JPH0768370A - Method for automatically pouring molten copper from refining furnace - Google Patents

Abstract

PURPOSE:To provide an aromatic molten copper pouring method from a refining furnace, by which the unevenness of thickness of a cast anode is reduced and the uniform anode can be cast. CONSTITUTION:The molten copper 6 in the refining furnace 2 is poured into a storing ladle 8 through a pipe 4 by tilting the refining furnace 2. Then the molten copper 6 in the storing ladle 8 is poured into a weighing ladle 10 by tilting the storing ladle 8 and at last, the molten metal is poured into a mold M conveyed while holding with a turn-table 12 from the weighing ladle 10. Further, the wt. just before discharging the molten metal from the storing ladle is memorized in order to stabilize the wt. just before discharging from the storing ladle 8 to near the aimed value. The value at the moment and the wt. variating gradiation are calculated based on the trend data and fuzzy inference is applied to each value and the suitable tilting speed is introduced and the tilting speed of the refining furnace is controlled.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention generally relates to a pouring method for pouring molten copper from a refining furnace into a mold through a pan, and particularly to measuring the weight of the pan. The present invention relates to an automatic pouring method characterized in that the tilting speed of a refining furnace is controlled by fuzzy control, and the amount of molten copper poured into a pan is kept constant.

[0002]

2. Description of the Related Art Conventionally, casting of an anode for copper electrolysis has been carried out by a Walker method in which molten copper from a refining furnace is poured into a horizontally arranged mold in a fixed amount.

To briefly explain with reference to FIG. 5, in order to inject a fixed amount of molten metal into a mold M having a predetermined shape, first, the refining furnace 2 is tilted, and molten copper in the refining furnace 2 is inserted through a gutter 4. 6 is poured into a pan 8 having an internal volume of about 10 times the internal volume of the mold, and then
Tilt the pan 8 so that the molten copper 6 in the pan 8 has a capacity of 2
The pouring water is poured into two measuring pans 10 each having a double internal volume, and finally, a fixed amount is poured into the two molds M carried on the turntable 12 and conveyed from the pan.

The weighing pan 8 and the weighing pan 10 have weighing equipment 14,
16 are installed, for example, the discharge amount of the pan 8 is 65
It is set to about 0 kg, and the discharge amount of each weighing pan 10 is 325 K.
It is designed to be about g.

With the above structure, the tilting of the refining furnace 2 is performed by the tilting electric motor 20 composed of a high-speed motor and a low-speed motor. Basically, the low-speed motor is used during casting, and the high-speed motor is cast. Used for tilting operations other than time. The high-speed motor is capable of normal rotation and reverse rotation at a furnace rotation speed of, for example, 310 × 10 ^-3 rpm at the maximum, and the low-speed motor is only for forward rotation at a furnace rotation speed of, for example, 1.43 × 10 ^-3 rpm at the maximum. It can rotate.

Since the molten metal continuously flows from the refining furnace 2 as shown in FIG. 6, the weight of the pan 8 gradually rises and then sharply drops when pouring it into the weighing pan. It is changing repeatedly. The sampling of the weight data takes in the point P which is the peak of the figure.

[0007] Further, the ladle 8 reverses the tilting direction before the pouring amount into the measuring ladle 10 reaches a predetermined value, and starts a returning operation.
Immediately after the start of the returning operation, the molten copper 6 in the pan 8 is still poured into the measuring pan 10, and the inflow of the molten copper 6 into the measuring pan 10 after the returning operation is referred to as "dripping".

The amount of drooping differs depending on whether the amount of hot water in the pan 8 is large or small. When the amount of hot water is small, the amount of drooping is small, and when the amount of hot water is large, the amount of drooping is large. .

A phenomenon similar to that of the sump 8 occurs in the measuring pan 10, and the amount of sag changes depending on the amount (weight) of hot water immediately before the discharging of the measuring pan 10.

Therefore, the variation of the pouring amount into the mold M is caused by the variation of the sagging amount, and as a result, the variation of the anode weight is caused.

For this reason, in order to perform stable anode casting in the Walker system, first of all, the weight of the pan 8 is kept constant, that is, the tilting speed v of the refining furnace 2 is controlled. It is empirically known to be the most important.

At present, this adjustment is performed by the operator operating the power control panel 24 while observing the weight of molten copper (weight display 22) in the pan 8 and the weighing pan 10 displayed on the weight display 22 or the like. This is performed by controlling the drive of 20 and manually adjusting the tilting speed v of the refining furnace 2 on the upstream side of the pan 8.

However, there are individual differences in the skill of the operator, and it is difficult to standardize the adjustment method, which is a factor that disturbs stable operation.

Although automation has been attempted, in the conventional automatic pouring method, the tilting speed v of the refining furnace 2 is programmed according to the pouring amount, and the refining furnace 2 is tilted at a speed according to the programmed tilting speed v.
Moreover, the correction was made based on the deviation from the actual weight.

However, in such an automatic pouring method, even if the tilting speed is the same, the pouring amount varies depending on the size of the sprue of the refining furnace 2 and the temperature of the molten copper 6 in the refining furnace 2. In the case where the tilt pattern is stored in the memory, the deviation increases and it is difficult to stabilize the weight of molten copper in the pan 8.

As described above, according to the prior art, since the weight of the anode is not stable as a result, the thickness of the anode also varies, and the handling step, which is a post-process,
It also causes troubles in the milling process. Further, it also causes a short circuit during electrolysis, resulting in deterioration of current efficiency.

Therefore, an object of the present invention is to provide an automatic pouring method of molten copper from a refining furnace, which reduces variations in the thickness of the cast anode and enables uniform casting of the anode. is there.

[0018]

The above object can be achieved by the method for automatically pouring molten copper from a refining furnace according to the present invention. In summary, the present invention is a method for pouring molten copper from a refining furnace into a mold through a pan, in which the weight of the pan is measured and stored, and the trend thereof is stored. It is characterized in that an instantaneous value and a weight fluctuation tendency value are calculated based on the data, and fuzzy inference is performed from each of them to calculate an appropriate tilting speed, and the tilting speed of the refining furnace is controlled based on the result. This is an automatic pouring method of molten copper from a refining furnace.

According to another aspect of the present invention, the molten copper in the refining furnace is continuously poured into the pan, the poured molten copper in the pan is intermittently poured into the measuring pan, and this measuring is further performed. In the molten copper pouring method of pouring from the pan to the mold, the peak value of the weight of the pan that increases and decreases with time is stored, and based on this stored data, an instantaneous value that is a deviation value from the target value. Then, the weight fluctuation tendency value is calculated by comparing it with the past peak weight value before pouring the pouring water, and fuzzy inference is performed based on these instantaneous values and the weight fluctuation tendency value indicating whether there is an upward tendency or a downward tendency. There is provided a method for automatically pouring molten copper from a refining furnace, which calculates a speed and controls a tilting speed of the refining furnace based on the result.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A method for automatically pouring molten copper from a refining furnace according to the present invention will be described below in more detail with reference to the drawings.

In the present invention, in order to stabilize the weight immediately before the discharge of the saucepan near the target value, the weight immediately before the discharge of the saucepan is stored, and based on the trend data, the instantaneous value and the weight change tendency are stored. And fuzzy inference from each,
The proper tilt speed is derived and fed back to the tilt speed of the refining furnace to obtain the proper speed.

FIG. 1 schematically illustrates an automatic pouring system constructed according to the present invention. According to this system,
The molten copper 6 in the refining furnace 2 is poured through a trough 4 into a pan 8 having an internal volume of about 10 times the internal volume of the mold by tilting the refining furnace 2. The pan 8 tilts and the pan 8
The molten copper 6 in the mold is poured into a measuring pan 10 which is usually provided with two machines and has an internal volume about twice the internal volume of the mold, and finally the measuring pan 1
A certain amount of molten metal is poured into the two molds M carried from 0 to the turntable 12.

Here, the pan 8 is a load cell 14 for weighing.
The weight is measured by, and the signal is transmitted to the weighing unit 18 and converted into a weight signal. This weight signal is input to the storage circuit 102 of the fuzzy control unit 100, and in the present embodiment, the data up to the past 15 times is stored. The sampling of the weight data is the peak value.

Further, according to the present invention, based on the data in the storage circuit 102, the arithmetic circuit 103 changes the deviation value (d ₁ ) from the set value of the pan weight immediately before the tilting operation of the refining furnace and the fluctuation of the pan weight. The trend value (d ₂ ) is calculated and each is subjected to fuzzy inference.

A speed command signal is output based on the result of this calculation, and the drive of the refining furnace tilting electric motor 20 is controlled via the power controller 24.

Next, the fuzzy inference according to the present invention will be described in more detail with reference to the embodiments.

Example 1 In this example, an instantaneous value in fuzzy reasoning, that is, a deviation value (d ₁ ), was defined as d ₁ = (weight target value)-(weight data immediately before), and It is a deviation from the set value of the saucepan weight immediately before the rolling operation, and the fluctuation tendency value (d ₂ ) is d ₂ = (average value of data from immediately before to 5 times before)-(6 times before to 10 times before) The average value of the data up to the previous time) is used to calculate how much the tendency of fluctuations in the weight of the ladle is increasing or decreasing, and this is taken as the weight fluctuation tendency.

The above deviation value (d ₁ ) and fluctuation tendency value (d ₂ )
Is used as the condition part membership function to infer the furnace tilting speed manipulated variable v (conclusion part membership function).

In order to obtain the inference data of fuzzy control, a good operator was interviewed and arranged, and the condition part membership function and the conclusion part membership function as shown in FIG. 2 were obtained. These functions are functionalized by substituting numerical values for the qualitative feeling felt by the operator.

Next, with respect to the production rules, Table 1 shows the results of hearing and summarizing with the operator in the same manner.

[0031]

[Table 1]

Term explanation Label name Meaning NL (Negative Large) Large in negative direction NM (Negative Medium) Medium in negative direction NS (Negative Small) Small in negative direction ZR (Approximately Zero) Approximately zero PS (Positive) Small) Small in the positive direction PM (Positive Midum) Medium in the positive direction PL (Positive Larga) Large in the positive direction

Explaining the procedure of fuzzy inference based on the above-mentioned membership function and production rule, as shown in FIG. 3, first, the deviation value (d ₁ ) and the fluctuation tendency value (d ₂ ) are input to the arithmetic circuit. It The arithmetic circuit calculates the grade of the condition part membership function in each production rule, and takes the minimum of the condition part membership functions as the grade of the conclusion part membership function. Next, the logical sum of the conclusion section membership functions is calculated to obtain the center of gravity, and the tilting operation amount v of the refining furnace is determined.

For example, when casting is performed while tilting at a low speed, when the weight of the ladle is 125 kg heavier than the target value and the tendency of the weight fluctuation is about 25 kg reduction, the refining furnace tilts. According to the present invention, a fuzzy inference was made as to how much the rolling speed should be adjusted with respect to the current speed. That is, it was inferred that the weight deviation (d ₁ ) was 125 kg and the weight variation tendency (d ₂ ) was −25 kg. Figure 4 shows the result of the logical sum of the conclusion function.

In FIG. 4, the operation amount (v) calculated from the center of gravity was -73 rpm. The value obtained from this inference result was almost the same as the operation actually performed by the operator.

Embodiment 2 In this embodiment, in order to more effectively cancel out the abnormal value peculiar to the saucepan, both the membership function and the production rule are trend-oriented, and the data acquisition is performed once. The instantaneous value acquisition was changed to the previous two average value acquisitions. That is, it is considered that this makes it possible to take an average deviation and cancel out the abnormal value. To improve the responsiveness to the deviation, the difference between the averages of 5 times and the difference between the averages of 2 times (previous 1 to 2 times weighted average)-(previous 10 to 1
Changed to (average once). It was thought that this would improve the responsiveness due to the difference between the two-time averages. The production rules have also been changed as shown in Table 2 in order to improve the responsiveness of the inference result and to prevent unnecessary operations when the result is within the allowable range.

That is, in this embodiment, as the deviation value (d ₁ ) and the fluctuation tendency value (d ₂ ), d ₁ = {(current weight immediately before discharge) + (weight immediately before that)} ÷ 2 d ₂ = {(weight immediately before the current discharge) + (weight immediately before the first discharge)} ÷ 2-{(weight immediately before the discharge 10 times before) + ( Weight)} / 2.

[0038]

[Table 2]

As for the operation result, better controllability than expected could be obtained.

[0040]

As described above, in the method for automatically pouring molten copper from the refining furnace according to the present invention, the weight immediately before the discharge of the pan is stored, and the instantaneous value is calculated based on the trend data. The weight change tendency is calculated, the fuzzy inference is performed from each of them, the proper tilting speed is derived, and the tilting speed of the refining furnace is controlled. The anode can be cast. Therefore, the anode manufactured according to the present invention does not require a handling step, a milling step or the like as a post-step, and can solve the problem of electrode short-circuiting in the next step, an electrolytic plant.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of an automatic pouring system for carrying out an automatic pouring method of molten copper from a refining furnace according to the present invention.

FIG. 2 is a diagram showing an example of a membership function.

FIG. 3 is an explanatory diagram showing a procedure of fuzzy inference.

FIG. 4 is an explanatory diagram showing an example of a logical sum obtained by fuzzy inference.

FIG. 5 is a schematic configuration diagram of an automatic pouring system for carrying out a method for automatically pouring molten copper from a conventional refining furnace.

FIG. 6 is a diagram showing a change in weight of a saucepan.

[Explanation of symbols]

 2 Refining furnace 8 Reservoir 10 Measuring pan 20 Electric motor for tilting 100 Fuzzy control section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location F27D 3/14 Z 7141-4K G05B 13/02 N 9131-3H

Claims (2)

[Claims] 1. A method of pouring molten copper from a refining furnace into a mold through a pan and a measuring pan, in which the weight of the pan is measured and stored, and its trend is stored. An instantaneous value and a weight fluctuation tendency value are calculated based on the data, and fuzzy inference is performed from each of them to calculate an appropriate tilting speed, and based on the result, the tilting speed of the refining furnace is controlled. Automatic pouring method of molten copper from refining furnace. 2. The molten copper of the refining furnace is continuously poured into a pan,
The molten copper in the pouring pot is poured into the measuring pan intermittently, and further, in the molten copper pouring method of pouring from this measuring pan into the mold, the weight of the lapping pan that increases and decreases with time The peak value is stored, and based on this stored data, the instantaneous value that is the deviation value from the target value and the weight fluctuation tendency value by comparing the past peak value before pouring the pouring water are calculated, and these instantaneous values are calculated. From a refining furnace characterized by fuzzy inference from a weight fluctuation tendency value indicating whether it is in an upward tendency or a downward tendency and calculating an appropriate tilting speed, and controlling the tilting speed of the refining furnace based on the result. Automatic molten copper pouring method.