RU2220231C2 - Process of control over feed of aluminum oxide into electrolytic cells to win aluminum - Google Patents

Abstract

FIELD: aluminum production. SUBSTANCE: process of control over feed of aluminum oxide into electrolytic cells to win aluminum is based on measurement of resistance between electrodes of cell which value is recorded in fixed time intervals. Resistance between electrodes is measured, values of resistance are recorded in fixed time intervals, concentration of aluminum oxide in electrolytic cell is evaluated and aluminum oxide is fed into cell in inadequate or excessive amount at fixed rate to prevent undesirable anode effect, undesirable high concentration of oxide, to detect unbalance between energy supply and decomposition of oxide fed into bath from outside and to look after state of cell. Time moment T is chosen on basis of number of resistance values recorded right before and after time moment T, lines E and Y, angle β included between lines E and Y are computed and angular coefficients are found for evaluation of concentration. On basis of several computed values one or several curves are constructed, value of amplitude for curve of angles β is recorded and recorded values of resistance and obtained curves are compared with time moments of beginning of excessive or insufficient feed. EFFECT: raised efficiency of process. 6 cl, 5 dwg

Description

 The present invention relates to a method for controlling the supply of alumina to electrolytic cells for producing aluminum, based on measuring the electrical resistance between the electrodes of an electrolytic furnace, the value of which is recorded at fixed time intervals.

 Known control algorithms for the point (i.e., localized in certain places of input) feed feed are based on the principle of an ideal match between the cell resistance and the oxide concentration represented by a U-shaped curve, on which the concentration is plotted along the x axis, and the cell resistance along y axis. Consider the upper curve in FIG. 1. The lower part of the curve usually corresponds to an oxide concentration of 4%. Such a control algorithm consists in controlling the cell by keeping the concentration low - below 4%. In this region, the cell resistance increases with decreasing concentration and, conversely, decreases with increasing concentration. In practice, this can be done in such a way that if, for example, measuring the resistance of a cell gives a value corresponding to an oxide concentration of 2%, an excess amount of oxide can be supplied to the cell, as a result of which the concentration of oxide in the bath increases. After a certain period of time, usually 30-60 minutes, the oxide concentration in the usual case will be approximately 3%, and the cell resistance will decrease. Subsequently, an insufficient amount of oxide is fed into the cell, so that the concentration drops again. When the oxide concentration is again 2%, a new period of over feed can be started. Using this control algorithm, the cell is excited by the oxide concentration in order to obtain a signal in the form of a cell resistance, which is used to control the oxide concentration in the bath for a given cell at a relatively low level. The term "R-signal" will be used hereinafter to mean the resistance of the cell in the periods of time surrounding the moments of change in the feed rate. The term "rapid feed" hereinafter refers to excessive feed at a fixed speed.

Definitions:
U - period of time with insufficient supply
N - time period with accelerated feed
UH period - time period U + H
start N - time moment of the start of accelerated feed
start U is the point in time when the insufficient feed starts.

 If the concentration of oxide in the bath becomes sufficiently low (approximately 1.8%), then the so-called anode effect takes place. With the anode effect, the voltage usually rises immediately to approximately 50 watts. This effect usually lasts approximately 5 minutes. To eliminate the anode effect, as a rule, special measures are necessary. Anode effects may be desirable or undesirable. One of the advantages of spot feed technology is that the frequency of anode effects can be drastically reduced.

 The energy supply to the electrolytic cell can be controlled by adjusting the position of the anode in the up and down directions during periods when there is no anode effect. Automatic adjustment is based on measuring the ohmic resistance of the cell. If the measurement gives a value outside the deadband around the reference resistance value, the anode is adjusted. The upper and lower dead zones are used. Two insensitivity and a reference value may change automatically depending on the state of the cell.

 In order to optimize operating conditions and increase efficiency, it is desirable to maintain a low concentration of oxide in the bath. In order to determine the point that corresponds to an oxide concentration of approximately 2%, i.e. concentration, which usually slightly exceeds the concentration created due to the anode effect in the cell, a change in the resistance level is used, for example, during the period of time preceding the achievement of this point, or the angular coefficient of resistance near this point. The angular coefficient of resistance can be determined based on the equation for a straight line in the xy coordinate system, i.e. y = ax + b, where a is an angular coefficient. The control algorithm can be based on a combination of both changing the resistance level and the angular coefficient of resistance. The decision that a point has been reached is called a prediction (i.e., a prediction of the anode effect).

 A problem for known prediction methods is that the ideal relationship between resistance and concentration on which these methods are based can be seriously compromised by other conditions in the cell — conditions that affect the change in resistance over time. Such disturbances are especially large in Söderberg cells (electrolyzers) and lead to many erroneous predictions. Erroneous forecasts are forecasts that occur at relatively high concentrations in the bath. True forecasts are forecasts that occur at sufficiently low concentrations of oxide in the bath.

 An example of a violation is a change in resistance due to absorption of an oxide deposit from the pallet and from the wall coating. The absorption of the precipitate causes a drop in the metal level with an increase as a result of this resistance. The change in resistance per unit time depends on many factors, for example, the energy supply for the decomposition of oxide, the amount of sediment in the cell, chemical and mechanical accessibility to the precipitate, the geometry of the wall coating, the number of bathtubs, etc. Mechanical accessibility in this regard includes, but is not limited to, fluid flows in the bath and metal. The flow paths and physical flow velocities are significant.

 In all electrolytic cells for producing aluminum, the concentration of oxide in the bath always depends on two sources of oxide: the oxide supplied to the bath from the outside, and the oxide formed in the bath from the inside. The oxide from the internal source comes from the sediment from the sump and from the wall coating. Due to the fact that the oxide is fed into the cell through a sintered layer in the bath located at the feed points (injection points) of the raw material, some of the oxide can pass through the phase in the bath without decomposition. This oxide becomes a precipitate. The amount of oxide that becomes a precipitate depends, among other things, on the concentration of oxide in the bath and the local energy supply to the oxide portion. High concentration and a slight "temperature rise" favor the formation of sediment. "Excess temperature" refers to the difference between the temperature in the bath and the melting temperature of the contents of the bath (liquidus temperature). The two aforementioned oxide streams can vary greatly in the Soderberg cell with point feed. In a cell with preliminary firing and with a point feed of raw materials, the oxide flow from the feed points is usually dominant.

 If there is a large amount of readily available sediment in the cell, a high oxide concentration in the bath can be maintained for a long period when sediment absorption occurs. A precipitate that disappears often leads to an increase in resistance. This increase in resistance often leads to erroneous forecasts. The forecast gives a relatively high feed rate of oxide through the feeders (feed mechanisms) for a certain period of time (over feed). Thus, oxide is supplied, which gradually forms a new precipitate; the concentration of oxide in the bath can remain high for a long time. Periods of high concentration reduce the efficiency of the process.

 Other variable parameters in the cell can also lead to false predictions, for example, changes in temperature, changes in the chemical composition of the bath, and changes in the shape of the wall coating.

 Other examples of conditions in which known control methods do not always work optimally are the period after the manifestation of the anode effect, the period after the release of metal, the period after deposition on the walls or final deposition, other periods with a high concentration of oxide in the bath, periods with an extremely low concentration of oxide in the bath, periods with a high level of noise of resistance in the cells, periods with high sediment absorption and periods with a high temperature in the bath.

 In relation to the above states, known control algorithms will be characterized to a greater or lesser extent by erroneous forecasts and a high oxide concentration, since a sufficiently good general idea of the oxide concentration is not available at any time. Soderberg cells are usually more prone to erroneous forecasts than cells with pre-firing.

 A known method of controlling the supply of alumina to electrolytic cells for producing aluminum is described, for example, in Norwegian patent 166821 or in Norwegian patent 172192. In addition, there is a method for controlling the supply of alumina to electrolytic cells for producing aluminum, including measuring the resistance between the electrodes of the electrolytic cell, registration of resistance values at fixed time intervals, estimation of the concentration of alumina in the electrolytic cell and the supply of alumina to the cell in insufficient or excessive amounts at a fixed speed (EP 044794 C 25 C 3/20, 01/27/1982).

 The objective of the invention is to develop an algorithm for controlling the supply of alumina to electrolytic cells to produce aluminum, which can improve the control of the oxide concentration in order to increase the reliability of forecasts.

 This problem is solved in a method for controlling the supply of alumina to electrolytic cells to produce aluminum, including measuring the resistance between the electrodes of the electrolytic cell, recording resistance values at fixed time intervals, estimating the concentration of alumina in the electrolytic cell, and supplying aluminum oxide to the cell in an insufficient or excessive amount with a fixed speed, due to the fact that to estimate the concentration, choose the time moment T based on a series of resistance values of the measurements recorded immediately before time T, calculate the first line E on the basis of a series of resistance values recorded immediately after time T, calculate the second line Y, calculate the angle β between the lines E and Y for several values of time T, for lines E and Y determine the angular coefficients on the basis of several calculated values of the angle β and, if necessary, the angular coefficient for the straight line Y for a number of moments of time T that coincide with the moments of resistance measurements, If one or more curves are observed, the amplitude value for the angle curve β is recorded, and the recorded resistance values and the obtained curves are compared with the times when the over or under feed starts.

According to preferred embodiments, control is used to eliminate concentrations that are lower than a predetermined value to prevent an undesired anode effect;
control is used to detect oxide concentrations that are higher than a predetermined value in order to prevent an undesirably high oxide concentration;
control is carried out to reduce the amount of oxide deposit in the cell with too much precipitate;
control is used to detect an imbalance between the supply of energy and the decomposition of oxide supplied to the bath from the outside;
the values recorded in connection with the control are used to monitor the state of the cell;
control is used when controlling an electrolytic cell based on a measurement of electrical resistance.

 According to the invention, cell resistance and ongoing processes are continuously monitored (monitoring). This tracking is carried out by the described methods in order to avoid incorrect interpretation of measurements. In those periods when the known control scheme can lead to undesirable development due to an incorrect interpretation, the oxide supply to the cell is controlled using fixed oxide supply algorithms, regardless of the resistance of the cell. At the same time, they are accompanied by fixed algorithms for supplying energy. These algorithms are created in such a way that the cell reaches a state in which a known control scheme is again introduced as quickly as possible. Algorithms are also defined in such a way that over a long period they systematically counteract the formation of sediment and lead to the collection of sediment.

 The R-waveform can be registered, for example, as follows (see FIG. 2): β represents the angle between two straight lines, where one line, namely the earlier line E, is drawn through 10 points on the resistance curve, which corresponds to in this case, a period of 20 minutes. Another line, namely the later line Y, is drawn through the next 10 points. The curve of β values is obtained by moving the lines one dimension forward in time, i.e. for 2 minutes. Further, in relation to curve β, the terms “curve β” and “angle curve” are used. The value β and the angular coefficients for the straight lines E and Y are examples of the so-called UH-parameters. The corresponding definition is given below.

The method is described in more detail below using examples of execution and graphs, where:
figure 1 shows a curve showing the dependence of the resistance of the cell R on the concentration of oxide in the bath;
in FIG. 2 on the same graph shows a typical change in the resistance R of the cell, the associated β curve and the feed rate for the cell. A coordinate transformation was carried out for the R value curve, so that it can be built on the same graph as the β curve;
in FIG. Figure 3 shows the fundamental relationships that are found between the β curve, the concentration of oxide in the bath, and feed rates;
figure 4 shows a typical curve β before and after the anode effect, and
figure 5 shows the registration of the anode effect.

 The method uses the accumulated information about the shape of the resistance curve for a period of time, which includes a period of insufficient supply of raw materials and the subsequent period of excessive supply of raw materials. This period is hereinafter referred to as the “R-signal period”. At low values of the concentration of oxide in the bath under ideal conditions for the resistance curve, a downward deviation occurs when switching from an insufficient supply of raw materials to an excess supply. At concentrations of approximately 5% and above, the curve deviates upward. At concentrations close to approximately 4%, the deviation of the curve becomes insignificant or absent. The effect of sediment absorption and other “disturbances” on the R signal is added to ideal curves. In Soderberg cells it is often observed that such effects lead to the fact that the R-signal rotates counterclockwise (see figure 1).

 At low concentrations, the R-signal breaks, so that its growth rate due to the excess feed is lower than the growth rate due to the insufficient feed. At higher concentrations, the gap of the R signal is less sharp. The gap may even be completely absent, in other words, the R-signal has characteristic forms that depend on the concentration of oxide in the bath.

 The period of the R-signal can be described by numerically selecting a suitable mathematical function, hereinafter referred to as the "UH-function". The UH function may be, for example, a parabola. In this method, two lines E and Y are used (see the above definition of β and FIG. 2). The mathematical function is adjusted to a resistance curve for a fixed period of time ranging from 30 to 60 minutes. The function is uniquely described by its parameters - UH-parameters. Β and the slope for Y are selected as UH parameters. By moving the fitted function forward in time by one measurement, new values for the UH parameters are obtained. Each individual UH parameter can be obtained as a curve that changes over time. The parameters change with a change in the concentration of oxide in the bath. UH parameters can vary from cell to cell, even if the cells have the same oxide concentration in the bath. In the same cell at a fixed level of oxide concentration, the parameters can also change with time, however, this change is slow. Using the data fitting mechanism, the expected value for the UH parameters of the cell close to the concentration corresponding to the anode effect can be determined. The influence of any anode regulation operations on the resistance curve can be eliminated using simple techniques before calculating the UH parameters.

 To find the expected values of the UH parameters, empirically obtained graphs can be used. Experimental plots can be the same for multiple cells. Experimental plots can also be individual for each cell. Which method is best depends on the uniformity and stability of the entire set of cells under consideration.

 The concentration of oxide in the bath can be estimated at any time by comparing the current UH parameters with the expected values of the parameters. The control program stores the values of the samples of the amount of raw material supplied through the feeders. Using this information and the course of the change in the UH parameters, it is possible to evaluate the supply of oxide to the bath, which does not come from the feeders. In the usual case, self-supply of raw materials from sediment and wall coverings will prevail. Thus, "self-feeding" is defined as the supply of oxide to the bath from the sediment and from the wall covering.

 It is advisable that the decision on the duration of the insufficient and excessive supply of raw materials in the next UH period be made in the previous UH period, so that the concentration of oxide in the bath changes optimally. During the UH period, the upper deadband for adjusting the anode is relatively high until the overfeeding period ends. Then the deadband decreases. Thus, the adjustment of the anode is consistent with the UH period. In the usual case, only special perturbations lead to the adjustment of the anode over a period in a different way, for example, metal release, final deposition, shutter extension, noise processing, etc. In the case of large disturbances of the R-signal (noise), the estimated oxide concentration in the bath is rejected as control information for a given UH period. In the case of regulation of the anode, the influence of the disturbance on the resistance curve is eliminated before calculating the variables.

 Regarding the assessment, information, for example, of the last two UH periods can be used and filtered so that the last accepted value is taken into account more than the previous value, etc.

 It is advisable that the UH periods are as short as possible. However, they must be long enough to ensure that the information contained in the resistance curve (R-signal) does not become outdated.

 The optimal change in the concentration of oxide in the bath is determined by the situation and adapts to the supply of raw materials from the sediment in the cell. With a low concentration of oxide in the bath and a low feed from the precipitate, most of the oxide should be dosed through point feeders in order to eliminate the anode effect. If the point feeders supply more oxide than the amount absorbed over a period of time, the excess oxide usually creates a relatively readily available precipitate and / or increases the concentration of oxide in the bath.

 Over a period of time immediately after the anode effect, the oxide is fed through the feeders in fixed quantities per unit time, so that the oxide concentration varies optimally. During this period, forecasts are not used in the management process. At the end of this period, the control mechanism described above may be applied.

 Tests were performed in the Soderberg electrolytic cells to study the relationship between resistance signals and feed rates at various oxide concentrations. During the tests, a “rhythmic supply” of raw materials was carried out in the cells, i.e. with a predetermined duration of insufficient supply of U and excessive supply of H during those periods during which forecasts can be used in the control mechanism. The supply of oxide was controlled so that the concentration of oxide in the bath was maintained at approximately the same level over long periods of time. The duration of U was changed during the tests. Duration H was set to 30 minutes.

 During testing, a "hysteresis effect" was recorded in some cells at oxide concentrations in the bath below 4%. For extended periods of time, the resistance signals were the opposite of those expected. The resistance of the cells decreased with insufficient supply of raw materials and increased with accelerated supply. The reaction of the cells during the test can be explained by the fact that the accelerated supply leads to an increase in the oxide layer that floats on the metal and increases the resistance of the cell, whereas with insufficient supply the oxide layer and resistance decrease. The above observation will generally be a disturbance in the control of the feeders supplying oxide to the cell.

 The registration of the hysteresis effect indicates a too low energy supply for the oxide, which is supplied to the cell. The oxide is not chemically decomposed. The combination of local flows in the bath and local energy conditions in the cell may be the cause of this condition. If this condition is recorded in connection with a rhythmic feed, the feed of oxide through the feeders can be temporarily reduced, and the reference value of the resistance in the cell can be temporarily increased. These measures are triggered automatically when other conditions are available. The extent and duration of these measures can be controlled using parameters that are set manually or automatically.

 However, it has been found that oxide concentration can be monitored using “angle recording”, which will be described later. In the case where the registration of the angle shows a high concentration, a rhythmic feed can be used until the registration of the angle shows that the concentration has decreased. At low concentrations, well-known forecasting mechanisms can be used to control the feed. At an extremely low concentration (close to the state of the anode effect), the energy pulsates, and a high feed rate can be used to exclude the anode effect. The resistance curve is monitored continuously and using angle recording in parallel.

 In FIG. Figure 3 shows the fundamental relationship between feed rate, angle curve and oxide concentration. This relationship is based on the test results, and at the top of FIG. 3, dashes are shown corresponding to the “start U” and “start H” and “angle β” moments as a function of time. At the bottom of FIG. 3, the corresponding oxide concentration is shown as a function of time. The latest R signals are ideal with respect to the standard control algorithm during the time preceding the anode effect. Dashes for the “start H” and “start U” moments are as they should be in relation to the curve β. In FIG. it is shown that the last 6 dashes are "inside" the curve. This means that the resistance increases with insufficient supply and decreases with accelerated supply of raw materials. In this case, the time difference between the dash and the corresponding maximum or minimum on the curve is small. In the case where the dashes are “inside” the β curve and the deviations are relatively large, it was reported that the oxide concentration is low. When calculating β in FIG. 3, the influence of the anode adjustments on the resistance R of the cell was not eliminated.

 However, the first lines are "outside" the curve. These measurements were made at a higher concentration, and in this state it was recorded that the bars are outside the curve, and the curve has large deviations. These measurements show that resistance increases with accelerated feed and decreases with insufficient feed, indicating a state in which the oxide does not decompose to the desired degree.

 At an average oxide concentration, the deviations of the angle curve are relatively small. Dashes may be inside or outside the curve β. Here, the time difference between the dash and the nearest maximum or minimum on the curve is relatively large.

2, the letter T denotes the time corresponding to the maximum and minimum values on the curve β. A denotes the amplitude β at time instants T. After a certain delay after instants of time T, some values are recorded that relate to the point (T, A). Delay is necessary for mathematical calculations. The following 4 values are logged:
- Angular coefficients for the lines E and Y at the point (T, A).

 - A (the value of β at the point (T, A) is calculated based on the lines E and Y).

 - The time difference between the corresponding change in feed rate and moment T.

 Immediately after registering the values, the state of the cell is evaluated in terms of the concentration of oxide in the bath. The assessment may be based on several sets of recorded values in a previous period of time, for example, on 3 sets.

 Controlling the oxide supply includes measuring the electrical resistance between the electrodes of the electrolytic cell. The resistance value is recorded at fixed time intervals when the oxide is fed into the electrolytic cell during periods of accelerated supply (N) and insufficient supply (U). At a given time (T), the first line (E) is calculated based on a series of resistance values recorded immediately before time (T). The second line (Y) is calculated on the basis of a series of resistance values recorded immediately after time (T), and then the angle between the first line (E) and the second line (Y) is determined, so that for several time instants (T), for β are recorded amplitude values (A), and for the lines (E) and (Y), angular coefficients are recorded. On the basis of several calculated values of β and, possibly, the angular coefficient for Y for a number of time points coinciding with the moments of resistance measurement, one or more curves are obtained, and the recorded values and curves are compared with time points for accelerated feed (start H) and insufficient feed (start U) in order to evaluate the concentration of oxide in the electrolytic bath.

 Control is used to exclude a concentration that is lower than a predetermined value, so that an undesired anode effect does not occur, or to detect oxide concentrations that are higher than a predetermined value, to exclude an undesirable high oxide concentration.

 The control can be used to reduce the amount of oxide deposit in the cell when the amount of precipitate is too large, or can be used to detect an imbalance between the energy supply and the decomposition of oxide fed into the bath from the outside.

 The values recorded during control can be used to monitor the state of the cell. In addition, the control can be an integral part of the known control circuit of the electrolytic cell, which is based on the measurement of electrical resistance.

 Based on the foregoing, to date, some variables have been developed that can be used to describe the concentration of oxide in the bath. This can be done using a standard control algorithm such that when recording the angle shows a high concentration, a "rhythmic feed" can be used until the concentration drops. The rhythmic flow in this case is set so that the amount of oxide supplied through the feeders is below the theoretical absorption value. At low concentrations, a standard forecast mechanism is used to control the feed. At an extremely low concentration, the energy pulsates, and a high feed rate can be used to eliminate the anode effect. The resistance curve is continuously monitored and angle recording is used in parallel. This mechanism will lead to a systematic absorption of sediment and in the long term give a lower degree of self-flow. Signals for good forecasts require a low degree of self-feed.

 At a high oxide concentration, the cell is controlled using a rhythmic feed until the concentration is again judged to be low or close to the anode effect. Feeding is continued by mixing long periods of underfeeding and rapid shifts of fast feed and underfeeding to produce R signals. The sequence of feed methods has such an average feed rate that leads to a decrease in the amount of sediment in the cell. This does not apply to cells that have an extremely low current efficiency.

 The above control can be carried out according to a known control scheme by performing continuous monitoring of the calculated concentration of oxide in the cell. If the cell is in a state close to the anode effect, the automatic system will try to prevent the anode effect using increased feed rate and power ripple. At low oxide concentrations, the cell is controlled using the predictions used in the usual way.

 Anode adjustment is usually performed at the end of the rapid feed period in order to obtain minimal disturbance of the resistance signals used for control. If, however, the resistance is far outside the deadband, the anode is adjusted to be within the deadband using various rules.

 Although the described mechanism is intended, in particular, for use in combination with Soderberg cells (electrolyzers), it goes without saying that it can also be used in combination with cells that have prebaked anodes.

 Figure 4 shows a typical β curve before and after the anode effect in cell B169, which is a Soderberg cell with a current of 130 kA, equipped with 4 point feeders. The set value for the excess aluminum fluoride in the bath is 11 wt.%.

 The concentration of oxide in the bath was measured approximately 4 hours before the anode effect and at that time it had a value of 3%. Over time, the concentration decreased until an anodic effect appeared, at approximately 17:20. Immediately before 17:20, the oxide concentration was approximately 1.8%. During the anode effect, operations on the cell are performed in such a way as to bring the concentration up to about 5%. Typically, the concentration will remain high for many hours after the anode effect due to self-feeding.

 The cell resistance (R) converted by coordinates, the dashes corresponding to the change in the feed rate, and β are shown in the same graph. Periods with accelerated feed are marked as H, and periods with insufficient feed are marked as U. The influence of the anode adjustments on R is eliminated. The zero line for β is shown. The amplitude β increases as the concentration decreases before the anode effect. The time moments corresponding to dashes and extremum points β coincide over the period with a low oxide concentration. The angular coefficient for R during periods of H increases with decreasing concentration.

 In the period with a high oxide concentration after the anode effect, the amplitude β is relatively small. The times corresponding to the dashes and the corresponding times of the points of extremum β do not coincide.

 In FIG. 5 shows the registration of the hysteresis effect. The curves and symbols in FIG. 5 correspond to the description for the above FIG. 4.

 In the period up to the mark of 6.30 a rhythmic flow was carried out in the cell for a long period of time. The hysteresis effect, as previously stated, is characterized by the fact that the cell resistance decreases with insufficient supply and increases with excessive supply. Known control methods are based on the opposite phenomenon, i.e. an increase in R with insufficient feed and a decrease in R with accelerated feed.

Claims (7)

1. A method of controlling the supply of alumina to electrolytic cells to produce aluminum, including measuring the resistance between the electrodes of the electrolytic cell, recording resistance values at fixed time intervals, estimating the concentration of alumina in the electrolytic cell, and supplying the alumina to the cell in an insufficient or excessive amount with a fixed speed, characterized in that for the assessment of concentration choose a point in time T, based on a series of resistance values, zaregist calculated immediately before time T, calculate the first line E, based on a series of resistance values recorded immediately after time T, calculate the second line Y, calculate the angle β between lines E and Y for several values of time T, for lines E and Y determine the angular coefficients, on the basis of several calculated values of the angle β and, if necessary, the angular coefficient for the straight line Y for a number of time points T that coincide with the moments of resistance measurements, build one or not How many curves recorded value of the amplitude for the curve of angles β and registered and obtained values of the resistance curves is compared with the start time instants of excessive or insufficient supply. 2. The method according to claim 1, characterized in that the control is used in order to exclude concentrations that are lower than a predetermined value in order to prevent an undesired anode effect. 3. The method according to claim 1, characterized in that the control is used to detect oxide concentrations that are higher than a predetermined value, in order to prevent an undesirable high oxide concentration. 4. The method according to claim 1, characterized in that the control is carried out to reduce the amount of oxide precipitate in the cell with too much precipitate. 5. The method according to claim 1, characterized in that the control is used to detect an imbalance between the supply of energy and the decomposition of oxide supplied to the bath from the outside. 6. The method according to claim 1, characterized in that the values recorded in connection with the control are used to monitor the state of the cell. 7. The method according to claim 1, characterized in that the control is used when controlling an electrolytic cell based on the measurement of electrical resistance.

Images