Processing apparatus, processing method, and program

Abstract

To improve the calculation accuracy of the surface profile of charged materials in a blast furnace.SOLUTION: A processing apparatus 400 calculates a correction parameter λ based on the difference in the Z coordinate of the surface position in a furnace based on the measured values of a plurality of measuring apparatuses A and B.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an apparatus, a processing method, and a program, and is particularly suitable for use in calculating the profile of raw materials charged into a blast furnace. [Background technology]

[0002] In the operation of a blast furnace for producing pig iron, sintered ore (made by sintering iron ore powder) and massive iron ore, along with coke, are generally charged alternately from a charging device installed at the top of the furnace and deposited inside. In the following explanation, sintered ore and massive iron ore will be referred to as iron ore or ore as needed. The raw materials charged into the blast furnace (iron ore and coke in the above example) will be referred to as blast furnace interior materials as needed. The profile of blast furnace interior materials refers to the shape of the deposition surface of the blast furnace interior materials.

[0003] As blast furnace contents accumulate, ore and coke layers form within the furnace. The iron ore is heated and reduced (indirectly reduced) by CO gas produced by the reaction between the hot air blown in from the tuyeres at the bottom of the furnace and the coke. In addition, some of the iron ore is directly reduced by the coke. This heating and reduction of the iron ore forms a softened and fused zone. The softened and fused zone becomes molten iron and passes through the coke layers, accumulating at the bottom of the furnace. Due to these furnace reactions, the ore and coke layers gradually descend within the blast furnace.

[0004] In the processes described above, adjusting the distribution of blast furnace internal components and obtaining an appropriate gas distribution is extremely important. Therefore, for example, obtaining a profile of the blast furnace internal components when charging iron ore and coke in order to adjust the distribution of blast furnace internal components is important in blast furnace operation.

[0005] Patent documents 1 to 3 describe techniques for calculating the profile of blast furnace interior containers. Patent documents 1 to 3 disclose a method for measuring the distance from each of two microwave distance meters, which are installed at the top of the blast furnace, aligned in the diametrical direction of the blast furnace and axially symmetric with respect to the central axis of the blast furnace, to the surface of the blast furnace interior container, and calculating the profile of the blast furnace interior container based on the measured distance.

[0006] Specifically, Patent Document 1 discloses that microwaves are output from a microwave distance meter in such a way that the incident angle of microwaves on the surface of the blast furnace interior container does not fall below 50°. Patent Document 2 discloses that, as distance data from the microwave distance meter to the surface of the blast furnace interior container, data from the side opposite the installation position of the microwave distance meter is used. Patent Document 3 discloses that, of the distance data from two microwave distance meters to the surface of the blast furnace interior container, the data from the microwave incident angle on the surface of the blast furnace interior container that is closer to 90° is used. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2010-174371 [Patent Document 2] Japanese Patent Publication No. 2011-2241 [Patent Document 2] Japanese Patent Publication No. 2014-219299 [Overview of the project] [Problems that the invention aims to solve]

[0008] The technologies described in Patent Documents 1 to 3 assume that the measuring device (a microwave distance meter in the examples described in Patent Documents 1 to 3) for measuring the distance to the surface of the blast furnace interior container always performs measurements as designed. However, in actual use, various factors may cause the measurement by the measuring device to not be as designed. For example, the measuring device has manufacturing tolerances (individual differences). Also, the measuring device may be installed in a position that is off from the position specified in the design drawings. Also, the measuring device may be installed in a direction that is off from the installation direction specified in the design drawings. Also, there may be a deviation in the measurement start point of the measuring device (a deviation in the zero point during so-called zero adjustment). Also, the blast furnace (furnace body) may become distorted over time. Also, the condition of the measuring device may change over time. Also, thermal expansion may occur in the furnace body or parts of the measuring device during actual operation. When the measurement by the measuring device does not behave as designed due to the various factors exemplified above, errors occur in the measurement data of the distance from the measuring device to the surface of the blast furnace interior container. As a result, the accuracy of calculating the profile of the blast furnace interior container decreases.

[0009] This invention has been made in view of the above-mentioned problems, and aims to improve the accuracy of calculating the profile of blast furnace interior contents. [Means for solving the problem]

[0010] The present invention relates to a processing apparatus for calculating the profile of a blast furnace interior container, comprising: an acquisition unit that acquires measurement values ​​from a plurality of measuring devices that measure the distance to the surface of the blast furnace interior container; a parameter calculation unit that calculates a value for a correction parameter for correcting the position of the surface calculated based on the measurement values, based on the difference in the height direction of the surface based on the measurement values ​​from the plurality of measuring devices that measure the same position on the surface; and a profile calculation unit that calculates the profile based on the measurement value from at least one of the measuring devices and the value of the correction parameter. The correction parameter includes at least one of the following: a physical quantity that defines the installation position of the measuring device and a physical quantity that defines the installation direction of the measuring device. .

[0011] The present invention provides a processing method for calculating the profile of a blast furnace interior container, comprising: an acquisition step of acquiring measurement values ​​from a plurality of measuring devices that measure the distance to the surface of the blast furnace interior container; a parameter calculation step of calculating a value for a correction parameter for correcting the position of the surface calculated based on the measurement values, based on the difference in the height direction of the surface based on the measurement values ​​from the plurality of measuring devices that measure the same position on the surface; and a profile calculation step of calculating the profile based on the measurement value from at least one of the measuring devices and the value of the correction parameter. The correction parameter includes at least one of the following: a physical quantity that defines the installation position of the measuring device and a physical quantity that defines the installation direction of the measuring device. .

[0012] The program of the present invention causes a computer to function as one of the components of the processing apparatus. [Effects of the Invention]

[0013] According to the present invention, the accuracy of calculating the profile of blast furnace interior filler can be improved. [Brief explanation of the drawing]

[0014] [Figure 1] This figure shows an example of a blast furnace configuration. [Figure 2] This figure shows an example of the configuration of a measuring device. [Figure 3A] This diagram illustrates an example of a measuring device installed according to the design specifications. [Figure 3B] This diagram illustrates an example of a measuring device that is not installed according to the design specifications. [Figure 4] This figure shows an example of the functional configuration of the processing unit. [Figure 5] This diagram illustrates an example of a method for calculating replacement values ​​for abnormal measurements by performing interpolation. [Figure 6] This diagram illustrates an example of a method for calculating the Z coordinate in a common Y coordinate system by performing linear interpolation. [Figure 7A] This flowchart illustrates one example of a process for calculating correction parameters. [Figure 7B]This flowchart illustrates one example of a processing method used to calculate the reactor internal surface profile. [Figure 8A] This figure shows an example of the furnace surface profile at measurement timing t1 when the correction parameters are fixed to the design values. [Figure 8B] This figure shows an example of the furnace surface profile at measurement timing t2 when the correction parameters are fixed to the design values. [Figure 8C] This figure shows an example of the furnace surface profile at measurement timing t3 when the correction parameters are fixed to the design values. [Figure 8D] This figure shows an example of the furnace surface profile at measurement timing t4 when the correction parameters are fixed to the design values. [Figure 8E] This figure shows an example of the furnace surface profile at measurement timing t5 when the correction parameters are fixed to the design values. [Figure 9A] This figure shows an example of the furnace surface profile at measurement timing t4 when correction parameters are calculated. [Figure 9B] This figure shows an example of the furnace surface profile at measurement timing t5 when correction parameters are calculated. [Modes for carrying out the invention]

[0015] One embodiment of the present invention will be described below with reference to the drawings. Furthermore, the comparison of objects in terms of length, position, size, spacing, etc., includes not only cases where they are strictly identical, but also cases where they differ within a range that does not deviate from the spirit of the invention (for example, differences within the tolerance range defined during the design phase). Also, for the sake of notation and explanation, each figure will show only the components necessary for the explanation, simplified as needed.

[0016] (Overview of facilities) Figure 1 shows an example of the configuration of blast furnace 1. In Figure 1, a bellless charging device 4 is installed at the furnace mouth (the area open at the top of the furnace) of the furnace body 2. The bellless charging device 4 is equipped with a swirling chute 5. The raw materials to be charged into the blast furnace 1 (for example, iron ore and coke) are charged into the blast furnace 1 through the swirling chute 5 and are deposited inside the blast furnace 1 as blast furnace interior containers 3. Measuring devices A and B measure the distance from the measuring devices A and B to the surface of the blast furnace interior containers 3 non-contact. In this embodiment, an example is given in which the two measuring devices A and B are installed outside the furnace body 2, near the top of the furnace body 2. In this embodiment, an example is given in which the two measuring devices A and B are arranged with a distance between them in the diametrical direction of the blast furnace 1 (parallel to the Y-axis in the example shown in Figure 1), and are installed at positions that are axially symmetric with respect to the central axis C1 of the blast furnace 1 (furnace body 2). As will be explained later with reference to Figure 3B, measuring devices A and B are not necessarily installed in the design positions. Also, the blast furnace 1 itself can be any known blast furnace and is not limited to the blast furnace 1 with the configuration shown in Figure 1. For example, Figure 1 illustrates the case where the blast furnace 1 is equipped with a bell-less charging device 4, but the blast furnace 1 may be equipped with a bell-type charging device instead of a bell-less charging device. Furthermore, the number of measuring devices is not limited to two. Also, the positional relationship between two measuring devices is not limited to being spaced apart in the diametrical direction of the blast furnace 1. Furthermore, the positional relationship between multiple measuring devices is not limited to an axially symmetric relationship with respect to the central axis C1 of the blast furnace 1 (furnace body 2). The reason why the number and position of multiple measuring devices are not limited will be explained later in the (Calculation Example) section.

[0017] Figure 2 shows an example of the configuration of measuring devices A and B. In this embodiment, we illustrate the case where measuring devices A and B have the same configuration. Furthermore, in this embodiment, we illustrate the case where measuring devices A and B are profile measuring devices described in Patent Document 3. Therefore, we will describe the general outline of measuring devices A and B here, and omit a detailed explanation.

[0018] In Figure 2, measuring devices A and B each include an antenna 11, a reflector 12, a waveguide 13, a microwave generator 14, a drive shaft 15, and a reflector drive device 16.

[0019] The antenna 11 and reflector 12 are housed inside the pressure vessel 20. An opening 21 is formed in the bottom surface of the pressure vessel 20, which communicates with the inside of the blast furnace 1. A partition plate 22, a shutter 23, and a protective net 24, etc., that allow microwaves to pass through are installed in the opening 21.

[0020] Antenna 11 is, for example, a parabolic antenna. Antenna 11 is connected to microwave generator 14 via waveguide 13. Microwave generator 14 generates microwaves whose frequency continuously changes over time within a certain range and performs microwave input and output. A data processing unit 18 is connected to microwave generator 14 via signal line 19.

[0021] The microwaves generated by the microwave generator 14 are emitted from the antenna 11, reflected by the reflector 12, and then irradiated toward the surface of the blast furnace container 3 inside the blast furnace 1. The microwaves irradiated toward the surface of the blast furnace container 3 are reflected by the surface of the blast furnace container 3 and received as reflected waves by the antenna 11 via the reflector 12. The microwave generator 14 detects these reflected waves as input. The data processing unit 18 calculates the distance from the antenna 11 to the measurement target (the surface of the blast furnace container 3) using, for example, the FMCW (Frequency Modulated Continuous Wave) method.

[0022] As shown in Figure 2, the drive shaft 15 is positioned opposite the antenna 11 at a distance from it, and rotates approximately coaxially (preferably coaxially) with the central axis C2 of the antenna 11. The reflector 12 is fixed to the drive shaft 15 such that the angle between its surface and the central axis C2 of the antenna 11 is approximately 45° (preferably 45°). The reflector drive device 16 rotates the drive shaft 15. The reflector 12 rotates in conjunction with the rotation of the drive shaft 15. With the reflector 12 rotating in this manner, microwaves are emitted from the antenna 11 in a direction parallel to its central axis C2, so that measuring devices A and B scan the microwaves reflected by the reflector 12 in the diametrical direction of the blast furnace 1. As the reflector 12 rotates, the direction of microwave reflection by the reflector 12 moves in a direction perpendicular to the plane of the paper in Figure 2 (in the Y-axis direction shown in Figure 1). In other words, Figure 2 illustrates the case where the scanning direction of the microwaves is perpendicular to the plane of Figure 2 (the Y-axis direction shown in Figure 1). It is preferable that the reflector 12 is installed so that the microwaves pass through the central axis C1 of the blast furnace 1 during the scanning process as described above.

[0023] As mentioned above, this embodiment exemplifies the case where measuring devices A and B are profile measuring devices described in Patent Document 3. However, measuring devices A and B are not limited to the profile measuring devices described in Patent Document 3. Measuring devices A and B may be, for example, profile measuring devices described in Patent Documents 1 and 2. Measuring devices A and B may also be known microwave distance meters different from those described in Patent Documents 1 to 3. Electromagnetic waves other than microwaves may also be used. To reduce the influence of dust inside the blast furnace 1 received during propagation, it is preferable to use microwaves, millimeter waves, or submillimeter waves as electromagnetic waves. However, depending on the environment inside the blast furnace 1, other electromagnetic waves may be used, or waves other than electromagnetic waves (that can propagate through air) may be used. As described above, measuring devices A and B should be devices capable of measuring the distance to the surface of the blast furnace interior container 3 in a non-contact manner. Measuring devices A and B may also be contact-type measuring devices. Furthermore, this embodiment exemplifies the case where measuring devices A and B wirelessly transmit the measured data to the processing device 400 described later. Therefore, measuring devices A and B are equipped with transmitters (not shown). However, the means by which measuring devices A and B output the measured data are not limited to wireless communication. For example, measuring devices A and B may output the measured data via a communication cable.

[0024] (Insights, Ideas, Overview) The following describes the knowledge and ideas gained by the inventors in realizing the processing apparatus and processing method of this embodiment, as well as an overview of the method for calculating the furnace surface profile in this embodiment.

[0025] First, referring to Figure 3A, we will explain an example of measuring devices A and B installed as designed. Note that "as designed" means not only that the information, such as numerical values, is as explicitly shown in the design drawings, but also that the information can be derived from the information, such as numerical values, that is explicitly shown in the design drawings.

[0026] In FIG. 3A, a case where the installation positions LPA and LPB of the measuring devices A and B are at the designed positions is illustrated. Therefore, in FIG. 3A, the Y-Z coordinates (Y OA , Z OA ) and (Y OB , Z OB ) of the installation positions LPA and LPB of the measuring devices A and B are the Y-Z coordinates shown in the design drawing. The installation positions LPA and LPB of the measuring devices A and B may be determined at predetermined positions of the measuring devices A and B. Here, as shown in FIG. 2, a case where the position of the intersection of the central axis C2 of the antenna 11 and the reflector 12 is determined as the installation positions LPA and LPB of the measuring devices A and B is illustrated. In the present embodiment, a case where the Y-Z coordinates (Y OA , Z OA ) and (Y OB , Z OB ) of the installation positions LPA and LPB of the measuring devices A and B are represented as physical quantities defining the installation positions of the measuring devices A and B is illustrated. In the following description, the Y coordinate and Z coordinate of the installation position of the measuring device are abbreviated as the Y coordinate and Z coordinate of the measuring device, respectively, as necessary.

[0027] As described in the (Overview of Equipment) section, in the present embodiment, a case where the scanning direction of the microwave is in the Y-axis direction (the diameter direction of the blast furnace 1) shown in FIG. 3A is illustrated. In FIG. 3A, the scanning direction of the microwave in the measuring device A is denoted as the scanning direction SA, and the scanning direction of the microwave in the measuring device B is denoted as the scanning direction SB. Note that the scanning directions SA and SB may be in directions opposite to the directions of the arrow lines shown in FIG. 3A (that is, the scanning directions SA and SB are in the positive direction and negative direction of the Y-axis, respectively). The scanning angles θ A , θ B are angles indicating the irradiation direction of the microwave. In the measuring devices A and B, the distances to the surface of the charge 3 in the blast furnace are measured (calculated) at the respective scanning angles θ A , θ B . Note that, as described in the (Overview of Equipment) section, in the present embodiment, a case where the scanning angles θ A , θ B are determined by the rotation angle of the reflector 12 (reflector driving device 16) is illustrated.

[0028] Furthermore, this embodiment illustrates a case where the angle from the scanning reference direction is expressed as the microwave scanning angle. Therefore, measuring devices A and B measure the microwave scanning angle θ when microwaves are irradiated in the scanning reference direction. A , θ B Assuming that is 0°, the scanning angle θ of the microwave A , θ B It detects the following. Specifically in this embodiment, measuring devices A and B detect the microwave scanning angle θ when the rotation angle of the reflector 12 is the reference angle. A , θ B Assuming that is 0°, the scanning angle θ of the microwave A , θ B It detects [something]. The reference angle is determined during the design phase.

[0029] Furthermore, in this embodiment, when measuring devices A and B are installed as designed, the scanning reference direction for measuring devices A and B is the direction extending vertically downward from the installation positions LPA and LPB of measuring devices A and B (the negative direction of the Z axis). Therefore, in this embodiment, when measuring devices A and B are installed as designed, when the rotation angle of the reflector 12 is the reference angle, microwaves are irradiated in the scanning reference direction as designed (the direction extending vertically downward from measuring devices A and B (the negative direction of the Z axis)).

[0030] Therefore, in this embodiment, we illustrate a case in which zero adjustment is performed by adjusting the installation direction of measuring devices A and B (for example, the orientation of the reflector 12) so that the rotation angle of the reflector 12 when microwaves are irradiated in the scanning reference direction as designed becomes the reference angle. Note that zero adjustment refers to the scanning angle θ of the microwaves. A , θ B The zero adjustment involves adjusting the installation direction of measuring devices A and B so that the microwave irradiation direction, where zero is 0°, is in the direction specified in the design. A , θ B This varies depending on the state of measuring devices A and B when the value is 0°, and is not limited to what was explained at the beginning of this paragraph.

[0031] In Figure 3A, the reference time deviation angles in measuring devices A and B are θ, respectively. OA , θ OB This is how it is written. Reference time deviation angle θ OA , θ OB This is the deviation angle between the actual scanning reference direction and the scanning reference direction as designed. As mentioned above, in this embodiment, we illustrate the case where the scanning reference direction as designed is the direction extending vertically downward from the installation positions LPA and LPB of the measuring devices A and B (the negative direction of the Z axis).

[0032] Figure 3A illustrates the case where measuring devices A and B are installed as designed. Therefore, the reference time deviation angle θ OA , θ OB These are both 0° (in Figure 3A, "θ" OA =0°, θ OB "=0°" represents this. In this embodiment, the reference time deviation angle θ OA , θ OB However, an example is given where it is expressed as a physical quantity that defines the installation direction of measuring devices A and B.

[0033] Furthermore, as shown in Figure 3A, in this embodiment, the measuring device A has a scanning start angle θ As From scan end angle θ Ae Within this range, a constant angular pitch Δθ A Each microwave scanning angle θ A Change the respective scanning angles θ A In this example, we will show a case where microwave transmission and reception are performed as described above to calculate the distance to the surface of the blast furnace interior container 3. Similarly, in this embodiment, the measuring device B sets the scanning start angle θ Bs From scan end angle θ Be Within this range, a constant angular pitch Δθ B Each microwave scanning angle θ B Change the respective scanning angles θ B The following example illustrates the case where microwave transmission and reception are performed as described above to calculate the distance to the surface of the blast furnace interior container 3. Note that the scanning start angle θ As , θ Bs Scanning end angle θ Ae , θ Be, and angular pitch Δθ A , Δθ B This is determined during the design phase.

[0034] Furthermore, in this embodiment, the scanning angle θ of the microwave A , θ B The scanning angle θ of the microwave when the microwave is irradiated towards the central axis C1 of the blast furnace 1 compared to when the angle is 0°. A , θ B The sign of is taken as positive, and the microwave scanning angle θ when microwaves are irradiated on the opposite side. A , θ B Let's take an example where the sign of is negative. Therefore, in the example shown in Figure 3A, the scanning start angle θ As , θ Bs This is a positive value, and the scanning end angle θ Ae , θ Be It is a negative value.

[0035] Figure 3B illustrates an example of measuring devices A and B when they are not installed as designed. In Figure 3B, the installation positions LPA and LPB of measuring devices A and B are ΔY in the Y-axis direction relative to the installation positions LPA and LPB as designed shown in Figure 3A. OA (>0), ΔY OB It shifts by (<0) and also moves ΔZ in the Z-axis direction. OA (>0), ΔZ OB Let's illustrate the case where the deviation is (>0). Also, in Figure 3B, the scanning reference direction of measuring devices A and B is Δθ from the direction as designed (the direction extending vertically downward from the installation positions LPA and LPB of measuring devices A and B). OA (>0), Δθ OB Let's look at an example where the shift is (>0). Note that (>0) and (<0) are preceded by the symbols shown before them (for example, ΔY). OA This indicates whether the value of ) is positive or negative.

[0036] As mentioned above, in Figure 3A, measuring devices A and B are installed as designed, and the reference time deviation angle θ OA , θ OB Let's take an example where the angle is 0° (as shown in Figure 3A, "θ"). OA =0°, θ OB(see "= 0°"). Therefore, in FIG. 3B, Δθ OA , Δθ OB are respectively the reference deviation angles θ OA , θ OB and are the same (in FIG. 3B, "θ OA (Δθ OA )", "θ OB (Δθ OB )" indicate this).

[0037] In FIG. 3B, L A , L B indicate the distances from the measuring devices A and B to the surface position SP of the charge 3 in the blast furnace. The distances L A , L B in the measuring devices A and B are measured. In the following description, the surface position of the charge 3 in the blast furnace is referred to as the inner surface position of the furnace as necessary.

[0038] The Y coordinates Y A , Y B and the Z coordinates Z A , Z B are represented by, for example, the following equations (1) to (4).

[0039] [Equation]

[0040] As described above, the scanning angles θ A , θ B detected by the measuring devices A and B are angles from the scanning reference direction. When the measuring devices A and B are installed as designed as shown in FIG. 3A, the scanning reference direction is the designed direction (the direction extending vertically downward from the installation positions LPA and LPB of the measuring devices A and B), and the reference deviation angles θ OA , θ OB are 0°. Therefore, the measuring devices A and B detect the angles of the microwave scanning angles θ A , θ B from the reference scanning direction as designed. On the other hand, as shown in FIG. 3B, the reference deviation angles θ OA , θOB When the value is other than 0 (zero), the scanning reference direction is θ from the designed direction OA (Δθ OA ) and is deviated by θ OB (Δθ OB ). Therefore, the measuring devices A and B detect the angles from the reference scanning directions deviated by θ A and θ B as the scanning angles θ of the microwave from the designed reference scanning direction. Thus, although the positions SP of the inner surface of the furnace to be measured are at the angles of θ OA ° and θ OB ° from the designed reference scanning direction, the measuring devices A and B detect the positions at the angles of θ A + θ OA ° and θ B + θ OB ° as the positions of the inner surface SP of the furnace to be measured. In other words, the measuring devices A and B should detect the position of the inner surface SP of the furnace to be measured as the positions at the angles of θ A ° and θ B °, but instead detect the position of the inner surface SP of the furnace to be measured as the positions at the angles of θ A + θ OA ° and θ B + θ OB °.

[0041] Also, as shown in Fig. 3B, when the installation positions LPA and LPB of the measuring devices A and B are deviated by ΔY OA and ΔY OB in the Y-axis direction and by ΔZ OA and ΔZ OB in the Z-axis direction with respect to the designed installation positions LPA and LPB shown in Fig. 3A, the Y OA , Y OB , Z OA , Z OB in equations (1) to (4) are the designed Y coordinates Y OA , Y OB and the Z coordinates Z OA , Z OB respectively, with respect to ΔY OA , ΔY OB , ΔZ​​​​OA , ΔZ OB The value will be shifted by that amount.

[0042] As described above, if measuring devices A and B are not installed as designed, even though measuring devices A and B are measuring the same internal furnace surface position SP, the YZ coordinates (Y A ,Z A ), (Y B ,Z B This will result in different YZ coordinates being calculated.

[0043] As described above, the inventors have found that if measuring devices A and B are not installed as designed, the profile of the blast furnace interior container 3 calculated based on the measurements of both devices will deviate. In the following description, the profile of the blast furnace interior container 3 will be referred to as the furnace surface profile, as needed.

[0044] Based on this finding, the inventors considered that if there were no deviations from the design drawings in the installation positions and directions of the multiple measuring devices A and B, then when the same furnace surface position SP is measured by multiple measuring devices A and B, the measurement results of both should naturally match.

[0045] Based on this idea, the inventors have found that the furnace surface profile can be calculated with high accuracy by calculating a correction parameter to correct the furnace surface position calculated based on the measurement value of at least one of the multiple measuring devices A and B, based on the difference in the Z coordinate of the furnace surface position (the position in the height direction of the surface of the blast furnace interior container 3) based on the measurement values ​​of multiple measuring devices A and B, which measure the same furnace surface position. The correction parameter may include, for example, a parameter for which a design value exists, which is used when calculating the furnace surface position based on the measurement value of at least one of the multiple measuring devices A and B. The design value is, for example, a value that can be determined without measurement when calculating the furnace surface position. Specifically, the design value is, for example, at least one of the values ​​explicitly shown in the design drawing and the value derived from the information explicitly shown in the design drawing. In this case, the correction parameter is usually a value different from the design value, but it is also possible that the design value (the value as designed) is calculated (i.e., it is sufficient that a value different from the design value can be calculated as the correction parameter).

[0046] Furthermore, in the technologies described in Patent Documents 1 to 3, a discrepancy occurs in the furnace surface position based on the measurements of measuring devices A and B, requiring data synthesis processing to select one of several furnace surface positions calculated based on the measurements of multiple measuring devices A and B. On the other hand, as mentioned above, if a correction parameter is calculated based on the difference in the Z coordinates of the furnace surface position based on measurements of multiple measuring devices A and B that target the same furnace surface position, the discrepancy in the furnace surface position based on the measurements of measuring devices A and B can be suppressed. Therefore, for example, it is not necessarily required to use the measurements of both multiple measuring devices A and B, as in the technologies described in Patent Documents 1 to 3, and even if the furnace surface position is calculated based on the measurement of one of the measuring devices A or B and the correction parameter, a decrease in the accuracy of the furnace profile calculation can be suppressed. However, it is also possible to use the measurements of both multiple measuring devices A and B, as in the technologies described in Patent Documents 1 to 3 (in this embodiment, this case is exemplified to facilitate comparison with the technology described in Patent Document 3).

[0047] Furthermore, the inventors have identified a physical quantity (Y) that determines the installation positions of measuring devices A and B as a correction parameter. OA ,Z OA ), (Y OB ,Z OB ) and the physical quantity θ that determines the installation direction of measuring devices A and B. OA , θ OB It was considered preferable to use at least one of the two such methods, as it would allow for the calculation of the furnace surface position while considering the deviation of the installation state of measuring devices A and B from the design drawings for each installation position and direction. In the examples of equations (1) to (4), the physical quantity that determines the installation position of measuring devices A and B is Y OA , Y OB , Z OA , Z OB The physical quantity that determines the installation direction of measuring devices A and B is θ. OA , θ OB Therefore, in this embodiment, the correction parameter is the Y coordinate of measuring devices A and B. OA , Y OB and Z coordinate Z OA , Z OB And the reference time deviation angle θ OA , θ OB The following is an example of when to use the above. Note that, for example, if a deviation in either the installation position or the installation direction does not cause problems due to the installation conditions of measuring devices A and B, then the one of the installation position and installation direction that does not cause problems does not need to be used as a correction parameter. In this embodiment, the Y coordinates of measuring devices A and B are exemplified as correction parameters. OA , Y OB and Z coordinate Z OA , Z OB And the reference time deviation angle θ OA , θ OB The following is an example of a case where a design value exists. In this way, when including parameters for which a design value exists in the correction parameters, it is not necessary to set all parameters for which a design value exists as correction parameters; at least one of the parameters for which a design value exists may be set as a correction parameter.

[0048] Herein, the following can be said about the measurement environment for the furnace surface profile. First, the manufacturing tolerances (individual differences) of measuring devices A and B, the errors in the installation positions of measuring devices A and B, and the deviations in the installation direction of measuring devices A and B during zero adjustment will basically not change after measuring devices A and B are installed, and even if they do change, the frequency will be low.

[0049] Furthermore, the time required for changes such as distortion of the furnace body 2 due to aging, changes in the components of measuring devices A and B due to aging, and expansion of the components of the furnace body 2 and measuring devices A and B to occur is sufficiently long compared to the measurement cycle of the furnace inner surface profile.

[0050] From the above, the scanning start angle θ As , θ Bs From scan end angle θ Ae , θ Be It is preferable to calculate the current correction parameter using a representative value of the correction parameter calculated based on the measurement values ​​of measuring devices A and B during multiple scans up to that point. This is because it is possible to suppress large temporary fluctuations in the correction parameter used to calculate the furnace surface profile in response to temporary operational fluctuations in the blast furnace 1. Therefore, in this embodiment, an example of how to calculate such a representative value of the correction parameter is provided. Furthermore, the scanning start angle θ As , θ Bs From scan end angle θ Ae , θ Be The difference in the Z coordinate of the furnace surface position calculated based on the measurements of measuring devices A and B during multiple scans up to that point may be evaluated collectively to calculate the correction parameter. However, in this embodiment, for the sake of simplicity, the scanning start angle θ As , θ Bs From scan end angle θ Ae , θ Be An example is given of evaluating the difference in the Z coordinate of the furnace surface position calculated based on the measurements of measuring devices A and B during a single scan up to that point.

[0051] For the sake of notation, in the following explanation, the measurement values ​​necessary to calculate the furnace surface profile are obtained (in this embodiment, the scanning start angle θ). As , θ Bs From scan end angle θ Ae , θ Be The measurement to obtain the measured values ​​of measuring devices A and B in a single scan up to that point will, if necessary, be simply referred to as the measurement of measuring devices A and B.

[0052] Measurements using measuring devices A and B are generally performed during the period between the charging of iron ore and coke. Therefore, the calculation of the furnace surface profile can be performed, for example, during the period between the charging of iron ore and coke, and there is no need to do it in a short time (in the following explanation, the period between the charging of iron ore and coke will be referred to as the charging preparation period as needed). In other words, there is a relatively large margin of time for calculating the furnace surface profile. Therefore, calculation methods that take a relatively long time, such as convergence calculations, can be employed.

[0053] Therefore, in this embodiment, we illustrate a case in which the difference in the Z coordinates of the furnace surface positions calculated based on the measurements of measuring devices A and B is used to calculate the value of a correction parameter that is minimized in the algorithm of a predetermined optimization method, during the charging preparation period. If measurements are taken multiple times by measuring devices A and B during a single charging preparation period, multiple furnace surface profiles may be calculated as the furnace surface profile for that charging preparation period. Furthermore, the timing of calculating the furnace surface profile is not limited to the charging preparation period. For example, if measurements can be taken by measuring devices A and B during the charging of iron ore and coke, the furnace surface profile may be calculated while the iron ore and coke are being charged.

[0054] Furthermore, the inventors have found that, based on the characteristics of the furnace surface profile, if there are no abnormalities in the measured values, the shape of the evaluation function J used in the optimization method will be unimodal. Therefore, even if a general steepest descent method is used as the optimization method, the calculation of local optima can be suppressed. Thus, in this embodiment, the case in which a general steepest descent method is used as the optimization method is illustrated. However, the optimization method is not limited to the steepest descent method. For example, an optimization method other than the steepest descent method (e.g., a metaheuristic method such as a genetic algorithm) may be used as the optimization method. Also, the method for calculating the values ​​of the correction parameters is not limited to the optimization method. For example, the correction parameters may be calculated using a machine learning model (e.g., a neural network) in which the difference in the Z coordinates of the furnace surface positions is used as the explanatory variable and the furnace surface profile is used as the objective variable. The outline of the method for calculating the furnace surface profile in this embodiment is as described above. An example of the processing apparatus and processing method of this embodiment will be described below.

[0055] (Processing device 400) Figure 4 shows an example of the functional configuration of the processing unit 400. The hardware of the processing unit 400 can be realized, for example, by using an information processing device equipped with a processor, main memory, auxiliary memory, and various interfaces. Alternatively, the hardware of the processing unit 400 may be realized by dedicated hardware such as an ASIC (Application Specific Integrated Circuit).

[0056] <Acquisition part 410> As explained in the (Knowledge, Ideas, Overview) section, in this embodiment, measuring devices A and B have a scanning start angle θ As , θ Be From scan end angle θ Ae , θ Be Within this range, a constant angular pitch Δθ A , Δθ B Each microwave scanning angle θ A , θ B Change the distance L to the surface of the blast furnace interior container 3. A, L B The scanning angle θ of the microwave A , θ B Examples of measurement (calculation) in each of these cases are given below. In the following explanation, the distance to the surface of the blast furnace interior container 3 will be abbreviated as distance as necessary, and the microwave scanning angle will be abbreviated as scanning angle as necessary.

[0057] The acquisition unit 410 acquires data from each scanning angle θ as an example of the measured values ​​of the measuring device A. A Distance L A The data is acquired. The acquisition unit 410 also acquires data for each scanning angle θ as an example of the measured value data of the measuring device B. B Distance L B The data is acquired. As explained in the (Equipment Overview) section, this embodiment exemplifies the case where measurement data is transmitted from transmitters equipped in measuring devices A and B. Therefore, the acquisition unit 410 acquires the measurement data of measuring devices A and B by receiving the measurement data transmitted in this manner from measuring devices A and B. However, as explained in the (Equipment Overview) section, the means by which measuring devices A and B output the measurement data is not limited to wireless communication. In addition, the acquisition unit 410 may acquire the measurement data of measuring devices A and B via external devices other than measuring devices A and B.

[0058] As explained in the (Knowledge, Ideas, Overview) section, in this embodiment, for the sake of simplicity, the acquisition unit 410 has a scanning start angle θ As , θ Be From scan end angle θ Ae , θ Be This example illustrates a case where measurement data from measuring devices A and B are acquired in a single scan.

[0059] <Parameter calculation unit 420, parameter storage unit 430> The parameter calculation unit 420 calculates the value of the correction parameter based on the measurement values ​​of measuring devices A and B acquired by the acquisition unit 410, which are based on measurement values ​​that target the same furnace surface position, and stores the value in the parameter storage unit 430.

[0060] As described in the (Findings, Ideas, Summary) section, in the present embodiment, the correction parameters are the Y coordinates Y OA 、Y OB and Z coordinates Z OA 、Z OB and the reference time deviation angles θ OA 、θ OB . In addition, in the present embodiment, the parameter calculation unit 420 calculates the in-furnace surface position SP based on the measurement values of one of the measuring devices A or B, and performs this for each of the measuring devices A and B, thereby calculating the in-furnace surface positions SP based on the measurement values of the measuring devices A and B respectively, and exemplifying the case where the value of the correction parameter is obtained using the steepest descent method based on the difference in the Z coordinates of the calculated in-furnace surface positions SP. Note that the difference in the Z coordinates of a plurality of in-furnace surface positions SP is calculated at the same Y coordinate.

[0061] An example of more detailed processing in the parameter calculation unit 420 will be described below. <<Coordinate Transformation>> First, the parameter calculation unit 420 calculates the Y coordinate Y A and Z coordinate Z A of the in-furnace surface position SP by equations (1) and (2) based on the measurement value of the measuring device A (the distance L OA 、Z OA 、θ OA ) and the value of the correction parameter stored in the parameter storage unit 430. As a result, the Y coordinate Y A and Z coordinate Z A of each in-furnace surface position SP corresponding to each scanning angle θ As from the scanning start angle θ Ae to the scanning end angle θ A in one scan are calculated. A and Z coordinate Z A are calculated.

[0062] Similarly, the parameter calculation unit 420 calculates the measurement value of the measuring device B (the distance L B at each scanning angle θ B ) and the value of the correction parameter stored in the parameter storage unit 430.OB , Z OB , θ OB Based on ) and , equations (3) and (4) give the Y coordinate of the furnace surface position SP Y B and Z coordinate Z B This calculates the scanning start angle θ. Bs From scan end angle θ Be Each scan angle θ in a single scan up to this point B Y coordinate of each furnace surface position SP B and Z coordinate Z B This is calculated.

[0063] Furthermore, the parameter calculation unit 420 calculates the value of the correction parameter (θ OA , θ OB , Y OA , Y OB , Z OA , Z OB If the correction parameter θ is not calculated, the parameter calculation unit 420 uses, for example, a predetermined value as the value of the correction parameter. In this embodiment, the correction parameter θ OA , θ OB , Y OA , Y OB , Z OA , Z OB Since a design value exists, we will illustrate the case where the design value is used as the predetermined value. The Y coordinate of measuring devices A and B is Y OA , Y OB and Z coordinate Z OA , Z OB The design values ​​are, for example, the Y and Z coordinates of measuring devices A and B in the design drawings. Reference time deviation angle θ OA , θ OB The design value is, for example, 0 (zero). Note that the predetermined value does not have to be the design value; for example, it may be a value set by the operator as an assumed value for the correction parameter.

[0064] Furthermore, as explained in the (Knowledge, Ideas, Overview) section, in this embodiment, the value of the correction parameter (θ) OA , θ OB , Y OA , Y OB , Z OA , Z OBAs an example, a representative value of the correction parameter calculated as the optimal solution in each of the R measurements is stored in the parameter storage unit 430 (details of this will be described later in the <<Parameter Calculation section>>).

[0065] <<Data Removal>> As described in Patent Document 3, the parameter calculation unit 420 calculates the YZ coordinates (Y A ,Z A ), (Y B ,Z B It is preferable to remove the YZ coordinates based on abnormal measurements from among them. For example, the parameter calculation unit 420 calculates the YZ coordinates (Y A ,Z A )((Y B ,Z B The slope (angle of inclination) of the straight line connecting the two points, or the distance L to the two furnace surface positions SP. A (L B Based on this, the YZ coordinates (Y A ,Z A )((Y B ,Z B It is also possible to determine whether the )) is an abnormal measurement value. Note that an example of a method for removing YZ coordinates based on such abnormal measurement values ​​is described in Patent Document 3, so a detailed explanation is omitted here. In the following explanation in this section (<Parameter Calculation Unit 420, Parameter Storage Unit 430> section), unless otherwise specified, the YZ coordinates (Y A ,Z A ), (Y B ,Z B ) shall not include YZ coordinates based on abnormal measurements.

[0066] Furthermore, if an abnormal measurement value is removed, the parameter calculation unit 420 will, for example, use the normal measurement values ​​before and after the abnormal measurement value (distance L to the furnace surface position SP). A , L B , scanning angle θ A , θ BBased on this, calculate a replacement measurement for the abnormal measurement, and use the YZ coordinate (Y A ,Z A ), (Y B ,Z B ) and scanning angle θ A , θ B The parameter calculation unit 420 may calculate the replacement measurement value by performing interpolation based on normal measurement values ​​before and after the abnormal measurement value.

[0067] Referring to Figure 5, an example of a method for calculating replacement values ​​for abnormal measurements by interpolation will be explained. For simplicity, the reference time deviation angle θ will be used here. OA is 0°(θ OA Let's take the case where =0° as an example.

[0068] In Figure 5, furnace surface positions 510 to 513 represent furnace surface positions SP based on measurements from measuring device A. Furthermore, in Figure 5, furnace surface positions 510 and 513 are assumed to be furnace surface positions SP based on normal measurements from measuring device A, while furnace surface positions 511 to 512 are assumed to be furnace surface positions SP based on abnormal measurements from measuring device A.

[0069] Furthermore, let Q be the number of abnormal measurements obtained between the timings in which two normal measurements were obtained. Figure 5 illustrates the case where Q=2. Also, assume that no other normal measurements were obtained between the timings in which the two normal measurements were obtained. In Figure 5, furnace surface position 510 is the furnace surface position based on the nth measurement, and furnace surface positions 511, 512, and 513 are the furnace surface positions based on the (n+1), (n+2), and (n+3) measurement, respectively.

[0070] In Figure 5, the distances from measuring device A to furnace surface positions 510, 511, 512, and 513 are respectively defined as L. n,A , L n+1,A , L n+2,A , L n+3,A ( =L n+Q+1,A) is denoted as, and the scanning angles at furnace surface positions 510 and 513 are θ, respectively. n,A , θ n+3,A (=θ n+Q+1,A ) is written as .

[0071] Then, distance L n+q,A Distance L' to replace n+q,A This is calculated by equation (5) below (where q is an integer between 1 and Q). Also, the scanning angle θ q+q,A This is calculated by the following equation (6). Note that the scanning angle will have the same value regardless of whether the measurement is abnormal or normal. Also, as explained in the (Knowledge, Ideas, Overview) section with reference to Figure 3A, in this embodiment, the scanning angle is constant, and the angular pitch Δθ A Let's illustrate the case where it is changed by one. Therefore, instead of equation (6), θ n+q,A =θ n,A +q × Δθ A You may also use the distance L' shown in equation (5). n+q,A Using this method, the Y-coordinate and Z-coordinate based on the abnormal measurement values ​​of measuring device A are given by the Y-coordinate Y' shown in equations (7) and (8) below, respectively. n+q,A , Z coordinate Z' n+q,A It will be replaced by...

[0072]

number

[0073] In Figure 5, the internal furnace surface positions 521 and 522, indicated by white circles, are examples of internal furnace surface positions that can be replaced by internal furnace surface positions 511 and 512, respectively. The YZ coordinates of internal furnace surface positions 521 and 522 are the Y coordinates Y' calculated by setting q=1 and q=2 in equations (7) and (8). n+1,A , Y' n+2,A and Z coordinate Z' n+1,A , Z' n+2,A These are the YZ coordinates determined by [the specified method / function].

[0074] Although Figure 5 uses measuring device A as an example, if A is replaced with B in the explanation of Figure 5 above, the same method can be used with measuring device B to calculate a replacement for the furnace surface position (YZ coordinate) based on the abnormal measurement value. Therefore, a detailed explanation of the method for calculating the replacement measurement value for the abnormal measurement value of measuring device B will be omitted.

[0075] <<Descent Correction>> Scanning start angle θ As , θ Bs From scan end angle θ Ae , θ Be During the period in which one scan is performed, the blast furnace interior contents 3 descend due to in-furnace reactions, etc. Therefore, the parameter calculation unit 420 calculates the Z coordinate Z of each in-furnace surface position SP by the amount of this descent of the blast furnace interior contents 3. A , Z B It is preferable to correct this. For example, the parameter calculation unit 420 calculates each scan angle θ A , θ B From the measurement time (each measurement time), the scan end angle θ Ae , θ Be Based on the elapsed time until the measurement time (measurement end time) and the descent velocity of the blast furnace interior container 3, each scanning angle θ A , θ B Y coordinate of each furnace surface position SP A , Y B The amount of fallout from the blast furnace interior container 3 is calculated.

[0076] In this case, it may be assumed that the descent rate of the blast furnace container 3 is constant regardless of the position on the surface of the blast furnace container 3. In this case, if the descent amount of the blast furnace container 3 is r (m), the descent rate of the blast furnace container 3 is d (m / s), and the difference between the end of measurement and the measurement time is t (s) (r≧0, d≧0, t≧0), then the descent amount r of the blast furnace container 3 can be calculated by the following equation (9). r = d × t ... (9)

[0077] In this embodiment, the parameter calculation unit 420 calculates the Y coordinate of each furnace surface position SP using equation (9). A , YB An example is given of calculating the amount r of the blast furnace interior contents 3. In this case, the parameter calculation unit 420 calculates, for example, the Z coordinate of each furnace interior surface position SP. A , Z B From there, the Y coordinate of the furnace surface position SP is Y A , Y B By subtracting the amount r of the blast furnace interior filler 3 from the Z coordinate of each furnace interior surface position SP, A , Z B Correct it.

[0078] In the following explanations in this section (the section for <parameter calculation unit 420, parameter storage unit 430>), unless otherwise specified, the Z coordinate Z of the reactor internal surface position SP will be used. A , Z B This is the Z coordinate corrected using the amount r of the fallout of the blast furnace interior contents 3 as described above.

[0079] <<Calculation of the furnace surface position in the common Y coordinate system>> In this embodiment, the difference in the Z coordinates of each furnace surface position SP is evaluated. To this end, this embodiment illustrates a case in which the difference between the Z coordinate of the furnace surface position SP based on the measurement value of measuring device A and the Z coordinate of the furnace surface position SP based on the measurement value of measuring device B is calculated at the same Y coordinate. The Y coordinate of each furnace surface position SP calculated as described above is Y A , Y B The coordinate values ​​are not necessarily the same. Therefore, the parameter calculation unit 420 calculates the YZ coordinates (Y A ,Z A ), (Y B ,Z B Based on this, the Z coordinate at the same Y coordinate of the furnace surface position SP is calculated for each of the measuring devices A and B. In the following description, the Y coordinate (the same Y coordinate mentioned above) used to calculate the difference in the Z coordinate of the furnace surface position SP based on the measurements of measuring devices A and B will be referred to as the common Y coordinate, if necessary. For example, the parameter calculation unit 420 may calculate the Z coordinate at the common Y coordinate by performing linear interpolation using the YZ coordinates of the furnace surface position SP that have Y coordinates before and after the common Y coordinate.

[0080] An example of a method for calculating the Z coordinate at a common Y coordinate by performing linear interpolation will be described below with reference to FIG. 6. In FIGS. 6(a) and 6(b), each Y coordinate separated by a predetermined distance ΔY in the Y-axis direction is the common Y coordinate Y i (···, Y n , Y n+1 , Y n+2 , Y n+3 , Y n+4 , ···) is illustrated. Here, for simplicity of explanation, a case where the predetermined distance ΔY is a constant value is illustrated. Here, i is an index number for identifying the common Y coordinate, and is assumed to be an integer of 1 or more and N or less (in FIGS. 6(a) and 6(b), the common Y coordinates Y i (Y n , Y n+1 , Y n+2 , Y n+3 , Y n+4 ) are illustrated). In FIGS. 6(a) and 6(b), the common Y coordinates with the same index number i have the same coordinate value (for example, the common Y coordinate Y n shown in FIG. 6(a) and the common Y coordinate Y n shown in FIG. 6(b) have the same coordinate value).

[0081] Also, in FIGS. 6(a) and 6(b), the in-furnace surface positions 611 to 615 and 621 to 625 are the respective in-furnace surface positions SP calculated by the process described in the <<Drop Amount Correction>> column. In the following description, each in-furnace surface position SP calculated by the process described in the <<Drop Amount Correction>> column is referred to as a base in-furnace surface position as necessary.

[0082] In FIG. 6(a), the parameter calculation unit 420 determines, among the base in-furnace surface positions 611 to 615 based on the measurement values of the measuring device A, the base in-furnace surface position having a Y coordinate greater than or equal to the common Y coordinate Y i and closest to the common Y coordinate Y i , and the Y coordinate less than or equal to the common Y coordinate Y i and closest to the common Y coordinate Y iThe base furnace surface position having the Y coordinate closest to is extracted. For example, in Figure 6(a), the common Y coordinate Y i Y n If this is the case, the parameter calculation unit 420 extracts two base furnace inner surface positions 611 and 612.

[0083] Then, the parameter calculation unit 420 linearly interpolates the two extracted base furnace inner surface positions to obtain the common Y coordinate Y i Z coordinate in Z i,A The parameter calculation unit 420 calculates the common Y coordinate Y as described above. i Z coordinate in Z i,A The calculation of all common Y coordinates i This process is carried out for (i=1 to N). Note that Z i,A The "A" after the comma indicates that the measurement is based on the measurement of measuring device A.

[0084] In Figure 6(a), the common Y coordinate Y is obtained by linear interpolation between the base furnace inner surface positions 611-612, 612-613, 613-614, and 614-615. n , Y n+1 , Y n+2 , Y n+3 , Y n+4 And the Z coordinate in the common Y coordinate Z n,A , Z n+1,A , Z n+2,A , Z n+3,A , Z n+4,A The following example illustrates how the internal furnace surface positions 630, 631, 632, 633, and 634, whose YZ coordinates are determined by the above, can be calculated. In the following explanation, the internal furnace surface positions whose YZ coordinates are determined by the common Y coordinate and the Z coordinate in the said common Y coordinate will be referred to as the common Y coordinate internal furnace surface positions as necessary.

[0085] Similarly, in Figure 6(b), the parameter calculation unit 420 calculates the common Y coordinate Y from the base furnace inner surface positions 621 to 625 based on the measurement values ​​of the measuring device B. i The above Y coordinates and the common Y coordinate Y i The base reactor inner surface position having the Y coordinate closest to the common Y coordinate Y iThe following Y coordinates, and the common Y coordinate Y i Extract the base Y coordinate that has the closest Y coordinate to and . In Figure 6(b), for example, the common Y coordinate Y i Y n If this is the case, the parameter calculation unit 420 extracts two base furnace inner surface positions 621 and 622.

[0086] Then, the parameter calculation unit 420 linearly interpolates the two extracted base furnace inner surface positions to obtain the common Y coordinate Y i Z coordinate in Z i,B The parameter calculation unit 420 calculates the common Y coordinate Y as described above. i Z coordinate in Z i,B The calculation of all common Y coordinates i This process is carried out for (i=1 to N). Note that Z i,B The "B" after the comma indicates that the measurement is based on the measurement of measuring device B.

[0087] In Figure 6(b), the common Y coordinate Y is obtained by linear interpolation of the base furnace inner surface positions 621-622, 622-623, 623-624, and 624-625. n , Y n+1 , Y n+2 , Y n+3 , Y n+4 And the Z coordinate in the common Y coordinate Z n,B , Z n+1,B , Z n+2,B , Z n+3,B , Z n+4,B The following is an example of how the common Y coordinate furnace surface positions 640, 641, 642, 643, and 644, which are determined by the YZ coordinates, can be calculated.

[0088] Note that the Common Y coordinate is Y i Z coordinate in Z i,A , Z i,B The method for calculating the base internal surface position is not limited to the methods described above. For example, based on the base internal surface positions 611-615 and 621-625, an approximate curve of the base internal surface position can be calculated using a method different from the linear interpolation described in this section (<<Calculation of internal surface position in common Y coordinate>>), and the common Y coordinate of the calculated approximate curve is Yi You may also calculate the Z coordinate at [location].

[0089] Furthermore, the Y coordinate of the furnace surface position SP, based on the measurements of measuring devices A and B, which is calculated at the time the processing of this section (<<Calculation of furnace surface position in common Y coordinate>>) is started. A , Y B If the coordinate values ​​are the same, the parameter calculation unit 420 does not need to perform the processing in this section.

[0090] <<Parameter Calculation>> In this embodiment, the parameter calculation unit 420 calculates the same common Y coordinate Y from the common Y coordinate furnace surface positions 630-634 based on the measurement values ​​of measuring device A and the common Y coordinate furnace surface positions 640-644 based on the measurement values ​​of measuring device B. i Z coordinate in Z i,A , Z i,B Extract the Z coordinate Z i,A , Z i,B An example is given of calculating the value of the correction parameter based on the difference. Here, the symbol for the correction parameter when referring to it collectively is denoted as λ, and the symbol for the small change in the correction parameter λ is denoted as Δλ, if necessary. In this embodiment, the correction parameter λ is the Y coordinate of measuring devices A and B. OA , Y OB and Z coordinate Z OA , Z OB And the reference time deviation angle θ of measuring devices A and B OA , θ OB Let's take an example of a case that includes the following. In this case, the correction parameter λ is the six components θ OA , θ OB , Y OA , Y OB , Z OA , Z OB It includes the small change in the correction parameter λ, Δλ, and the six components Δθ. OA , Δθ OB ΔY OA ΔY OB , ΔZ OA , ΔZ OBIt includes. Also, a symbol representing the number of repeated calculations in the steepest descent method shall be denoted as k (with a subscript) as necessary.

[0091] Then, the evaluation function J in the steepest descent method is represented by, for example, the following equation (10). Also, the gradient ∂J / ∂λ of the evaluation function J k is represented by, for example, the following equation (11).

[0092] [Number]

[0093] In this embodiment, an example is given where the parameter calculation unit 420 determines whether to satisfy the update end condition of the correction parameter λ k based on the value of the gradient ∂J / ∂λ of the evaluation function J k Hereinafter, in the following description, the update end condition of the correction parameter λ k shall be abbreviated as the update end condition as necessary. The update end condition may be a known condition used in the algorithm of the steepest descent method. The parameter calculation unit 420 determines that the update end condition is satisfied when at least one of the following conditions is met: the value of the gradient ∂J / ∂λ of the evaluation function J k is less than or equal to a predetermined value p (∂J / ∂λ k ≦p (or <p)), and the number of repeated calculations k is greater than or equal to a predetermined value C (k≧C (or >C)). Otherwise, it may be determined that the update end condition is not satisfied. Note that the former update end condition (|∂J / ∂λ k |≦p (or <p)) is an example of the update end condition when the update end condition is that the gradient ∂J / ∂λ of the evaluation function J k is close to 0 (zero).

[0094] As described above, in this embodiment, the correction parameter λ is composed of six components Δθ OA , Δθ OB , ΔY OA , ΔY OB , ΔZ OA , ΔZ OBThis includes the gradient of the evaluation function J, ∂J / ∂λ. k Six values ​​are calculated as follows. Therefore, the parameter calculation unit 420 calculates the gradient ∂J / ∂λ of these six evaluation functions J. k Based on this, it is determined whether or not the update termination conditions are satisfied. For example, the parameter calculation unit 420 calculates the gradient ∂J / ∂λ of the six evaluation functions J. k All of these are the gradients ∂J / ∂λ of the respective evaluation function J. k The update termination condition may be determined to be satisfied if the value is less than or equal to a predetermined value set individually for each of the parameters. Furthermore, the parameter calculation unit 420 calculates the gradient ∂J / ∂λ of the six evaluation functions J. k The update completion condition may be determined to be satisfied if the cumulative value is less than or equal to a predetermined value.

[0095] If the parameter calculation unit 420 determines that the update termination condition is not met, it corrects the parameter λ k Update the correction parameter λ. k+1 This can be expressed, for example, by equation (12) below.

[0096]

number

[0097] Here, α (>0) is the learning rate. If the value of the learning rate α is too large, the accuracy of calculating the optimal solution decreases, but the computation time decreases. Conversely, if the learning rate α is too small, the computation time increases, but the accuracy of calculating the optimal solution increases. The learning rate α is set in advance from this perspective. A value commonly used in the gradient descent algorithm may be used as the value of the learning rate α. As mentioned above, in this embodiment, the correction parameter λ has six components θ OA , θ OB , Y OA , Y OB , Z OA , Z OB It includes. Therefore, the learning rate α also has 6 components αθ. A , αθ B , α YA , α YB , αZA , α ZB This includes the correction parameter θ. OA , θ OB , Y OA , Y OB , Z OA , Z OB The learning rates for each are αθ. A , αθ B , α YA , α YB , α ZA , α ZB This is how it is written.

[0098] If the update termination conditions are not met, the parameter calculation unit 420 updates the number of iterations by adding 1 to the number of iterations k (i.e., k = k + 1), performs the processes described in the sections on <<Coordinate Transformation>>, <<Data Removal>>, <<Descent Correction>>, and <<Calculation of In-Furnace Surface Position in Common Y Coordinates>>, and calculates the gradient ∂J / ∂λ of the evaluation function J as described in this section (the <<Parameter Calculation>> section). k This process involves calculating the value and determining whether the update completion conditions are met.

[0099] Furthermore, if the update completion conditions are not met and the processes described in the sections above, namely <<Coordinate Transformation>>, <<Data Removal>>, <<Descent Correction>>, and <<Calculation of In-Factor Surface Position in Common Y-Coordinate>>, are performed again, the correction parameter λ calculated in equation (12) will be used in those processes. k+1 is, λ k It is used as such. For example, the correction parameter λ calculated in equation (12) k+1 In the calculations of equations (1) to (4) below, λ k It is used as a value.

[0100] The parameter calculation unit 420 repeats the above process until it determines that the update termination condition is satisfied. The parameter calculation unit 420 then calculates the correction parameter λ obtained at the time the update termination condition is satisfied. k (θ OA , θ OB , Y OA , Y OB , Z OA, Z OB This value will be considered the optimal solution for this measurement.

[0101] In this embodiment, the correction parameter λ(θ OA , θ OB , Y OA , Y OB , Z OA , Z OB As an example, a representative value of the correction parameter λ calculated as the optimal solution in each of the R measurements (R≧2) is stored in the parameter storage unit 430. In this way, it is possible to suppress large temporary fluctuations in the correction parameter λ used to calculate the furnace surface profile in response to temporary operational fluctuations in the blast furnace 1.

[0102] In this embodiment, we specifically illustrate the case where the representative value is a moving average. Therefore, the value of the correction parameter λ stored in the parameter storage unit 430 is the moving average of the correction parameter λ calculated as the optimal solution in each of the R measurements. However, this is not always necessary. As a representative value, for example, a weighted average may be used instead of the moving average. In this case, the closer the correction parameter λ was calculated to the current time, the larger or smaller the weight coefficient may be, or other weight coefficients may be used. Furthermore, the parameter calculation unit 420 does not need to calculate a representative value of the correction parameter λ calculated as the optimal solution in each of the R measurements. For example, the parameter calculation unit 420 may simply store the correction parameter λ calculated as the optimal solution in the current measurement in the parameter storage unit 430. For example, this may be done when the operational fluctuations in the blast furnace 1 are gradual.

[0103] <Profile Calculation Unit 440> The profile calculation unit 440 calculates the furnace surface profile based on the measurement value of at least one of the multiple measuring devices A and B, and the correction parameter λ calculated by the parameter calculation unit 420. As explained in the (Knowledge, Ideas, Overview) section, this embodiment exemplifies the case where measurement values ​​from multiple measuring devices A and B are used, similar to Patent Documents 1 to 3.

[0104] Furthermore, in this embodiment, the profile calculation unit 440 uses the measured values ​​of measuring devices A and B acquired by the acquisition unit 410 and the correction parameter λ(θ) stored in the parameter storage unit 430. OA , θ OB , Y OA , Y OB , Z OA , Z OB This example illustrates a case where the furnace surface profile is calculated during the charging preparation period based on the above. In this embodiment, it is also illustrated that the same measurement values ​​from measuring devices A and B are simultaneously output from the acquisition unit 410 to both the parameter calculation unit 420 and the profile calculation unit 440. Therefore, in this embodiment, at the timing when the profile calculation unit 440 calculates the furnace surface profile (i.e., the timing when the measurement values ​​from measuring devices A and B are output from the acquisition unit 410), the correction parameter λ based on the measurement values ​​from measuring devices A and B has not been calculated by the parameter calculation unit 420, and the correction parameter λ stored in the parameter storage unit 430 is the correction parameter λ calculated up to the previous measurement, not the current measurement. Therefore, the profile calculation unit 440 reads such a correction parameter λ from the parameter storage unit 430.

[0105] Furthermore, the acquisition unit 410 does not necessarily have to output the same measurement values ​​as the measurement values ​​of measuring devices A and B to both the parameter calculation unit 420 and the profile calculation unit 440 simultaneously. For example, the acquisition unit 410 may output the same measurement values ​​as the measurement values ​​of measuring devices A and B to the parameter calculation unit 420 and the profile calculation unit 440 at different timings.

[0106] The profile calculation unit 440 uses the measured values ​​of measuring devices A and B acquired by the acquisition unit 410 and the correction parameter λ(θ) stored in the parameter storage unit 430. OA , θ OB , Y OA , Y OB , Z OA , Z OB Based on the value of ) and , the scanning start angle θ is obtained by equations (1) to (4), as explained in the section on <<Coordinate Transformation>>. As , θ Bs From scan end angle θ Ae , θ Be Each scanning angle θ up to A , θ B Y coordinate of each furnace surface position SP A , Y B and Z coordinate Z A , Z B The following is calculated. As explained in the (Knowledge, Ideas, Overview) section, in this embodiment, the scanning start angle θ As , θ Bs From scan end angle θ Ae , θ Be This example illustrates how to calculate a single furnace surface profile based on the measurements taken by measuring devices A and B during a single scan.

[0107] Also, as explained in the section on <<Data Removal>>, the profile calculation unit 440 calculates the YZ coordinates (Y A ,Z A ), (Y B ,Z B It is preferable to remove the YZ coordinates based on abnormal measurements from among them. Also, as explained in the section on <<Descent Amount Correction>>, the profile calculation unit 440 calculates each scan angle θ A , θ B Z coordinate of the furnace surface position SP corresponding to Z A , Z B From there, the Y coordinate of the furnace surface position SP is Y A , Y B By subtracting the amount r of the blast furnace interior filler 3 in each scanning angle θ, A , θ B Z coordinate of the furnace surface position SP corresponding to Z A , ZB It is preferable to correct this.

[0108] Furthermore, the profile calculation unit 440 may further calculate, for each measuring device A and B, the Z coordinate of the furnace surface position SP at a Y coordinate different from the Y coordinate included in the YZ coordinate, based on the YZ coordinate of each furnace surface position SP whose Z coordinate has been corrected using the amount r of descent of the blast furnace interior material 3, thereby increasing the number of YZ coordinates of the furnace surface position SP. For example, the profile calculation unit 440 calculates one or more YZ coordinates of the furnace surface position SP between two furnace surface positions SP by linearly interpolating the YZ coordinates of two furnace surface positions SP whose Z coordinates have been corrected using the amount r of descent of the blast furnace interior material 3. Note that this process corresponds to the averaging process referred to as smoothing in Patent Document 3, so a detailed explanation of this process is omitted here.

[0109] The profile calculation unit 440 calculates the YZ coordinates of each furnace surface position SP for each measuring device A and B as described above. Then, based on the YZ coordinates of each furnace surface position SP calculated for each measuring device A and B as described above, the profile calculation unit 440 determines which of the furnace surface profile YZ coordinates to adopt for each Y coordinate—the furnace surface profile based on the measurement value of measuring device A or the furnace surface profile based on the measurement value of measuring device B—and calculates the combined furnace surface profile based on the determined YZ coordinates.

[0110] The furnace surface profile based on the measurements from measuring devices A and B is calculated, for example, by drawing a straight line connecting two adjacent Y-coordinates at each furnace surface position SP for each measuring device A and B. However, this is not always necessary. For example, an approximate curve of the Y-coordinates of each furnace surface position SP can be calculated using a method other than drawing a straight line, and this can also be used as the furnace surface profile based on the measurements from measuring devices A and B.

[0111] Furthermore, the choice of which furnace surface profile's YZ coordinates to adopt—the one based on measurements from measuring device A, or the one based on measurements from measuring device B—may be determined, for example, based on the relative magnitudes of the microwave incidence angles at each Y coordinate of the furnace surface profiles based on measurements from measuring devices A and B. For example, the YZ coordinate with the larger microwave incidence angle (closer to 90°) at each Y coordinate may be adopted.

[0112] Furthermore, the choice of which furnace surface profile's YZ coordinates to adopt—the one based on measurements from measuring device A, or the one based on measurements from measuring device B—may be determined, for example, by the distance from each YZ coordinate of the furnace surface profile based on measurements from measuring devices A and B to the YZ coordinates before and after that YZ coordinate. For example, for each Y coordinate, the YZ coordinate from the furnace surface profile based on measurements from measuring devices A and B that minimizes the distance from that YZ coordinate to the YZ coordinates before and after it may be adopted.

[0113] Furthermore, the synthesized furnace surface profile may be calculated, for example, by drawing a straight line connecting two adjacent YZ coordinates among the adopted YZ coordinates. Alternatively, for example, an approximation curve based on the adopted YZ coordinates may be calculated using a method other than drawing such a straight line, and this may be used as the synthesized furnace surface profile.

[0114] Note that an example of the process for calculating the furnace surface profile based on the measurements of measuring devices A and B, the process for determining which YZ coordinates to adopt from the YZ coordinates of the furnace surface profile based on the measurements of measuring devices A and B, and the process for calculating the furnace surface profile after synthesis corresponds to the incident angle calculation (calculation of distance between measurement points), adoption point determination, and two-sided data synthesis process described in Patent Document 3, so a detailed explanation of this process is omitted here.

[0115] Furthermore, the profile calculation unit 440 may calculate YZ coordinates such that the distance between two adjacent Y coordinates is a predetermined distance ΔY, based on the synthesized furnace surface profile calculated as described above, and calculate a furnace surface profile for display based on the calculated YZ coordinates. The furnace surface profile for display is calculated, for example, by connecting two adjacent YZ coordinates with a straight line. However, this is not necessarily required. For example, an approximation curve of YZ coordinates at predetermined distances ΔY may be calculated using a method other than connecting them with straight lines, and this may also be used as the furnace surface profile for display.

[0116] The profile calculation unit 440 may calculate the furnace surface profile using the method described in any one of the Patent Documents 1 to 3. However, the profile calculation unit 440 calculates the furnace surface profile using a correction parameter λ stored in the parameter storage unit 430 as the correction parameter λ, rather than a predetermined value (e.g., a design value).

[0117] <Output section 450> The output unit 450 outputs information about the furnace surface profile calculated by the profile calculation unit 440. For example, if the profile calculation unit 440 calculates a furnace surface profile for display, the output unit 450 outputs information about the furnace surface profile for display. If the profile calculation unit 440 does not calculate a furnace surface profile for display, the output unit 450 may output information about the furnace surface profile after synthesis.

[0118] In this embodiment, an example is given where the information of the furnace surface profile is displayed on a computer display. However, the form of output of the information of the furnace surface profile by the output unit 450 is not limited to display on a computer display. As a form of output of the information of the furnace surface profile, for example, in addition to or instead of display on a computer display, at least one of the following may be adopted: transmission to an external device (e.g., an information processing device for managing the operation of the blast furnace 1) and storage in an internal or external storage medium of the processing device 400.

[0119] (Processing method) Next, an example of the processing method of this embodiment will be described with reference to the flowcharts in Figures 7A and 7B. The flowcharts in Figures 7A and 7B are realized, for example, by a process in which the processor of the processing unit 400 executes a computer program stored in the auxiliary storage device using the main memory as a work area.

[0120] First, we will explain an example of the process for calculating correction parameters, referring to the flowchart in Figure 7A.

[0121] First, in step S701, the acquisition unit 410 acquires the measurement values ​​of measuring devices A and B. In this embodiment, the acquisition unit 410 acquires the scanning start angle θ in one step S701 process. As , θ Be From scan end angle θ Ae , θ Be This example illustrates a case where measurement data from measuring devices A and B are acquired together in a single scan.

[0122] Next, in step S702, the parameter calculation unit 420 reads the correction parameter λ stored in the parameter storage unit 430. In this embodiment, it is illustrated that the parameter storage unit 430 stores a moving average value, which is an example of a representative value of the correction parameter λ calculated as the optimal solution in each of the R measurements. In addition, in this embodiment, it is illustrated that the process in step S702 is omitted if the correction parameter λ is not stored in the parameter storage unit 430. In this case, the parameter calculation unit 420 uses, for example, a predetermined value of the correction parameter λ (for example, a design value) instead of the correction parameter λ stored in the parameter storage unit 430.

[0123] Next, in step S703, the parameter calculation unit 420 sets the number of iterations k to 1. In the example shown in Figure 7A, the number of iterations k represents the number of iterations of the process described later in steps S704 to S708.

[0124] Next, in step S704, the parameter calculation unit 420 calculates the furnace surface position SP based on the measured values ​​of one measuring device A and one measuring device B, for each of the measuring devices A and B. In this embodiment, we will illustrate the case in which the parameter calculation unit 420 performs the processing described in the sections "Coordinate Transformation", "Data Removal", "Descent Correction", and "Calculation of Furnace Surface Position in Common Y Coordinate". Specific examples of these processes have been described in those sections, so a detailed explanation is omitted here.

[0125] Next, in step S705, the parameter calculation unit 420 calculates the gradient ∂J / ∂λ of the evaluation function J. k The value of is calculated (see equation (11)). In this embodiment, the correction parameter λ is used in equation (11). k The correction parameter λ calculated in step S707, described later, when the number of iterations k is k-1. k+1 (θ OA , θ OB , Y OA , Y OB , Z OA , Z OBThe following is an example of when ) is used. Also, when the number of iterations k is 1, the correction parameter λ in equation (11) k For example, a predetermined value (e.g., a design value) is used.

[0126] Next, in step S706, the parameter calculation unit 420 determines whether or not the update completion condition is satisfied. As explained in the section on "Parameter Calculation," for example, the parameter calculation unit 420 calculates the gradient ∂J / ∂λ of the evaluation function J. k The value of is less than or equal to a predetermined value p and (∂J / ∂λ k The update termination condition is determined to be satisfied if at least one of the following conditions is met: (≤p), the number of iterations k is greater than or equal to a predetermined value C, and (k≧C). Otherwise, the update termination condition is determined to be unsatisfied.

[0127] If the result of the determination in step S706 does not satisfy the update termination condition (if NO is found in step S706), the process in step S707 is performed. In step S707, the parameter calculation unit 420 calculates the correction parameter λ k The parameter calculation unit 420 performs the calculation of equation (12) to update the correction parameter λ. k+1 By calculating the correction parameter λ, k An example of updating this will be given.

[0128] Next, in step S708, the parameter calculation unit 420 updates the number of iterations k by adding 1. Then, the iterative processing of steps S704 to S708 is performed again as processing for the updated number of iterations k. Note that in step S708, by adding 1 to the number of iterations k, the correction parameter λ calculated in the preceding step S707 is updated. k+1 The correction parameter λ is used in the iterative processing for the number of iterations k after the update. k It will be treated as such.

[0129] As a result of the determination in step S706, if the update end condition is satisfied (YES in step S706), the process of step S709 is performed. In step S709, the parameter calculation unit 420 stores the correction parameter λ k (θ OA 、θ OB 、Y OA 、Y OB 、Z OA 、Z OB ) as the optimal solution in the current measurement. Note that the parameter calculation unit 420 may store the correction parameter λ in the parameter storage unit 430 so as to be distinguished from the moving average value of the correction parameter λ to be stored in the next step S710, or may store it in a storage unit different from the parameter storage unit 430.

[0130] Next, in step S710, the parameter calculation unit 420 calculates the moving average value of the correction parameter λ calculated as the optimal solution in each of the R measurements and stores it in the parameter storage unit 430. When the process of step S710 ends, the process according to the flowchart of FIG. 7A ends.

[0131] Note that after calculating the moving average value of the correction parameter λ calculated as the optimal solution in each of the R measurements, the parameter calculation unit 420 may discard the oldest correction parameter λ among the correction parameters λ calculated as the optimal solution in each of the R measurements.

[0132] Also, when the current iteration count k is R'(R' < R) less than R, there is no correction parameter λ calculated as the optimal solution in each of the R measurements. In this case, for example, the parameter calculation unit 420 may calculate the moving average value of the correction parameter λ calculated as the optimal solution in each of the R' measurements. Also, for example, the parameter calculation unit 420 may store the current correction parameter λ in the parameter storage unit 430 without calculating the moving average value of the correction parameter λ.

[0133] Next, we will explain an example of the processing method used to calculate the furnace surface profile, referring to the flowchart in Figure 7B.

[0134] First, in step S711, the acquisition unit 410 acquires the measurement values ​​from measuring devices A and B. In this embodiment, the acquisition unit 410 processes the scanning start angle θ in one step S711. As , θ Be From scan end angle θ Ae , θ Be This example illustrates a case where measurement data from measuring devices A and B are acquired together in a single scan.

[0135] Furthermore, in this embodiment, the processes shown in the flowcharts of Figures 7A and 7B are performed in parallel at the same timing, and the example given is that the same measured value is used as the measured value of measuring devices A and B in the processes shown in the flowcharts of Figures 7A and 7B that are performed in parallel at the same timing. Therefore, although step S701 in Figure 7A and step S711 in Figure 7B are shown and described as separate steps, step S701 in Figure 7A and step S711 in Figure 7B are the same step. However, as explained in the section on <Profile Calculation Unit 440>, it is not necessary for the same measured value to be output simultaneously from the acquisition unit 410 to both the parameter calculation unit 420 and the profile calculation unit 440 as the measured value of measuring devices A and B. In this case, step S701 in Figure 7A and step S711 in Figure 7B are processed as separate steps. In this case, step S701 in Figure 7A and step S711 in Figure 7B start at different timings.

[0136] Next, in step S712, the profile calculation unit 440 reads the correction parameter λ stored in the parameter storage unit 430. In this embodiment, the profile calculation unit 440 reads a moving average value from the parameter storage unit 430, which is an example of a representative value of the correction parameter λ calculated as the optimal solution for each of the R measurements.

[0137] In this embodiment, we illustrate a case where, in the flowcharts of Figures 7A and 7B which start at the same timing, the processing of step S710 in Figure 7A has not yet finished when the processing of step S712 in Figure 7B begins. Therefore, the correction parameter λ read from the parameter storage unit 430 in step S712 is the correction parameter λ stored in the parameter storage unit 430 in step S710 of the flowchart of Figure 7A (the previous step S710), which started at the timing immediately preceding that timing. However, in the flowcharts of Figures 7A and 7B which start at the same timing, the processing of step S710 in Figure 7A may have already finished when the processing of step S712 in Figure 7B begins. In this case, in step S712, the profile calculation unit 440 may read the correction parameter λ stored in the parameter storage unit 430 in that step S710. As mentioned above in the explanation of the processing of step S711, the flowcharts of Figures 7A and 7B do not need to start at the same timing.

[0138] Next, in step S713, the profile calculation unit 440 calculates the measurement values ​​of measuring devices A and B acquired by the acquisition unit 410 in step S711, and the correction parameter λ(θ) read from the parameter storage unit 430 in step S712. OA , θ OB , Y OA , Y OB , Z OA , Z OB Based on the above, the furnace surface profile is calculated. In this embodiment, when calculating the furnace surface profile, the profile calculation unit 440 calculates the scanning start angle θ using equations (1) to (4). As , θ Bs From scan end angle θ Ae , θ Be Each scanning angle θ up to A , θ B Y coordinate of each furnace surface position SP A , Y B and Z coordinate Z A , Z B Calculate the YZ coordinates (Y A ,ZA ), (Y B ,Z B An example of calculating the furnace surface profile based on the above is given. A specific example of the method for calculating the furnace surface profile has been explained in the section on <Profile Calculation Unit 440>, so a detailed explanation is omitted here. Alternatively, for example, the profile calculation unit 440 may calculate the furnace surface profile using the method described in any one of the patent documents 1 to 3. However, the profile calculation unit 440 calculates the furnace surface profile using a correction parameter λ stored in the parameter storage unit 430 as the correction parameter λ, rather than a predetermined value (e.g., a design value).

[0139] Next, in step S714, the output unit 450 outputs the furnace surface profile information calculated in step S713. Once the processing in step S714 is complete, the process shown in the flowchart in Figure 7B is finished.

[0140] (Calculation example) Next, we will explain a calculation example. In this calculation example, the furnace surface profile was calculated for each measuring device A and B based on simulated data of the measured values ​​from measuring devices A and B. The simulated data was created by processing actual operating data so that the calculated furnace surface profile has a simpler shape than the furnace surface profile based on actual operating data. The same results as in this calculation example can be obtained even if actual operating data is used. In addition, in this calculation example, the actual operating data used was data from several years after measuring devices A and B were installed.

[0141] Figures 8A to 8E show examples of furnace surface profiles at each measurement timing t1 to t5 when the value of the correction parameter λ is fixed to the design value. Measurement timings t1 to t5 are the start timings of measurements by measuring devices A and B. The order of measurement timings t1 to t5 is t1, t2, t3, t4, and t5 (measurement timing t1 is the first measurement timing, and measurement timing t5 is the last measurement timing).

[0142] In Figures 8A to 8E, furnace surface profiles 811 to 815 are furnace surface profiles based on measurements from measuring device A, and furnace surface profiles 821 to 825 are furnace surface profiles based on measurements from measuring device B.

[0143] As shown in Figures 8A to 8E, it can be seen that a clear discrepancy has emerged between the furnace surface profiles 811 to 815 based on the measurements of measuring device A and the furnace surface profiles 821 to 825 based on the measurements of measuring device B, after several years have passed since the installation of measuring devices A and B.

[0144] Figures 9A and 9B show examples of furnace surface profiles at each measurement timing t4 to t5 when the correction parameter λ is calculated as described in this embodiment. Figures 8D and 9A show the furnace surface profiles at the same measurement timing t4, and Figures 8E and 9B show the furnace surface profiles at the same measurement timing t5.

[0145] In Figures 9A and 9B, furnace surface profiles 911 to 912 are furnace surface profiles based on measurements from measuring device A, and furnace surface profiles 921 to 922 are furnace surface profiles based on measurements from measuring device B.

[0146] In calculating the furnace surface profiles 911-912 and 921-922, in this calculation example, as explained in this embodiment, the same common Y coordinate Y i Z coordinate in Z i,A , Z i,B The difference is the correction parameter θ that is minimized in the steepest descent algorithm. OA , θ OB , Y OA , Y OB , Z OA , Z OB This was calculated as the optimal solution for each measurement. The learning rate αθ was used in this calculation. A , αθ B , α YA , α YB , α ZA , α ZB As, αθA =αθ B =20, α YA =α YB =0.1, α ZA =α ZB We used a value of =0.01 for the small change in the correction parameter Δθ. OA , Δθ OB ΔY OA ΔY OB , ΔZ OA , ΔZ OB As, Δθ OA =Δθ OB =0.1, ΔY OA =ΔY OB =ΔZ OA =ΔZ OB We used =0.01. Also, Σ|∂J / ∂λ k The update termination condition was that at least one of the following conditions was satisfied: |≦0.000001 (p=0.000001) and k>80 (C=80).

[0147] Furthermore, in calculating the furnace surface profiles 911-912 and 921-922, in this calculation example, the moving average of the correction parameter λ calculated as the optimal solution in each of the three consecutive measurements (R=3) was used as the correction parameter λ used to calculate the furnace surface profiles. Therefore, the correction parameter λ used to calculate the furnace surface profiles 911 and 921 at measurement timing t4 shown in Figure 9A was calculated as follows. First, based on the measured values ​​obtained from the measurements of measuring devices A and B at measurement timings t1-t3 shown in Figures 8A-8C, the optimal solution for the correction parameter λ in each measurement at measurement timings t1-t3 was calculated. Then, the moving average of the calculated optimal solution for the correction parameter λ was calculated as the correction parameter λ used to calculate the furnace surface profiles 911 and 921 at measurement timing t4. Similarly, the correction parameter λ used to calculate the furnace surface profiles 912 and 922 at measurement timing t5 shown in Figure 9B was the moving average of the optimal solution of the correction parameter λ for each measurement at measurement timings t2 to t4.

[0148] The furnace surface profiles 911 and 921 shown in Figure 9A show no discrepancy with the furnace surface profiles 814 and 824 shown in Figure 8D, which were calculated at the same measurement timing t4 as in Figure 9A, indicating that they have been corrected to match. Similarly, the furnace surface profiles 912 and 922 shown in Figure 9B also show no discrepancy with the furnace surface profiles 814 and 824 shown in Figure 8D, which were calculated at the same measurement timing t5 as in Figure 9B, indicating that they have been corrected to match.

[0149] As described above, this embodiment illustrates the case in which the furnace surface profile is calculated using the measured values ​​of multiple measuring devices A and B. However, as explained in this section (the (Calculation Example) section), the correction parameter λ is calculated using the method of this embodiment to obtain the YZ coordinates (Y A ,Z A ), (Y B ,Z B By calculating this, it is possible to suppress large discrepancies between the YZ coordinates of each furnace surface position SP based on the measurements of each measuring device A and B (see Figures 9A and 9B). Therefore, even if the measurement of one measuring device A or B is used instead of using the measurement of multiple measuring devices A and B as in the techniques described in Patent Documents 1 to 3, it is possible to suppress a large decrease in the accuracy of calculating the furnace surface profile. Thus, the furnace surface profile may be calculated based on the measurement of one measuring device A or B (without using the measurement of multiple measuring devices A and B).

[0150] Furthermore, in this embodiment, the correction parameter λ should be calculated so as to suppress the discrepancy between the furnace surface profiles 811-815 based on the measurements of measuring device A and the furnace surface profiles 821-825 based on the measurements of measuring device B, as shown in Figures 8A-8E. Therefore, the number and positions of multiple measuring devices are not limited as long as it is possible to measure the same position SP on the furnace surface of the target to be measured. In other words, the number of multiple measuring devices is not limited to 2. Also, the positional relationship between the two measuring devices is not limited to being spaced apart in the diametrical direction of the blast furnace 1. Furthermore, the positional relationship between the multiple measuring devices is not limited to being axially symmetric with respect to the central axis C1 of the blast furnace 1 (furnace body 2).

[0151] (summary) As described above, in this embodiment, the processing device 400 calculates a correction parameter λ based on the difference in the Z coordinates of the furnace surface positions based on the measurements of multiple measuring devices A and B. Therefore, the accuracy of calculating the furnace surface profile can be improved.

[0152] Furthermore, in this embodiment, at least one of the physical quantities that define the installation positions of measuring devices A and B, and the physical quantity that defines the installation direction of measuring devices A and B, is used as the correction parameter λ. More specifically, the YZ coordinates (Y OA ,Z OA ), (Y OB ,Z OB ) and the reference time deviation angle θ OA , θ OB At least one of these is used as the correction parameter λ. Therefore, the deviation of the installation state of measuring devices A and B from the design drawings can be taken into account for each installation position and direction to calculate the furnace surface position.

[0153] Furthermore, in this embodiment, the processing device 400 calculates the value of a correction parameter SP that minimizes the difference in the Z coordinates of the furnace surface positions SP calculated based on the measurements of measuring devices A and B, according to the algorithm of a predetermined optimization method. Therefore, it is possible to quantitatively calculate the value of the correction parameter λ that reduces the difference in the Z coordinates of the furnace surface positions calculated based on the measurements of measuring devices A and B.

[0154] Furthermore, in this embodiment, the processing device 400 uses the steepest descent method as a predetermined optimization technique. Therefore, without using a complex optimization technique, it is possible to quantitatively calculate the value of the correction parameter λ that reduces the difference in the Z coordinates of the furnace surface positions calculated based on the measurements of measuring devices A and B.

[0155] Furthermore, in this embodiment, the processing device 400 calculates the value of the correction parameter λ used when calculating the furnace surface profile based on the correction parameter λ calculated at multiple timings. Therefore, it is possible to suppress fluctuations in the correction parameter λ used to calculate the furnace surface profile in response to temporary operational fluctuations in the blast furnace 1.

[0156] (Other embodiments) Furthermore, the embodiments of the present invention described above can be realized by a computer executing a program. A computer-readable recording medium on which the program is recorded, and a computer program product such as the program itself, can also be applied as embodiments of the present invention. Examples of recording media that can be used include flexible disks, hard disks, optical disks, magneto-optical disks, CD-ROMs, magnetic tapes, non-volatile memory cards, ROMs, and the like. Furthermore, the embodiments of the present invention described above are merely examples of how the invention can be implemented, and the technical scope of the invention should not be interpreted as being limited by them. In other words, the present invention can be implemented in various forms without departing from its technical concept or its main features.

[0157] Furthermore, the disclosure of the above embodiments is as follows, for example. (Disclosure 1) A processing device for calculating the profile of blast furnace interior contents, An acquisition unit that acquires measurement values ​​from multiple measuring devices that measure the distance to the surface of the blast furnace interior container, A parameter calculation unit calculates a value for a correction parameter to correct the position of the surface calculated based on the measurement value, based on the difference in the height direction position of the surface based on measurement values ​​from multiple measuring devices, where the same position on the surface is the target of measurement. A processing apparatus comprising: a profile calculation unit that calculates the profile based on the measurement value of at least one of the measuring devices and the value of the correction parameter. (Disclosure 2) The processing apparatus according to disclosure 1, wherein the correction parameter includes at least one of a physical quantity that defines the installation position of the measuring device and a physical quantity that defines the installation direction of the measuring device. (Disclosure 3) The apparatus according to disclosure 1 or 2, wherein the parameter calculation unit calculates the value of the correction parameter that minimizes the difference in position in the height direction of the surface in the algorithm of a predetermined optimization method. (Disclosure 4) The apparatus according to Disclosure 3, wherein the predetermined optimization method is the steepest descent method. (Disclosure 5) The processing apparatus according to any one of disclosures 1 to 4, wherein the parameter calculation unit calculates the value of the correction parameter to be used when the profile calculation unit calculates the profile, based on the value of the correction parameter calculated at multiple timings. (Disclosure 6) The processing apparatus according to any one of disclosures 1 to 5, wherein the number of the plurality of measuring devices is 2. (Disclosure 7) A method for calculating the profile of blast furnace interior containers, A process of acquiring measurement values ​​from multiple measuring devices that measure the distance to the surface of the blast furnace interior container, A parameter calculation step of calculating a value for a correction parameter to correct the position of the surface calculated based on the measurement value, based on the difference in the height direction position of the surface based on measurement values ​​from multiple measuring devices, where the same position on the surface is the target of measurement; A profile calculation step of calculating the profile based on the measurement value of at least one of the measuring devices and the value of the correction parameter, A processing method comprising: (Disclosure 8) A program for causing a computer to function as a part of the processing apparatus described in any one of disclosures 1 to 6. [Explanation of symbols]

[0158] 1 blast furnace 2 Furnace body 3 Contents inside blast furnace 4 Bell-less charging device 5. Swinging Shot 11 Antennas 12 Reflector 13 Waveguide 14 Microwave generator 15 Drive shaft 16 Reflector drive device 20 Pressure vessels 21 Opening 22 partition plates 23 Shutter 24 Protective net 400 Processing Units 410 Acquisition Department 420 Parameter Calculation Unit 430 Parameter Storage Unit 440 Profile Calculation Unit 450 Output section 510~513 Furnace surface position 521~522 Core surface position that replaces the core surface position based on the (n+1)th and (n+2)th measurements 611-615 Base furnace inner surface position based on measurements from measuring device A 621-625 Base furnace inner surface position based on measurements from measuring device B 630-634 Common Y-coordinate of furnace inner surface position based on measurements from measuring device A 640-644 Common Y-coordinate furnace surface position based on measurements from measuring device B 811-815 Furnace surface profile based on measurements from measuring device A when the correction parameter values ​​are set to design values. 821-825 Furnace surface profile based on measurements from measuring device B when the correction parameter values ​​are set to design values. 911-912 Furnace surface profile based on measurements from measuring device A when the correction parameter values ​​are changed from the design values. 921-922 Furnace surface profile based on measurements from measuring device B when the correction parameter values ​​are changed from the design values. A, B Measuring device C1 Blast furnace central axis C2 Antenna central axis L A , L B Distance from the measuring device to the furnace surface position Installation location of LPA and LPB measuring devices SA, SB microwave scanning direction SP Furnace surface position Y n ~Y n+4 Common Y-coordinate Y OA , Y OB The amount of deviation of the Y coordinate of the installation position of the measuring device. Z n,A ~Z n+4,A Z coordinate in the common Y coordinate system based on the measurement device of measuring device A Z n,B ~Z n+4,B Z coordinate in the common Y coordinate system based on the measurement device of measurement device B Z OA , Z OB Z coordinate of the installation position of the measuring device t1~t5 Measurement timing θ A , θ B Scanning angle θ OA , θ OB Reference time deviation angle θ As , θ BsScanning start angle θ Ae , θ Be Scanning end angle θ n,A Scanning angle of the furnace surface position based on the nth measurement. θ n+3,A Scanning angle of the furnace surface position based on the (n+3)th measurement. ΔY: Distance along the Y-axis in the Common Y coordinate system. ΔY OA ΔY OB The amount of deviation in the Y-axis direction of the installation position of the measuring device. ΔZ OA , ΔZ OB Z-axis displacement of the installation position of the measuring device Δθ A , Δθ B Angle pitch Δθ OA , Δθ OB Angle of deviation from the scanning reference direction in the measuring device.

Claims

1. A processing device for calculating the profile of blast furnace interior contents, An acquisition unit that acquires measurement values ​​from multiple measuring devices that measure the distance to the surface of the blast furnace interior container, A parameter calculation unit calculates a value for a correction parameter to correct the position of the surface calculated based on the measurement value, based on the difference in the height direction position of the surface based on measurement values ​​from multiple measuring devices, where the same position on the surface is the target of measurement. A profile calculation unit that calculates the profile based on the measurement value of at least one of the measuring devices and the value of the correction parameter, Equipped with, The processing apparatus wherein the correction parameter includes at least one of a physical quantity that defines the installation position of the measuring device and a physical quantity that defines the installation direction of the measuring device.

2. The apparatus according to claim 1, wherein the parameter calculation unit calculates the value of the correction parameter that minimizes the difference in position in the height direction of the surface in the algorithm of a predetermined optimization method.

3. The apparatus according to claim 2, wherein the predetermined optimization method is the steepest descent method.

4. The processing apparatus according to any one of claims 1 to 3, wherein the parameter calculation unit calculates the value of the correction parameter to be used when the profile calculation unit calculates the profile, based on the value of the correction parameter calculated at multiple timings.

5. The processing apparatus according to any one of claims 1 to 3, wherein the number of the plurality of measuring devices is 2.

6. A method for calculating the profile of blast furnace interior containers, Acquisition process: Obtaining measurement values ​​from multiple measuring devices that measure the distance to the surface of the blast furnace interior container. and, A parameter calculation step of calculating a value for a correction parameter to correct the position of the surface calculated based on the measurement value, based on the difference in the height direction position of the surface based on measurement values ​​from multiple measuring devices, where the same position on the surface is the target of measurement; A profile calculation step that calculates the profile based on the measurement value of at least one of the measuring devices and the value of the correction parameter, Equipped with, A processing method wherein the correction parameter includes at least one of a physical quantity that defines the installation position of the measuring device and a physical quantity that defines the installation direction of the measuring device.

7. A program for causing a computer to function as each part of the processing apparatus according to any one of claims 1 to 3.

Images