JPH0774396B2 - Method and apparatus for controlling water addition of blended raw material - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial application] The present invention relates to a method and an apparatus for controlling the addition of water to a raw material mixture for supplying water to the raw material mixture for sintering under optimum conditions.

[Prior Art] As a sintering raw material for an iron mill, various powdered ores, cokes, limestone, and the like are mixed, mixed, and mixed with an appropriate amount of water. The mixed raw material is mixed with a mixer by dropping, rolling, and colliding particles with each other, and mixing and rerolling are performed. At the same time, water is added to obtain a wet state containing moisture, which is sent to a sintering machine and burned. In the firing process,
Smelting ore with many pores is generated by melting and diffusion bonding. The influence of moisture is very large in the process of producing sinter, and there are various factors such as air permeability, firing temperature, quality, and retention,
The water action is implemented based on a comprehensive judgment based on the past results.

By the way, the factors that influence the air permeability on the sintering bed include the charging layer thickness, the raw material particle size, the charging density, and the like. For example, variations in the water content of the blended raw materials cause instability of air permeability, resulting in a reduction in yield, a reduction in strength, and an influence on the quality of the powder ratio map and the like. Therefore, it is desired that the water content of the raw materials to be blended be maintained at a desired water content quickly and appropriately.

FIG. 14 is a conceptual diagram for explaining a conventional method for controlling water addition of blended raw materials.

As described above, the blended raw material obtained by mixing the three raw materials is supplied to the mixer 1, and an appropriate amount of moisture is added under the control of the moisture controller 2. The mixed raw materials that have exited the mixer 1 are carried into a charging hopper 3 and sent to a sintering machine (not shown) via a drum feeder 4 where they are sintered to form a sinter ore into a furnace. Is brought in.

Upon exiting the drum feeder 4, some of the blended ingredients are sampled by the operator for sampling. This sampling is performed, for example, about once every 2 to 4 hours. Then, the sampling is performed by the dry moisture meter 6
Moisture is measured by and this becomes the actual moisture value. This actual moisture value is displayed on an analog display. Therefore, the operator 7 operates the deviation amount between the actual water value and the target water value by the water content controller 2 so that the target water value displayed on the display device 8 is reached, and the mixer 1 increases or decreases the amount of water corresponding to the deviation amount. Do (tons / hour).

[Problems to be Solved by the Invention] However, in the above-described conventional method for controlling the water content addition of the blended raw material, the sampling frequency by the dry-type water meter is about once every 2 to 4 hours, so that the method is quick and suitable. It was difficult to make such a response. Further, since the control is performed manually, it is difficult to control with high accuracy.

An object of the present invention is to provide a method and an apparatus for controlling the water addition of a blended raw material, which enables highly accurate control and automation of the water addition control of the blended raw material.

[Means for Solving the Problems] In order to achieve the above-mentioned object, the present invention measures the water content in the blended raw material at a constant cycle in the subsequent stage of a mixer for adding and mixing water to the blended raw material, and measuring the same. Calculate the actual moisture and the amount of moisture change based on the result and the preset target moisture value,
The fuzzy membership function determined in advance is used to calculate the fuzzy control amount from the fuzzy membership function of the water added to the mixer, and the amount of water added to the mixer is controlled by the fuzzy control amount.

In addition, a moisture meter for measuring the water content of the blended raw material after the addition of water, a data storage means for storing the measurement data by the moisture meter at a constant cycle, and the data stored in the means are used to measure the current blended raw material. A present blending raw material moisture value calculation means for calculating the moisture content value, and the actual moisture content and moisture variation quantity calculated by the means and the actual moisture content and moisture variation quantity calculated based on a preset target moisture content value, using a fuzzy membership function. And an inference means for calculating the amount of water added to the mixer by performing fuzzy inference using a fuzzy membership function of the controlled amount of water added to the mixer, and a conversion for converting the inferred value by the means to the set value of the water added to the mixer The device can also be configured by providing a means.

[Operation] According to the above-mentioned means, the amount of water control to be added to the mixer is obtained using the fuzzy membership function and the fuzzy inference means for the data based on the measurement result by the moisture meter and the filed empirical rule. Therefore, it becomes possible to perform highly accurate water addition control by fuzzy control instead of the manual control that depends only on the empirical rule as in the conventional case, and it is possible to improve the stability of baking and the yield.

[Embodiment] An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing an example of a water addition control apparatus for blended raw materials and a manufacturing facility for blended raw materials according to the present invention. Note that, in FIG. 1, the same reference numerals are used for the same elements as those used in FIG. 14, and thus duplicated description will be omitted below.

The compounded raw materials are stored in the storage tank 10, and the small amount is stored in the conveyor 11
It drops to the top and is continuously conveyed to the mixer side. In this embodiment, two mixers 1a and 1b are used, and water addition control is performed for each. Further, as the moisture meter, an infrared moisture meter 13 having excellent measurement accuracy and capable of increasing the sampling frequency is used.

A sinter 14 is disposed below the drum feeder 4 to continuously charge the blended raw material from the drum feeder 4. The blended raw material charged is surface-ignited by a combustion gas in an ignition furnace (not shown) and combusted by forcibly sucking air downward by an exhaust fan (not shown).

In addition to the control system according to the present invention, the water addition system of the mixers 1a and 1b is configured to be switched to a conventional manual control system, and switches 15a and 15b are provided in the control circuits of the mixers 1a and 1b, respectively.
Are connected. The switches 15a and 15b have fuzzy (F
UZZY: ambiguous) A mixed raw material moisture control unit 16 using logic is connected. This blended raw material moisture control unit 16 includes an infrared moisture meter 13, an operator 7 to set target moisture values (air permeability, air volume, wind pressure, burn-through point (BTP) temperature,
An input device 17 for inputting (determined in consideration of position etc.) and a display device 18 for displaying analysis data are connected to each other.

In the above configuration, when automatically controlling the water addition of the blended raw material, the switches 15a and 15b are switched to the automatic side, and the operator 7 inputs the target moisture value to the blended raw material moisture control unit 16 using the input device 17. .

The mixed raw material from the storage tank 10 is carried into the mixer 1a by the conveyors 11 and 12, and mixed with water supplied based on the control output of the mixed raw material water control unit 16. The mixed raw material containing water in the mixer 1a is mixed by the conveyor 12 into the mixer 1b.
It is carried in to and is mixed with the added water under a fixed amount of added water. After that, it is carried into the charging hopper 3 by the conveyor 12, and the moisture in the blended raw material is measured by the infrared moisture meter 13 in the middle thereof. The measurement result is sent to the blended raw material moisture control unit 16 and used as data for inferring the amount of water to be supplied to the mixer 1a next time. The mixed raw material discharged from the charging hopper 3 is sent to the sinter device 14 via the drum feeder 4, is subjected to a sintering process, and is transferred to the blast furnace as a sinter.

FIG. 2 is a block diagram showing details of the blended raw material water content control unit 16, and the processing is shown in FIGS. 3 and 4.

An input interface 21 is provided to receive outputs from the infrared moisture meter 13 and the input device 17, and a bus 22 is connected to the input interface 21. On bus 22,
CPU (Central Processing Unit) 23, first ROM (Read Only Memory) 24 (ROM1) storing a processing program for managing the control of the entire system in addition to the control of sintering,
The second ROM25 (ROM that stores the program for executing the water addition control of the blended raw material based on the fuzzy logic
2), RAM that temporarily stores input information and processing results
(Random access memory) 26, magnetic storage device 27 that stores (files) various parameters and data,
A display device 18 for displaying input data and processing contents on a CRT (cathode ray tube), etc., and an output interface 29 for outputting a water addition control command obtained by executing the program of the ROM 25 to the switches 15a, 15b are respectively connected. ing.

As the magnetic storage device 27, a storage device using a floppy disk or a magnetic disk as a storage medium can be used.

In FIG. 2, a Neumann system digital computer is used, and various languages (for example, Fortran language, Lis
Although the p language and Prolog language) are adopted to execute the fuzzy control, a dedicated fuzzy computer may be constructed.

For example, a fuzzy memory for storing setting contents and detected contents, a plurality of inference engines for executing fuzzy inference, and a fuzzy output obtained by combining the inference results as fuzzy words by the inference engine to obtain a fuzzy output C-MAX (Cor.
A fuzzy computer having a responding MAX circuit and a defuzzifier for converting a fuzzy output by the C-MAX circuit into a non-fuzzy output can be considered. However, as in the present embodiment, for a controlled object that does not require such a high-speed response, even if a conventionally used digital computer is used as hardware, no particular problem occurs.

Next, the water addition control of the blended raw materials will be described with reference to FIG.

The measured value (analog voltage value) of the infrared moisture meter 13 is, for example, fetched by the CPU 23 via the input interface 21 and the bus 22 at 1-minute intervals (S31) and stored in the RAM 26.
The taken-in voltage value (1V to 5V) is converted into a percentage value (0.0 to 10%) indicating the water content by the CPU 23 (S32), and this value is stored in the magnetic storage device 27 as 1-minute instantaneous data. (S33), which becomes the 1-minute instantaneous data file F1. Further, the 1-minute cycle data is stored in time series (S34), and this becomes a 1-minute data time series file F2 as shown in FIG. 4 (the above is the data storage means).

Next, it is determined whether or not the execution cycle n of fuzzy control (determined in consideration of the time required from the installation position of the mixer 1a to the detection position of the infrared moisture meter 13; in this embodiment, 20 minutes). Make a determination (S35). If it has not reached n, the processing is terminated, and if it has reached n, the 1-minute data time-series file F2 is used to calculate the present blended raw material moisture value (calculation of the latest m-minute average value) as shown in FIG. (S36). Here, m is a time value preset in the moisture control parameter file F3, which is 10 minutes in this embodiment. That is, here, the water content is calculated in a cycle of 20 minutes, but it means that the average value calculation at that time is performed only for the past 10 minutes from the present time.

After that, the fuzzy membership control value is calculated (S37). This process is performed by the operator 7 using the input device 1
The following values are obtained using the data input from 7 and stored in the magnetic storage device 27 as the moisture control parameter file F3.

Actual moisture = Current blended raw material moisture value−Target moisture value Moisture change amount = Current blended raw material moisture value−Previous blended raw material moisture value Next, the mixer added water control amount is calculated based on this actual moisture and the moisture change amount. Fuzzy logic is used in this calculation process. The basic control rules are the following three.

(A) The difference from the target is large- → the target is quickly approached (a large amount of action).

(B) Tendency to deviate from the target-> suppress.

(C) The difference from the target is not large, but the difference is not reduced --- approach the target (small action amount).

This basic control law can be further expressed as shown in FIG. Here, the actual water content means the difference between the average value 10 minutes before and the target water value, and the water content change represents the comparison with the previous time, that is, the difference between 20 minutes before and 10 minutes before. As can be seen from this table, the actual moisture has 5 levels, and the actual moisture change is 5
Similarly, there are five levels of control amounts for the actual moisture.

However, excluding cases where actions have been taken in advance or where normal operation is not possible, the water content change and the control amount are only three stages each. As a result, the fuzzy control law is as shown in FIG. 7, and this data is prepared as the moisture control fuzzy constant file F4. Here, the actual moisture is represented by M, the moisture change is represented by DM, and the control amount is represented by A. And the numbers after the alphabetic symbols have the following meanings.

Actual Moisture M (deviation from target) M1: Low M2: Low M3: Normal M4: High M5: High Moisture change DM DM1: Decrease DM2: Slight decrease DM3: No change DM4: Slight increase DM5: Increase Trend Controlled amount A (Controlled amount of added water amount set value) A1: Decrease A2: Slightly decrease A3: No change A4: Slightly increase A5: Increase As described above, in this embodiment, since there are 15 rules in total, each of these Fuzzy reasoning about.

Next, the actual water content M and the water content change D are calculated using the control law of FIG.
A method of obtaining the membership functions of M and the controlled variable A will be described.

Here, the grade is set to 0 to 1.0. For this grade, in the case of the actual moisture M, the degree of the above moisture level of each of M1 to M5 is determined from the deviation from the target. Then, five trapezoidal or triangular functions are formed with overlapping portions (matching portions) between adjacent ones. The width (spreading) of this function varies depending on the specific gravity that its meaning occupies, and when each has the same weight, for example, in the case of the control amount A, all have the same shape. In addition, in FIG. 8, the horizontal axis of the control amount A indicates the meter scale of the added water amount controller (the scale of the instrument viewed when the operator operates it).

Now, consider the meaning of FIG. Assuming a case of manual control, the M1 and DM1 parts are
It is considered that the “actual moisture content is low” and the “actual moisture content change tends to decrease”, and the control amount A corresponding to this is A5. Therefore, the operator may perform an action of increasing the control amount of the added water amount set value, that is, an operation of increasing the scale of about 0.2. However, in this embodiment, automatic control can be performed by using this operation as fuzzy inference.

Next, the added water control amount of the mixer 1a is calculated using FIG. 9 (S38).

Here, as one inference example, the difference from the target water content is −0.
The case where the amount of change in water is 5% and the amount of change in water is −0.075 is shown. Looking at the membership function diagram in Fig. 8, "difference from target water content"
The M whose is above the -0.05 is at the intersection of M2 and M3. On the other hand, DM with the "water change" above -0.075 is DM2
And DM3 overlap.

First, if we infer M3, as shown in FIG.
A corresponding to the combination of and DM3 is A3. Therefore, as shown in the upper diagram of FIG. 9, the membership of A3 is arranged adjacent to each membership of M3 and DM3.
And -0.05% and -0.075 respectively and M3 and D of FIG.
Read the grade value at the intersection with M3 and draw the lower grade value horizontally to A3.

On the other hand, in M2, as is clear from Fig. 7, DM3
The combination with and is A4. So, as shown in the middle figure of FIG. 9, put it next to each membership of M2 and DM3.
Place A4 membership. Then, read the grade value at the intersection of −0.05% and −0.075 with M2 and DM3 in FIG. 8, and determine whichever grade value is lower than A4.
Pull horizontally to.

Further, DM2 exists in the region where M2 is located. Then, looking at A when combining M2 and DM2, from Fig. 7 to A
It turns out to be 4. Therefore, as shown in the lower diagram of FIG. 9, the membership of A4 is arranged adjacent to each membership of M2 and MD2. And -0.05% and -0.07
Read the grade value at the intersection of each of 5 and M2 and DM2 in FIG. 8, and draw the lower grade value horizontally to A4.

From the above inference results, one A3 and two A4 are obtained, but for A4 that has a plurality of A for one rule, take the larger one. Then, the control amount A is obtained from the selected A4 and A3 by the center of gravity method. This is the added water control amount (S38).

Next, the added water control amount obtained in step 38 is converted into the added water set value of the mixer 1a (converting means). Further, [control amount A × 10] is added to the current added water set value obtained from the outside to obtain the added water set value (S39). Then, it is determined whether or not the setting control mode is instructed (S40). When the setting control mode exists, it is considered that the control is performed manually, and the process is ended, and the switches 15a and 15b are switched to the manual side. The manual control is performed when the blended material moisture control unit 16 cannot be used for some reason.

On the other hand, when there is no setting control mode, the CPU 23 regards it as the automatic control mode, moves the process to step 41, and switches the switches 15a and 15b to the automatic side. Then, the additive water control amount of FIG. 9 obtained by the center of gravity method (a value at which the area of the membership function is halved is taken and the core of the fuzzy expression is represented by a certain numerical value) is used. It is sent to the pulse setting device (not shown) provided in (1) through the output interface 29 (S41, S42).

As described above, by adopting the fuzzy control, the empirical rule of the operator can be used for the control side, and the automation of the water addition control of the blended raw material can be achieved. As a result, it is possible to stabilize the firing and prevent the yield from decreasing.

The water addition control is intended only for the mixer 1a, and the mixer 1b is usually fixed at a fixed amount of water.

FIG. 10 is a flow chart showing another embodiment of the present invention.

This embodiment deals with the case where the target value is changed during the operation (for example, because of heavy rain during the operation). As described above, the present invention updates the control every 20 minutes, and no action can be taken during this time. Therefore, when the target value is changed without waiting for the next control timing, the fuzzy function is first disconnected and the absolute value is changed according to the change, and then the fuzzy function is returned to perform processing. To do.

First, the target value is changed to an arbitrary value (S50). And
The previous target value is read from file F3, and (this time target value −
Previous target value) = target deviation value is calculated (S51). next,
The control amount (scale) is determined by associating the target deviation value with the control amount determination table set in advance as shown in FIG. 11 (S5
2). Thereafter, step 39 and step 40 shown in FIG.
The processing of 1 is sequentially executed.

After step 40, the control calculation cycle counter is reset (S53), and the moisture control parameters are filed in the moisture control parameter file F3. After that, the processes of step 41 and step 42 shown in FIG. 3 are sequentially executed.

The details of the processing of steps 40, 41 and 42 have already been described and will be omitted here.

Since the target value is changed while the control timing is performed every 20 minutes, the process of step 53 is performed to make the control timing start point of 20 minutes from this change.

[Test Example] FIGS. 12 and 13 show a test example of the present invention.

FIG. 12 shows the actual operation test of the moisture fuzzy control of the three sintering compounded raw materials, and the actual moisture value% with respect to the target value (measured value by the infrared moisture meter 13) is described. The measurement is performed every 20 minutes, and the time point is indicated by a black circle.

On the other hand, FIG. 13 shows an actual operation test corresponding to the measurement result of FIG. 12, and shows the state of the added water amount Ton / hour executed every 20 minutes for the same time period as FIG. 12. . As is clear from FIG. 13, according to the present invention, the automatic control based on the moisture measurement value performs the optimum control while taking into consideration the empirical rule of the operator.

In the above description, the water addition control of the blended raw material is taken as an example, but the present invention is not limited to this. Empirical rules are important and control is performed so that the detected (measured) value follows the set value. The present invention can be widely applied when the following is required.

[Effects of the Invention] Since the present invention is configured as described above, it has the effects described below.

In the method for controlling the addition of water to a blended raw material according to claim 1, the water content in the blended raw material is measured at a fixed period after the mixer for adding and mixing water to the blended raw material, and the measurement result and the preset target moisture are set. Calculate the actual water content and the amount of water change based on the value, and calculate the fuzzy control amount from each fuzzy membership function and the fuzzy membership function of the water added to the mixer, and the fuzzy control amount to the mixer Since I tried to control the amount of added water,
It is possible to automate the water addition control of the blended raw material, which has been difficult to control by the conventional digital computer, stabilize the firing, and improve the yield.

In the blending raw material moisture addition control device according to claim 2, in the blending raw material moisture addition control device that performs water adjustment by adding water according to a target moisture value with a mixer to the blending raw material carried into the sintering machine, A moisture meter for measuring the water content of the blended raw material after adding water, a data storage means for storing the measured data by the moisture meter at a constant cycle, and a current moisture value of the blended raw material using the data stored in the means. The present blended raw material moisture value calculation means and the actual moisture content and moisture variation amount calculated based on the calculation result of the means and the preset target moisture content value are used to calculate the actual moisture content and moisture content using respective fuzzy membership functions. An inference means for determining the degree of change and further performing a fuzzy inference using a fuzzy membership function of the control amount of water added to the mixer to calculate the amount of water added to the mixer, and the inference value by the means are mixed. Is provided with the means for converting the set value of the added water, it becomes possible to perform highly accurate moisture addition control with a simple configuration.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing an embodiment of a water addition control apparatus for blended raw materials and a production facility for blended raw materials according to the present invention.
FIG. 3 is a block diagram showing the details of the blended raw material water content control unit.
FIG. 4 is a flow chart showing the control means of the present invention, FIG. 4 is an explanatory view showing the 1-minute instantaneous data file processing in FIG. 3, FIG. 5 is an explanatory view showing the processing of the n-minute cycle in FIG. 3, and FIG. FIG. 7 is an explanatory diagram showing an operator's control law, FIG. 7 is an explanatory diagram showing a fuzzy control law in the present invention, FIG. 8 is an explanatory view showing a membership function process in the present invention, and FIG.
FIG. 10 is an explanatory diagram showing an inference example according to the present invention, FIG. 10 is a flowchart showing another control means of the present invention, FIG. 11 is a table diagram used for determining the control amount in FIG. 10, and FIG. Moisture value measurement characteristic diagram in actual operation test of water content fuzzy control of solid compound material, Fig. 13 shows additive addition rate (Ton / Ton /
FIG. 14 is a conceptual diagram illustrating a conventional method for controlling water addition of blended raw materials. In the figure. 1a, 1b: Mixer 3: Charging hopper 13: Infrared moisture meter 14: Sintering machine 15a, 15b: Switch 10: Moisture controller for blended raw material 17: Input device 18: Display device 23: CPU 24, 25: ROM 26: RAM 27: Magnetic storage device 29: Output interface

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Akira Harada 1 Kimitsu, Kimitsu-shi, Chiba Nippon Steel Corp. (72) Inventor Hiromi Koda 1 Kimitsu, Chiba Shin-Nihon Steel Co., Ltd. Company Kimitsu Works (72) Inventor Eiichi Shimozawa 1 Kimitsu, Kimitsu City, Chiba Shin Nippon Steel Co., Ltd. Kimitsu Works

Claims (2)

[Claims] 1. A moisture content in a blended raw material is measured at a fixed cycle after a mixer for adding and mixing water to the blended raw material, and the actual moisture content and the water content change based on the measurement result and a preset target moisture value. It is characterized in that the amount of water is calculated, a fuzzy control amount is calculated from each of the fuzzy membership functions determined in advance and the fuzzy membership function of the water added to the mixer, and the amount of water added to the mixer is controlled by this fuzzy control amount. A method for controlling water addition of compounded raw materials. 2. A water content control apparatus for adding a water content according to a target moisture value with a mixer to a material mixture carried into a sintering machine to adjust the water content. Moisture meter for measuring, a data storage means for storing measured data by the moisture meter at a constant cycle, and a present blended raw material moisture value calculation for calculating the present moisture value of the blended raw material using the data stored in the means Based on the means, the calculation result of the means, and the actual moisture and the amount of change in moisture calculated based on the preset target moisture value, the degree of the actual moisture and the amount of change in moisture is determined using each fuzzy membership function, and the mixer is further added. An inference means for performing fuzzy inference using a fuzzy membership function of the water control amount to calculate the amount of water added to the mixer, and a converter for converting the inferred value by the means into the set value of the water added to the mixer. Moisture addition control device of mixed material, characterized in that it comprises and.