JPH0938518A - Grain sizing-crushing control device for grain sizing plant - Google Patents

Abstract

PROBLEM TO BE SOLVED: To keep the quality of a grain sized product in the good and suitable state and improve the crushing efficiency of raw material lumps by carrying out the grain sizing of the raw material lumps by the fuzzy control method in a grain sizing plant for crushing the raw material lumps such as iron ores and grain sizing. SOLUTION: Crushers CR1-CR4 and screens SC1-SC8 are accumulated in the cascade shape in the constitution, and raw material iron ores are crushed and screened into lump iron ores as they are transferred from the upstream side to the downstream side of the flow. At that time, the current values of respective crushers CR1-CR4 and the like are computed by a master controller MC, and computation results are passed through an index meaning module 1, a grade computation module 2, a fuzzy logical inference module 3, an OR module 4, a weighted meaning module 5 and a primary delayed module 6 successively to find out the crushing amount set value AFW and the clearance set values S1-S4, and operation signals are output in compliance with the output values by a direct digital controller 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a particle size control and crushing control apparatus for a particle size control plant for crushing raw material agglomerates to size control,
In particular, the present invention relates to a control device of a sizing plant for sizing iron ore, which is composed of a combination of a plurality of crushers and a sieving machine.

[0002]

2. Description of the Related Art For example, in a raw material yard factory of the steel industry, several sizing processes for the purpose of sizing the raw ore are installed. The purpose of sizing the ore is to increase the porosity of the charge in the blast furnace by increasing the porosity of the charge in the blast furnace by aligning the ore to be the insert of the blast furnace with a particle group of appropriate size (+10 to +25 mm). At the same time, by increasing the contact area between the reducing gas and the ore, the descending rate of the charging raw material is increased, the amount of tapping is increased, and the coke consumption rate is decreased.

The smaller the particle size of iron ore, the larger the surface area,
Reducibility by reducing gas is good, but when viewed as a blast furnace insert, if there is a lot of powder steel, it impairs air permeability, and
The gas distribution becomes non-uniform, causing shelves and slips, and the amount of tapped iron decreases. Therefore, it is necessary to adjust the grain size of the ore charge from the both sides of the reducibility and the air permeability to make the ore grain size within the optimum range and remove the powder steel.

Recently, the lower limit of the particle size of the charged ore is 8 to 1.
0 mm and the upper limit is 25 to 30 mm. Strengthening ore sizing and powder steel removal are extremely important, which
The gas distribution in the blast furnace is uniform, and the contact between the charge and the gas is good. As a result, by sizing the ore, the heat exchange is performed well, the coke ratio is reduced, and the ore having a large grain size is reduced, so that the time required for the reduction of the steel particles can be reduced.

Conventionally, the ore mined in the mine is usually only coarsely powdered at the base of the mountain to the extent that it is not inconvenient to transport, so it is crushed in a sizing process having a plurality of crushers and sieving machines. And sieving is performed. In the conventional sizing process, the pulverizer has been classified into various types such as coarse powder, intermediate pulverization and fine pulverization depending on the sizing step. Generally, coarse powder is called primary crush, intermediate crush is called secondary crush, and fine crush is called tertiary and quaternary crush.

On the other hand, the sieving machine is for sieving the crushed ore into a size of mesh or more and less than mesh, and the sieving machine of the sizing process is divided into a crusher supply and a product sieving machine. There are a target intermediate sieving machine and a final sieving machine for the purpose of separating the final product, ingot steel and powder steel. FIG. 3 shows a sectional view of the structure of the crusher, and the operation of crushing iron ore will be described.

The crusher comprises a disk-shaped mantle 31 fixed to a rotary shaft 30 and a cone cave 32 arranged around the mantle 31. The crushing of the raw ore 33 is performed by the eccentric movement of the mantle 31 and the squeezing action of the cone cave 32 in the crusher chamber. The crushing of the raw ore progresses as the raw ore 33 progresses from the crusher inlet to the outlet.

On the other hand, the particle size of the sized product and the crushing ability of the crusher change depending on what value is set for the distance L (clearance) between the mantle 31 and the cone cave 32 on the outlet side of the crusher. The adjustment of the clearance value is determined by the position control of the vertical movement of the mantle 31 due to the hydraulic movement. Therefore, for each crusher, it is necessary to operate the above clearance value to the optimum value.
It becomes important in the sizing process.

Another important control symmetry in the sizing process is the control of the amount of steel feed, that is, the cutting amount, based on the control of the rotation number of the feeder, which determines the throughput in the sizing process. In carrying out the control of the clearance value and the control of the cutting amount, it is necessary to consider the grain size of the ore for grinding and its brand. Of these, regardless of brand, it must be said that the particle size is constantly changing due to segregation in the piles and hoppers that are the pre-steps of the particle size regulating process. When the particle size of the raw ore changes, in order to bring the particle size of the sized product into the specified value range, it is necessary to change the clearance value and the cut amount accordingly.

Therefore, conventionally, an operator at the site considers the change in the particle size of the raw ore, and uses the sampling result of the crushing sound of the crusher as a judgment factor to balance the entire sizing process while maintaining the balance of the crusher. The clearance value and the cutout amount were delicately controlled based on my own experience and intuition. In addition, conventionally, there is a conventional example in which a feedforward control is performed on the above-mentioned sizing process by using a mathematical model in principle according to a modern control theory or an optimal control theory (for example, Japanese Patent Publication No. 62-55899, Japanese Patent Laid-Open No. Sho 58-55899). 57-1
No. 0347).

[0011]

However, in a sizing plant, the operating states of a plurality of crushers, feeders, sieving machines, etc., which are stacked in a cascade, change every moment, and the grain size and brand of the raw ore are also changed. There is also a change disturbance
When trying to represent the behavior of the process by a mathematical model, the mathematical model was too complicated, and in reality, it was difficult to completely automate the sizing process. Therefore,
In the field of such raw material agglomerates, the sizing process cannot be completely automated, and in practice, the sizing process must be operated based on the experience and intuition of a skilled operator. Was there.

Therefore, it is no exaggeration to say that the operation of the sieving process is in an uncontrolled state, and the ratio of the agglomerated steel stones with a specified grain size to the powdered steel stones and the respective production amounts depend on the operation results. On the other hand, as a recent trend, the crushing of ore at Yamamoto is progressing, and the import ratio of powdered ore is increasing. This means that the ore does not need to be made unnecessarily for the sizing process, whereas the required amount of agglomerate must conventionally be produced in the sizing process. Therefore, in order to fully automate the production of agglomerated steel stones with a specified grain size, it is necessary to accurately detect the status of each facility during operation, and to quickly acquire and control the sizing product information under the current status. Therefore, a higher control method was required.

The present invention has been made in view of such conventional problems, and it is an object of the present invention to provide a sizing and crushing control device for a sizing plant capable of efficiently producing a lumped sizing product having a specified particle size. To aim.

[0014]

[Means for Solving the Problems] The inventors of the present invention have conducted various studies on a method for controlling sizing and sizing in a sizing plant. In the sizing plant, a multi-input, multi-output control system The sizing plan is based on the experience and intuition of a skilled operator whose control contents are multipurpose. It is operated and it is necessary to operate the plant by incorporating this experience and intuition, as mentioned above, it is difficult to create a precise model of the controlled object, etc. We have found that it is sufficient to apply the fuzzy control method to a sizing plant to achieve the above object.

The fuzzy inference describes a vague judgment method based on the intuition and experience of a skilled operator in the form of a control rule, and easily incorporates a situation judgment (large, small, etc.) in which ordinary measurement is difficult into a control calculation. It is possible to carry out control operations based on a lot of information. That is, in order to achieve the above object, the present invention is configured by combining a plurality of pulverizers and a sieving machine, in the sizing and pulverizing control device of the sizing plant, the amount of raw material agglomerate supplied to the sizing process. Raw material lump amount measuring means for measuring, sized lump amount measuring means for measuring lump amount after sizing, sized powder amount measuring means for sizing powder after sizing, current of the crusher Value, electric power value, vibration value, clearance value, etc. of the pulverizer operation state detecting means for detecting the operation state of the pulverizer, and sieve for detecting the operation state of the sieving machine such as current value, power value of the sieving machine Separator operating state detection means, input means for inputting operation pattern, raw material brand information, measurement value detected by each of the measuring means, detection value and operation pattern of the input means, raw material brand input information Take in,
Fuzzy inference means for fuzzy inferring the required amount of raw material agglomerates supplied to the sizing plant and the required value of the clearance of the crusher, and is supplied to the sizing plant based on the inference result in the fuzzy inference means A control means for controlling the feed amount of the raw material agglomerate and the clearance value of the crusher is provided.

Further, according to the present invention, the measured value of the raw material lump amount measuring means, the measured value of the sized lump material amount measuring means, the measured value of the sized powder material amount measuring means, and the crusher operating state detecting means The detected value, the detected value of the sieving machine operating state detecting means, the operation pattern of the input means, and the material brand input information are supplied to the fuzzy inference means, and the fuzzy inference is performed based on the supplied data.

In this fuzzy reasoning means, in order to increase the amount of the lump sized product and increase the pulverization efficiency of the raw lump, the required amount of the raw lump supplied to the sizing plant and the clearance of the crusher are increased. Fuzzy reasoning about required quantity.
Then, the control means controls the feed amount of the raw material lumps supplied to the sizing plant and the clearance value of the crusher based on the inference result.

As a result, it is possible to maintain the quality of the sized product in a suitable state and efficiently produce a lumped product with a specified particle size.

[0019]

Next, an embodiment of the present invention will be described. FIG. 2 is a flow sheet for explaining the sizing process of one embodiment of the sizing plant to which the present invention is applied. In FIG. 2, the crushers CR1 to CR4 are configured by accumulating in a cascade form, and similarly, the sieving machine SC1
~ SC8 are also configured by accumulating in a cascade fashion.

The supply of the raw material ore (lump) to the sizing plant constructed in this way is cut out from the raw material ore mountain P by the reclaimer RE, temporarily retained in the surge hopper SH, and then cut out by the apron feeder AF. It is controlled and done. This cutout amount is the belt wear BW
It is measured by 1. Further, the lump steel stone 40 and the fine ore 41, which are the sized products of the raw lumps, are belt wear BW2, B
Each is measured by W3.

Further, the +25 mm ratio and the -5 mm ratio in the agglomerated steel stone 40 and the powdered ore 41 are sampler SA, respectively.
Evaluated by As shown in FIG. 2, the crusher, the sieving machine and the like are connected by a belt conveyor shown by a solid line or a dotted line. The above process is composed of a cascade-like complicated flow including a recycle flow, and this flow can be variously changed depending on the operation pattern.

The raw material ore is crushed and sieved into agglomerated steel stone having a predetermined grain size range as it moves from the upstream side to the downstream side of the flow described in FIG. The sizing process described with reference to FIG. 2 includes various detectors that detect the operating states of various devices that make up the process, and inputs detection signals from the detectors to perform predetermined arithmetic control. And a controller for outputting the cutting amount of the apron feeder and the clearance value (L).

The block configuration of the controller is shown in FIG. 1 and will be described. Current measurement signals CR1A to CR4A of each crusher CR1 to CR4, power measurement signals CR1W to CR4W, vibration measurement signals CR1V to CR4V by a vibration sensor provided on the outer shell of each crusher, and the above of each crusher Clearance value (L) measurement signal CR1C to CR4
C is input from the data input module IN, and the master controller MC calculates the current value or the like of each crusher from this measurement signal. As described above, the current measurement value, power measurement value, vibration measurement value, and clearance measurement value of each crusher were used as factors for detecting the operating state of the crusher. The operating state of each crusher can be detected by using one or more of these measured values.

Similarly, the current measuring signals SC1A to SC8A and the power measuring signals SC1W to SC of the sieving machines SC1 to SC8 are similarly measured.
8W, weighing value measurement signals BW1W to BW3W of each belt wear BW1 to BW3, +25 mm amount ratio measurement signal SAU of the sampler SA, similarly, -5 mm amount ratio measurement signal S
AD, the operation pattern input signal SPP determined by the operation state of the sizing process, the raw material brand input signal SPM determined by the raw material brand are supplied to the data input module IN, and the master controller MC receives the supplied signal from the supplied signal. The current value of the sieving machine is calculated to determine the operation pattern and raw material brand.

The calculation result of the master controller MC is output to each control module 1 to 6 through the input interface circuit IN. The calculation result is finally set as the cut-out amount set value AFW of the apron feeder AF and the clearance set values S1 to S4 of the crushers CR1 to CR4 by sequentially passing through the control modules 1 to 6, and the data is obtained. It is output from the output module OUT.

The direct digital controller 8 determines the moving position of the mantle 31 and the feeder A according to the output value.
The set value of the rotation amount of F is calculated, and the operation signal is output so as to match the position and the rotation speed. An actuator (not shown) controls the number of rotations of the apron feeder AF and the movement amount of the mantle 32 of the crusher based on the operation signal determined from these set values.

The control module comprises an exponential averaging module 1, a grade calculation module 2, a fuzzy inference module 3, an OR module 4, a weighted averaging module 5, a first-order lag module 6, and the like. Since each of the control modules 1 to 6 is subjected to a subroutine processing, suffixes a to f are provided for each intended use. a to f = 1, 2, 3−, and the set value for each suffix differs depending on the intended use in the control module.

Next, regarding the operation of each control module,
explain. When the input value to this module is I, the exponential averaging module 1

[0029]

[Equation 1]

The exponential averaging process (1) is performed. The reason for performing such processing is that current values, electric power values, weighing values, vibration values, etc. input to the input module IN change drastically with time. This is because the true value of the change (n) is captured. The grade calculation module 2 outputs the grade of the membership function of ambiguous concept from the exponential average value of each input value (I) obtained by the exponential averaging module 1.

Next, the fuzzy inference module 3 performs fuzzy inference based on the grade value obtained by the grade calculation module 2 and outputs a membership function which is an inference result. In this fuzzy inference module, the cut-out amount AFW of the apron feeder AF and each of the pulverizers C are adjusted so that the ratio of the lump production amount to the sized product is large and the pulverization efficiency of the raw material lumps is high.
Infer the clearance values of R1-CR4.

Here, in this fuzzy inference module 3, when the sized product is out of the grain size control range,
A quality control logic that controls this to maintain a predetermined particle size, and when the sized product is within the particle size control range, prevents excessive crushing by each crusher CR1 to CR4, and also provides a suitable load balance and load factor. It is controlled by energy-saving logic that keeps the value.

The operation of the fuzzy inference module 3 will be described by taking the above energy saving logic as an example.
Is the powder ratio large by making the over-crush prevention control the first fuzzy control rule? (IF ₁₁ ), Is the load of the second crusher CR2 heavy? (IF ₁₂ ) is the antecedent part of the membership function, and the consequent part is the second crusher C based on this.
Large or small inference result of R2 clearance increase (THEN ₁ )
Get. Next, with load balance control as the second fuzzy control rule, is the load of the second crusher CR2 large? (IF
₂₁ ) is used as the antecedent part to obtain the membership function, and similarly, the inference result (THEN ₂ ) of the size of the increase in the clearance of the second crusher CR2 is obtained. Furthermore, if the load factor control is the third fuzzy control rule, does the load of the second crusher CR2 become too small? With (IF ₃₁ ) as the antecedent part, the inference result (THEN ₃ ) of the size of the increased clearance of the second crusher CR2 is obtained in the same manner as the inferences of the previous two cases.

In addition to the fuzzy inference shown in this example, in parallel with this, is the above-mentioned quality control logic defined in one of the control rules, and is the granularity largely out of the standard? As an antecedent part, fuzzy inference for inferring the increase amount of the apron feeder cutting amount AFW is also performed. Next, the OR module 4 includes an apron feeder AF and a second crusher CR.
The OR (logical sum) of the inference results obtained by the fuzzy inference module 3 is output for the control values of the second and third crushers.

[0035] That is, if you take the clearance increase control of the second grinder CR2 described above as an example, the inference result THEN ₁ ~THEN ₃ obtained by the above-mentioned fuzzy inference module 3
Is OR-added by this OR module 4 to specify the grade characteristic for each control value (inference value). Next, the weighted averaging module 5 calculates (inverse fuzzy) the weighted average (center of gravity) of the inferred values which are guided by the fuzzy inference and are OR-added.

This calculation processing will be described based on the above-mentioned example. The grade characteristics of each control value obtained by the OR module 4 are weighted and averaged to obtain the center of gravity, and the clearance of the second crusher CR2 is obtained. Will be determined as a specific control value. Then, the control value calculated as described above is input to the first-order delay module 6, and the same-order delay module 6 performs the first-order delay processing of the transfer function shown in the following expression (2) to clear the clearance of the crusher CR. It prevents sharp changes in the value output and the apron feeder AF cutout amount output.

K / (1 + sT) (2) where k is a gain constant, s is a Laplace operator, and T is a time constant. The first-order delay module 6 finally outputs the clearance setting values S1 to S4 of the crusher and the cut-out amount setting value AFW of the apron feeder from the data output module OUT.

First, the master control MS is
Each module of 6 is programmed to control the entire processing flow. That is, Master Control M
In S, each module 1 to 6, a data input module IN, and a data output module OUT are converted into a general-purpose subroutine and arbitrarily combined and processed. For example, as an example,

[0039]

[Equation 2]

A general-purpose arbitrary combination processing flow such as is possible. In the above processing flow, CR1W ~
CR4W shows the power measurement for each pulverizer, IN represents the data input module, C _4, C ₅ represents a fuzzy inference module 6, A ₆ to A ₉ represents an exponential averaging module 1, D ₇ Represents an OR calculation module 4, E ₂ represents a weighted averaging module 5, F ₄ represents a first-order lag filter module 6, and AFW represents a cutout amount set value of the apron feeder AF.

Of course, the operation pattern SPP,
It is also possible to appropriately change this combination by changing the input of the brand SPM. That is, changing the operation pattern SPP means changing the cascade configuration of the sieving machine and the crusher, and changing the brand is different from the brand of the ore to be milled and pulverized (the grain size and hardness are different). And to say.

The master controller MS also manages the time of execution control of the entire main control according to the above embodiment. The sizing process requires a very long process safety time. It takes about 15 minutes in a normal sizing plant with four crushers until the raw ore in the apron feeder reaches the final sizing product, so frequent control operation output causes control disturbance. .

Therefore, the master controller MS controls the execution of the data input interface IN, the data output interface OUT, and the control modules 1 to 6 by appropriate time management that does not cause control disturbance. Therefore, in the present invention, these consideration intervals and sampler SA
Is controlled to be executed as a kick, and the cut-out amount of the apron feeder AF and the clearance values S1 to S4 of the plurality of crushers are sequentially output according to the time progress of the process.

Also in the first-order delay module (F _f ) 6, F ₁ , F ₂ , F ₃ , F ₄ , and F ₅ are apron feeder cut-out amount setting value AFW and crusher clearance value (CR1C to CR4C). If it corresponds to the above ~
The time constant T of the transfer function of F _{1 to} F ₅ (Equation (2)) is set in consideration of the time delay of the downstream feeling. In the above-described embodiment, the disk 7 for storing and accumulating input and output data is provided for development for optimal design of each module. The control level can be increased by analyzing the data stored in the disk 7.

In the above embodiment, the case of sizing iron ore has been described, but the present invention is not limited to this, and the present invention can be applied to the case of sizing coal, for example. Further, in the above-described embodiment, the factor for detecting the operating state of the crusher sieving machine is not limited to the one described in this specification, and other factors can be used.

In the above-described embodiment, the ratio of the lumps and the powders to the sized product is obtained by the +25 mm amount ratio measurement signal SAU of the sampler SA and the -5 mm amount ratio measurement signal SAD as well. BW1 to BW
The ratio can be obtained from the measured amount of 3.

[0047]

As described above, according to the present invention, since the raw material agglomerates are sized by the fuzzy control method, the quality of the sized products can be maintained in an appropriate state. The ratio of the mass production to the sized product is increased, and the pulverization efficiency of the raw material lump is also increased, so that the sized product having the specified particle size can be efficiently produced.

Therefore, it is possible to obtain an agglomerated particle-sized product by avoiding unnecessary powder-sized particles of the raw material by using as little electric power as possible.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of an embodiment of a particle size control and crushing control apparatus for a particle size control plant according to the present invention.

FIG. 2 is a flow sheet illustrating a sizing process of an embodiment of a sizing plant to which the present invention is applied.

FIG. 3 is a sectional view showing the structure of a crusher.

[Explanation of symbols]

 1 Exponential averaging module 2 Grade calculation module 3 Fuzzy inference module 4 OR calculation module 5 Weighted averaging module 6 1st-order lag filter module 7 Disk 8 Direct digital controller

Claims (1)

[Claims] 1. A raw material agglomerate amount measuring means for measuring the amount of raw material agglomerates supplied to a particle size regulating process in a particle size control and pulverization control device of a particle size control plant, which comprises a plurality of crushers and a sieving machine in combination. A sized lump amount measuring means for measuring the sized lump amount, a sized powder amount measuring means for measuring the sized powder substance, a current value, a power value of the crusher,
Crusher operating state detecting means for detecting the operating state of the pulverizing machine such as vibration value and clearance value, and sieving machine operating state detecting means for detecting the operating state of the sieving machine such as current value and electric power value of the sieving machine And input means for inputting the operation pattern and raw material brand information, and the measurement values and detection values measured and detected by the respective measuring and detecting means, the operation pattern of the input means and the raw material brand input information, and adjusted. Fuzzy inference means for fuzzy inferring the required amount of raw material lumps supplied to the grain plant and the required value of the clearance of the crusher, and the raw material supplied to the sizing plant based on the inference result in the fuzzy inference means A sieving and pulverizing control device for a sieving plant, comprising: a control unit that controls a mass supply amount and a clearance value of the pulverizer.