Continuous prebaked anode aluminum electrolysis production method and system

Abstract

The invention discloses a continuous prebaked anode aluminum electrolysis production method and system, and relates to the technical field of anode aluminum electrolysis production, and the method comprises the following steps: firstly, collecting current density distribution data and electrolyte distribution state in an electrolytic cell through a real-time monitoring system, and reflecting current and electrolyte conditions of each area in the electrolytic cell in real time; and data support is provided for subsequent adjustment measures. The electrolyte distribution in the electrolytic cell is monitored in real time and intelligently evaluated, intelligent evaluation is performed in combination with the deep learning model, current distribution is automatically adjusted, and it is ensured that the current density is uniform. According to the scheme, the problems of local overload, excessive consumption of the anode and overheating of equipment are effectively avoided, the energy efficiency and the stability of the electrolysis process are improved, the service life of the equipment is prolonged, the maintenance cost is reduced, the downtime is shortened, and therefore the production efficiency and the economic benefits are improved. And meanwhile, human intervention is reduced, the automation level is improved, and the intelligent management capacity of aluminum electrolysis production is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of anode aluminum electrolysis production, and particularly relates to a continuous pre-baked anode aluminum electrolysis production method and system. Background Art

[0002] Continuous pre-baked anode aluminum electrolysis production refers to the production of electrolytic aluminum by using pre-baked anodes during the aluminum electrolysis process. The pre-baked anodes are made of petroleum coke, clay and other raw materials through high-temperature roasting, and have good electrical conductivity and corrosion resistance. During this production process, bauxite is decomposed into aluminum and oxygen through an electrolytic reaction. The aluminum is deposited at the bottom of the electrolytic cell, while the oxygen is released at the anode surface. The "continuous" feature of pre-baked anode aluminum electrolysis production means that during the production process, the stable operation of the electrolytic cell is maintained by continuously replacing anodes, replenishing electrolytes, etc. The whole process requires a high degree of automation and precise control to ensure the high efficiency and stability of the aluminum electrolysis process, thereby improving the production efficiency of aluminum and reducing energy consumption.

[0003] The prior art has the following deficiencies: In the aluminum electrolysis reaction, the uniform distribution of the electrolyte in the electrolytic cell is crucial. If the electrolyte is unevenly distributed, current concentration may occur in some areas, which will accelerate anode consumption and even cause local overheating or overload problems. Since this non-uniformity is often difficult to directly observe, it is usually only detected when excessive wear or melting occurs in some areas of the electrolytic cell, resulting in equipment damage or a significant decrease in energy efficiency.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a continuous pre-baked anode aluminum electrolysis production method and system. By real-time monitoring and intelligent evaluation of the electrolyte distribution in the electrolytic cell, combined with a deep learning model for intelligent evaluation, the current distribution is automatically adjusted to ensure uniform current density. This solution effectively avoids local overload, excessive anode consumption and equipment overheating problems, improves the energy efficiency and stability of the electrolysis process, extends the equipment life, reduces maintenance costs and downtime, thereby improving production efficiency and economic benefits. At the same time, it reduces human intervention, improves the automation level, and enhances the intelligent management ability of aluminum electrolysis production to solve the problems in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solution: A continuous pre-baked anode aluminum electrolysis production method, comprising the following steps:

[0007] First, the real-time monitoring system collects the current density distribution data and the distribution status of the electrolyte in the electrolytic cell, reflecting the current and electrolyte conditions in each area of ​​the electrolytic cell in real time, providing data support for subsequent adjustment measures;

[0008] Aggregate the collected data to form a complete data set, and pre-process the data in the data set;

[0009] Extract key features reflecting the uneven distribution of electrolyte in the electrolytic cell from the preprocessed data, analyze the extracted key features within the detection window, and quantify the degree of uneven distribution of electrolyte in the electrolytic cell;

[0010] The analyzed key features are input into a pre-trained deep learning model, and the deep learning model is used to intelligently evaluate whether there is uneven distribution of electrolyte in the electrolytic cell;

[0011] When uneven electrolyte distribution in the electrolytic cell is detected, dynamic current regulation measures are taken to reduce current input in areas with high current density to prevent excessive consumption of the anode and avoid local overheating; while in areas with low current density, the current flow is automatically increased to ensure uniform current distribution in the entire electrolytic cell.

[0012] Preferably, the specific steps of collecting the current density distribution data and the distribution state of the electrolyte in the electrolytic cell through the real-time monitoring system and reflecting the current and electrolyte conditions in each area of ​​the electrolytic cell in real time are as follows:

[0013] First, multiple sensors are arranged in the electrolytic cell. The sensors are distributed in different positions of the electrolytic cell to ensure full coverage of all areas of the cell.

[0014] Then, the sensor collects the current density, temperature and electrolyte concentration data of each area in real time, and transmits the acquired data to the central control unit through the data acquisition system;

[0015] Finally, the central control unit processes, stores and analyzes the received real-time data to ensure that the operating status of each area in the electrolyzer can be dynamically monitored and provide real-time feedback for subsequent optimization control.

[0016] Preferably, key features reflecting the uneven distribution of electrolyte in the electrolytic cell are extracted from the preprocessed data, including the degree of current density difference in different regions of the electrolytic cell and the coverage of electrolyte in different regions of the electrolytic cell. Under the detection window, the degree of current density difference in different regions of the electrolytic cell and the coverage of electrolyte in different regions of the electrolytic cell are analyzed to generate a local current density difference reference value and an electrolyte coverage reference value respectively. Through the local current density difference reference value and the electrolyte coverage reference value, the electrolyte distribution in each region of the electrolytic cell is quantified, thereby revealing the unevenness of electrolyte distribution and its influence on the current density.

[0017] Preferably, the specific steps for analyzing the degree of current density difference in different regions of the electrolytic cell under the detection window to generate a local current density difference reference value are as follows:

[0018] The current density in different regions of the electrolytic cell is analyzed in detail. First, the electrolytic cell is divided into several sub-regions, and the current density of each sub-region is calculated by collecting the data of current sensors. The calculation formula is as follows:

[0019] ,

[0020] In the formula, J i is the current density, representing the current intensity in the i-th region, I i is the current value of the i-th region, A i is the surface area of the current region;

[0021] Next, the degree of current density difference in different regions of the electrolytic cell is quantified. The unevenness of electrolyte distribution is reflected by calculating the difference between the current density of each region and the current density of its adjacent regions. The non-uniformity measurement formula is as follows:

[0022] ,

[0023] In the formula, ΔJ i is the current density non-uniformity measurement of the i-th region, used to quantify the difference between the current density of the i-th region and the current density of its adjacent regions, reflecting the unevenness of the current density in this region. N(i) is the set of neighborhood regions of the i-th region, J j is the current intensity in the j-th region, ∈ is a smoothing factor, which is a small constant;

[0024] Finally, to comprehensively consider the influence of current density difference on the operation of the electrolytic cell, the current density difference value of each region is weighted with the coverage of electrolyte concentration in the current region to generate the final local current density difference reference value. The generation formula is as follows:

[0025] ,

[0026] wherein, LCDD is the reference value of the local current density difference, n is the total number of regions into which the electrolytic cell is divided, C i is the electrolyte concentration of the i-th region, C k is the electrolyte concentration of the k-th region.

[0027] Preferably, the specific steps for analyzing the coverage degree of electrolytes in different regions of the electrolytic cell under the detection window to generate the reference value of the electrolyte coverage rate are as follows:

[0028] First, it is necessary to calculate the coverage degree difference of electrolytes in each region of the electrolytic cell. The coverage degree difference reflects the non-uniformity of the electrolyte distribution in each region. The local coverage degree function is used to represent the local coverage degree of the electrolyte concentration at the coordinate point. The calculation expression is as follows:

[0029] ,

[0030] wherein, C i (x, y) is the local coverage degree of the electrolyte in the i-th region of the electrolytic cell, (x, y) is the coordinate point, E i (x, y) is the electrolyte concentration in the i-th region of the electrolytic cell, E max is the maximum theoretical concentration of the electrolyte in the current region;

[0031] After obtaining the local coverage degree C i (x, y) of the electrolyte, further calculate the local coverage degree difference index between each region and its surrounding regions. The local coverage degree difference index measures the degree of difference in the electrolyte distribution between regions. The calculation expression is as follows:

[0032] ,

[0033] wherein, ΔC i is the local coverage degree difference index, representing the electrolyte coverage degree difference of the i-th region, N(i) is the set of neighborhood regions of the i-th region, C j (x, y) is the local coverage degree of the electrolyte in the j-th region of the electrolytic cell, |C i (x, y)-C j (x, y)| α is the coverage degree difference weighting, and α is the control local difference weight coefficient;

[0034] Finally, by comprehensively considering the local differences of each region and calculating with weights, the reference value of the electrolyte coverage rate is obtained. The calculation expression is as follows:

[0035] The formula is as follows:

[0036] ,

[0037] where ECR is the reference value of the electrolyte coverage rate, n is the total number of regions divided in the electrolytic cell, max(ΔC) is the maximum value of the local coverage difference index, w i is the weight of region i, and β is the weighting index that controls the importance of the global difference.

[0038] Preferably, the locally generated current density difference reference value and the electrolyte coverage rate reference value after analysis are input into a pre-trained deep learning model. An electrolyte distribution non-uniformity state coefficient is generated through the deep learning model, and the electrolyte distribution state in the electrolytic cell is evaluated through the electrolyte distribution non-uniformity state coefficient, so as to intelligently evaluate the electrolyte distribution state in the electrolytic cell.

[0039] Preferably, the electrolyte distribution non-uniformity state coefficient generated when evaluating the electrolyte distribution state in the electrolytic cell through a pre-trained deep learning model is compared and analyzed with a pre-set reference threshold of the electrolyte distribution non-uniformity state coefficient, so as to intelligently evaluate the electrolyte distribution state in the electrolytic cell. The specific steps are as follows:

[0040] If the electrolyte distribution non-uniformity state coefficient is greater than the pre-set reference threshold of the electrolyte distribution non-uniformity state coefficient, the electrolyte distribution state in the electrolytic cell is classified as non-uniform; if the electrolyte distribution non-uniformity state coefficient is less than or equal to the pre-set reference threshold of the electrolyte distribution non-uniformity state coefficient, the electrolyte distribution state in the electrolytic cell is classified as uniform.

[0041] Preferably, when it is detected that the electrolyte distribution in the electrolytic cell is non-uniform, the specific steps for taking dynamic current adjustment measures are as follows:

[0042] First, the electrolyte distribution non-uniformity state coefficient EDUS generated by the deep learning model is compared and analyzed with a pre-set reference threshold. When the electrolyte distribution non-uniformity state coefficient EDUS is greater than the pre-set reference threshold, it indicates that the electrolyte distribution is non-uniform. At this time, the dynamic current adjustment mechanism is triggered to calculate the current adjustment amplitude. The calculation expression is as follows:

[0043] ,

[0044] where ΔI is the current adjustment amplitude, θ is the adjustment coefficient, and EDUS ref is the reference threshold of the electrolyte distribution non-uniformity state coefficient.

[0045] Based on the evaluation results, the current input to the regions with high current density in the electrolytic cell is reduced to prevent excessive consumption of the anode and avoid local overheating. The calculation expression for the current reduction amplitude is as follows:

[0046] ,

[0047] Where ΔI high is the current adjustment value in the high current density region, ω is the current regulation coefficient, representing the sensitivity of current adjustment, J high is the actual current density in the high current density region, J ref is the reference value of the current density, is the current density adjustment index, γ is the heat load regulation coefficient, T high is the temperature in the high current density region, T ref is the temperature in the reference region of the current density;

[0048] In the region with low current density, the current flow is automatically increased to ensure the uniformity of the current distribution in the electrolytic cell. The amplitude of the current increase is calculated by the following formula:

[0049] ,

[0050] Where ΔI low is the current adjustment value in the low current density region, J low is the actual current density in the low current density region, δ is the temperature regulation coefficient, representing the influence of temperature on current adjustment, T low is the temperature in the low current density region.

[0051] A continuous pre-baked anode aluminum electrolysis production system includes a real-time monitoring and data acquisition module, a data summary and preprocessing module, a key feature extraction and analysis module, a deep learning intelligent evaluation module, and a dynamic current regulation and optimization module:

[0052] The real-time monitoring and data acquisition module. First, it collects the current density distribution data in the electrolytic cell and the distribution state of the electrolyte through the real-time monitoring system, and reflects the current and electrolyte conditions in each region of the electrolytic cell in real time, providing data support for subsequent adjustment measures;

[0053] The data summary and preprocessing module. It summarizes the collected data to form a complete data set and preprocesses the data in the data set;

[0054] The key feature extraction and analysis module. It extracts the key features reflecting the uneven distribution of the electrolyte in the electrolytic cell from the preprocessed data, analyzes the extracted key features within the detection window, and quantifies the degree of uneven distribution of the electrolyte in the electrolytic cell;

[0055] The deep learning intelligent evaluation module. It inputs the analyzed key features into a pre-trained deep learning model, and uses the deep learning model to intelligently evaluate whether there is uneven distribution of the electrolyte in the electrolytic cell;

[0056] The dynamic current regulation and optimization module, when detecting uneven electrolyte distribution in the electrolytic cell, takes dynamic current regulation measures. In the area with high current density, the current input is reduced to prevent excessive consumption of the anode and avoid local overheating; while in the area with low current density, the current flow is automatically increased to ensure uniform current distribution throughout the electrolytic cell.

[0057] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0058] By implementing real-time monitoring, intelligent evaluation of uneven electrolyte distribution, and dynamic current regulation measures, the present invention can effectively solve the problem of uneven electrolyte distribution in aluminum electrolysis production. Through precise monitoring of the current density and electrolyte distribution in the electrolytic cell, combined with the intelligent evaluation of the electrolyte distribution state by the deep learning model, the system can timely detect the area with uneven electrolyte and automatically adjust the current distribution to ensure uniform current density in the electrolytic cell. This solution effectively prevents problems such as local overload, excessive anode consumption, and equipment overheating, not only improving the energy efficiency and stability of the electrolysis process, but also extending the service life of the equipment, reducing the maintenance cost and downtime during production, and enhancing the production efficiency and economic benefits. At the same time, by reducing manual intervention and real-time adjustment, while optimizing the electrolysis process, it also improves the automation level and enhances the intelligent management ability of aluminum electrolysis production. Brief Description of the Drawings

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0060] Figure 1 It is a method flow chart of a continuous pre-baked anode aluminum electrolysis production method of the present invention.

[0061] Figure 2 It is a module schematic diagram of a continuous pre-baked anode aluminum electrolysis production system of the present invention. Detailed Embodiments

[0062] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples described herein; on the contrary, these exemplary embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art.

[0063] The present invention provides a continuous pre-baked anode aluminum electrolysis production method as Figure 1 shown, including the following steps:

[0064] First, the current density distribution data in the electrolytic cell and the distribution state of the electrolyte are collected through a real-time monitoring system, which can reflect the current and electrolyte conditions in each area of the electrolytic cell in real time, providing data support for subsequent adjustment measures.

[0065] Real-time monitoring can provide accurate feedback and promptly identify abnormalities in the electrolytic cell, especially the uneven distribution of the electrolyte and the abnormal distribution of the current density, providing necessary data support for subsequent adjustment measures.

[0066] First, a variety of sensors are arranged in the electrolytic cell, such as current sensors, temperature sensors, and electrolyte concentration sensors. These sensors are distributed at different positions in the electrolytic cell to ensure that all areas of the cell body can be comprehensively covered. Then, the sensors collect data such as the current density, temperature, and electrolyte concentration in each area in real time, and transmit these data to the central control unit through a data acquisition system. The central control unit processes, stores, and analyzes the received real-time data to ensure that the operating state of each area in the electrolytic cell can be dynamically monitored and provide real-time feedback for subsequent optimal control. This process can promptly detect the uneven distribution of the electrolyte or the abnormality of the current density in the electrolytic cell, providing data support for further adjustment.

[0067] The collected data is aggregated to form a complete data set, and the data in the data set is preprocessed.

[0068] These data sets include parameters such as the current density, electrolyte concentration, and temperature in each area of the electrolytic cell. By establishing the data set, the system can comprehensively understand the operating state inside the electrolytic cell. The role of this step is to integrate all the original data into a unified format, facilitating subsequent processing and analysis, and laying a data foundation for subsequent feature extraction and deep learning model training.

[0069] The main task of preprocessing is to process noise, missing values, and outliers, and standardize or normalize the data. The purpose of this step is to improve the data quality and ensure that the data in subsequent analysis is clean and meets statistical requirements. In addition, by removing abnormal data or filling in missing values, it can be ensured that the system will not make incorrect decisions due to inaccurate data in subsequent analysis, thus ensuring the stability and accuracy of the model.

[0070] Key features reflecting the uneven distribution of the electrolyte in the electrolytic cell are extracted from the preprocessed data, and the extracted key features are analyzed within a detection window to quantify the degree of uneven distribution of the electrolyte in the electrolytic cell.

[0071] These features may include the spatial distribution of current density, the gradient of temperature change, electrolyte concentration differences, etc. The extracted features will enable the system to accurately identify whether the electrolyte is evenly distributed. The extraction of key features is the core of problem identification, which enables the system to make accurate judgments based on the actual operating conditions inside the electrolytic cell, rather than relying solely on single sensor data. The extracted features will be the basis for model training and subsequent analysis.

[0072] Extract key features from the preprocessed data that reflect the uneven distribution of the electrolyte inside the electrolytic cell, including the degree of current density difference in different regions of the electrolytic cell and the coverage of the electrolyte in different regions of the electrolytic cell. Under the detection window, analyze the degree of current density difference in different regions of the electrolytic cell and the coverage of the electrolyte in different regions of the electrolytic cell, and generate local current density difference reference values and electrolyte coverage reference values respectively. Through the local current density difference reference values and electrolyte coverage reference values, quantify the electrolyte distribution in each region of the electrolytic cell, thereby revealing the unevenness of the electrolyte distribution and its impact on the current density.

[0073] An overly large degree of current density difference in different regions of the electrolytic cell usually indicates uneven electrolyte distribution inside the electrolytic cell. The distribution of the electrolyte in the electrolytic cell directly affects the flow and distribution of current. If the electrolyte concentration is low in certain regions, the current will concentrate in these regions, resulting in too high current density, while in regions with higher electrolyte concentration, the current density is lower. An overly large current density difference usually means there is obvious unevenness in the distribution of the electrolyte inside the electrolytic cell, and this unevenness will lead to local overheating, rapid anode consumption, or low electrolysis efficiency of aluminum. Therefore, the increase in current density difference can be used as an important indicator of uneven electrolyte distribution.

[0074] The specific steps for analyzing the degree of current density difference in different regions of the electrolytic cell under the detection window to generate local current density difference reference values are as follows:

[0075] Conduct a detailed analysis of the current density in different regions of the electrolytic cell. First, divide the electrolytic cell into several sub-regions, and the current density of each sub-region is calculated by collecting data from current sensors. The calculation formula is as follows:

[0076] ,

[0077] In the formula, J i is the current density, representing the current intensity in the i-th region, I i is the current value in the i-th region, A i is the surface area of the current region;

[0078] By analyzing the current density of multiple sub-regions, the current density distribution state in different regions of the electrolytic cell can be determined. If the current density in some regions is significantly higher or lower, it may indicate uneven distribution of the electrolyte.

[0079] Next, quantify the degree of difference in current density in different regions of the electrolytic cell. The unevenness of the electrolyte distribution is reflected by calculating the difference between the current density of each region and that of its adjacent regions. The non-uniformity measurement formula is as follows:

[0080] ,

[0081] In the formula, ΔJ i is the non-uniformity measurement of the current density in the i-th region, used to quantify the degree of difference in current density between the i-th region and its adjacent regions, reflecting the unevenness of the current density in this region. N(i) is the set of neighboring regions of the i-th region. That is to say, all regions adjacent to region i are taken into consideration. J j is the current intensity in the j-th region, ∈ is a smoothing factor, which is a small constant, usually taking the value of ∈→0, aiming to avoid the denominator in the division operation being zero;

[0082] The function of this step is to measure the relative difference between the current density of each region and that of the surrounding regions. The greater the difference, the more significant the non-uniformity of the current distribution in the electrolytic cell, which also means uneven distribution of the electrolyte.

[0083] Finally, to comprehensively consider the impact of the current density difference on the operation of the electrolytic cell, the current density difference value of each region is weighted with the electrolyte concentration coverage degree of the current region to generate the final local current density difference reference value. The generation formula is as follows:

[0084] ,

[0085] In the formula, LCDD is the local current density difference reference value, n is the total number of regions divided in the electrolytic cell, C i is the electrolyte concentration in the i-th region, C k is the electrolyte concentration in the k-th region.

[0086] The function of the above steps is to weight the current density difference to more accurately reflect the impact of the region with lower electrolyte concentration on the unevenness of the current distribution. Through this weighted index, the uneven degree of the electrolyte distribution in the electrolytic cell can be more precisely evaluated, providing a basis for subsequent adjustment.

[0087] The greater the reference value of the local current density difference generated after analyzing the degree of difference in current density in different regions of the electrolytic cell under the detection window, generally, it means that the electrolyte distribution in the electrolytic cell is uneven. The current density difference index is used to evaluate the uniformity of electrolyte distribution by analyzing the current density difference in each region of the electrolytic cell. When the electrolyte distribution is uneven, the current will concentrate in the regions with lower electrolyte concentration, resulting in higher current density in these regions and forming obvious differences. Therefore, the increase in the reference value of the local current density difference reflects the unevenness of electrolyte distribution. If the reference value is small, it means that the current density difference in the electrolytic cell is small, indicating that the electrolyte is more evenly distributed in each region and the electrolysis process is more stable.

[0088] Excessive differences in the coverage degree of electrolytes in different regions of the electrolytic cell indicate uneven electrolyte distribution in the electrolytic cell. During the aluminum electrolysis process, the uniform distribution of electrolytes is crucial for the efficiency and stability of the reaction. If the electrolyte coverage in some regions is too thin or too concentrated, the current density in these regions will be affected unevenly, resulting in current concentration or overload in local areas, which will accelerate the consumption of the anode or cause problems such as overheating. Excessive differences in the coverage degree of electrolytes in the electrolytic cell will affect the balance of the aluminum electrolysis reaction, reduce energy efficiency, and may even cause equipment damage. Therefore, the uniform distribution of electrolytes must be fully controlled and optimized.

[0089] The specific steps to generate the electrolyte coverage reference value by analyzing the coverage degree of electrolytes in different regions of the electrolytic cell under the detection window are as follows:

[0090] First, it is necessary to calculate the difference in the coverage degree of electrolytes in each region of the electrolytic cell. The difference in coverage degree reflects the unevenness of electrolyte distribution in each region. The local coverage degree function is used to represent the local coverage degree of electrolyte concentration at the coordinate point, and the calculation expression is as follows:

[0091] ,

[0092] In the formula, C i (x, y) is the local coverage degree of electrolyte in the i-th region of the electrolytic cell, (x, y) is the coordinate point, E i (x, y) is the electrolyte concentration in the i-th region of the electrolytic cell, E max is the maximum theoretical concentration of electrolyte in the current region;

[0093] In this way, the electrolyte concentration can be mapped to a standardized coverage degree range, thus avoiding errors caused by differences in the absolute magnitude of the concentration itself. The difference in coverage degree of each region can be measured by studying the change rate of the coverage degree of adjacent regions.

[0094] After obtaining the local electrolyte coverage degree C iAfter (x, y), the local coverage difference index between each region and its surrounding regions is further calculated. The local coverage difference index measures the degree of difference in electrolyte distribution between regions. The local difference is represented by the change in electrolyte coverage within the spatial neighborhood. The calculation expression is as follows:

[0095] ,

[0096] In the formula, ΔC i is the local coverage difference index, representing the electrolyte coverage difference of the i-th region, quantifying the change in electrolyte coverage between region i and its neighboring regions. N(i) is the set of neighborhood regions of the i-th region. That is, all regions adjacent to region i are taken into consideration. C j (x, y) is the local electrolyte coverage of the j-th region in the electrolytic cell. |C i (x, y) - C j (x, y)| α is the weighted coverage difference, calculating the electrolyte coverage difference between region i and region j and weighting it. α is the weight coefficient controlling the local difference;

[0097] This step evaluates the electrolyte inhomogeneity of each region and its surrounding regions by calculating the difference in electrolyte coverage of adjacent regions. A larger ΔC i value indicates that the electrolyte distribution in this region is quite different from that of the surrounding regions, indicating that there is a phenomenon of uneven electrolyte distribution in this region.

[0098] Finally, the reference value of electrolyte coverage rate is obtained by comprehensively considering the local differences of each region and weighted calculation. The calculation expression is as follows:

[0099] The formula is as follows:

[0100] ,

[0101] In the formula, ECR is the reference value of electrolyte coverage rate, n is the total number of regions divided in the electrolytic cell, max(ΔC) is the maximum value of the local coverage difference index, w i is the weight of region i, and β is the weighting index, controlling the importance of the global difference.

[0102] Through the above steps, the reference value of electrolyte coverage rate ECR will reflect the uniformity of electrolyte distribution in the entire electrolytic cell. If the value is large, it indicates strong inhomogeneity of electrolyte distribution; if it is small, it indicates relatively uniform electrolyte distribution. This reference value can be used as the basis for optimization control to adjust the operating parameters of the electrolytic cell.

[0103] The greater the reference value of the electrolyte coverage rate generated after analyzing the coverage degree of the electrolyte in different regions of the electrolytic cell under the detection window, the greater the difference in the coverage degree of the electrolyte in the electrolytic cell, indicating uneven electrolyte distribution. When the electrolyte coverage in some regions is too thin or too thick, the reference value of the electrolyte coverage rate will be on the high side, indicating an uneven distribution. This unevenness will lead to problems such as concentrated current density, uneven anode consumption, and excessive heat load, thus affecting the efficiency and stability of the electrolytic cell. On the contrary, when the reference value of the electrolyte coverage rate is low, it indicates that the electrolyte distribution in the electrolytic cell is relatively uniform, the entire electrolysis process is more balanced, and the equipment operates more stably.

[0104] Input the key features after analysis into a pre-trained deep learning model, and use the deep learning model to intelligently evaluate whether there is uneven electrolyte distribution in the electrolytic cell;

[0105] Input the reference value of the local current density difference and the reference value of the electrolyte coverage rate generated after analysis into a pre-learned deep learning model, generate the electrolyte distribution unevenness state coefficient through the deep learning model, and evaluate the electrolyte distribution state in the electrolytic cell through the electrolyte distribution unevenness state coefficient to intelligently evaluate the electrolyte distribution state in the electrolytic cell.

[0106] The pre-learned deep learning model refers to a neural network model trained through a large amount of historical data and actual operation experience. This model has mastered the complex relationship between the uneven pattern of electrolyte distribution in the electrolytic cell and the change of current density. During the training process, the model learns various electrolytic cell performance problems caused by uneven electrolyte distribution through a large amount of input data (such as parameters like electrolyte concentration, temperature, current density, anode consumption, etc.) and actual results (such as equipment failures, energy efficiency losses, etc.). Through the backpropagation algorithm, the model optimizes its internal weights so that when facing new input data, it can make accurate predictions about the state of electrolyte distribution. This "pre-learning" is the basis for the deep learning model in practical applications. It can understand how the unevenness of electrolyte distribution affects the overall performance of the aluminum electrolysis reaction based on the previous training data and make intelligent evaluations based on these patterns.

[0107] In deep learning models, common architectures include Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), etc. Through these models, the system can extract complex features from parameters such as the reference value of local current density difference and the reference value of electrolyte coverage rate in the input, conduct in-depth analysis, and finally generate the coefficient of uneven electrolyte distribution state. This state coefficient can quantify the degree of unevenness of electrolyte distribution and evaluate the overall operating state of the electrolytic cell according to the set threshold. For example, if the electrolyte distribution is too concentrated or sparse, the deep learning model can use the state coefficient to warn of possible problems such as overheating and excessive anode consumption, thus providing decision-making support for adjustment operations. During the continuous operation of the model, it will also self-optimize according to new data feedback, thereby improving its evaluation accuracy and adaptability, and gradually achieving intelligent state monitoring and adjustment.

[0108] The deep learning model is not limited here. Any deep learning model that can comprehensively analyze the reference value of local current density difference LCDD and the reference value of electrolyte coverage rate ECR to generate the coefficient of uneven electrolyte distribution state EDUS can be used. To implement the technical solution of the present invention, the present invention provides a specific implementation method;

[0109] The formula for generating the coefficient of uneven electrolyte distribution state EDUS is as follows:

[0110] ,

[0111] In the formula, d1 and d2 are respectively the preset proportionality coefficients of the reference value of local current density difference LCDD and the reference value of electrolyte coverage rate ECR, and both d1 and d2 are greater than 0.

[0112] The preset proportionality coefficient refers to the fixed proportionality coefficients d1 and d2 set for the reference value of local current density difference LCDD and the reference value of electrolyte coverage rate ECR in the formula for generating the coefficient of uneven electrolyte distribution state EDUS. These coefficients are used to weight the importance of each parameter to ensure that the influence of the reference value of local current density difference LCDD and the reference value of electrolyte coverage rate ECR on the final calculation result is balanced during comprehensive analysis. d1 and d2 are respectively the weighting coefficients for local current density difference and electrolyte coverage rate. The design of the preset proportionality coefficient is usually determined based on experimental data or actual production experience, aiming to ensure that the relative importance of the two conforms to the actual situation. By adjusting these two proportionality coefficients, the evaluation ability of the model for the uneven electrolyte distribution state can be optimized, thereby improving the stability and efficiency of the overall electrolysis process.

[0113] It can be seen from the electrolyte distribution non-uniformity coefficient that the greater the reference value of the local current density difference generated after analyzing the difference degree of the current density in different regions of the electrolytic cell under the detection window, and the greater the reference value of the electrolyte coverage rate generated after analyzing the coverage degree of the electrolyte in different regions of the electrolytic cell under the detection window, the greater the electrolyte distribution non-uniformity coefficient generated when evaluating the electrolyte distribution state in the electrolytic cell through a pre-trained deep learning model, indicating that the electrolyte distribution in the electrolytic cell is uneven. On the contrary, it indicates that the electrolyte distribution in the electrolytic cell is uniform.

[0114] Compare and analyze the electrolyte distribution non-uniformity coefficient generated when evaluating the electrolyte distribution state in the electrolytic cell through a pre-trained deep learning model with the pre-set reference threshold of the electrolyte distribution non-uniformity coefficient, and intelligently evaluate the electrolyte distribution state in the electrolytic cell. The specific steps are as follows:

[0115] If the electrolyte distribution non-uniformity coefficient is greater than the pre-set reference threshold of the electrolyte distribution non-uniformity coefficient, the electrolyte distribution state in the electrolytic cell is classified as uneven distribution; if the electrolyte distribution non-uniformity coefficient is less than or equal to the pre-set reference threshold of the electrolyte distribution non-uniformity coefficient, the electrolyte distribution state in the electrolytic cell is classified as uniform distribution.

[0116] When it is detected that the electrolyte distribution in the electrolytic cell is uneven, take dynamic current adjustment measures. In the area with high current density, reduce the current input to prevent excessive consumption of the anode and avoid local overheating; while in the area with low current density, automatically increase the current flow to ensure uniform current distribution throughout the electrolytic cell;

[0117] When it is detected that the electrolyte distribution in the electrolytic cell is uneven, the specific steps for taking dynamic current adjustment measures are as follows:

[0118] First, compare and analyze the electrolyte distribution non-uniformity coefficient EDUS generated by the deep learning model with the pre-set reference threshold to judge the uniformity of the electrolyte distribution in the electrolytic cell. When the electrolyte distribution non-uniformity coefficient EDUS is greater than the pre-set reference threshold, it indicates that the electrolyte distribution is uneven, and the current density is too high or too low in some regions. At this time, trigger the dynamic current adjustment mechanism and calculate the current adjustment amplitude. The calculation formula is as follows:

[0119] ,

[0120] In the formula, ΔI is the current adjustment amplitude, θ is the adjustment coefficient, representing the sensitivity of current adjustment, controlling the range of current adjustment amplitude, EDUS ref is the reference threshold of the electrolyte distribution non-uniformity coefficient,

[0121] Through the above steps, the uneven state of electrolyte distribution can be quantified, providing a quantitative basis for subsequent current adjustment.

[0122] Based on the evaluation results, the current input to the areas with high current density in the electrolytic cell is reduced to prevent excessive consumption of the anode and avoid local overheating. Specifically, when the current density is higher than the set threshold, the current input to this area is automatically reduced to adjust the current density to be more uniform. The calculation expression for the current reduction amplitude is as follows:

[0123] ,

[0124] In the formula, ΔI high is the current adjustment value in the area with high current density, ω is the current adjustment coefficient, representing the sensitivity of current adjustment, J high is the actual current density in the area with high current density, J ref is the reference value of the current density, is the current density adjustment index, usually less than 1, used to adjust the amplitude of current reduction to avoid over-adjustment. γ is the heat load adjustment coefficient, representing the influence of heat load on current adjustment, T high is the temperature in the area with high current density, T ref is the temperature in the reference area of current density;

[0125] By comprehensively considering the current density and temperature, the current value that needs to be reduced is calculated, thereby avoiding anode loss and overheating in the area with too high current density. The amount of current reduction is not only related to the current density difference but also closely related to the temperature difference and heat load. In this way, through the dual regulation of temperature and current, the current can be controlled more precisely to ensure the efficiency and stability of the electrolysis process.

[0126] In the area with low current density, the current flow is automatically increased to ensure the uniformity of current distribution in the electrolytic cell. The core of this process is to increase the current input to the area with low current density to avoid too low electrolyte distribution and affect the aluminum electrolysis efficiency. The amplitude of current increase is calculated by the following formula:

[0127] ,

[0128] In the formula, ΔI low is the current adjustment value in the area with low current density, J low is the actual current density in the area with low current density, δ is the temperature adjustment coefficient, representing the influence of temperature on current adjustment, T low is the temperature in the area with low current density.

[0129] By calculating the additional current required in the low current density area, ensure that the current distribution in the electrolytic cell tends to be uniform, and at the same time make fine-tuning according to the temperature difference. The temperature difference may affect the fluidity of the electrolyte and the current density. Therefore, considering the temperature effect can adjust the current more precisely, improving the adaptive ability and accuracy of the system.

[0130] When it is detected that the electrolyte distribution in the electrolytic cell is uneven, the role of taking dynamic current regulation measures is to ensure uniform current density distribution during the aluminum electrolysis process, thereby optimizing the electrolysis efficiency and extending the service life of the equipment. Uneven electrolyte distribution in the electrolytic cell may cause too high current density in some areas, which will accelerate the consumption of the anode and cause local overheating, thus affecting the stability and energy efficiency of the electrolytic cell; while areas with low current density will lead to insufficient utilization of the electrolyte, reducing the reaction efficiency and affecting the overall output and energy efficiency. Therefore, through dynamic current regulation measures, reduce the current input in areas with high current density, effectively avoiding the risks of excessive anode consumption and overheating. At the same time, increase the current flow in areas with low current density to ensure the uniformity of the electrolysis process. This regulation mechanism can precisely control the current distribution, maintain the balanced state in the electrolytic cell, thereby avoiding equipment failures and performance degradation caused by local overload or insufficiency. In this way, the overall operating efficiency of the electrolytic cell is improved, energy consumption is optimized, and the production process is more stable and efficient. At the same time, this intelligent current regulation can reduce human intervention, improve the automation level, reduce production risks, extend the service life of the equipment, and ultimately promote the sustainable development of the aluminum electrolysis process.

[0131] The present invention can effectively solve the problem of uneven electrolyte distribution in aluminum electrolysis production by implementing real-time monitoring, intelligent evaluation of uneven electrolyte distribution, and dynamic current regulation measures. By accurately monitoring the current density and electrolyte distribution in the electrolytic cell and combining the intelligent evaluation of the electrolyte distribution state by the deep learning model, the system can timely detect the uneven electrolyte areas and automatically adjust the current distribution to ensure uniform current density in the electrolytic cell. This solution effectively prevents problems such as local overload, excessive anode consumption, and equipment overheating, not only improving the energy efficiency and stability of the electrolysis process, but also extending the service life of the equipment, reducing the maintenance cost and downtime during the production process, and improving the production efficiency and economic benefits. At the same time, by reducing human intervention and making real-time adjustments, while optimizing the electrolysis process, it also improves the automation level and enhances the intelligent management ability of aluminum electrolysis production.

[0132] The present invention provides a Figure 2 continuous pre-baked anode aluminum electrolysis production system as shown, including a real-time monitoring and data acquisition module, a data summary and preprocessing module, a key feature extraction and analysis module, a deep learning intelligent evaluation module, and a dynamic current regulation and optimization module:

[0133] Real-time monitoring and data acquisition module. First, the real-time monitoring system collects the current density distribution data in the electrolytic cell and the distribution state of the electrolyte, which can reflect the current and electrolyte conditions in each area of the electrolytic cell in real time, providing data support for subsequent adjustment measures;

[0134] Data aggregation and preprocessing module. Aggregate the collected data to form a complete data set and preprocess the data in the data set;

[0135] Key feature extraction and analysis module. Extract the key features reflecting the uneven distribution of the electrolyte in the electrolytic cell from the preprocessed data, analyze the extracted key features within the detection window, and quantify the degree of uneven distribution of the electrolyte in the electrolytic cell;

[0136] Deep learning intelligent evaluation module. Input the key features after analysis into a pre-trained deep learning model, and use the deep learning model to intelligently evaluate whether there is uneven distribution of the electrolyte in the electrolytic cell;

[0137] Dynamic current regulation and optimization module. When it is detected that the electrolyte distribution in the electrolytic cell is uneven, take dynamic current regulation measures. In the area with high current density, reduce the current input to prevent excessive consumption of the anode and avoid local overheating; while in the area with low current density, automatically increase the current flow to ensure uniform current distribution throughout the electrolytic cell.

[0138] A continuous pre-baked anode aluminum electrolysis production method provided by an embodiment of the present invention is realized through the above-mentioned continuous pre-baked anode aluminum electrolysis production system. The specific methods and processes of a continuous pre-baked anode aluminum electrolysis production system are detailed in the embodiments of the above-mentioned continuous pre-baked anode aluminum electrolysis production method, which will not be elaborated here.

[0139] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0140] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

[0141] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0142] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0143] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0144] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0145] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0146] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0147] As described above, this is only the specific implementation of the present application. However, the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the said claims.

[0148] Only some exemplary embodiments of the present invention have been described by way of illustration above. Without doubt, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

Claims

1. A method for producing a continuous prebaked anode aluminum electrolysis, characterized in that: The following steps are involved: First, the real-time monitoring system collects the current density distribution data and the distribution status of the electrolyte in the electrolytic cell, reflecting the current and electrolyte conditions in each area of ​​the electrolytic cell in real time, providing data support for subsequent adjustment measures; Aggregate the collected data to form a complete data set, and pre-process the data in the data set; Extract key features reflecting the uneven distribution of electrolyte in the electrolytic cell from the preprocessed data, analyze the extracted key features within the detection window, and quantify the degree of uneven distribution of electrolyte in the electrolytic cell; The analyzed key features are input into a pre-trained deep learning model, and the deep learning model is used to intelligently evaluate whether there is uneven distribution of electrolyte in the electrolytic cell; When uneven electrolyte distribution in the electrolytic cell is detected, dynamic current regulation measures are taken to reduce current input in areas with high current density to prevent excessive consumption of the anode and avoid local overheating; while in areas with low current density, the current flow is automatically increased to ensure uniform current distribution in the entire electrolytic cell.

2. A method for producing a continuous prebaked anode aluminum electrolysis according to claim 1, characterized in that: The specific steps of collecting the current density distribution data and the distribution status of the electrolyte in the electrolytic cell through the real-time monitoring system and reflecting the current and electrolyte conditions in each area of ​​the electrolytic cell in real time are as follows: First, multiple sensors are arranged in the electrolytic cell. The sensors are distributed in different positions of the electrolytic cell to ensure full coverage of all areas of the cell. Then, the sensor collects the current density, temperature and electrolyte concentration data of each area in real time, and transmits the acquired data to the central control unit through the data acquisition system; Finally, the central control unit processes, stores and analyzes the received real-time data to ensure that the operating status of each area in the electrolyzer can be dynamically monitored and provide real-time feedback for subsequent optimization control.

3. The method for producing a continuous prebaked anode aluminum electrolysis according to claim 1, characterized in that: Key features reflecting the uneven distribution of electrolyte in the electrolytic cell are extracted from the preprocessed data, including the degree of difference in current density in different areas of the electrolytic cell and the degree of electrolyte coverage in different areas of the electrolytic cell. Under the detection window, the degree of difference in current density in different areas of the electrolytic cell and the degree of electrolyte coverage in different areas of the electrolytic cell are analyzed, and local current density difference reference values ​​and electrolyte coverage reference values ​​are generated respectively. The electrolyte distribution in each area of ​​the electrolytic cell is quantified through the local current density difference reference values ​​and the electrolyte coverage reference values, thereby revealing the uneven distribution of electrolyte and its influence on the current density.

4. A method for producing a continuous prebaked anode aluminum electrolysis according to claim 3, characterized in that: The specific steps for analyzing the difference in current density between different areas in the electrolytic cell under the detection window to generate a reference value for the local current density difference are as follows: The current density in different areas of the electrolytic cell is carefully analyzed. First, the electrolytic cell is divided into several sub-areas. The current density of each sub-area is calculated by collecting data from the current sensor. The calculation formula is as follows: ， In the formula, J i is the current density, which indicates the current intensity in the i-th region, I i is the current value of the ith region, A i is the current region surface area; Next, we quantify the difference in current density between different regions in the electrolytic cell. The difference between the current density of each region and the current density of its adjacent regions is calculated to reflect the inhomogeneity of the electrolyte distribution. The inhomogeneity measurement formula is as follows: ， Where, ΔJ i is the current density non-uniformity measure of the ith region, which is used to quantify the difference between the current density of the ith region and its adjacent regions, reflecting the non-uniformity of the current density in the region. Ni is the set of neighboring regions of the ith region, and J j is the current intensity in the jth region, ∈ is the smoothing factor, which is a small constant; Finally, in order to comprehensively consider the impact of current density differences on the operation of the electrolyzer, the current density difference value of each area is weighted with the electrolyte concentration coverage of the current area to generate the final local current density difference reference value. The generation formula is as follows: ， Where LCDD is the reference value of local current density difference, n is the total number of regions divided by the electrolytic cell, C i is the electrolyte concentration in the ith region, C k is the electrolyte concentration in the kth region.

5. The method for producing a continuous prebaked anode aluminum electrolysis according to claim 3, characterized in that: The specific steps for analyzing the coverage of electrolyte in different areas of the electrolytic cell under the detection window to generate the reference value of electrolyte coverage are as follows: First, it is necessary to calculate the coverage difference of the electrolyte in each area of ​​the electrolytic cell. The coverage difference reflects the uneven distribution of the electrolyte in each area. The local coverage degree of the electrolyte concentration at the coordinate point is expressed by the local coverage function. The calculation expression is as follows: ， In the formula, C i x, y are the local electrolyte coverage of the ith region in the electrolytic cell, x, y are the coordinate points, E i x, y are the electrolyte concentrations in the ith region of the electrolytic cell, E max is the maximum theoretical concentration of electrolyte in the current region; In order to obtain the electrolyte local coverage C i After x and y, the local coverage difference index between each region and its surrounding areas is further calculated. The local coverage difference index measures the degree of difference in electrolyte distribution between regions. The calculation expression is as follows: ， Where, ΔC i is the local coverage difference index, which indicates the electrolyte coverage difference of the ith region, Ni is the set of neighboring regions of the ith region, and C j x, y is the local electrolyte coverage of the jth region in the electrolytic cell, |C i x,yC j x,y| α is the coverage difference weighting, α is the weight coefficient controlling the local difference; Finally, the reference value of electrolyte coverage is obtained by comprehensively considering the local differences in each area and performing weighted calculation. The calculation expression is as follows: The formula is as follows: ， Where ECR is the reference value of electrolyte coverage, n is the total number of regions divided into the electrolytic cell, maxΔC is the maximum value of the local coverage difference index, and w i is the weight of region i, and β is a weighting exponent that controls the importance of global differences.

6. The method for producing a continuous prebaked anode aluminum electrolysis according to claim 3, characterized in that: The local current density difference reference value and electrolyte coverage reference value generated after analysis are input into the pre-learned deep learning model, and the electrolyte uneven distribution state coefficient is generated by the deep learning model. The electrolyte distribution state in the electrolytic cell is evaluated by the electrolyte uneven distribution state coefficient, and the electrolyte distribution state in the electrolytic cell is intelligently evaluated.

7. A method for producing continuous prebaked anode aluminum electrolysis according to claim 6, characterized in that: The electrolyte distribution uneven state coefficient generated when evaluating the electrolyte distribution state in the electrolytic cell through the pre-learned deep learning model is compared and analyzed with the pre-set electrolyte distribution uneven state coefficient reference threshold, and the electrolyte distribution state in the electrolytic cell is intelligently evaluated. The specific steps are as follows: If the electrolyte uneven distribution state coefficient is greater than a preset electrolyte uneven distribution state coefficient reference threshold, the electrolyte distribution state in the electrolytic cell is classified as uneven distribution; if the electrolyte uneven distribution state coefficient is less than or equal to a preset electrolyte uneven distribution state coefficient reference threshold, the electrolyte distribution state in the electrolytic cell is classified as uniform distribution.

8. The method for producing continuous prebaked anode aluminum electrolysis according to claim 7, characterized in that: When uneven electrolyte distribution in the electrolytic cell is detected, the specific steps for taking dynamic current regulation measures are as follows: First, the electrolyte uneven distribution state coefficient EDUS generated by the deep learning model is compared and analyzed with the preset reference threshold. When the electrolyte uneven distribution state coefficient EDUS is greater than the preset reference threshold, it means that the electrolyte is unevenly distributed. At this time, the dynamic current regulation mechanism is triggered and the current adjustment amplitude is calculated. The calculation expression is as follows: ， Where ΔI is the current adjustment range, θ is the adjustment coefficient, and EDUS ref is the reference threshold value of electrolyte inhomogeneous distribution coefficient, Based on the evaluation results, the current input is reduced in the areas with high current density in the electrolytic cell to prevent excessive anode consumption and avoid local overheating. The current reduction is calculated as follows: ， In the formula, ΔI high is the current adjustment value in the high current density area, ω is the current adjustment coefficient, which indicates the sensitivity of current adjustment, J high is the actual current density in the high current density region, J ref is the reference value of current density, is the current density adjustment index, γ is the heat load adjustment coefficient, T high is the temperature of the region with high current density, T ref is the temperature of the current density reference region; In areas with low current density, the current flow rate is automatically increased to ensure uniform current distribution in the electrolyzer. The magnitude of the current increase is calculated by the following formula: ， In the formula, ΔI low is the current adjustment value in the low current density region, J low is the actual current density in the low current density area, δ is the temperature adjustment coefficient, which indicates the effect of temperature on current adjustment, T low is the temperature in the area of ​​low current density.

9. A continuous prebaked anode aluminum electrolysis production system, used to implement a continuous prebaked anode aluminum electrolysis production method as described in any one of claims 1 to 8, characterized in that: It includes real-time monitoring and data acquisition module, data aggregation and preprocessing module, key feature extraction and analysis module, deep learning intelligent evaluation module and dynamic current regulation and optimization module: Real-time monitoring and data acquisition module: First, the real-time monitoring system collects the current density distribution data and the distribution status of the electrolyte in the electrolytic cell, reflects the current and electrolyte conditions in each area of ​​the electrolytic cell in real time, and provides data support for subsequent adjustment measures; The data aggregation and preprocessing module aggregates the collected data to form a complete data set, and preprocesses the data in the data set; The key feature extraction and analysis module extracts the key features reflecting the uneven distribution of electrolyte in the electrolytic cell from the preprocessed data, analyzes the extracted key features within the detection window, and quantifies the degree of uneven distribution of electrolyte in the electrolytic cell; The deep learning intelligent evaluation module inputs the analyzed key features into the pre-trained deep learning model, and uses the deep learning model to intelligently evaluate whether there is uneven distribution of electrolytes in the electrolytic cell; Dynamic current regulation and optimization module: When uneven electrolyte distribution in the electrolytic cell is detected, dynamic current regulation measures are taken to reduce current input in areas with high current density to prevent excessive consumption of the anode and avoid local overheating; in areas with low current density, the current flow is automatically increased to ensure uniform current distribution in the entire electrolytic cell.

Images