A method and device for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells.

Abstract

This application relates to the field of aluminum electrolysis technology, and provides a method and apparatus for online measurement of anode current and early warning of anode effects in aluminum electrolysis cells. The method includes: acquiring regional current data; calculating the deviation of regional current data for each divided anode current acquisition zone and marking candidate abnormal regions; acquiring real-time operation plan data, determining whether the candidate abnormal regions are within the execution period of the operation plan data, and if so, determining whether the current fluctuation is caused by normal process operation based on the characteristic waveform corresponding to that period, and unmarking the candidate abnormal regions that match the normal process operation; for the marked and unmarked candidate abnormal regions, acquiring the current data at their non-adjacent spatially symmetrical positions in the electrolysis cell, calculating the deviation value of the current change between the candidate abnormal region and the spatially symmetrical position, and identifying the true precursors of local anode effects. This effectively improves the current efficiency and operational stability of aluminum electrolysis production.

Description

Technical Field

[0001] This invention relates to the field of aluminum electrolysis technology, and in particular to a method and device for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells. Background Technology

[0002] Aluminum electrolysis production employs the Hall-Heroo process, where alumina is dissolved in a high-temperature molten cryolite electrolyte, and direct current is applied to initiate the electrolysis reaction. Metallic aluminum is deposited at the cathode, and carbon dioxide gas is generated at the anode. During electrolysis, alumina concentration is a key parameter affecting current efficiency and cell stability. Existing technologies employ a control strategy of alternating underfeed and overfeed cycles to maintain the alumina concentration in the electrolytic cell. Specifically, by monitoring the overall cell resistance (pseudo-resistance) and using the pseudo-resistance-alumina concentration relationship curve, the average alumina concentration of the entire cell is indirectly estimated, thereby controlling the amount of alumina added. When the cell resistance increases to a preset upper limit, the alumina concentration is considered too low, and the system switches to overfeed mode; when the cell resistance decreases to a preset lower limit, the alumina concentration is considered too high, and the system switches to underfeed mode, thus maintaining the overall alumina concentration within a target range.

[0003] In actual large-scale aluminum electrolysis cells, the anode current distribution varies significantly across different areas due to factors such as anode replacement, aluminum tapping operations, and the busbar conductive structure, resulting in inconsistent alumina consumption rates in different areas. When the alumina concentration in a localized area prematurely drops below a critical value, the current path in that area is blocked by an anode effect, creating a gas curtain. However, because the existing technology relies on the average cell resistance signal, which only reflects the overall average alumina concentration of the electrolysis cell, the average cell resistance may not show a significant change when the aforementioned anomaly occurs in a localized area. Therefore, existing control methods cannot promptly identify this localized precursor, causing the localized anode effect to rapidly spread into a high-voltage anode effect across the entire cell. This leads to a sharp increase in cell voltage, a significant decrease in current efficiency, a dramatic increase in energy consumption, and even the generation of large amounts of perfluorocarbons (PFCOs), a greenhouse gas. Summary of the Invention

[0004] This invention provides an online measurement method and device for the anode current of an aluminum electrolytic cell and an early warning method for the anode effect. It is used to solve the problem that the lack of accurate perception of abnormal alumina concentration in local areas and precursors of the anode effect leads to the inability to provide effective early warning and intervention before the anode effect occurs.

[0005] The first aspect of this application provides a method for online measurement of the anode current in an aluminum electrolytic cell and early warning of the anode effect, including: Obtain the regional current data corresponding to each anode conductor after calibration with the reference calibration electrode; The multiple current sensing units corresponding to the regional current data are divided into a preset number of anode current acquisition partitions. For each acquisition partition, the deviation of the regional current data is calculated based on the historical sliding baseline. If the deviation exceeds a first preset threshold and the duration is greater than a first preset duration, it is marked as a candidate abnormal region. Obtain real-time operation plan data of aluminum electrolysis cell, determine whether the candidate abnormal area is currently within the execution period of the operation plan data; if so, determine whether the candidate abnormal area is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and remove the mark of the candidate abnormal area that matches the normal process operation. For candidate abnormal regions that have been marked but not unmarked, obtain the current data of the candidate abnormal regions at non-adjacent spatially symmetrical positions in the electrolytic cell; calculate the deviation between the current change in the candidate abnormal regions and the current change at the spatially symmetrical positions; if the deviation is greater than a second preset threshold, it is determined to be a precursor to a real local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. When the presence of precursors to a local anodic effect is confirmed, an anodic effect warning signal is generated and output.

[0006] Optionally, acquiring the region current data corresponding to each anode conductor after calibration with the reference calibration electrode includes: The calibration poles are measured using a calibration probe to obtain the calibration reference parameters; Each anode guide rod is self-calibrated online according to the calibration reference parameters to eliminate thermal drift and installation tolerances. The measurement action after self-calibration acquires the regional current data corresponding to each anode conductor.

[0007] Optionally, the online self-calibration of each anode guide rod according to the calibration reference parameters includes: After the pole-changing operation is completed or during periodic inspection, the calibration probe is triggered to measure the reference calibration pole. A correction coefficient matrix is ​​generated based on the calibration reference parameters; The correction coefficient matrix is ​​used to correct the regional current data corresponding to each anode conductor in real time.

[0008] Optionally, calculating the deviation of the regional current data based on a historical sliding baseline includes: Acquire current data from multiple historical regions for each acquisition partition within a preset time window; The dynamic baseline value of the acquisition zone is calculated based on the current data from the multiple historical regions; Calculate the relative deviation rate between the current area current data and the dynamic baseline value as the deviation magnitude; wherein, the relative deviation rate is the ratio of the difference between the current area current data and the dynamic baseline value to the dynamic baseline value.

[0009] Optionally, determining whether the candidate abnormal region is currently within the execution period of the job plan data includes: Obtain the physical location identifier of the candidate anomaly region; Based on the physical location identifier, match the corresponding job plan item from the real-time job plan data; Determine whether the current time is within the planned execution time period of the job plan item; if so, determine that the candidate abnormal region is within the execution time period of the job plan data.

[0010] Optionally, determining whether the candidate abnormal region is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period includes: The execution period is divided into at least two consecutive judgment sub-periods, where different judgment sub-periods correspond to different expected current fluctuation characteristics; the judgment sub-periods are divided based on the execution stage of the work plan item, including an initial stabilization stage, a mid-term fluctuation stage, and a final recovery stage; Determine the sub-time period to which the current moment belongs; Obtain the normal current fluctuation characteristic parameters corresponding to the judgment sub-period to which the current time belongs; Calculate the current fluctuation characteristic value of the candidate abnormal region and compare it with the normal current fluctuation characteristic parameter; If the current fluctuation characteristic value matches the normal current fluctuation characteristic parameter, it is determined to be a current fluctuation caused by normal process operation.

[0011] Optionally, the step of calculating the current fluctuation characteristic value of the candidate abnormal region and comparing it with the normal current fluctuation characteristic parameter includes: Obtain the preset threshold range corresponding to the current sub-period of judgment; Determine whether the current change amplitude of the candidate abnormal region is within the preset threshold range; If yes, it is determined to be a match, and the current fluctuation is determined to be caused by normal process operation; if no, it is determined to be a mismatch, and the candidate abnormal area is retained as a state to be verified.

[0012] Optionally, the spatially symmetrical position is an anode current acquisition zone that is spatially symmetrical with respect to the symmetrical center line or symmetrical center point of the aluminum electrolytic cell and has no adjacent relationship with the candidate abnormal region.

[0013] Optionally, calculating the deviation between the current change in the candidate abnormal region and the current change at the spatially symmetrical location includes: Obtain the historical current difference between the candidate abnormal region and the spatially symmetrical position within a preset historical time window; The baseline deviation value is determined based on the historical current difference. Calculate the current current difference between the candidate anomaly region and the spatially symmetrical position at the current moment; The absolute value of the difference between the current current difference and the reference deviation value is taken as the deviation value.

[0014] The second aspect of this application provides an online measurement device for the anode current of an aluminum electrolytic cell and an anode effect early warning device, comprising: The regional current data acquisition module is used to acquire the regional current data corresponding to each anode conductor after calibration by the reference calibration electrode; The deviation magnitude calculation and judgment module is used to divide the multiple current sensing units corresponding to the regional current data into a preset number of anode current acquisition partitions. For each acquisition partition, the deviation magnitude of the regional current data is calculated based on the historical sliding baseline. If the deviation magnitude exceeds a first preset threshold and the duration is greater than a first preset duration, it is marked as a candidate abnormal region. The execution period determination module is used to acquire real-time operation plan data of aluminum electrolysis cells, determine whether the candidate abnormal area is currently within the execution period of the operation plan data; if so, it determines whether the candidate abnormal area is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and removes the mark of the candidate abnormal area that matches the normal process operation. The deviation value calculation and judgment module is used to obtain the current data of the candidate abnormal region at a non-adjacent spatially symmetrical position in the electrolytic cell for the marked and unremoved candidate abnormal region; calculate the deviation value between the current change of the candidate abnormal region and the current change at the spatially symmetrical position; if the deviation value is greater than a second preset threshold, it is determined to be a precursor to a real local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. The anode effect early warning signal generation module is used to generate and output an anode effect early warning signal when the presence of local anode effect precursors is confirmed.

[0015] As can be seen from the above technical solutions, this application has the following advantages: This application performs online self-calibration of the regional current data of each anode conductor using a reference calibration electrode, eliminating the influence of thermal drift and installation tolerances on measurement accuracy. Multiple current sensing units are divided into a preset number of anode current acquisition zones. For each acquisition zone, the deviation of the regional current data is calculated based on a historical sliding baseline, and candidate abnormal regions are initially marked. Real-time operation plan data of the aluminum electrolysis cell is acquired to determine whether the candidate abnormal regions are within the execution period of the operation plan data. If so, the current fluctuations are determined based on the characteristic waveform corresponding to that period to determine whether they are caused by normal process operations, and the marking of candidate abnormal regions matching normal process operations is removed. For marked candidate abnormal regions that have not been removed from the marking, the current data at their non-adjacent spatially symmetrical positions in the electrolysis cell is acquired. By calculating the deviation value of the current change between the candidate abnormal region and the spatially symmetrical position, the true precursors of local anode effects are accurately identified, and global current fluctuations are excluded. Compared with existing technologies, this application achieves rapid localization of abnormal areas by calculating and marking deviation amplitudes in different regions; by introducing real-time operation plan data and time-segmented characteristic waveform matching, it accurately eliminates interference caused by normal process actions such as electrode switching and aluminum tapping, significantly reducing the false alarm rate; and by using double-ended verification based on spatially symmetrical position current deviation values, it achieves high-precision identification of precursors to local anode effects, effectively distinguishing between local anomalies and global current fluctuations. In summary, this application can accurately sense local precursor signals before the occurrence of anode effects, significantly improving the accuracy and timeliness of anode effect early warning, thereby effectively preventing local anode effects from spreading to full-cell high-voltage anode effects, reducing energy consumption, reducing perfluorocarbon emissions, and improving the current efficiency and operational stability of aluminum electrolysis production. Attached Figure Description

[0016] Figure 1 A flowchart illustrating an online measurement method for the anode current of an aluminum electrolytic cell and an early warning method for the anode effect provided in this application; Figure 2 A flowchart of online self-calibration and data acquisition of anode current based on a reference calibration electrode provided for this application; Figure 3 A flowchart for calculating the regional current deviation magnitude based on a historical sliding baseline is provided for this application. Figure 4 A flowchart for matching candidate abnormal areas with work plan time periods provided in this application; Figure 5 A flowchart for determining normal process actions based on segmented feature matching provided in this application; Figure 6 The flowchart for calculating the deviation value based on the spatial symmetry position difference provided in this application. Detailed Implementation

[0017] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding to,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] Example 1 This application provides an online measurement method for the anode current of an aluminum electrolytic cell and an early warning method for anode effects. This method can be applied to the field of cell condition monitoring and intelligent control in the aluminum electrolysis production process, and is particularly suitable for large prebaked anode aluminum electrolytic cells. The method in this application embodiment can be executed by an intelligent control system for the electrolytic cell integrating a data acquisition module, a processing module, and a control output module. This system can be a standalone industrial control computer or a software module or logic circuit embedded in the electrolytic cell control system. This method combines hardware measurement with software algorithms to achieve accurate measurement of the anode current and early identification of precursors to local anode effects, thereby providing reliable data support and decision-making basis for the intelligent control of aluminum electrolytic cells. Please refer to [link to relevant documentation]. Figure 1 The online measurement of anode current and early warning method for anode effect in aluminum electrolysis cells provided in this application embodiment includes the following steps: S1. Obtain the area current data corresponding to each anode conductor after calibration with the reference calibration electrode; This step involves setting a reference calibration electrode at the top of the electrolytic cell. Using this electrode as a reference, the current sensing unit mounted on the anode rod is self-calibrated online to eliminate measurement deviations caused by factors such as high temperature, strong magnetic field interference, and installation tolerances, thereby obtaining high-precision regional current data. Please refer to [link / reference]. Figure 2 The implementation process includes the following: S11. Use the calibration probe to measure the reference calibration pole and obtain the calibration reference parameters; This calibration probe is a high-precision current measuring device with a measurement accuracy better than 0.1%. It can be mounted on the upper part of the electrolytic cell in either a movable or fixed manner. In calibration mode, the calibration probe is driven close to or in contact with the reference calibration electrode to measure it and obtain the calibration reference parameters under the current environment. The calibration reference parameters include, but are not limited to, the magnetic field strength, temperature value, and the corresponding data of the theoretical current value at the location of the reference calibration electrode. These calibration reference parameters are used as a reference for subsequent correction of the measurement data of each anode conductor.

[0019] S12. Perform online self-calibration of each anode guide rod according to the calibration reference parameters to eliminate thermal drift and installation tolerances; 1. After the pole-changing operation is completed or during periodic inspection, trigger the calibration probe to measure the reference calibration pole; 2. Generate a correction coefficient matrix based on the calibration reference parameters; 3. The current data of the corresponding area of ​​each anode conductor is corrected in real time using the correction coefficient matrix.

[0020] Specifically, after the electrode replacement operation is completed and the new anode is just put into the electrolytic cell, there may be slight deviations in the installation position and contact state of its guide rod; during periodic inspections, due to long-term operation, the sensor may drift due to the cumulative effect of temperature. The above two times are the optimal time windows for calibration.

[0021] It should be noted that the correction coefficient matrix is ​​a diagonal matrix, which is used to correct the measurement system error of each current sensing unit. Since the installation tolerance and thermal drift characteristics of each current sensing unit are position-independent, the sensor error of each guide rod is determined only by its own installation state and temperature environment, and is independent of the current changes of adjacent guide rods. Therefore, a diagonal matrix is ​​used to independently correct each sensor.

[0022] The diagonal elements of the correction coefficient matrix correspond to the correction coefficients for each anode conductor. Each correction coefficient is calculated as follows: based on the calibration reference parameters measured by the calibration probe at the reference calibration electrode, and combined with the known theoretical current value of that reference calibration electrode, a reference correction factor is determined for the current environment; for each current sensing unit on the anode conductor, its original measurement value under the same calibration environment is obtained; the reference correction factor is compared with the original measurement value of the sensor to calculate the correction coefficient for that sensor. This correction coefficient reflects the relative degree to which the sensor deviates from its ideal state due to installation tolerances or thermal drift.

[0023] By using the aforementioned correction coefficient matrix, the original measured current value of each anode conductor is multiplied by the correction coefficient at the corresponding position to obtain the corrected current data, thereby eliminating measurement errors caused by installation tolerances or thermal drift.

[0024] S13. Obtain the area current data corresponding to each anode conductor through the self-calibrated measurement action.

[0025] After completing the above calibration steps, the corrected current sensing unit continuously collects the anode conductor current. The regional current data includes the real-time current value, current change rate, and current waveform characteristic values ​​for each anode conductor. This data is transmitted from the data acquisition module to the processing module for subsequent anomaly detection and early warning analysis.

[0026] S2. Divide the multiple current sensing units corresponding to the regional current data into a preset number of anode current acquisition partitions. For each acquisition partition, calculate the deviation of the regional current data based on the historical sliding baseline. The preset number of anode rods is determined based on the total number of anode rods and the structure of the aluminum electrolysis cell. Multiple anode rods are grouped by physical location, with each group containing 8 to 12 consecutively arranged anode rods, thus forming a preset number of anode current acquisition zones. For example, an electrolysis cell with 48 anode rods can be divided into 6 acquisition zones, each corresponding to 8 rods. The principle for dividing the acquisition zones is that the anode rods within the same zone are physically adjacent and their current changes have a high correlation; different zones are spatially independent, facilitating subsequent regional comparative analysis. Please refer to [link / reference]. Figure 3 After the partitioning is completed, perform the following steps for each acquisition partition: S21. Acquire current data from multiple historical regions within a preset time window for each acquisition partition; Specifically, the preset time window is set to a length of 30 minutes to 2 hours based on the typical feeding cycle of the aluminum electrolysis cell and the rate of change in alumina concentration. This window length can cover at least two complete feeding cycles, thus effectively reflecting the normal current fluctuation range of the sampling area. It should be noted that the multiple historical area current data obtained here are different from the area current data corresponding to each anode rod obtained in step S1; the data obtained in S1 is the real-time current data of each anode rod at the current moment, used for real-time monitoring; while the historical area current data obtained here is the historical current data of the sampling area over a period of time, used to establish a baseline reference for normal operation. The relationship between the two is that historical data provides a benchmark for comparison with real-time data, while real-time data provides current status information for anomaly judgment.

[0027] S22. Calculate the dynamic baseline value of the acquisition zone based on current data from multiple historical regions; The dynamic baseline value is a current reference value for each data acquisition zone, calculated statistically based on current data from multiple historical regions within a preset time window. Specifically, the dynamic baseline value is determined by calculating the arithmetic mean or median of the current data from multiple historical regions. The arithmetic mean is calculated by adding the historical current data and dividing by the number of data points; the median is calculated by sorting the historical current data by their numerical values ​​and taking the middle value. Compared to a fixed threshold, the dynamic baseline value can adaptively adjust to the gradual changes in the electrolyzer's operating conditions. For example, when the electrolyzer is in different production stages or the anode is replaced, the dynamic baseline value can be updated accordingly, thus maintaining the sensitivity and accuracy of anomaly detection.

[0028] S23. Calculate the relative deviation rate between the current area current data and the dynamic baseline value as the deviation magnitude; where the relative deviation rate is the ratio of the difference between the current area current data and the dynamic baseline value to the dynamic baseline value.

[0029] The relative deviation rate is a dimensionless indicator used to quantify the degree of deviation of current data in a given area from the normal operating baseline. The formula for calculating the relative deviation rate is: Relative Deviation Rate = (Current Area Current Data - Dynamic Baseline Value) / Dynamic Baseline Value. This relative deviation rate eliminates the absolute value differences caused by baseline value variations between different acquisition zones, making the deviation degrees of different zones comparable. A positive relative deviation rate indicates that the current current is higher than the dynamic baseline value; a negative relative deviation rate indicates that the current current is lower than the dynamic baseline value.

[0030] S3. If the deviation exceeds the first preset threshold and the duration is greater than the first preset duration, it is marked as a candidate abnormal region; The first preset threshold is determined using statistical analysis based on historical relative deviation rate data under normal operating conditions. For example, relative deviation rate data is collected over 24 hours under normal electrolytic cell operating conditions, and its mean and standard deviation are calculated. The first preset threshold is set to the mean plus three times the standard deviation, ensuring that the false alarm rate is below 0.3% within the normal fluctuation range. Exceeding the first preset threshold means that the absolute value of the relative deviation rate is greater than the first preset threshold. Since the relative deviation rate can be positive or negative, whether the current is too high or too low, as long as the deviation exceeds the threshold range, it is considered an abnormal deviation. The first preset duration is set to 3 to 5 seconds based on the typical duration of anode effect precursors and the current fluctuation characteristics of normal operations such as electrode switching and aluminum tapping. The purpose of setting this duration is to avoid false alarms caused by instantaneous current disturbances or acquisition noise. A duration greater than the first preset duration means that the relative deviation rate continuously exceeds the first preset threshold for a period longer than the first preset duration. In other words, the system will continuously monitor each collection partition. Only when the relative deviation rate of the partition exceeds the first preset threshold for a continuous time period, and the length of the continuous time period is greater than the first preset duration, will it be determined as an abnormal state that meets the conditions.

[0031] When the relative deviation rate of a certain acquisition zone exceeds a first preset threshold for a continuous time period, and the length of this continuous time period is greater than a first preset duration, the zone is determined to have experienced a significant abnormal current deviation and is marked as a candidate abnormal region. Candidate abnormal regions are acquisition zones that, after initial screening, may exhibit precursory signs of an anode effect. Subsequent steps will further verify these regions to confirm whether they are genuine precursors.

[0032] S4. Obtain real-time operation plan data of aluminum electrolysis cells and determine whether the candidate abnormal area is currently within the execution period of the operation plan data; Real-time operation plan data is a pre-defined set of data containing the time arrangements for various process operations during aluminum electrolysis production. This data includes, but is not limited to, electrode replacement operation plans, aluminum tapping operation plans, and cell condition adjustment operation plans. The electrode replacement operation plan records the planned replacement time, replacement location, and operation duration for each anode guide rod; the aluminum tapping operation plan records the time, tapping location, and expected aluminum tapping quantity for each operation. This real-time operation plan data is acquired in real-time through the aluminum electrolysis cell production management system or periodically retrieved from the database. Please refer to [link to relevant documentation]. Figure 4 After obtaining this data, the following judgment steps are performed on the candidate abnormal regions: S41. Obtain the physical location identifier of the candidate anomaly region; The physical location identifier is an identification information used to uniquely determine the physical location of candidate anomaly areas within the electrolytic cell. This physical location identifier can be the number of the anode guide rod or the number of the sampling zone; for example, "A-03" indicates the 3rd guide rod in region A. This number corresponds one-to-one with the physical layout of the electrolytic cell.

[0033] S42. Match the corresponding job plan item from the real-time job plan data based on the physical location identifier; A work plan item is a specific operational task associated with a particular physical location within real-time work plan data. For example, a record in the real-time work plan data stating "Perform a pole-changing operation on guide rod B-05 between 14:00 and 14:10" is a work plan item. Specifically, the physical location identifier of a candidate anomaly area is compared with the operational location recorded in the work plan item. If they match, it is determined that a corresponding work plan item exists for that candidate anomaly area during that time period.

[0034] S43. Determine whether the current time is within the planned execution time period of the work plan item; Specifically, the current system time is obtained and compared with the planned start and end times of the matched job schedule. If the current time is greater than or equal to the planned start time and less than or equal to the planned end time, it is determined that the current time is within the planned execution period of the job schedule; otherwise, it is determined that it is not.

[0035] S44. If so, then determine that the candidate abnormal region is within the execution period of the job plan data.

[0036] If the current timeframe falls within the planned execution period of the matched job schedule item, then the candidate anomaly region is determined to be within the execution period of the job schedule data. This conclusion indicates that the current fluctuation in the candidate anomaly region may be caused by normal process operations, and further judgment is needed in subsequent steps to confirm whether the current fluctuation is caused by normal process operations. Conversely, if the judgment result of S43 is negative, then the candidate anomaly region is determined to be outside the execution period of the job schedule data. This conclusion indicates that the current fluctuation in the candidate anomaly region is unrelated to normal process operations and needs to be handled as an anomaly.

[0037] S5. If so, determine whether the candidate abnormal region is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and remove the mark from the candidate abnormal region that matches the normal process operation. When a candidate anomaly region falls within the execution period of the work plan data, its current fluctuations may be caused by normal process operations or by precursors of localized anodic effects. By dividing the execution period into different judgment sub-periods and setting corresponding normal current fluctuation characteristic parameters for each sub-period, the current fluctuation characteristic values ​​of the candidate anomaly region are compared with the normal characteristic parameters, thereby accurately distinguishing between current fluctuations caused by normal process operations and precursors of anomalies. Please refer to [link / reference]. Figure 5 The implementation process includes the following: S51. Divide the execution period into at least two consecutive judgment sub-periods, where different judgment sub-periods correspond to different expected current fluctuation characteristics; the judgment sub-periods are divided based on the execution stage of the work plan item, including the initial stabilization stage, the intermediate fluctuation stage, and the final recovery stage; It should be noted that the relationship between "two consecutive judgment sub-periods" and "initial stabilization phase, intermediate fluctuation phase, and final recovery phase" is as follows: The execution period can be divided into two consecutive judgment sub-periods, and each judgment sub-period can be further subdivided into multiple execution phases. For example, the first judgment sub-period can include the initial stabilization phase, and the second judgment sub-period can include the intermediate fluctuation phase and the final recovery phase; alternatively, the execution period can be divided into three consecutive judgment sub-periods, corresponding to the initial stabilization phase, the intermediate fluctuation phase, and the final recovery phase, respectively. In this embodiment, the division into three consecutive judgment sub-periods is preferred, that is, each judgment sub-period corresponds to one execution phase, so as to more precisely match the different current fluctuation characteristics of each phase.

[0038] Expected current fluctuation characteristics refer to the typical patterns of current fluctuations caused by normal process operations within different judgment sub-periods. For example: Initial stabilization phase: When the pole switching operation is just completed, the new anode has not yet been fully turned on. The expected current fluctuation characteristics during this phase are that the current rises slowly and the fluctuation amplitude is small. Mid-term fluctuation stage: During the pole switching operation, the anode conductor is raised or lowered. The expected current fluctuation characteristics during this stage are violent current fluctuations with large fluctuation amplitude. Final recovery phase: The pole-switching operation is nearing completion, and the current gradually stabilizes. The expected current fluctuation characteristics in this phase are that the current slowly recovers to normal levels, and the fluctuation amplitude gradually decreases.

[0039] S52. Determine the sub-time period to which the current time belongs; The current time is the point in time when the system is monitoring in real time, specifically the point at which it is judging candidate anomaly areas. Specifically, the difference between the current time and the planned start time of the work schedule item is obtained, and the corresponding judgment sub-period is determined based on the range of this difference. For example, if the total duration of the planned execution period is 30 minutes, the first 10 minutes are defined as the initial stabilization phase, the middle 10 minutes as the intermediate fluctuation phase, and the last 10 minutes as the final recovery phase. If the difference between the current time and the start time is 5 minutes, the current time is determined to belong to the judgment sub-period corresponding to the initial stabilization phase.

[0040] S53. Obtain the normal current fluctuation characteristic parameters corresponding to the judgment sub-period to which the current time belongs; Normal current fluctuation characteristic parameters are quantitative parameters pre-defined for each judgment sub-period to characterize the current fluctuation characteristics caused by normal process operations within that period. These parameters include, but are not limited to, the preset threshold range, the expected current change amplitude range, and the expected current change slope range corresponding to that judgment sub-period. The corresponding normal current fluctuation characteristic parameters are retrieved from a pre-stored characteristic parameter database based on the identifier of the current judgment sub-period.

[0041] S54. Calculate the current fluctuation characteristic value of the candidate abnormal region and compare it with the normal current fluctuation characteristic parameter; 1. Obtain the preset threshold range corresponding to the current judgment sub-period; 2. Determine whether the current change amplitude in the candidate abnormal region is within the preset threshold range; 3. If yes, it is determined to be a match, and the current fluctuation is determined to be caused by normal process operation; if no, it is determined to be a mismatch, and the candidate abnormal area is retained as a state to be verified.

[0042] Specifically, the preset threshold range is based on historical normal process operation data. For each judgment sub-period, current change amplitude data under normal operation is collected multiple times, and its mean and standard deviation are calculated. The preset threshold range is set to the mean plus or minus three times the standard deviation, thus covering 99.7% of fluctuations under normal operation. The current current change amplitude is the absolute value of the difference between the real-time current value of the candidate abnormal region at the current moment and its baseline value before entering the judgment sub-period. It is determined whether the current current change amplitude is within the preset threshold range, that is, whether the absolute value of the difference is greater than the lower limit of the preset threshold range and less than the upper limit. If it is within the range, the current fluctuation characteristic value of the candidate abnormal region is determined to match the normal current fluctuation characteristic parameter, and the current fluctuation is determined to be caused by normal process operation; if it is not within the range, it is determined to be a mismatch, and the candidate abnormal region is reserved as a pending verification state. The pending verification state indicates that the candidate abnormal region has not been eliminated after this layer of judgment and needs to enter the subsequent steps for further verification.

[0043] S55. If the current fluctuation characteristic value matches the normal current fluctuation characteristic parameter, it is determined to be a current fluctuation caused by normal process operation.

[0044] When a match is determined in S54, it is confirmed that the current fluctuation in the candidate abnormal region is caused by normal process operations. At this point, the candidate abnormal region is removed from the candidate abnormal region list and is no longer subject to subsequent anode effect precursor verification. This step effectively eliminates current fluctuation interference caused by normal process operations such as electrode switching and aluminum tapping, reducing the false alarm rate.

[0045] S6. For candidate abnormal regions that have been marked but not unmarked, obtain the current data of the candidate abnormal regions at non-adjacent spatially symmetrical positions in the electrolytic cell; calculate the deviation between the current change in the candidate abnormal regions and the current change at the spatially symmetrical positions. In this embodiment, the spatially symmetrical position refers to the anode current acquisition zone that is spatially symmetrical with respect to the symmetrical center line or symmetrical center point of the aluminum electrolysis cell, and where there is no adjacent relationship between the two. The meaning of the spatially symmetrical position is as follows: Aluminum electrolytic cells typically have a symmetrical physical structure, such as being symmetrical about the longitudinal centerline. Assuming a candidate anomaly region is located in a sampling zone on the left side of the electrolytic cell, its spatially symmetrical location is the sampling zone on the right side of the cell that is symmetrical about the centerline, and this symmetrical zone is not physically adjacent to the candidate anomaly region (there are other zones separating them). For example, in an electrolytic cell with six sampling zones, numbered A, B, C, D, E, and F from left to right, if the candidate anomaly region is zone B, its spatially symmetrical location is zone E (B and E are symmetrical about the centerline, and B and E are not adjacent); if the candidate anomaly region is zone C, its spatially symmetrical location is zone D (C and D are symmetrical about the centerline, and C and D are not adjacent). By comparing the current change between the candidate anomaly region and its spatially symmetrical location, local anomalies and global current fluctuations can be effectively distinguished. Please refer to [link to relevant documentation]. Figure 6 The implementation process includes the following: S61. Obtain the historical current difference between the candidate anomaly region and the spatially symmetrical location within a preset historical time window; S62. Determine the reference deviation value based on the historical current difference; S63. Calculate the current current difference between the candidate anomaly region and the spatially symmetrical position at the current moment; S64. The absolute value of the difference between the current current difference and the reference deviation value is taken as the deviation value.

[0046] The preset historical time window is set to a length of 30 minutes to 2 hours based on the normal operating cycle of the electrolytic cell and the typical time scale of alumina concentration changes. Within this preset historical time window, the current values ​​of the candidate anomaly area and the spatially symmetrical position are collected at a fixed sampling frequency. The difference between the two is calculated at each sampling moment, resulting in a series of historical current difference data. The reference deviation value is a reference benchmark for the current difference between the candidate anomaly area and its spatially symmetrical position under normal operating conditions. The reference deviation value is determined by statistically calculating the series of historical current difference data obtained in S61 and taking the average or median as the reference deviation value. This reference deviation value reflects the inherent deviation level between the candidate anomaly area and the spatially symmetrical position during normal operation, such as the stable offset of the current difference due to structural asymmetry or differences in process parameters. At the current moment, the real-time current value of the candidate anomaly area and the real-time current value of the spatially symmetrical position are collected simultaneously, and the difference between the two is calculated. A positive difference indicates that the current in the candidate anomaly area is higher than that in the spatially symmetrical position, and a negative difference indicates that the current in the candidate anomaly area is lower than that in the spatially symmetrical position. Subtract the reference deviation value from the current current difference and take the absolute value of the result. This deviation value reflects the degree of deviation of the current difference between the candidate anomaly region and the spatially symmetrical position at the current moment relative to the reference deviation value under normal operating conditions.

[0047] S7. If the deviation value is greater than the second preset threshold, it is determined to be a precursor to a real local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. The second preset threshold is determined using statistical analysis based on deviation data from historical normal operating conditions. Specifically, deviation data of the electrolytic cell under normal operating conditions over 24 hours are collected, and its mean and standard deviation are calculated. The second preset threshold is set as the mean plus three times the standard deviation. This setting ensures that the probability of the deviation exceeding the second preset threshold under normal operating conditions is less than 0.3%, thus effectively controlling the false alarm rate while maintaining sensitivity. If the deviation value calculated in step S64 is greater than the second preset threshold, it is determined that the current symmetry relationship between the candidate abnormal region and its spatially symmetrical position is significantly disrupted. The current fluctuation in this candidate abnormal region has obvious local characteristics and is confirmed as a true precursor to a local anode effect. If the deviation value is less than or equal to the second preset threshold, it is determined that the current symmetry relationship between the candidate abnormal region and its spatially symmetrical position remains within the normal operating range. The current fluctuation in this candidate abnormal region is consistent with the global current fluctuation and belongs to the global current fluctuation rather than a precursor to a local anode effect. At this time, the candidate abnormal region is excluded, and the anode effect warning is no longer triggered. This step effectively distinguishes between local anomalies and global fluctuations, avoiding false alarms caused by global current fluctuations.

[0048] S8. When the presence of a precursor to a local anodic effect is confirmed, an anodic effect warning signal is generated and output.

[0049] When the deviation value is determined to be greater than the second preset threshold in step S7, confirming the existence of a real precursor to a local anodic effect in the candidate abnormal area, the system immediately performs the following actions: First, it generates an anodic effect warning signal, which includes, but is not limited to, the identifier of the candidate abnormal area where the precursor has occurred, the deviation value at the current moment, and the precursor level. Second, it outputs the generated warning signal to the human-machine interface of the electrolytic cell control system through the data communication interface, notifying the operator through sound or visual prompts. Simultaneously, the warning signal can also be transmitted to the automatic control actuator through the industrial control network, triggering a local feeding device for the candidate abnormal area to precisely replenish the area, increasing the alumina concentration in that area and thus nipping the anodic effect in the bud. Furthermore, the system records this warning event in the historical database for subsequent statistical analysis and model optimization.

[0050] Through the above steps, this application achieves accurate identification and timely early warning of precursors to local anode effects, effectively preventing the spread of local anode effects into full-cell high-voltage anode effects, significantly reducing power consumption and perfluorocarbon emissions, and improving the current efficiency and operational stability of aluminum electrolysis production.

[0051] Example 2 An embodiment of the online measurement and anode effect early warning device for the anode current of an aluminum electrolytic cell disclosed in this application includes the following: The regional current data acquisition module is used to acquire the regional current data corresponding to each anode conductor after calibration by the reference calibration electrode; The deviation magnitude calculation and judgment module is used to divide the multiple current sensing units corresponding to the regional current data into a preset number of anode current acquisition partitions. For each acquisition partition, the deviation magnitude of the regional current data is calculated based on the historical sliding baseline. If the deviation magnitude exceeds the first preset threshold and the duration is greater than the first preset duration, it is marked as a candidate abnormal region. The execution period determination module is used to obtain real-time operation plan data of aluminum electrolysis cells and determine whether the candidate abnormal area is currently within the execution period of the operation plan data; if so, it determines whether the candidate abnormal area is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and removes the mark of the candidate abnormal area that matches the normal process operation. The deviation value calculation and judgment module is used to obtain the current data of the candidate abnormal area at a non-adjacent spatially symmetrical position in the electrolytic cell for the marked and unremoved candidate abnormal area; calculate the deviation value between the current change of the candidate abnormal area and the current change at the spatially symmetrical position; if the deviation value is greater than the second preset threshold, it is determined to be a true precursor to a local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. The anode effect early warning signal generation module is used to generate and output an anode effect early warning signal when the presence of local anode effect precursors is confirmed.

[0052] For specific limitations regarding the device, please refer to the limitations of the method above, which will not be repeated here. Each module in the above device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0053] It is understood that those skilled in the art can combine various implementation methods in the above embodiments under the guidance of the above examples to obtain technical solutions with multiple implementation methods.

[0054] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for online measurement of anode current and early warning of anode effect in an aluminum electrolytic cell, characterized in that, include: Obtain the regional current data corresponding to each anode conductor after calibration with the reference calibration electrode; The multiple current sensing units corresponding to the regional current data are divided into a preset number of anode current acquisition partitions. For each acquisition partition, the deviation of the regional current data is calculated based on the historical sliding baseline. If the deviation exceeds the first preset threshold and the duration is greater than the first preset duration, it is marked as a candidate abnormal region; Obtain real-time operation plan data of aluminum electrolysis cells, and determine whether the candidate abnormal area is currently within the execution period of the operation plan data; If so, then determine whether the candidate abnormal region is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and remove the marking of the candidate abnormal region that matches the normal process operation. For candidate abnormal regions that have been marked but not unmarked, obtain current data of the non-adjacent spatially symmetrical positions of the candidate abnormal regions in the electrolytic cell; Calculate the deviation between the current change in the candidate abnormal region and the current change at the spatially symmetrical location; If the deviation value is greater than the second preset threshold, it is determined to be a precursor to a real local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. When the presence of precursors to a local anodic effect is confirmed, an anodic effect warning signal is generated and output.

2. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 1, characterized in that, The acquisition of the region current data corresponding to each anode conductor after calibration with the reference calibration electrode includes: The calibration poles are measured using a calibration probe to obtain the calibration reference parameters; Each anode guide rod is self-calibrated online according to the calibration reference parameters to eliminate thermal drift and installation tolerances. The measurement action after self-calibration acquires the regional current data corresponding to each anode conductor.

3. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 2, characterized in that, The online self-calibration of each anode guide rod according to the calibration reference parameters includes: After the pole-changing operation is completed or during periodic inspection, the calibration probe is triggered to measure the reference calibration pole. A correction coefficient matrix is ​​generated based on the calibration reference parameters; The correction coefficient matrix is ​​used to correct the regional current data corresponding to each anode conductor in real time.

4. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 1, characterized in that, The calculation of the deviation of the regional current data based on the historical sliding baseline includes: Acquire current data from multiple historical regions for each acquisition partition within a preset time window; The dynamic baseline value of the acquisition zone is calculated based on the current data from the multiple historical regions; Calculate the relative deviation rate between the current area current data and the dynamic baseline value as the deviation magnitude; wherein, the relative deviation rate is the ratio of the difference between the current area current data and the dynamic baseline value to the dynamic baseline value.

5. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 1, characterized in that, The step of determining whether the candidate abnormal region is currently within the execution period of the job plan data includes: Obtain the physical location identifier of the candidate anomaly region; Based on the physical location identifier, match the corresponding job plan item from the real-time job plan data; Determine whether the current time is within the planned execution time period of the job plan item; if so, determine that the candidate abnormal region is within the execution time period of the job plan data.

6. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 5, characterized in that, The step of determining whether the candidate abnormal region is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period includes: The execution period is divided into at least two consecutive judgment sub-periods, where different judgment sub-periods correspond to different expected current fluctuation characteristics; the judgment sub-periods are divided based on the execution stage of the work plan item, including an initial stabilization stage, a mid-term fluctuation stage, and a final recovery stage; Determine the sub-time period to which the current moment belongs; Obtain the normal current fluctuation characteristic parameters corresponding to the judgment sub-period to which the current time belongs; Calculate the current fluctuation characteristic value of the candidate abnormal region and compare it with the normal current fluctuation characteristic parameter; If the current fluctuation characteristic value matches the normal current fluctuation characteristic parameter, it is determined to be a current fluctuation caused by normal process operation.

7. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 6, characterized in that, The calculation of the current fluctuation characteristic value of the candidate abnormal region and the comparison with the normal current fluctuation characteristic parameter includes: Obtain the preset threshold range corresponding to the current sub-period of judgment; Determine whether the current change amplitude of the candidate abnormal region is within the preset threshold range; If yes, it is determined to be a match, and the current fluctuation is determined to be caused by normal process operation; if no, it is determined to be a mismatch, and the candidate abnormal area is retained as a state to be verified.

8. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 1, characterized in that, The spatially symmetrical position is the anode current acquisition zone that is spatially symmetrical with respect to the symmetrical center line or symmetrical center point of the aluminum electrolysis cell and has no adjacent relationship with the candidate abnormal region.

9. The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to claim 8, characterized in that, The calculation of the deviation between the current change in the candidate abnormal region and the current change at the spatially symmetrical location includes: Obtain the historical current difference between the candidate abnormal region and the spatially symmetrical position within a preset historical time window; The baseline deviation value is determined based on the historical current difference. Calculate the current current difference between the candidate anomaly region and the spatially symmetrical position at the current moment; The absolute value of the difference between the current current difference and the reference deviation value is taken as the deviation value.

10. An online measurement device for the anode current of an aluminum electrolytic cell and an early warning device for the anode effect, characterized in that, The method for online measurement of anode current and early warning of anode effect in aluminum electrolysis cells according to any one of claims 1 to 9 includes: The regional current data acquisition module is used to acquire the regional current data corresponding to each anode conductor after calibration by the reference calibration electrode; The deviation magnitude calculation and judgment module is used to divide the multiple current sensing units corresponding to the regional current data into a preset number of anode current acquisition partitions. For each acquisition partition, the deviation magnitude of the regional current data is calculated based on the historical sliding baseline. If the deviation magnitude exceeds a first preset threshold and the duration is greater than a first preset duration, it is marked as a candidate abnormal region. The execution period determination module is used to acquire real-time operation plan data of aluminum electrolysis cells, determine whether the candidate abnormal area is currently within the execution period of the operation plan data; if so, it determines whether the candidate abnormal area is a current fluctuation caused by normal process operation based on the characteristic waveform corresponding to the execution period, and removes the mark of the candidate abnormal area that matches the normal process operation. The deviation value calculation and judgment module is used to obtain the current data of the candidate abnormal region at a non-adjacent spatially symmetrical position in the electrolytic cell for the marked and unremoved candidate abnormal region; calculate the deviation value between the current change of the candidate abnormal region and the current change at the spatially symmetrical position; if the deviation value is greater than a second preset threshold, it is determined to be a precursor to a real local anode effect; otherwise, it is determined to be a global current fluctuation and is excluded. The anode effect early warning signal generation module is used to generate and output an anode effect early warning signal when the presence of local anode effect precursors is confirmed.

Images