Aluminum electrolysis anode effect identification method and device based on multiple time windows

Abstract

The invention discloses an aluminum electrolysis anode effect identification method and device based on multiple time windows, and relates to the technical field of aluminum electrolysis, the method comprises the steps that a target aluminum electrolysis cell is divided into multiple areas, and each area comprises at least one pair of anodes; sampling the cell voltage and the region current of each region according to a set sampling period, traversing each region in sequence in a circulating manner, and when traversing the current region, obtaining a region current time sequence and a cell voltage time sequence at the current moment and within a set duration before the current moment; on the basis of multi-time window data, whether the aluminum electrolysis anode effect occurs in the current area or not is identified according to the resistance difference of the equivalent area, the cell voltage variation coefficient and the cell voltage average value, if yes, anode effect early warning is given out, the number of times of aluminum oxide blanking is set for the current area and the area adjacent to the current area, and if not, the next area is traversed; and all the regions are traversed. The aluminum electrolysis production stability can be improved.

Description

Technical Field

[0001] This application relates to the field of aluminum electrolysis technology, and in particular to a method and apparatus for identifying the anode effect in aluminum electrolysis based on multiple time windows. Background Technology

[0002] The cryolite-alumina molten salt electrolysis method is currently the only industrial method for producing metallic aluminum. The anode effect is an unavoidable and frequent phenomenon in this industrial production process. When the anode effect occurs, it causes a sharp increase and violent fluctuation in the cell voltage, increased energy loss, reduced current efficiency, and leads to a hot cell, resulting in the large-scale emission of perfluorocarbons, a strong greenhouse gas, which seriously damages normal industrial production and the ecological environment.

[0003] The fundamental cause of the anode effect is the excessively low alumina concentration near the anode in the electrolytic cell. The anode effect typically begins at individual anodes, known as a localized anode effect or low-voltage anode effect. When a localized anode effect occurs, the anode bubble resistance increases, causing a significant decrease in the anode current and the regional current in that anode area, although the cell voltage does not change significantly at this time. If the alumina replenishment in this area is not timely or sufficient, or if the electrolyte mass transfer behavior in the electrolytic cell is insufficient to adequately replenish the alumina concentration in this low-alumina area, this localized anode effect will spread to other anodes or other areas, eventually triggering a whole-cell effect or flashing effect, known as a high-voltage anode effect. In this case, the cell voltage rises sharply to above 8V and fluctuates violently before being detected by the cell control system. The system then issues an alarm after several seconds, alerting on-site workers to manually extinguish the anode effect, a process that takes 1-5 minutes. The above process of detecting, alarming, and extinguishing the anode effect based on cell voltage is a technical challenge currently faced by all aluminum electrolysis plants. It not only consumes manpower and resources, but also causes the anode effect to continue for several minutes before it can be extinguished, resulting in a significant reduction in current efficiency, large-scale emissions of perfluorocarbons (PFCO), a strong greenhouse gas, and increased power consumption, among other problems.

[0004] More seriously, the occurrence of localized anode effects can trigger the emission of perfluorocarbons, a strong greenhouse gas. However, current detection technologies and methods based on tracking cell voltage cannot detect localized anode effects, which means that the problem of perfluorocarbon emissions during aluminum electrolysis has not been effectively solved. Summary of the Invention

[0005] The purpose of this application is to provide a method and apparatus for identifying the anode effect in aluminum electrolysis based on multiple time windows, which can reduce power loss and emissions of perfluorocarbons, a strong greenhouse gas, and improve the stability of aluminum electrolysis production.

[0006] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for identifying the anode effect in aluminum electrolysis based on multiple time windows, the method comprising: The target aluminum electrolytic cell is divided into multiple regions according to the number and spatial distribution characteristics of the feeding ports, and each region includes at least one pair of anodes; The tank voltage and tank current are sampled in each region according to a set sampling period. When the sampling duration exceeds the set duration, each region is traversed sequentially in a loop. When the current region is reached, the region current time series and tank voltage time series within the set duration prior to the current time are obtained. The set duration is divided into three time windows: a first time window, a second time window, and a third time window. The time length between the start of the third time window and the end of the first time window is the set duration. The third time window and the second time window overlap, while the second time window and the first time window do not overlap and have a set time interval. The equivalent region resistance at the current time is calculated based on the region current time series and tank voltage time series within the first time window. The tank voltage variation coefficient is calculated based on the region current time series and tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

[0007] Optionally, the first time window is from the current time to the time before the current time. The time period, the second time window is Time to Before the moment The time period, wherein the length of the overlapping portion of the third time window and the second time window is [missing information]. The time length of the non-overlapping portion of the third time window and the second time window is The time length between the end of the second time window and the current time is , The sampling period is , , and All are constants. greater than ( nPositive integers that are +1 and less than 60. n greater than or equal to 3 and less than positive integers, A positive integer greater than 1 and less than 120, 0 < p <1, , This is a rounding function.

[0008] Optionally, based on the current time and The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region, specifically including: like If so, it is determined that no anode effect will occur in the current region at the current time; where, For the current moment and The difference in equivalent region resistance at time . The threshold value is the resistance difference of the equivalent region. R is the cell resistance of the target electrolytic cell; like Then, it is determined whether the average value of the first slot voltage is greater than or equal to the slot voltage threshold, and the determination result is obtained; the average value of the first slot voltage is calculated based on the slot voltage time series within the first time window, and the slot voltage threshold is determined based on the average value of the second slot voltage, the average value of the third slot voltage, and the slot voltage variation coefficient; the average value of the second slot voltage is calculated based on the slot voltage time series within the second time window, and the average value of the third slot voltage is calculated based on the slot voltage time series within the third time window; If the judgment result is yes, it is determined that the anode effect has occurred or is about to occur in the current region at the current time, and the number of times alumina is fed into the current region and the adjacent regions of the current region is set. If the judgment result is negative, it is determined that no anode effect will occur in the current region at the current time.

[0009] Optionally, the range of values ​​for the number of times can be set as follows: , To set the number of times, This represents the number of anodes contained in the current region.

[0010] Optionally, the slot voltage threshold is expressed as: ; in, The slot voltage threshold is... This represents the average voltage of the second slot. The average voltage of the third slot. The voltage variation coefficient of the slot is denoted as . To set the threshold for the coefficient of variation, , The first set voltage value, , The average cell voltage of the target aluminum electrolytic cell over 30 days.

[0011] Optionally, the equivalent region resistance at the current moment is calculated based on the region current time series and the tank voltage time series within the first time window, specifically including: Calculate the average value of the first region current and the average value of the first tank voltage within the first time window based on the region current time series and the tank voltage time series within the first time window, respectively. Calculate the equivalent region resistance at the current moment based on the average current in the first region and the average voltage in the first tank. The formula for calculating the equivalent region resistance at the current moment is: ; in, Let be the equivalent region resistance at the current moment. The average voltage of the first slot. This represents the average current in the first region. Set the second voltage value. .

[0012] Optionally, the calculation is based on the region current time series and the tank voltage time series within the second time window. The equivalent region resistance at time t, specifically includes: Calculate the average value of the second region current and the average value of the second tank voltage within the second time window based on the region current time series and the tank voltage time series within the second time window, respectively. Calculated based on the average current in the second region and the average voltage in the second tank. The equivalent region resistance at time t; The formula for calculating the equivalent region resistance at time t is: ; in, for The equivalent region resistance at time t. This represents the average voltage of the second slot. This represents the average current in the second region. Set the second voltage value. .

[0013] Optionally, the formula for calculating the coefficient of variation of the slot voltage is: ; ; in, The voltage variation coefficient of the slot is denoted as . The standard deviation of the slot voltage within the second time window. for x Slot voltage at time, This represents the number of sampling periods within the second time window.

[0014] Optionally, the target aluminum electrolytic cell is divided into multiple regions according to the number and spatial distribution characteristics of the feed ports, each region including at least one pair of anodes, specifically including: With the centerline of the target aluminum electrolytic cell along its length as the axis, the installation position of the anode is divided into two axially symmetrical sides, namely side A and side B. An anode on side A and its axially symmetrical anode on side B form a pair of anodes. The target aluminum electrolytic cell is divided into multiple regions according to the number and spatial distribution characteristics of the feed ports, and each region includes at least one pair of anodes.

[0015] Secondly, this application provides a multi-time-window-based aluminum electrolysis anode effect identification device, the multi-time-window-based aluminum electrolysis anode effect identification device comprising: The region division module is used to divide the target aluminum electrolysis cell into multiple regions according to the number and spatial distribution characteristics of the feed ports, and each region includes at least one pair of anodes; The anode effect identification and processing module samples the tank voltage and region current of each region according to a set sampling period. When the sampling duration exceeds the set duration, it iterates through each region sequentially. When it reaches the current region, it acquires the region current time series and tank voltage time series within the set duration prior to the current time. The set duration is divided into three time windows: a first time window, a second time window, and a third time window. The set duration is defined as the time between the start of the third time window and the end of the first time window. The third and second time windows overlap, while the second and first time windows do not overlap and have a set time interval. The equivalent region resistance at the current time is calculated based on the region current time series and tank voltage time series within the first time window. The tank voltage variation coefficient is calculated based on the region current time series and tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and apparatus for identifying the anode effect in aluminum electrolysis based on multiple time windows. It utilizes the regional current time series and cell voltage time series within multiple time windows, and determines the effect based on the current time and... The difference in equivalent region resistance, cell voltage variation coefficient, and average voltage of the third cell at any given time are used to identify whether an aluminum electrolysis anode effect is occurring in the current region. By using multiple time dimensions and multiple physical quantities to identify the anode effect in each region, accurate identification of local anode effects is achieved. When an anode effect is determined to occur, alumina is fed into the current region and its adjacent regions a set number of times, which effectively extinguishes the local anode effect and prevents the local anode effect from spreading to the surrounding areas and eventually developing into a flashing effect or a full-cell anode effect. This can reduce energy loss and strong greenhouse gas emissions of perfluorocarbons, and improve the stability of aluminum electrolysis production. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a method for identifying the anode effect in aluminum electrolysis based on multiple time windows, provided as an embodiment of this application.

[0019] Figure 2 This is a schematic diagram of the time series data collected in the current region (the j-th region) and the time axis corresponding to multiple time windows, provided in an embodiment of this application.

[0020] Figure 3 This is a detailed flowchart illustrating a method for identifying the anode effect in aluminum electrolysis based on multiple time windows, provided as an embodiment of this application.

[0021] Figure 4 This is a schematic diagram of the functional modules of an aluminum electrolysis anode effect identification device based on multiple time windows, provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] In one exemplary embodiment, this application discloses a method for identifying the anode effect in aluminum electrolysis based on multiple time windows, such as... Figure 1 As shown, the aluminum electrolysis anode effect identification method based on multiple time windows includes steps 101-102.

[0025] Step 101: Divide the target aluminum electrolytic cell into multiple regions according to the number and spatial distribution characteristics of the feed ports, with each region including at least one pair of anodes.

[0026] Step 102: Sample the tank voltage and area current of each area according to the set sampling period. When the sampling duration is longer than the set duration, traverse each area in a loop. When traversing to the current area, obtain the area current time series and tank voltage time series within the set duration before the current time.

[0027] The set duration is divided into three time windows: the first time window, the second time window, and the third time window. The start time of the third time window is ( From the end of the first time window () t 1) The time length between them is the set duration, wherein the third time window and the second time window overlap, the second time window and the first time window do not overlap and have a set time interval; the equivalent region resistance at the current moment is calculated based on the region current time series and the tank voltage time series within the first time window, and the tank voltage variation coefficient is calculated based on the region current time series and the tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

[0028] After step 102 is completed, the process continues to traverse each region sequentially to perform the next round of aluminum electrolysis anode effect identification.

[0029] This application identifies the anodic effect in each region through multiple time dimensions and multiple physical quantities, achieving accurate identification of local anodic effects. When an anodic effect is determined to occur, alumina is added to the current region and its adjacent regions a set number of times, effectively extinguishing the local anodic effect and preventing it from spreading to the surrounding areas and eventually developing into a flashing effect or a full-tank anodic effect. This reduces energy loss and perfluorocarbon emissions, improving the stability of aluminum electrolysis production.

[0030] In an exemplary embodiment, step 101 specifically includes steps 201-202.

[0031] Step 201: Using the centerline of the target aluminum electrolytic cell along its length as the axis, divide the anode installation position into two symmetrical sides, namely side A and side B. An anode on side A and its symmetrical anode on side B form a pair of anodes.

[0032] Step 202: Divide the target aluminum electrolytic cell into multiple regions according to the number and spatial distribution characteristics of the feed ports. Each region includes at least one pair of anodes. That is, each side includes at least one anode. Specifically, divide the target aluminum electrolytic cell into M regions, where the number of anodes in the j-th region is... , , , Let N be the number of anodes in the j-th region, and N be the total number of anodes. The M regions of the target electrolytic cell are connected in parallel (therefore, the cell voltages in the M regions are equal, but the region currents in each region are not equal).

[0033] In an exemplary embodiment, step 102 specifically includes: using a cyclical approach, with a sampling period... Iterate through each region sequentially and obtain the current region (region j) at the current time. The current time series data of multiple regions and the corresponding voltage time series data of multiple tanks within the preceding time period are collected to obtain the region current time series and the tank voltage time series. This process is repeated until the collection duration T is greater than or equal to... Subsequent calculations will begin at the specified time. Each sampling period... Data on tank voltage and area current were collected for each region.

[0034] like Figure 2 As shown, the first time window is from the current time ( (Time) up to the current time The time period, i.e., the length of the first time window, is The second time window is Time to Before the moment The time period, i.e., the length of the second time window, is The time length of the overlapping portion of the third time window and the second time window is The time length of the non-overlapping portion of the third time window and the second time window is The time length between the end of the second time window and the current time is , The sampling period is ≤60s, , , and All are constants. greater than ( n Positive integers that are +1 and less than 60. n greater than or equal to 3 and less than integers, It is a positive integer greater than 1 and less than 120. More specifically, n This represents the number of sampling periods within the first time window. This represents the number of sampling periods within the second time window. Figure 2 middle Before the time ( The moment to the left of the current moment is a historical moment. This is a rounding function. There are two non-default parameters. The first parameter is the number to be rounded, and the second parameter is the number of digits to be rounded. When the second parameter is 0, it means that the number is rounded to an integer.

[0035] in, p The range of values ​​for is: , p•β Rounding to the nearest integer; the principle for determining the duration of the non-overlapping portion of windows w2 and w3. The range of values ​​for is: , Round to the nearest integer.

[0036] The set time interval between the second time window and the first time window is .

[0037] Calculate and store the j-th region according to formula (1). t The average current value of the first region within window w1 before time 1 (inclusive) I j, w1 According to the j-th region in t Regional current value at time 1 and in t The n region current values ​​of the j-th region obtained consecutively before time 1 , ... The maximum value is removed from a total of (n+1) consecutive region current values ​​in the j-th region. Remove a minimum value The average value I after j, w1 The duration of the w1 window is... Second.

[0038] (1) Calculate and store the j-th region according to formula (2). t 2 moments ( t 1-ɑ• s Time, see Figure 2 The average regional current I within the w2 window (before and after) j, w2 According to the j-th region in t The regional current value I at time 2 j, t2 and in t The j-th region obtained consecutively before time 2 β Current value of each region , ... , total consecutive ( β The average value of the regional current in the j-th region (+1) is I j, w2 This time window is named w2 and has a duration of [duration missing]. β• T s .

[0039] (2) Select Collection , , ... The tank voltages collected simultaneously are calculated according to formula (3) and stored. t 1 moment and I j, w1 The corresponding average slot voltage U avg, w1 According to the j-th region in t Regional current value at time 1 and int 1 moment ago and Corresponding slot voltage value , ... A total of (n+1) consecutive slot voltage values ​​are processed, with the maximum value removed. Remove a minimum value The average value U after avg, w1 .

[0040] (3) Select Collection , , ... Simultaneously, the corresponding tank voltage is collected, calculated according to formula (4), and stored. t 2 moments ( t 1-ɑ•T s Time, see Figure 1 ) and I j, w2 The corresponding average slot voltage U avg, w2 According to the j-th region in t Regional current value at time 2 and in t The j-th region obtained consecutively before time 2 β Current value of each region , ... Corresponding slot voltage value , ... , total continuous β +1 slot voltage average value U avg, w2 .

[0041] (4) In an exemplary embodiment, calculating the tank voltage variation coefficient based on the region current time series and tank voltage time series within the second time window specifically includes: The formula for calculating the coefficient of variation of the slot voltage is: (5) (6) in, The voltage variation coefficient of the slot is denoted as . The standard deviation of the slot voltage within the second time window. for x Slot voltage at time, This represents the number of sampling periods within the second time window.

[0042] Calculate and store according to formula (7) Figure 2 The image shown [ t 2-(1- p )· β ·T s The average voltage of the third slot within the w3 window before (and including) the specified time. w3 time window duration (p+l)•β• T s Furthermore, windows w3 and w2 must have some overlap in time, but not complete overlap; the length of the overlapping time window is... p•β• T s .

[0043] (7) In an exemplary embodiment, calculating the equivalent region resistance at the current moment based on the region current time series and the tank voltage time series within the first time window specifically includes: Calculate the average value of the first region current and the average value of the first tank voltage within the first time window based on the region current time series and the tank voltage time series within the first time window, respectively.

[0044] Calculate the equivalent region resistance at the current moment based on the average current in the first region and the average voltage in the first slot.

[0045] The formula for calculating the equivalent region resistance at the current moment is: (8) in, Let be the equivalent region resistance at the current moment. The average voltage of the first slot. This represents the average current in the first region. Set the second voltage value. .

[0046] Calculate and store the j-th region according to formula (8). The equivalent region resistance at time t.

[0047] In one exemplary embodiment, calculation is performed based on the region current time series and the tank voltage time series within the second time window. The equivalent region resistance at time t, specifically includes: Calculate the average value of the second region current and the average value of the second tank voltage within the second time window based on the region current time series and the tank voltage time series within the second time window, respectively. Calculated based on the average current in the second region and the average voltage in the second tank. The equivalent region resistance at time t; The formula for calculating the equivalent region resistance at time t is: (9) in, for The equivalent region resistance at time t. This represents the average voltage of the second slot. This represents the average current in the second region. Set the second voltage value. .

[0048] Calculate and store the j-th region according to formula (9). The equivalent region resistance at time t.

[0049] Calculate the current time of the j-th region according to formula (10). t 1 moment and t The difference in equivalent region resistance Δ between two time points R j .

[0050] (10) Determine the current state of region j. △R j The threshold value for the difference in resistance between the two equivalent regions. △R th Size relationship: when △R j Less than 0.95 △R th ( △R j <0.95 △R th ), determine the area in t At time 1, the anode effect will not occur. Calculate the equivalent regional resistance difference Δ of the next region (region j+1) according to formulas (1) to (12). R j+1 .when △R j Greater than 1.05 △R th ( △R j >1.05 △ R th ), determine the area in t At time 1, the anode effect has most likely already occurred or is about to occur; when △R j The value ranges from 0.95. △R th -1.05 △R th Within (0.95) △R th ≤ △R j ≤1.05△R th ), determine the area in t At time 1, there is a certain probability that the anode effect has already occurred or is about to occur. From the perspective of industrial production (the factors that can cause fluctuations in measurement data in engineering are numerous, complex and unpredictable), a clear conclusion cannot be given at this time.

[0051] In actual electrolytic aluminum production, the measured cell voltage and current values ​​used to calculate zone resistance both fluctuate. This fluctuation is not primarily due to the accuracy of the measuring equipment, but rather to the various electrochemical reactions and unavoidable interfacial movements and fluctuations that occur during production. Therefore, such fluctuations are objectively present and are a normal phenomenon.

[0052] This normal fluctuation will occur. Value, and can also cause The value changes. The occurrence of the anode effect can also trigger... The generation and changes of [something]. Although caused by normal fluctuations. The changes, theoretically not as significant as the anode effect, still require identification of the two causes mentioned above. Otherwise, and It's easy to make mistakes in determining whether the size comparison is due to normal phenomena or anodizing effects. For example, if... If the value is too large, it will lead to missed detections, and what should be an anode effect will be mistaken for normal fluctuations. Choosing the wrong size can lead to misjudgment, where a normal fluctuation is mistaken for an anode effect. These are the problems that this application aims to solve.

[0053] In one exemplary embodiment, based on the current time and The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region, specifically including: like If so, it is determined that no anode effect will occur in the current region at the current time; where, For the current moment and The difference in equivalent region resistance at time . The threshold value is the resistance difference threshold of the equivalent region.

[0054] like Then, it is determined whether the average value of the first slot voltage is greater than or equal to the slot voltage threshold, and the determination result is obtained; the average value of the first slot voltage is calculated based on the slot voltage time series within the first time window, and the slot voltage threshold is determined based on the average value of the second slot voltage, the average value of the third slot voltage, and the slot voltage variation coefficient; the average value of the second slot voltage is calculated based on the slot voltage time series within the second time window, and the average value of the third slot voltage is calculated based on the slot voltage time series within the third time window.

[0055] If the judgment result is yes, it is determined that an anode effect has occurred or is about to occur in the current region at the current time, and an anode effect warning is issued for the current region to achieve advanced identification of the anode effect. The number of times alumina is fed into the current region and the adjacent regions of the current region is set; the anode effect is automatically and accurately pre-treated as needed to extinguish, weaken or suppress the anode effect.

[0056] If the judgment result is negative, it is determined that no anode effect will occur in the current region at the current time, and the equivalent region resistance difference of the next region ((j+1) region) is calculated according to formulas (1) to (10). △R j+1 .

[0057] In other words, this application uses a voltage adaptive algorithm that dynamically adjusts and changes voltage across multiple time windows (w1, w2, and w3) to determine whether the actual tank voltage exceeds the tank voltage threshold, thereby determining whether the anode effect has occurred or is about to occur, reducing the probability of false positives and false negatives. The method involves determining the coefficient of variation (CV) of the tank voltage value in window w2. w2 With the set coefficient of variation threshold CV th The size relationship is determined by the following formula (11). t The slot voltage threshold at time 1.

[0058] The slot voltage threshold is expressed as: (11) in, The slot voltage threshold is... This represents the average voltage of the second slot. The average voltage of the third slot. The voltage variation coefficient of the slot is denoted as . To set the threshold for the coefficient of variation, The first set voltage value, , The average cell voltage of the target aluminum electrolytic cell over 30 days, in mV; It varies depending on the type, stability, condition, and production status of the target electrolytic cell. The value range is 5×10-4 ~0.1.

[0059] This application uses the first time window and the second time window to calculate the average as the current time and... Real-time data reduces the impact of normal fluctuations on the data. If you find... If it is determined that an anode effect may occur, but to prevent this from being caused by normal fluctuations, further investigation is needed based on whether the average value of the first cell voltage is greater than or equal to the cell voltage threshold. This needs to be confirmed. However, choosing the cell voltage threshold presents another challenge. A completely fixed value would fail to reflect the influence of various factors on the production site, so it cannot be a fixed value. Therefore, a third time window is introduced. The approach of this application is to consider the data before the anode effect occurs, that is, to consider the data in the third time window as data from the normal production process, which already contains the most and most up-to-date influence of various factors on the current production site. Thus, the data in the third time window is used to calculate the cell voltage threshold. The basis for this.

[0060] Once the anodic effect occurs, the alumina feeding process is initiated to extinguish it. However, during this process, the cell voltage and zone current data will fluctuate drastically. This is because the dissolution of alumina added to the electrolyte requires a process, and therefore, completely extinguishing the anodic effect also requires a process; it is not a sudden change. Therefore, the cell voltage fluctuations during the anodic effect are abnormal. Using the cell voltage data at this time to calculate the cell voltage threshold would be inappropriate. This will obviously affect the determination of when the anodic effect has been successfully extinguished. Therefore, the latest data from the third time window cannot be used in this case; instead, the data from the previous third time window should be used to calculate the tank voltage threshold. The question then arises regarding whether to use the latest third time window or the previous third time window. This is specifically determined by the coefficient of variation (the magnitude of the reaction fluctuation). The biggest difference between the cell voltage and zone current fluctuations during the anodic effect quenching process and normal production lies in the magnitude of the fluctuations; therefore, the coefficient of variation is used to differentiate them.

[0061] The above tests and calculations will be performed according to Figure 3 The loop shown continues continuously, meaning that each sampling period is performed according to... Figure 3 The process shown is used for calculation and judgment.

[0062] This application determines that the target area (current area) is within Once the current time indicates that an anode effect is unlikely to occur or may have already occurred / is about to occur, a voltage adaptive algorithm based on multiple time windows (w1, w2, and w3) is used to determine whether the actual cell voltage (average value of the first cell voltage) exceeds the cell voltage threshold, thus confirming that the anode effect has occurred or is about to occur. When it is determined that the anode effect has already occurred / is about to occur, k feeding commands are issued to the target area and adjacent areas to preprocess the anode effect. This application can perform anode effect preprocessing based on timely advance identification of the anode effect, effectively extinguishing local effects, reducing the frequency of anode effect occurrence, and shortening the duration of the anode effect, thereby reducing energy loss and perfluorocarbon (PFC) emissions.

[0063] Among them, w1, w2 and w3 are the first time window, the second time window and the third time window, respectively.

[0064] Sampling period T s The unit is seconds (s); the unit for area current is kiloamperes (kA); voltage (slot voltage, voltage threshold) is... U th The unit for (etc.) is millivolts (mV); the unit for resistance (area resistance, area resistance difference threshold, etc.) is microohms (μΩ). Acquisition period T s Less than or equal to 60 seconds.

[0065] Equivalent region resistance difference threshold △R th The value range is 0.25•R to 0.9•R (where R is the cell resistance of the target electrolytic cell, which varies depending on the cell type, cell condition, cell voltage, and series current). U ext The value range is 1600mV~1800mV, varying depending on the cell type, cell condition, cell voltage, and series current; after determining that the anode effect has occurred or is about to occur and issuing an anode effect warning, the number of times materials are added to the anode effect warning area and adjacent areas. k The range of values ​​for is, ( N j The number of anodes contained in one region varies depending on the tank type, series current, and number of feed ports, and is rounded up to the nearest integer (i.e., not rounded off, such as 1.1 which is directly rounded to 2).

[0066] In an exemplary embodiment, the determination principle for simultaneously sending alumina feeding commands to the j-th region and the adjacent region of a target electrolytic cell with a total of M regions is as follows: when j = 1, simultaneously send alumina feeding commands to the 1st region and the adjacent 2nd region; when j = M, simultaneously send alumina feeding commands to the M-th region and the adjacent (M - 1)-th region; when 1 < j < M (j is an integer), simultaneously send alumina feeding commands to the (j - 1)-th region, the j-th region, and the (j + 1)-th region.

[0067] This application divides the target aluminum electrolytic cell into regions, sequentially traverses each region in a cyclic manner, and pre-identifies whether an anode effect has occurred or is about to occur based on the region current of each region and the corresponding target electrolytic cell voltage, and gives a warning signal. It can pre-identify the anode effect, automatically and accurately add alumina as needed, extinguish local anode effects, suppress flashing effects and cell-wide effects, reduce power loss, improve current efficiency, reduce emissions of strong greenhouse gases such as perfluorocarbons, and improve the stability of aluminum electrolysis production.

[0068] In Example 1, the cell voltage and region current of the 2130# aluminum electrolytic cell in a 400kA aluminum smelter can be synchronously and online collected. The region current is collected by a fiber optic loop current sensor, and the cell voltage is collected by a voltage collection device. The two collectors can achieve synchronous and online collection. The collection period is 1 second.

[0069] This aluminum electrolytic cell has a total of 6 alumina feeding ports. The feeding ports are numbered starting from the tapping end. The feeding port closest to the tapping end is numbered 1#, and the feeding port closest to the flue end at the other end is numbered 6. Therefore, this electrolytic cell can be evenly divided into 6 regions according to the number of feeding ports. The 1# feeding port corresponds to the 1st region, and so on.

[0070] Through the fiber optic loop current sensor, the region current of the 6th region at 22:43:26 on June 25, 2025 ( t Time 1) was collected as 55.0kA, and the corresponding cell voltage was 3956mV. At this time, the cell control machine of the factory did not issue an anode effect warning for this cell.

[0071] Create multiple time windows w1, w2, and w3. The time lengths of w1, w2, and w3 are 10s, 30s, and 30s respectively. Calculate and store the average region current I 6, w1 ≈57.5kA of the 6th region within the w1 window (10s) before this moment, and the corresponding cell voltage ≈3905mV; calculate and store the average region current I t ≈61.1kA of the 6th region within the w2 window (30s) before 22:42:46 on June 25, 2025 ( 6, w2 Time 2), and the corresponding cell voltage ≈3865mV; Calculate and store the standard deviation of the slot voltage s for the w2 window (30s). w2 and coefficient of variation CV w2 The values ​​are 15.9mV and 0.0041, respectively; calculate and store the average slot voltage U of window w3 (30s) before 22:42:32 on June 25, 2025. avg, w3 ≈3857mV, with an overlap time window length of 15 seconds.

[0072] Calculate and store the equivalent region resistance of region 6 at 22:43:26 on June 25, 2025, based on the above data. ≈39.2 μΩ, equivalent region resistance of region 6 at 22:42:46 on June 25, 2025. ≈36.3μΩ, and thus the equivalent regional resistance difference Δ of region 6 between the two times above can be calculated. R 6 It is 2.9 μΩ.

[0073] Based on the above data, determine the △ in region 6 at this time. R 6 The value of 2.9 μΩ is greater than the threshold of equivalent region resistance difference ( △R th =1.05 times that of 1.8μΩ), it is determined that the anodic effect in the 6th region has most likely occurred or is about to occur at 22:43:26 on June 25, 2025. In order to reduce the risk of misjudgment, a voltage adaptive algorithm is further used to determine whether the actual tank voltage exceeds the tank voltage threshold to determine whether the anodic effect has occurred or is about to occur.

[0074] Due to the coefficient of variation (CV) of the slot voltage value in window w2 w2 =0.0041) is greater than the set coefficient of variation threshold CV th (CV) th =0.003), then at 22:43:26 on June 25, 2025 ( t Slot voltage threshold at time 1 = U avg, w3 +△U=3857+35=3892mV, and the slot voltage at this moment (U avg, w1 ≈3905mV) is greater than the current slot voltage threshold ( =3892mV), it is determined that the anodic effect must have occurred or is about to occur in the 6th region at 22:43:26 on June 25, 2025, and needs to be dealt with. At this time, an anodic effect warning is issued to the 6th region, and material feeding orders are issued twice to the 6th region and the adjacent 5th region. After 1 minute and 35 seconds, that is, at 22:45:01 on June 25, 2025, the current value of the 6th region is 63.1kA, which has recovered to the normal level of the average regional current of the 6th region (the average regional current of the 6th region on that day is 61.5kA).

[0075] The method successfully identified the anodic effect (or local anodic effect) that occurred only in region 6 and automatically and accurately preprocessed the local anodic effect as needed, thus effectively extinguishing the local anodic effect and preventing it from spreading to the surrounding areas and eventually developing into a scintillation effect or a whole-tank anodic effect, thereby contributing to reducing the effect coefficient.

[0076] Example 2 differs from Example 1 only in that the current time of data collection is 22:29:43 on June 25, 2025. t At time 1, the current was collected in region 6 at 22:29:43 on June 25, 2025 using a fiber optic loop current sensor. t At time 1, the region current is 55.4kA, and the corresponding tank voltage is 3909mV.

[0077] Create multiple time windows w1, w2, and w3, calculate and store the average regional current I of region 6 within window w1 (10s) before that moment. 6, w1 ≈56.1kA, corresponding to the slot voltage ≈3906mV; Calculate and store 22:29:03 on June 25, 2025 ( t The average regional current I in region 6 within the window w2 (30s) before time 2) 6, w2 ≈58.4kA, corresponding to the slot voltage ≈3890mV; Calculate and store the standard deviation of the slot voltage s for the w2 window (30s). w2 and coefficient of variation CV w2 The values ​​are 9.9mV and 0.0025, respectively; calculate and store the average slot voltage U of window w3 (30s) before 22:28:49 on June 25, 2025. avg, w3 ≈3886mV, with an overlap time window length of 15 seconds.

[0078] Calculate and store the equivalent region resistance of region 6 at 22:29:43 on June 25, 2025, based on the above data. Equivalent region resistance of region 6 at 22:29:03 on June 25, 2025, approximately 40.2 μΩ. ≈38.4μΩ, and thus the equivalent regional resistance difference Δ of region 6 between the two times above can be calculated. R 6 It is 1.8 μΩ.

[0079] Based on the above data, determine the △ in region 6 at this time. R 6 The value of 1.8 μΩ is less than the threshold value of the equivalent region resistance difference. △R th =1.8μΩ) is 1.05 times that of the equivalent region resistance difference threshold ( △R th =1.8μΩ) 0.95 times, it is judged that the anodic effect may have occurred or is about to occur in the 6th region at 22:43:26 on June 25, 2025. From the perspective of industrial production, a clear conclusion cannot be given at the moment. Further, through the voltage adaptive algorithm, it is determined whether the actual tank voltage exceeds the tank voltage threshold to determine whether the anodic effect has occurred or is about to occur.

[0080] Due to the coefficient of variation (CV) of the slot voltage value in window w2 w2 =0.0025) is less than the set coefficient of variation threshold CV th (CV) th =0.003), then at 22:29:43 on June 25, 2025 ( t Slot voltage threshold at time 1 = U avg, w2 +△U=3890+35=3925mV, and the slot voltage at this moment (U avg, w1 ≈3906mV) is less than the current slot voltage threshold ( =3925mV), then it is determined that the anodic effect has not occurred in region 6 at 22:29:43 on June 25, 2025. Region 1 is monitored according to the same logic loop and calculation as described above.

[0081] This example illustrates that although the 6th region is at the current 22:29:43 ( tAt time 1, the regional current value (55.4 kA) was significantly lower than the normal level of the average value of the 6th region (the daily average regional current of the 6th region was 61.5 kA). However, according to the advanced identification method in this application, it was not identified as an anode effect (or local anode effect) of the 6th region, indicating that there was no misjudgment. Monitoring of the regional current data afterwards showed that no local anode effect occurred in the 6th region in the following 10 minutes, and no flickering effect or whole-cell effect occurred in the entire 2130# aluminum electrolysis cell. Therefore, the advanced identification result of the anode effect in this application is correct and did not lead to missed judgment or misjudgment (misjudgment). In summary, it shows that the advanced identification result of the anode effect in this application is correct.

[0082] Example 3 differs from Example 1 only in that the current signal collected is from the second region. The current signal from the second region was collected using a fiber optic loop current sensor at 22:43:26 on June 25, 2025. t At time 1, the region current is 76.4kA, and the corresponding tank voltage is 3956mV.

[0083] Create multiple time windows w1, w2, and w3, and calculate and store the average regional current I of region 2 within window w1 (10s) before that moment. 2, w1 ≈76.4kA, corresponding to the slot voltage ≈3905mV; Calculate and store 22:42:46 on June 25, 2025 ( t The average regional current I in region 2 within the window w2 (30s) before time 2) 2, w2 ≈76.0kA, corresponding to the slot voltage ≈3865mV; Calculate and store the standard deviation of the slot voltage s for the w2 window (30s). w2 and coefficient of variation CV w2 The values ​​are 15.9mV and 0.0041, respectively; calculate and store the average slot voltage U of window w3 (30s) before 22:42:32 on June 25, 2025. avg, w3 ≈3857mV, with an overlap time window length of 15 seconds.

[0084] Calculate and store the equivalent region resistance of region 2 at 22:43:26 on June 25, 2025, based on the above data. ≈29.5μΩ, equivalent region resistance of region 2 at 22:42:46 on June 25, 2025. ≈29.1μΩ, and thus the equivalent regional resistance difference Δ of the second region between the two times above can be calculated. R 6 It is 0.4 μΩ.

[0085] Based on the above data, determine the △ in region 2 at this time. R 2 The value of 0.4 μΩ is less than the equivalent region resistance difference threshold ( △R th =1.8μΩ) 0.95 times, it is determined that the second region will not or will soon experience an anode effect at 22:43:26 on June 25, 2025. The same logic loop and calculation described above are used to monitor the next region (region 3).

[0086] This example illustrates that although the second region is at the current 22:29:43 ( t The current value of the region at time 1 (76.4 kA) shows a certain fluctuation deviation from the normal level of the average current of the second region on that day (the average current of the second region on that day was 75.0 kA). This fluctuation phenomenon of regional current is common in the normal production process of aluminum electrolysis and may lead to misjudgment (misdiagnosis) of the anode effect. However, through the advanced identification method of the anode effect in this application, it is possible to quickly determine whether the regional current of each region is normal in the normal aluminum electrolysis production process. That is, it is possible to quickly determine that the fluctuation deviation of the regional current of a certain region at the current moment is not caused by the anode effect, and there is no need to perform on-demand automatic and precise pre-processing intervention for that region.

[0087] Based on the same inventive concept, this application also provides an embodiment for implementing the above-mentioned aluminum electrolysis anode effect identification device based on multiple time windows. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the aluminum electrolysis anode effect identification device based on multiple time windows provided below can be found in the limitations of the aluminum electrolysis anode effect identification method based on multiple time windows above, and will not be repeated here.

[0088] In one exemplary embodiment, such as Figure 4 As shown, a multi-time-window-based aluminum electrolysis anode effect identification device is provided, comprising: The region division module is used to divide the target aluminum electrolytic cell into multiple regions according to the number and spatial distribution characteristics of the feed ports, and each region includes at least one pair of anodes.

[0089] The anode effect identification and processing module samples the tank voltage and region current of each region according to a set sampling period. When the sampling duration exceeds the set duration, it iterates through each region sequentially. When it reaches the current region, it acquires the region current time series and tank voltage time series within the set duration prior to the current time. The set duration is divided into three time windows: a first time window, a second time window, and a third time window. The set duration is defined as the time between the start of the third time window and the end of the first time window. The third and second time windows overlap, while the second and first time windows do not overlap and have a set time interval. The equivalent region resistance at the current time is calculated based on the region current time series and tank voltage time series within the first time window. The tank voltage variation coefficient is calculated based on the region current time series and tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for identifying the anode effect in aluminum electrolysis based on multiple time windows, characterized in that, The method for identifying the anode effect in aluminum electrolysis based on multiple time windows includes: The target aluminum electrolytic cell is divided into multiple regions according to the number and spatial distribution characteristics of the feeding ports, and each region includes at least one pair of anodes; The tank voltage and tank current are sampled in each region according to a set sampling period. When the sampling duration exceeds the set duration, each region is traversed sequentially in a loop. When the current region is reached, the region current time series and tank voltage time series within the set duration prior to the current time are obtained. The set duration is divided into three time windows: a first time window, a second time window, and a third time window. The time length between the start of the third time window and the end of the first time window is the set duration. The third time window and the second time window overlap, while the second time window and the first time window do not overlap and have a set time interval. The equivalent region resistance at the current time is calculated based on the region current time series and tank voltage time series within the first time window. The tank voltage variation coefficient is calculated based on the region current time series and tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

2. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 1, characterized in that, The first time window is from the current time to the time before the current time. The time period, the second time window is Time to Before the moment The time period, wherein the length of the overlapping portion of the third time window and the second time window is [missing information]. The time length of the non-overlapping portion of the third time window and the second time window is The time length between the end of the second time window and the current time is , The sampling period is , , and All are constants. greater than ( n Positive integers that are +1 and less than 60. n greater than or equal to 3 and less than positive integers, A positive integer greater than 1 and less than 120, 0 < p <1, , This is a rounding function.

3. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 1, characterized in that, Based on the current time and The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region, specifically including: like If so, it is determined that no anode effect will occur in the current region at the current time; where, For the current moment and The difference in equivalent region resistance at time . The threshold value is the resistance difference of the equivalent region. R is the cell resistance of the target electrolytic cell; like Then, it is determined whether the average value of the first slot voltage is greater than or equal to the slot voltage threshold, and the determination result is obtained; the average value of the first slot voltage is calculated based on the slot voltage time series within the first time window, and the slot voltage threshold is determined based on the average value of the second slot voltage, the average value of the third slot voltage, and the slot voltage variation coefficient; the average value of the second slot voltage is calculated based on the slot voltage time series within the second time window, and the average value of the third slot voltage is calculated based on the slot voltage time series within the third time window; If the judgment result is yes, it is determined that the anode effect has occurred or is about to occur in the current region at the current time, and the number of times alumina is fed into the current region and the adjacent regions of the current region is set. If the judgment result is negative, it is determined that no anode effect will occur in the current region at the current time.

4. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 3, characterized in that, The range of values ​​for the set number of times is: , To set the number of times, This represents the number of anodes contained in the current region.

5. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 3, characterized in that, The slot voltage threshold is expressed as: ； in, The slot voltage threshold is... This represents the average voltage of the second slot. The average voltage of the third slot. The voltage variation coefficient of the slot is denoted as . To set the threshold for the coefficient of variation, , The first set voltage value, , The average cell voltage of the target aluminum electrolytic cell over 30 days.

6. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 1, characterized in that, The equivalent region resistance at the current moment is calculated based on the region current time series and the tank voltage time series within the first time window, specifically including: Calculate the average value of the first region current and the average value of the first tank voltage within the first time window based on the region current time series and the tank voltage time series within the first time window, respectively. Calculate the equivalent region resistance at the current moment based on the average current in the first region and the average voltage in the first tank. The formula for calculating the equivalent region resistance at the current moment is: ; in, Let be the equivalent region resistance at the current moment. The average voltage of the first slot. This represents the average current in the first region. Set the second voltage value. .

7. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 1, characterized in that, Calculate based on the region current time series and tank voltage time series within the second time window The equivalent region resistance at time t, specifically includes: Calculate the average value of the second region current and the average value of the second tank voltage within the second time window based on the region current time series and the tank voltage time series within the second time window, respectively. Calculated based on the average current in the second region and the average voltage in the second tank. The equivalent region resistance at time t; The formula for calculating the equivalent region resistance at time t is: ; in, for The equivalent region resistance at time t. This represents the average voltage of the second slot. This represents the average current in the second region. Set the second voltage value. .

8. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 7, characterized in that, The formula for calculating the coefficient of variation of the slot voltage is: ; ； in, The voltage variation coefficient of the slot is denoted as . The standard deviation of the slot voltage within the second time window. for x Slot voltage at time, This represents the number of sampling periods within the second time window.

9. The method for identifying the anode effect of aluminum electrolysis based on multiple time windows according to claim 1, characterized in that, The target aluminum electrolysis cell is divided into multiple regions according to the number and spatial distribution characteristics of the feeding ports. Each region includes at least one pair of anodes, specifically including: With the centerline of the target aluminum electrolytic cell along its length as the axis, the installation position of the anode is divided into two axially symmetrical sides, namely side A and side B. An anode on side A and its axially symmetrical anode on side B form a pair of anodes. The target aluminum electrolytic cell is divided into multiple regions according to the number and spatial distribution characteristics of the feed ports, and each region includes at least one pair of anodes.

10. A device for identifying the anode effect in aluminum electrolysis based on multiple time windows, characterized in that, The aluminum electrolysis anode effect identification device based on multiple time windows includes: The region division module is used to divide the target aluminum electrolysis cell into multiple regions according to the number and spatial distribution characteristics of the feed ports, and each region includes at least one pair of anodes; The anode effect identification and processing module samples the tank voltage and region current of each region according to a set sampling period. When the sampling duration exceeds the set duration, it iterates through each region sequentially. When it reaches the current region, it acquires the region current time series and tank voltage time series within the set duration prior to the current time. The set duration is divided into three time windows: a first time window, a second time window, and a third time window. The set duration is defined as the time between the start of the third time window and the end of the first time window. The third and second time windows overlap, while the second and first time windows do not overlap and have a set time interval. The equivalent region resistance at the current time is calculated based on the region current time series and tank voltage time series within the first time window. The tank voltage variation coefficient is calculated based on the region current time series and tank voltage time series within the second time window. The equivalent region resistance at time t. The time is the end time of the second time window. The average value of the third slot voltage is calculated based on the slot voltage time series within the third time window. Based on the current time and... The difference in equivalent region resistance at any given time, the coefficient of variation of the cell voltage, and the average value of the third cell voltage are used to identify whether the aluminum electrolysis anode effect occurs in the current region. If so, an anode effect warning is issued and a set number of alumina feeds are applied to the current region and its adjacent regions. Otherwise, the next region is traversed until all regions have been traversed.

Images