KR102449020B1 - System and Method for Predicting Strength of Sintered Ore - Google Patents

Abstract

A sintered ore intensity prediction system according to an aspect of the present invention that can predict the strength of sintered ore in real time includes: a sintering machine bogie in which a mixture of a main raw material, a secondary raw material, and a fuel is ignited and burned to generate sintered ore; a cooler for cooling the sintered ore; And Characteristics information of the blending raw material, the blending condition information of the blending raw material, and the blending raw material characteristic information at the target time point where the intensity prediction of the sintered ore is required in the sintered ore intensity prediction model using the operating condition information for generating the sintered ore as a variable, It characterized in that it comprises a sintered ore intensity prediction unit for predicting the intensity of the sintered ore to be generated at the target time by inputting the mixing condition information and the operation condition information.

Description

System and Method for Predicting Strength of Sintered Ore

The present invention relates to a steel production process, and more particularly, to measuring the strength of sintered ore.

The ironmaking process is the first step in the production of various types of steel products in the steel mill, and in this ironmaking process, sinter and coke are burned in a blast furnace to produce liquid molten iron called molten iron.

At this time, the sintered ore is produced to have a certain size by mixing iron ore (magnetite, hematite, limonite) with auxiliary materials such as limestone or serpentine, heating it to a high temperature, and cooling and crushing it.
For example, sintered ore may be manufactured as described in Korean Patent Application Laid-Open No. 10-2019-0052380 (Sintered ore manufacturing apparatus and manufacturing method, published on May 16, 2019).

When the generation of the sintered ore is completed through this sintering process, the strength of the sintered ore, which is one of the criteria for evaluating the quality of the generated sintered ore, is measured. The intensity of the sintered ore indicates the degree of differentiation of the sintered ore during the process of transporting the sinter to the blast furnace or the charging process into the blast furnace, and is also called the falling strength of the sintered ore.

In general, in order to measure the strength of sintered ore, a sampler that can collect a sample is installed at the end of the sintering process to collect a sample of the sintered ore, and the collected sample is dropped in units of a certain time (about 4 hours) By doing so, the intensity of the sintered ore is measured. At this time, the larger the measured number indicates that the strength of the sintered ore is high, and the higher the strength of the sintered ore is, the less the probability of differentiation occurring while the sintered ore is transported until it is charged into the blast furnace, so the breathability of the sintered layer in the blast furnace can be improved. .

However, in the case of the existing method for measuring the strength of sintered ore, it is possible to measure the sintered ore only after a lot of time (approximately 6 hours) has elapsed from the completion of the production of the sintered ore. According to the strength of the sintered ore, there is a problem that the operating conditions or mixing ratio cannot be adjusted in real time during the sintering operation.

In addition, in the case of the conventional method for measuring the intensity of sintered ore, there is a problem in that the intensity of the sintered ore cannot be accurately measured because there is a deviation depending on the location and time of collecting the sample of the sintered ore.

The present invention is to solve the above-described problems, the present invention has its technical features to provide a sintered ore intensity prediction system and method that can predict the intensity of the sintered ore in real time.

In addition, the present invention is another technical feature to provide a sintered ore intensity prediction system and method capable of guiding the operating conditions for the sintered ore to have a predetermined target intensity based on the predicted intensity of the sintered ore.

In addition, the present invention is another technical feature to provide a sintered ore strength prediction system and method that can guide the optimal mixing conditions that can have the target strength at the minimum cost using the sinter strength prediction model.

A sintering ore intensity prediction system according to an aspect of the present invention for achieving the above object is a sintering ore bogie in which the blending raw material in which the main raw material, auxiliary raw material, and fuel are mixed for generating sintered ore is ignited and burned to generate sintered ore; a cooler for cooling the sintered ore; And the characteristic information of the compounding raw material, the compounding condition information of the compounding raw material, and the sintered ore intensity prediction model using the operation condition information for generating the sintered ore as variables. It characterized in that it comprises a sintered ore intensity prediction unit for predicting the intensity of the sintered ore to be generated at the target time by inputting the mixing condition information and the operation condition information.

The method for predicting the intensity of sintered ore according to an aspect of the present invention for achieving the above object is characteristic information of the compounding raw material used when generating sintered ore for each time in the past in the sintering machine, information on the compounding conditions of the compounding material, the sintered ore generation Collecting operating condition information for, and intensity data of the sintered ore at that time; generating a sintered ore intensity prediction model using the characteristic information of the compounding raw material, compounding condition information of the compounding raw material, and operating condition information for generating the sintered ore as variables using the collected data; And predicting the intensity of the sintered ore to be generated at the target time by inputting the blending raw material characteristic information, the blending condition information, and the operating condition information at the target time point where the intensity prediction of the sintered ore is required to the sintered ore intensity prediction model characterized in that

According to the present invention as described above, since the strength of the sintered ore can be predicted during the sintering operation, there is an effect that the predicted strength of the sintered ore can be reflected in the sintering operation in real time.

In addition, according to the present invention, it is possible to guide the operator to the operating conditions for making the sintered ore have the target strength based on the predicted strength value of the sintered ore, so that even if the operator is not an expert in the operation, not only can the sintered ore having the target strength be easily produced, but also the operator There is an effect that it is possible to remove the variation in the intensity of the sintered ore that may occur according to the

In addition, by varying the operating conditions based on the predicted value of the intensity of the sintered ore, the sintered ore can have a target strength, so that the strength of the sintered ore can be kept constant, which has the effect of securing the old yellow stability.

In addition, according to the present invention, it is possible to guide the operator to the optimal mixing conditions for the generation of sintered ore by using the sintered ore strength prediction model, thereby reducing the manufacturing cost by reducing the raw material mixing cost, thereby improving the productivity of the sintered ore. It has the effect that it can.

1 is a block diagram schematically showing the configuration of a sintered ore intensity prediction system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing the configuration of the control device shown in FIG. 1 .
3 is a table showing an example of characteristic information of a compounding raw material, an example of compounding condition information, and an example of operating conditions according to an embodiment of the present invention.
4 is a flowchart showing a method for predicting the intensity of sintered ore according to an embodiment of the present invention.

Like reference numerals refer to substantially identical elements throughout. In the following description, a detailed description of configurations and functions known in the art and cases not related to the core configuration of the present invention may be omitted. The meaning of the terms described herein should be understood as follows.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in a variety of different forms, only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the technical field to which the present invention belongs It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims.

The shapes, sizes, proportions, angles, numbers, etc. disclosed in the drawings for explaining the embodiments of the present invention are illustrative and the present invention is not limited to the illustrated matters. Like reference numerals refer to like elements throughout. In addition, in describing the present invention, if it is determined that a detailed description of a related known technology may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

When 'including', 'having', 'consisting', etc. mentioned in this specification are used, other parts may be added unless 'only' is used. When a component is expressed in the singular, the case in which the plural is included is included unless specifically stated otherwise.

In interpreting the components, it is interpreted as including an error range even if there is no separate explicit description.

In the case of a description of the positional relationship, for example, when the positional relationship of two parts is described as 'on', 'on', 'on', 'beside', etc., 'right' Alternatively, one or more other parts may be positioned between two parts unless 'directly' is used.

In the case of a description of a temporal relationship, for example, 'immediately' or 'directly' when a temporal relationship is described as 'after', 'following', 'after', 'before', etc. It may include cases that are not continuous unless this is used.

Although the first, second, etc. are used to describe various components, these components are not limited by these terms. These terms are only used to distinguish one component from another. Accordingly, the first component mentioned below may be the second component within the spirit of the present invention.

The term “at least one” should be understood to include all possible combinations from one or more related items. For example, the meaning of “at least one of the first, second, and third items” means 2 of the first, second, and third items as well as each of the first, second, or third items. It may mean a combination of all items that can be presented from more than one.

Each feature of the various embodiments of the present invention may be partially or wholly combined or combined with each other, technically various interlocking and driving are possible, and each of the embodiments may be implemented independently of each other or may be implemented together in a related relationship. may be

Hereinafter, an embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram schematically showing the configuration of a sintered ore intensity prediction system according to an embodiment of the present invention. As shown in FIG. 1 , the sintered ore intensity prediction system 100 includes a sintering machine 110 and a sintering machine controller 130 .

The sintering machine 110 is a facility for generating sintered ore by using a compounding raw material, and injecting the generated sintered ore into a blast furnace where the ironmaking process is performed. As shown in FIG. 1 , the sintering machine 110 is a main blower 112 , a sintering wind supply line 114 , a windshield 116 , a sintering machine cart 118 , a supply unit 120 , and a discharge unit 122 . ), a traveling rail 124 , and an ignition unit 126 .

The main blower 112 supplies the sintering wind to the sintering machine bogie 118 through the sintering wind supply line 114 .

The sintering wind supply line 114 is communicated with the main blower 112 and supplies the sintering wind supplied from the main blower 112 to the sintering machine bogie 118 through a plurality of wind chambers 116 .

The plurality of wind boxes 116 communicate with the sintering wind supply line 114 , and suck the sintering wind supplied through the sintering wind supply line 114 and supply it to the sintering machine cart 118 .

The sintering machine bogie 118 is charged with the compounding raw material supplied from the supply unit 120 to a predetermined thickness to produce sintered ore. Specifically, the sintering machine bogie 118 moves along the running rail 124 installed on the plurality of windshields 116 after the ignition process by the ignition unit 126 . At this time, an opening (not shown) for supplying sintering wind is formed in the lower portion of the sintering machine bogie 118 , so that the sintering wind supplied from the inlet of the plurality of windsang 116 is passed through the opening of the sintering machine bogie 118 . It is supplied to the blended raw material loaded inside the bogie 118 .

According to the supply of the sintering wind, the blended raw material charged to the sintering machine bogie 118 and oxygen in the sintering wind are burned, thereby sintering the blended raw material. When complete combustion is made to the lower layer of the blended raw material charged to the sintering machine cart 118, the sintering of the blended raw material is terminated, and only the agglomerated sintered ore remains in the sintering machine cart 118. The sintered ore that has been sintered is discharged through the discharge unit 122 .

The supply unit 120 supplies the blended raw material into the sintering machine bogie 118 . In one embodiment, the blend material is produced by mixing moisture into the main raw material, the auxiliary raw material, and the fuel. In an embodiment, the main raw material may include iron ore, the auxiliary raw material may include quicklime, limestone, or silica stone, and the fuel may include coke or coal (anthracite).

The discharge unit 122 discharges sintered ore from the sintering machine bogie 118 .

The ignition unit 126 ignites the blended raw material supplied to the sintering machine bogie 118 . The compounding raw material charged into the sintering machine bogie 118 starts to be ignited from the upper layer by the ignition unit 126 and then the lower layer is ignited.

The sintered ore discharged to the discharge unit 122 is cooled by the cooler 128 and then crushed to a predetermined size by a crusher (not shown) and injected into the blast furnace.

The control device 130 controls the operation of the sintering machine 110 . In particular, the control device 130 according to the present invention generates a sintered ore intensity prediction model for predicting the intensity of the sintered ore, and predicts the intensity of the sintered ore at a target time using the generated sintered ore intensity prediction model. In addition, when the predicted value of the sintered ore intensity calculated through the sintered ore intensity prediction model is different from the predetermined target value of the sintered ore intensity, the control device 130 controls the sintering machine 110 so that the predicted value of the sintered ore intensity can follow the target value of the sintered ore intensity. control the action.

Hereinafter, the configuration of the control device 130 according to the present invention will be described in more detail with reference to FIG. 2 .

2 is a block diagram showing the configuration of a control device according to an embodiment of the present invention. As shown in FIG. 2 , the control device 130 according to an embodiment of the present invention includes a sintered ore intensity prediction unit 210 , a sintered ore intensity prediction model 215 , and an operating condition guide unit 220 . In addition, the control device 120 may further include a data collection unit 230 , a model generation unit 240 , and a compounding condition guide unit 250 .

First, the sintered ore intensity prediction unit 210 is to be generated at the target time by inputting the compounding raw material characteristic information, the compounding condition information, and the operating condition information at the target time at which the intensity prediction of the sintered ore is required to the sintered ore strength prediction model 215 Predict the strength of sintered ore. In one embodiment, the sintered ore intensity prediction model 215 is a model for outputting the predicted value of the intensity of the sintered ore at the target point by using the characteristic information of the compounding raw material, the compounding condition information of the compounding raw material, and the operation condition information for generating the sintered ore as a variable to be.

According to this embodiment, the sintered ore intensity prediction model 215 may be defined as in Equation 1 below.

In Equation 1, Y represents the predicted value of the intensity of the sintered ore predicted at the target time, X1 is a first variable representing the characteristic information of the compounding raw material set at the target time, W1 represents the first weight given to the first variable, X2 is a second variable indicating information on the compounding conditions set at the target time, W2 is a second weight assigned to the second variable, X3 is a third variable indicating information about operating conditions set at the target time, W3 is the third Indicates the third weight given to the variable.

In this case, the first weight W1, the second weight W2, and the third weight W3, which are given to each of the first to third variables, have the first weight W1 within a range where the sum of the first weight W1 is equal to 1. It may be set such that it is greater than the second weight W2 and the second weight W2 is greater than the third weight W3 . This means that the characteristic information of the compounding raw material to which the first weight (W1) is applied has the greatest effect on the strength of the sintered ore, and the operating condition information to which the third weight (W3) is applied is the characteristic information of the compounding material or the compounding condition information. This is because the effect on the strength of sintered ore is small compared to that of the sintered ore.

3 shows an example of characteristic information of a compounding raw material according to an embodiment of the present invention, an example of compounding condition information, and an example of operating conditions. As can be seen from FIG. 3, the characteristic information of the blending raw material includes the total Fe content content value of the blending raw material, the FeO element content value of the blending raw material, the SiO2 content content value of the blending raw material, the Al2O3 content content value of the blending raw material, the CaO component of the blending raw material It may include at least one of a content rate value and a content rate value of the MgO component of the blending raw material.

In addition, information on the mixing conditions includes the amount of semi-gloss used in the ingredients, the total content of C component included in the ingredients, the total content of FeO ingredients included in the ingredients, the input ratio of the main and auxiliary ingredients, the use ratio of the main ingredient in the blending raw materials, and sintered ore may include at least one of basicity information of

In addition, the operation condition information may include at least one of the thickness of the compounding raw material charged to the sintering machine bogie, the moving speed of the sintering machine bogie, the rotational speed of the cooler, and the air permeability in the sintering machine bogie.

As described above, according to the present invention, since the sintered ore intensity prediction model 215 is generated using the characteristic information of the blending raw material, the blending condition information, and the operating condition information, the sintered ore intensity prediction unit 210 is the intensity of the sintered ore at the target time. When the prediction is requested, the characteristic information of the compounding raw material, the compounding condition information, and the operation condition information set at the target time are obtained from the sintering machine 110, and the obtained characteristic information of the compounding material, the compounding condition information, and the operation condition By inputting the information into the sintered ore intensity prediction model 215 expressed by Equation 1 above, it is possible to generate a predicted value of the sintered ore intensity at the target time point.

The sintered ore intensity prediction unit 210 outputs the predicted intensity value of the generated sintered ore to the operation condition guide unit 220 .

The operation condition guide unit 220 is based on the intensity predicted value of the sintered ore at the target time output from the sintered ore intensity prediction unit 210 and the predetermined intensity target value of the sintered ore. Operation conditions for the generation of sintered ore decide

Specifically, the operation condition guide unit 220 compares the intensity predicted value of the sintered ore with the target intensity of the sintered ore. As a result of the comparison, when the predicted intensity value of the sintered ore and the target intensity of the sintered ore are the same, it is determined to maintain the currently set operating conditions.

However, when the predicted intensity value of the sintered ore and the target intensity of the sintered ore are different, the operating condition guide unit 220 determines the target operating conditions required for the predicted intensity of the sintered ore to follow the target intensity of the sintered ore.

In one embodiment, the operating condition guide unit 220 is at least one of the moving speed of the sintering machine bogie 118 and the rotation speed of the cooler 128 required for the intensity predicted value of the sintered ore to follow the target intensity of the sintered ore. can be determined as the target operating condition.

As an example, the operating condition guide unit 220 sets the moving speed of the sintered ore lower than the current moving speed of the sintering machine trolley 118 when the predicted intensity value of the sintered ore is smaller than the target operating condition. can be decided This is, the strength of the sintered ore is inversely proportional to the speed of the sintering machine bogie 118. When the speed of the sintering machine's bogie 118 is too slow, the sintered ore's strength is increased more than necessary, but the productivity of the sintered ore is decreased, and vice versa If the speed of the sintering machine bogie 118 is too fast, the productivity is increased, but the strength of the sintered ore is decreased.

Accordingly, the operating condition guide unit 220 may adjust the moving speed of the sintering machine bogie 118 according to the predicted intensity of the sintered ore so that the intensity of the sintered ore becomes a target value.

In the above-described embodiment, the operating condition guide unit 220 has been described as controlling the moving speed of the sintering machine bogie 118 or the rotational speed of the cooler 128, but this is only an example of the operating condition guide unit 220 may be able to adjust the amount of sintering wind supplied by the main blower 112 by varying the operating conditions of the main blower 112.

The operating condition guide unit 220 provides the determined target operating condition to the operator so that the operator can change the operating condition of the sintering machine 110 to the target operating condition determined by the operating condition guide unit 220 .

In the above-described embodiment, the operating condition guide unit 220 guides the determined target operating condition to the operator, and it has been described that the operator changes the operating condition of the sintering machine 110 to the target operating condition, but in another embodiment In, the operating condition guide unit 220 may directly change the operating conditions of the sintering machine 110 according to the determined target operating conditions. According to this embodiment, the operating condition guide unit 220 may continuously change the operating conditions until the intensity predicted value of the sintered ore follows the target intensity value of the sintered ore.

The data collection unit 230 collects the characteristic information of the compounding raw material, the compounding condition information, the operating condition information, and the intensity data of the sintered ore obtained at the time for each time in the past. Specifically, the data collection unit 230 is the total Fe component content value of the compounding raw material, the FeO component content value of the compounding raw material, the SiO2 component content rate value of the compounding raw material, the Al2O3 component content rate value of the compounding material, the CaO component content rate value of the compounding raw material , MgO component content value of the compounding raw material, the amount of raw material cut out by bin (BIN) of the compounding raw material, the amount of moisture contained in the raw material for each compounding raw material bin, the amount of semi-gloss in the compounding raw material, the amount of the compounding material charged to the sintering machine cart It is possible to collect data such as thickness, moving speed of the sintering machine bogie, or the rotational speed of the cooler, air permeability in the sintering machine bogie.

When data used for generating sintered ore is collected in the sintering machine 110, the data collection unit 230 uses a machine learning algorithm (eg, a regression based model (MLR, Lasso) or a tree-based model (eg, Decision Tree) for the collected data). or Random Forest, etc.) can be used to derive effective factors affecting the strength of sintered ore.

Meanwhile, the data collection unit 230 may additionally generate a derived variable by using the collected data. As an example, the data collection unit 230 includes the total content of C component included in the raw material, the total content of FeO contained in the raw material, the input ratio of the main raw material and the auxiliary raw material, the use ratio of the main raw material among the blending raw materials, and the basicity information of the sintered ore. At least one can be created as a derived variable.

Among the ingredients, the higher the Fe component, the better, and the SiO2 component and Al2O3 component can increase the amount of slag, so the smaller the better. Derivative variables can be created by using factors such as the ratio of the input amount and the amount of coke to be mixed. In addition, when the particle size of the compounding material charged in the sintering machine truck increases, air permeability in the sintering machine truck improves and the sinter ore productivity increases due to good combustion, thereby generating a particle size derivative variable of the compounding material.

At this time, the total content of C component included in the raw material can be calculated using the extracted amount of each bin (BIN) of the blending raw material and the value of the C content in the blending raw material, and the total content of FeO component included in the blending raw material Silver can be calculated using the extracted amount of raw materials for each bin (BIN) and the content of FeO in the blending raw materials. In addition, the basicity information of the sintered ore can be calculated based on the extracted amount of each bin (BIN) of the raw material, the CaO content value in the blending raw material, and the SiO2 content rate value in the blending raw material.

On the other hand, the data collection unit 230 groups the effective factors or derivative variables related to the characteristics of the compounding raw material among the derived effective factors and the derived variables to generate the characteristic information of the compounding raw materials, and the effective factors or derivatives related to the compounding conditions. Variables are grouped to generate compounding condition information, and operating condition information can be generated by grouping effective factors or derived variables related to operating conditions.

The model generating unit 240 generates the sintered ore intensity prediction model 215 as described above by using the characteristic information, the compounding condition information, and the operating condition information of the compounding raw materials grouped by the data collecting unit 230 . In one embodiment, the model generation unit 240, based on the effective factors and derived variables collected by the data collection unit 230, the relationship between factors affecting the intensity of the sintered ore is a big data-based machine learning technique. It is possible to generate the sintered ore intensity prediction model 215 by complexly reflecting the . At this time, the sintered ore intensity prediction model generator 240 may give the same weight to the effective factors or derived variables included in the same group when the sintered ore intensity prediction model 215 is generated, but each effective factor and the derived variable Different weights may be applied to them.

On the other hand, the model generator 240 may determine the weight in consideration of this when generating the sintered ore intensity prediction model because the sintering strength and productivity are in inverse proportion to each other.

The model generating unit 240 compares the sintered ore intensity predicted value generated by the sintered ore intensity prediction unit 210 with the sintered ore intensity actually generated at that time, and uses the comparison result to learn the sintered ore intensity prediction model 215 have.

The blending condition guide unit 250 compares the predicted intensity value of the sintered ore with the target intensity value of the sintered ore, and the predicted intensity value of the sintered ore may guide the blending conditions required to follow the intensity target value of the sintered ore.

In one embodiment, the mixing condition guide unit 250 may determine the target intensity value of the sintered ore as the maximum value among the intensity values that the sintered ore can have under the corresponding operating conditions, and the mixing condition guide unit 250 is the corresponding operating condition. In the condition, it is possible to determine the information of the mixing condition required for the sintered ore to have the corresponding intensity target value. The blending condition guide unit 250 may determine a blending condition having the lowest raw material cost as an optimal blending condition by using a linear programming method (LP) among the determined blending conditions.

The reason for guiding the optimal mixing conditions through the mixing condition guide unit 250 in the present invention is to use an excessive amount of raw materials for component-oriented mixing in order to ensure that the sintered ore has a target strength value, or to set excessive operating conditions for quality and productivity. This is because doing so may result in wastage. Therefore, the present invention proposes a blending condition that can minimize the raw material cost while allowing the sintered ore to maintain the target strength value under the corresponding operating conditions by utilizing the optimization algorithm based on the linear programming method (LP) by the blending condition guide unit 250 be able to do

Hereinafter, a method for predicting the intensity of sintered ore according to the present invention will be described.

4 is a flowchart illustrating a method for predicting the intensity of sintered ore according to an embodiment of the present invention. The method for predicting the intensity of the sintered ore shown in FIG. 4 may be performed by the control device shown in FIG. 2 .

First, the control device collects the characteristic information of the compounding raw material used when generating the sintered ore for each time in the past in the sintering machine, the compounding condition information of the compounding raw material, the operating condition information for generating the sintered ore, and the intensity data of the sintered ore at the time Collect (S400).

In one embodiment, the data collected by the control device is the total Fe component content value of the compounding material, the FeO component content value of the compounding material, the SiO2 component content value of the compounding material, the Al2O3 component content rate value of the compounding material, CaO of the compounding material Component content value, MgO component content value of the compounding material, the amount of raw material cut out for each bin (BIN) of the compounding raw material, the amount of moisture contained in the raw material for each compounding raw material bin, the amount of semi-gloss in the compounding material, the amount charged to the sintering machine bogie It may include the thickness of the compounding material, the moving speed of the sintering machine bogie, or the rotational speed of the cooler, the air permeability in the sintering machine bogie.

When the data as described above are collected, the control device derives effective factors affecting the intensity of the sintered ore using the collected data using a machine learning algorithm.

On the other hand, the control device generates a derived variable using the collected data or effective factors. As an example, the control device derives at least one of the total content of C component included in the blending raw material, the total content of FeO ingredient included in the blending raw material, the input ratio of the main raw material and the subsidiary raw material, the use ratio of the main raw material in the blending raw material, and the basicity information of the sintered ore It can be created as a variable.

Among the ingredients, the higher the Fe component, the better, and the SiO2 component and Al2O3 component can increase the amount of slag, so the smaller the better. Derivative variables can be created by using factors such as the ratio of the input amount and the amount of coke to be mixed. In addition, when the particle size of the compounding material charged in the sintering machine truck increases, air permeability in the sintering machine truck improves and the sinter ore productivity increases due to good combustion, thereby generating a particle size derivative variable of the compounding material.

At this time, the total content of C component included in the raw material can be calculated using the extracted amount of each bin (BIN) of the blending raw material and the value of the C content in the blending raw material, and the total content of FeO component included in the blending raw material Silver can be calculated using the extracted amount of raw materials for each bin (BIN) and the content of FeO in the blending raw materials. In addition, the basicity information of the sintered ore can be calculated based on the extracted amount of each bin (BIN) of the raw material, the CaO content value in the blending raw material, and the SiO2 content rate value in the blending raw material.

The control device generates the characteristic information of the compounding raw material by grouping the effective factors or derived variables related to the characteristics of the compounding raw material among the derived effective factors and derived variables, and by grouping the effective factors or derived variables related to the compounding conditions, the compounding conditions Information is created, and operating condition information can be generated by grouping effective factors or derived variables related to operating conditions. An example of the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information is shown in FIG. 3 . Since the description of the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information has been described in the part related to FIG. 3 , a detailed description thereof will be omitted.

After that, the control device generates a sintered ore intensity prediction model using the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information collected in S400, the characteristic information of the compounding material, the compounding condition information, and the operating condition information as variables. do (S410).

Specifically, the control device complexly reflects the relationship between factors affecting the strength of the sintered ore based on the characteristic information of the compounding raw material collected in S400, the compounding condition information, and the operating condition information using a big data-based machine learning technique. By doing so, it is possible to generate a sintered ore intensity prediction model.

An example of the sintered ore intensity prediction model generated by the control device is described in Equation 1 above. Since the description of the sintered ore intensity prediction model has been described in the description of Equation 1 above, a detailed description thereof will be omitted.

The control device may give the same weight to the effective factors or derived variables included in the same group when generating the sinter intensity prediction model, but different weights may be applied to each effective factor and the derived variables . On the other hand, since the sintering strength and productivity are in inverse proportion to the control device, the weight may be determined in consideration of this when generating the sintered ore strength prediction model.

When the generation of the sintered ore intensity prediction model is completed, the control device inputs the compounding raw material characteristic information, the mixing condition information, and the operating condition information at the target time at which the intensity prediction of the sintered ore is required to the sintered ore strength prediction model to be generated at the target time. A predicted value of the intensity of the sintered ore is generated (S420).

Thereafter, the control device compares the predicted intensity value of the sintered ore generated in S420 with a predetermined target intensity of the sintered ore (S430). As a result of the comparison, when the predicted intensity value of the sintered ore is different from the intensity target value of the sintered ore, a target operating condition required for the intensity predicted value of the sintered ore to follow the target intensity value of the sintered ore is set ( S440 ).

In an embodiment, the target operating condition may include at least one of a movement speed of the sintering machine bogie and a rotation speed of a cooler required for the intensity predicted value of the sintered ore to follow the target intensity of the sintered ore.

As an example, when the predicted value of the intensity of the sintered ore is less than the target intensity of the sintered ore, the control device may determine the target operating condition for the moving speed of the sintering machine bogie reduced than the moving speed of the current sintering machine bogie. This is, the strength of the sintered ore is inversely proportional to the speed of the sintering machine bogie, and when the speed of the sintering machine's bogie is too slow, the productivity of the sintered ore is reduced instead of increasing the strength of the sintered ore more than necessary. If it is too fast, productivity will increase, but the strength of the sintered ore will decrease.

Therefore, the control device can make the intensity of the sintered ore a target value by adjusting the moving speed of the sintering machine bogie according to the predicted value of the intensity of the sintered ore.

In the above-described embodiment, the control device has been described as adjusting the moving speed of the sintering machine bogie or the rotational speed of the cooler, but this is only an example. It may be possible to adjust the amount of sintering wind supplied by the blower.

The control device provides the determined target operating condition to the operator (S445), so that the operator can change the operating condition of the sintering machine to the target operating condition.

In the above-described embodiment, it has been described that the control device guides the operator to the determined target operating condition and the operator changes the operating condition of the sintering machine to the target operating condition. Depending on the operating conditions, it may be possible to directly change the operating conditions of the sintering machine. According to this embodiment, the control device may continuously change the operating conditions until the intensity predicted value of the sintered ore follows the target intensity value of the sintered ore.

As a result of the comparison in S430, when the predicted value of the intensity of the sintered ore is the same as the target intensity of the sintered ore, the control device determines to maintain the currently set operating conditions (S450). In this case, the same is a concept including the same within the error range.

On the other hand, although not shown in FIG. 4 , when the predicted intensity value of the sintered ore generated in S420 is different from the target intensity of the sintered ore, the intensity predicted value of the sintered ore sets the mixing conditions required to follow the target intensity value. may be

Specifically, the control device may determine the intensity target value of the sintered ore as the maximum value among the intensity values that the sintered ore can have under the corresponding operating conditions, and information on the mixing conditions required for the sintered ore to have the corresponding target intensity under the corresponding operating conditions. can be decided The control device may use a linear programming method (LP) among the determined blending conditions to determine a blending condition having the lowest raw material cost as an optimal blending condition, and provide the determined optimal blending condition to the operator.

In the present invention, the reason that the control device determines the optimal mixing conditions and provides them to the operator is that excessive raw materials are used for component-centered mixing in order to ensure that the sintered ore has a target strength value, or excessive operating conditions are set for quality and productivity. This is because waste may occur. Therefore, in the present invention, the control device utilizes a linear programming (LP)-based optimization algorithm, so that the sintered ore maintains the target strength value under the corresponding operating conditions, and it is possible to present the mixing conditions that can minimize the raw material cost.

In addition, although not shown in FIG. 4 , the control device may continuously learn the sintered ore intensity prediction model using the comparison result by comparing the sintered ore intensity predicted value generated in S420 with the sintered ore intensity actually generated at the time point.

Those skilled in the art to which the present invention pertains will understand that the above-described present invention may be embodied in other specific forms without changing the technical spirit or essential characteristics thereof.

Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: sinter strength prediction system 110: sintering machine
112: main blower 114: sinter wind supply line
116: wind 118: sintering machine bogie
120: supply unit 122: discharge unit
124: running rail 126: ignition unit
128: cooler 130: control unit
2210: sintered ore intensity prediction unit 215: sintered ore intensity prediction model
220: operating condition guide unit 230: data collection unit
240: model generation unit 250: mixing condition guide unit

Claims (15)

a sintering machine bogie in which the main raw material, the auxiliary raw material, and the mixed raw material for the production of sintered ore is ignited and burned to produce sintered ore;
a cooler for cooling the sintered ore;
Characteristic information of ingredients, mixing condition information, operation condition information, and strength data of sintered ore are collected for each time in the past, and the collected data is used to influence the strength of sintered ore using at least one of a machine learning algorithm and a tree-based model Deriving effective factors affecting data collection unit;
a predictive model generation unit for generating a sintered ore intensity prediction model using the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information; and
The property information of the blending raw material, the blending condition information, and the blending raw material property information at the target time when the intensity prediction of the sintered ore is required in the sintered ore intensity prediction model using the operational condition information as variables, blending condition information, and operating conditions Sintered ore intensity prediction system comprising a sintered ore intensity prediction unit for predicting the intensity of the sintered ore to be generated at the target time by inputting information. According to claim 1,
Comparing the intensity predicted value of the sintered ore with the target intensity value of the sintered ore, the intensity predicted value further comprises an operation condition guide for guiding the target operation conditions required to follow the intensity target value Sintered ore intensity prediction system . 3. The method of claim 2,
The operation condition guide unit determines as the target operation condition at least one of the moving speed of the sintering machine bogie and the rotation speed of the cooler required for the predicted strength value to follow the target strength value, and providing it to the operator Sintered ore intensity prediction system. delete According to claim 1,
The sintered ore intensity prediction model is

is defined as
In the above formula, Y represents the predicted value of the intensity of the sintered ore predicted at the target time, X1 is a first variable representing the characteristic information of the compounding raw material set at the target time, W1 is the first weight given to the first variable , X2 is a second variable indicating the compounding condition information set at the target time, W2 is a second weight assigned to the second variable, and X3 is a second variable indicating the operation condition information set at the target time 3 variables, W3 represents a third weight assigned to the third variable, and the first to third weights have a first weight greater than a second weight within a range in which the sum of the first to third weights is 1, and the second weight is greater than the second weight. Sintered ore intensity prediction system, characterized in that it is set to be greater than the third weight. According to claim 1,
The property information of the raw material is the total Fe component content value of the raw material, the FeO component content value of the raw material, the SiO2 component content value of the raw material, the Al2O3 component content rate value of the raw material, the CaO component content rate value of the raw material, and the MgO component content rate value of the raw material Sintered ore intensity prediction system, characterized in that it comprises at least one of. According to claim 1,
The compounding condition information is, the amount of semi-gloss used in the compounding raw material, the total content of component C included in the compounding raw material, the total content of FeO component included in the compounding raw material, the input rate of the auxiliary material between the main raw material and the above, among the compounding raw materials Sintered ore intensity prediction system, characterized in that it comprises at least one of the use ratio of the main raw material, and the basicity information of the sintered ore. According to claim 1,
The operation condition information, the thickness of the compounding raw material charged to the sintering machine bogie, the moving speed of the sintering machine bogie, the rotational speed of the cooler, and air permeability in the sintering machine bogie sintered ore, characterized in that it comprises at least one Intensity Prediction System. According to claim 1,
Comparing the intensity predicted value of the sintered ore and the target intensity of the sintered ore, and the intensity predicted value further comprises a blending condition guide for guiding the blending conditions required to follow the intensity target value Sintered ore intensity prediction system. At each point in the past in the sintering machine, the characteristic information of the compounding raw material, the mixing condition information, the operating condition information, and the intensity data of the sintered ore obtained at that time are collected, and the collected data is used at least one of an algorithm and a tree-based model. to derive effective factors affecting the strength of the sintered ore, and group the effective factors to generate the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information, and collect the intensity data of the sintered ore at the time step;
generating a sintered ore intensity prediction model using the characteristic information of the compounding raw material, the compounding condition information, and the operating condition information as a variable, the characteristic information of the compounding material, the compounding condition information, and the operating condition information as variables; and
Predicting the intensity of the sintered ore to be generated at the target time by inputting the blending raw material characteristic information, the blending condition information, and the operating condition information at the target time point where the intensity prediction of the sintered ore is required to the sintered ore intensity prediction model A method for predicting the strength of sintered ore. 11. The method of claim 10,
When the intensity predicted value of the sintered ore is different from the intensity target value of the sintered ore, the method further comprising the step of setting target operating conditions required for the intensity predicted value to follow the intensity target value. 12. The method of claim 11,
The target operating condition is a sintered ore intensity prediction method, characterized in that it includes at least one of a moving speed of a sintering machine bogie and a rotation speed of a cooler required for the intensity predicted value to follow the intensity target value. 11. The method of claim 10,
The sintered ore intensity prediction model is

is defined as
In the above formula, Y represents the predicted value of the intensity of the sintered ore predicted at the target time, X1 is a first variable representing the characteristic information of the compounding raw material set at the target time, W1 is the first weight given to the first variable , X2 is a second variable indicating the compounding condition information set at the target time, W2 is a second weight assigned to the second variable, and X3 is a second variable indicating the operation condition information set at the target time 3 variables, W3 represents a third weight assigned to the third variable, and the first to third weights have a first weight greater than a second weight within a range in which the sum of the first to third weights is 1, and the second weight is greater than the second weight. Sintered ore intensity prediction method, characterized in that it is set to be greater than the third weight. 11. The method of claim 10,
The property information of the raw material is the total Fe component content value of the raw material, the FeO component content value of the raw material, the SiO2 component content value of the raw material, the Al2O3 component content rate value of the raw material, the CaO component content rate value of the raw material, and the MgO component content rate value of the raw material comprising at least one of
The blending condition information is, the amount of semi-gloss used in the blending raw material, the total content of component C included in the blending raw material, the total content of FeO ingredient included in the blending raw material, the input ratio of the main raw material and the auxiliary material between, the main raw material in the blending raw material contains at least one of the usage rate of the sintered ore, and basicity information of the sintered ore,
The operating condition information, the thickness of the compounding material charged to the sintering machine bogie, the moving speed of the sintering machine bogie, the rotational speed of the cooler, and air permeability in the sintering machine bogie, characterized in that it includes at least one of sintered ore intensity prediction Way. 11. The method of claim 10,
When the intensity predicted value of the sintered ore is different from the target intensity value of the sintered ore, the method further comprising the step of setting a mixing condition required for the intensity predicted value to follow the intensity target value.

Images