltage correction values, optimized fluoride salt feeding amounts, NB interval adjustment values, etc., to provide specific adjustment directions, optimize the production process, and improve production efficiency and product quality.

[0111] Step S408, based on the dynamic parameter adjustment strategy and the adjustment suggestions, generate and execute the control instructions of the green power regulation system for aluminum electrolysis.

[0112] In an alternative embodiment, step S408 includes:

[0113] According to the dynamic parameter adjustment strategy and the adjustment suggestion, dynamically match the changes in the green power supply on the grid side, and output a control command to the electrolytic cell control system through the green power regulation system for aluminum electrolysis to meet the production regulation requirements under different green power access ratios;

[0114] Real-time monitor the cell condition data after execution and feedback it to the multi-dimensional energy regulation model and the target deep learning model to form a closed-loop optimization.

[0115] Furthermore, based on the dynamic parameter adjustment strategy and the adjustment suggestion, this embodiment generates and executes the control command of the green power regulation system for aluminum electrolysis, ensures the effectiveness of the control command, continuously optimizes the production process, and improves the adaptive capacity and intelligent level of the system. The content of the command includes dynamically matching the changes in the green power supply on the grid side and outputting a control command to the electrolytic cell control system. The control command indicates meeting the production regulation requirements under different green power access ratios to ensure the stability and efficiency of the production process; it also indicates real-time monitoring of the cell condition data after execution and feedback to the multi-dimensional energy regulation model and the target deep learning model to form a closed-loop optimization.

[0116] In summary, when the external power supply fluctuates, this embodiment can, according to the actual working conditions of the electrolysis series collected in real time, through an intelligent algorithm model, promptly give the optimal matching process parameters to cope with the changes in the external power supply, ensure the safe and stable production, and reduce losses. First of all, this embodiment realizes the efficient utilization of green power energy through an intelligent algorithm. It can, according to the green power supply situation on the grid side and in combination with the actual needs of electrolytic aluminum production, dynamically adjust the production control strategy in real time, so as to maximize the utilization of green power energy, reduce the dependence on traditional energy sources, and lower the production cost. Moreover, in view of the instability and unpredictability of green power energy, this embodiment realizes the fine control of the production process by introducing advanced control algorithms and models, can respond to grid fluctuations, and promptly adjust production parameters to ensure the smooth progress of electrolytic aluminum production. This flexible regulation ability not only improves the production stability, but also guarantees the product quality and production capacity. Secondly, this embodiment makes full use of industrial Internet technology to collect and analyze production data in real time. Through the in-depth mining and analysis of these data, enterprises can more accurately understand the problems and bottlenecks in the production process and provide a more scientific basis for decision-making. This data-driven decision-making method not only improves the decision-making efficiency, but also reduces the decision-making risk. In addition, in view of the characteristics and requirements of electrolytic aluminum production, this embodiment has developed customized algorithm models (multi-dimensional energy regulation model and target deep learning model). These models can more accurately reflect the actual situation in the production process and improve the accuracy and practicability of the algorithms. By continuously training and optimizing these models, enterprises can continuously improve the effect of the production control strategy and achieve more efficient production. Furthermore, the implementation of this embodiment helps to promote the sustainable development of the electrolytic aluminum industry. By efficiently using green power energy, improving production stability and data-driven decision-making, etc., enterprises can reduce production costs, improve resource utilization efficiency, and reduce environmental pollution. This not only conforms to the national environmental protection policies, but also brings long-term economic and social benefits to enterprises.

[0117] In summary, this embodiment dynamically matches the changes in the green power supply on the grid side by establishing a multi-dimensional energy regulation model, and combines machine learning algorithms to optimize production parameters in real time (such as adjusting current efficiency and setting voltage), maximizes the green power consumption ratio, reduces the dependence on traditional fossil energy sources, lowers the direct current power consumption per unit of aluminum product, and significantly reduces the energy cost.

[0118] This embodiment uses a target deep learning model to predict the operation trend of the electrolytic cell in real time (such as voltage swing and abnormal cell temperature), and dynamically adjusts the parameters through a closed-loop optimization mechanism, responds to green power fluctuations in seconds, reduces the number of unplanned cell stops, avoids abnormal effects and out-of-control cell conditions caused by power supply fluctuations, and guarantees the continuity of product quality and production capacity.

[0119] This embodiment integrates multi-source data and provides real-time monitoring and multi-dimensional analysis functions through a visual interface. It enables enterprises to quickly locate production bottlenecks based on data, optimize process paths, and improve decision-making efficiency. This embodiment combines digital twin simulation to verify the effectiveness of the dynamic parameter adjustment strategy and iteratively updates model parameters through real-time feedback data.

[0120] Through intelligent algorithm drive, multi-model collaboration, and closed-loop optimization mechanism, this embodiment realizes the efficient, stable, and low-carbon operation of electrolytic aluminum production under green power fluctuations, while reducing the enterprise operation cost and providing a feasible technical path for the green transformation of the industry.

[0121] Please refer to Figure 6 , Figure 6 which is a schematic structural diagram of a computer device provided by an alternative embodiment of the present invention. As Figure 6 shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (such as an array of servers, a set of blade servers, or a multi-processor system). Figure 6 In

[0122] Figure, one processor 10 is taken as an example.

[0123] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0124] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device and the like. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely disposed relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0125] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memories.

[0126] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0127] The embodiments of the present invention also provide a computer-readable storage medium. The methods according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and to be stored in a local storage medium, so that the methods described herein can be stored in such software processes on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0128] A part of the present invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can call or provide the methods and / or technical solutions according to the present invention through the operations of the computer. Those skilled in the art should understand that the forms of existence of computer program instructions in a computer-readable medium include but are not limited to source files, executable files, installation package files, etc. Correspondingly, the ways for a computer to execute computer program instructions include but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0129] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the defined scope.

Claims

1. An aluminum electrolysis green power regulation method based on a power fluctuation regulation model, characterized in that The method includes: Collecting multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set; Performing data fusion processing and blind area filling processing on the initial data set, and constructing a multi-dimensional energy regulation model based on the processed initial data set; Based on the multi-dimensional energy regulation model, using machine learning algorithms to generate a dynamic parameter adjustment strategy; Predicting the operation trend of the electrolytic cell through a target deep learning model, diagnosing the abnormal state of the electrolytic cell, and generating adjustment suggestions for the dynamic parameter adjustment strategy; Based on the dynamic parameter adjustment strategy and the adjustment suggestions, generating and executing the regulation instructions of the green power regulation system for aluminum electrolysis.

2. The method according to claim 1, characterized in that The multi-source data includes the operation parameters of the electrolytic cell, the production data of the rectifier station system, chemical analysis data, and the real-time operation data of intelligent devices.

3. The method according to claim 1, wherein The performing data fusion processing and blind area filling processing on the initial data set, and constructing a multi-dimensional energy regulation model based on the processed initial data set includes: Successively performing data cleaning, association analysis, and interpolation algorithm processing on the initial data set to complete the data blind area required for the analysis and diagnosis of the electrolytic cell; Based on the processed initial data set and combined with simulation technology, constructing a multi-dimensional energy regulation model; the multi-dimensional energy regulation model is used to dynamically optimize the production parameters during the aluminum electrolysis process of the electrolytic cell.

4. The method according to claim 1, characterized in that The based on the multi-dimensional energy regulation model, using machine learning algorithms to generate a dynamic parameter adjustment strategy includes: Based on the multi-dimensional energy regulation model, using a target deep learning model to analyze the processed initial data set to generate a dynamic parameter adjustment strategy; the dynamic parameter adjustment strategy includes adjusting the NB interval, the fluoride salt feeding amount, and the set voltage.

5. The method according to claim 1, characterized in that The predicting the operation trend of the electrolytic cell through a target deep learning model, diagnosing the abnormal state of the electrolytic cell, and generating adjustment suggestions for the dynamic parameter adjustment strategy includes: Predicting the operation trend of the electrolytic cell through the target deep learning model to evaluate whether there is an abnormal cell condition; the abnormal cell condition includes voltage abnormality and cell temperature abnormality; Performing real-time diagnosis on the energy supply fluctuation and process parameter deviation through the target deep learning model to determine the influence of the energy supply fluctuation and the process parameter deviation on the operation of the electrolytic cell; According to the prediction and diagnosis results of the target deep learning model, generating adjustment suggestions for the dynamic parameter adjustment strategy; the adjustment suggestions for the dynamic parameter adjustment strategy include the set voltage correction value, the fluoride salt feeding amount optimization value, and the NB interval adjustment value.

6. The method according to claim 1, characterized in that, The based on the dynamic parameter adjustment strategy and the adjustment suggestions, generating and executing the regulation instructions of the green power regulation system for aluminum electrolysis includes: According to the dynamic parameter adjustment strategy and the adjustment suggestions, dynamically matching the change of green power supply on the grid side, and outputting regulation instructions to the electrolytic cell control system through the green power regulation system for aluminum electrolysis to meet the production regulation requirements under different green power access ratios; Real-time monitoring the cell condition data after execution and feeding it back to the multi-dimensional energy regulation model and the target deep learning model to form a closed-loop optimization.

7. An aluminum electrolysis green power regulation system based on a power fluctuation regulation model, characterized in that, The system is used to execute an aluminum electrolysis green power regulation method based on a power fluctuation regulation model according to any one of claims 1 to 6. The system includes: A data acquisition unit, configured to acquire multi-source data during the aluminum electrolysis process of the electrolytic cell to form an initial data set; A data processing unit, connected to the data acquisition unit, configured to perform data fusion processing and blind area filling processing on the initial data set, and construct a multi-dimensional energy regulation model based on the processed initial data set; An electrolysis big data platform, connected to the data processing unit, configured to store, manage, and analyze the data processed by the data processing unit, and provide it to the intelligent flexible regulation system to generate a dynamic parameter adjustment strategy; An intelligent flexible regulation system, connected to the electrolysis big data platform, configured to generate a dynamic parameter adjustment strategy based on the multi-dimensional energy regulation model and using a machine learning algorithm; predict the operation trend of the electrolytic cell through a target deep learning model, diagnose the abnormal state of the electrolytic cell, and generate adjustment suggestions for the dynamic parameter adjustment strategy; generate and execute a regulation instruction of the aluminum electrolysis green power regulation system based on the dynamic parameter adjustment strategy and the adjustment suggestions.

8. A computer device, characterized in that, Including: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute an aluminum electrolysis green power regulation method based on a power fluctuation regulation model according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute an aluminum electrolysis green power regulation method based on a power fluctuation regulation model according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Including computer instructions, and the computer instructions are used to cause a computer to execute an aluminum electrolysis green power regulation method based on a power fluctuation regulation model according to any one of claims 1 to 6.

Images