Blast furnace operation abnormality estimation method, blast furnace operation abnormality estimation device, blast furnace operation abnormality estimation program, and blast furnace operation abnormality estimation system

Abstract

To provide a blast furnace operation abnormality estimation method capable of estimating an abnormal factor without depending on the absolute value of an index.SOLUTION: A blast furnace operation abnormality estimation method includes a process of acquiring measured values of a plurality of indicators during operation of a blast furnace, calculating first estimated values of the plurality of indicators using predetermined measured values among the acquired measured values, calculating for each indicator a first degree of coincidence indicating for each indicator whether the acquired measured values and the calculated first estimated values coincide with each other, and estimating an abnormal factor of the blast furnace based on the first degree of coincidence calculated for each indicator and abnormal factor information indicating a relationship between an abnormal factor of the blast furnace and the indicator of the blast furnace.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to a blast furnace operation abnormality estimation method, a blast furnace operation abnormality estimation device, a blast furnace operation abnormality estimation program, and a blast furnace operation abnormality estimation system. [Background technology]

[0002] In recent years, in order to prevent global warming, there has been a demand for reducing the amount of carbon dioxide emitted in blast furnace operation, and it is desirable to lower the reducing agent rate. However, in blast furnace operation, lowering the reducing agent rate may cause the furnace condition to become unstable, such as a decrease in furnace heat, and may cause the production volume to fluctuate, resulting in unstable operation of the blast furnace.

[0003] In order to prevent the furnace condition from becoming unstable, various methods for detecting abnormalities in a blast furnace are known. For example, Patent Document 1 describes a technique for determining the presence or absence of an abnormality in the operation of a blast furnace based on the difference between a furnace heat index predicted using a physical model and a furnace heat index calculated based on an actual measurement value in a blast furnace during operation. The technique described in Patent Document 1 can estimate the cause of an abnormality in a blast furnace by automatically determining and classifying disturbances in the temperature distribution in the blast furnace. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2022-91600 Summary of the Invention [Problem to be solved by the invention]

[0005] In the technology described in Patent Document 1, an abnormality factor is estimated based on the magnitude of the difference between a predicted value and an actual measurement value, so that a furnace heat index with a large absolute value has a large difference between the predicted value and the actual measurement value, and is therefore more likely to be estimated as an abnormality factor. On the other hand, a furnace heat index with a small absolute value has a small difference between the predicted value and the actual measurement value, and is less likely to be estimated as an abnormality factor. In the technology described in Patent Document 1, an abnormality factor is estimated based on the magnitude of the furnace heat index, so that the absolute value of the furnace heat index may affect the estimation of the abnormality factor, and the estimation accuracy of the abnormality factor may decrease.

[0006] The present invention is devised to solve such problems, and aims to provide a blast furnace operation abnormality estimation method, a blast furnace operation abnormality estimation device, a blast furnace operation abnormality estimation program, and a blast furnace operation abnormality estimation system that are capable of estimating abnormality factors without relying on the absolute value of an index. [Means for solving the problem]

[0007] The present invention, which solves such problems, is summarized as a blast furnace operation abnormality estimation method, a blast furnace operation abnormality estimation device, a blast furnace operation abnormality estimation program, and a blast furnace operation abnormality estimation system described below. (1) Obtaining measurements of multiple indicators during operation of a blast furnace; Calculating a first estimate of a plurality of indices using predetermined measurement values ​​from among the acquired measurement values; calculating a first degree of agreement for each index, the first degree of agreement indicating whether or not the acquired measurement value and the calculated first estimated value match for each index; An abnormality factor of the blast furnace is estimated based on the first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace. A method for estimating abnormalities in blast furnace operation. (2) Information on the causes of blast furnace abnormalities The abnormality factor and a second degree of agreement indicating whether or not the second estimated value of the index in a stable state of the blast furnace coincides with the third estimated value of the index in the event of an abnormality factor occurring in the blast furnace are created based on the abnormality factor and the second degree of agreement indicating whether or not the second estimated value of the index in a stable state of the blast furnace coincides with the third estimated value of the index in the event of an abnormality factor occurring in the blast furnace The blast furnace operation abnormality estimation method according to (1). (3) the first estimated value is calculated using a first set of operational information consisting of a plurality of pieces of operational information obtained during operation of the blast furnace and a blast furnace mathematical model; The second estimate is A first operational information group consisting of a plurality of operational information in a stable state of the blast furnace and a blast furnace mathematical model are used to calculate the The third estimate is The blast furnace operation abnormality estimation method according to (2), characterized in that it is calculated using a second group of operation information consisting of a plurality of pieces of operation information when the blast furnace is in a stable state, and disturbance information indicating that a predetermined piece of operation information among the plurality of pieces of operation information when the blast furnace is in a stable state, and a blast furnace mathematical model. (4) the first degree of agreement is binary information indicating whether the first estimated value and the measured value agree with each other, The second degree of agreement is The blast furnace operation abnormality estimating method according to (3), characterized in that the second estimated value and the third estimated value are binary information indicating whether or not they match. (5) The first estimate is The blast furnace operation abnormality estimation method according to (4), characterized in that the estimated value is calculated using a plurality of pieces of acquired operation information that are acquired at a time closest to the time at which the abnormality cause is estimated and a blast furnace mathematical model. (6) calculating the material balance of the blast furnace from the obtained measurements; The method for estimating an abnormality in a blast furnace operation according to any one of (1) to (5), characterized in that it is determined whether or not the acquired measured values ​​include abnormal values ​​based on the calculated value of the material balance. (7) a measurement value acquisition unit that acquires measurement values ​​of a plurality of indicators during operation of the blast furnace; a first estimate calculation unit that calculates first estimates of a plurality of indexes using a predetermined measurement value among the acquired measurement values; a coincidence calculation unit that calculates a first coincidence for each index, the first coincidence indicating whether or not the acquired measurement value and the calculated first estimated value coincide with each index; an abnormality factor estimation unit that estimates an abnormality factor of the blast furnace based on a first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace; A blast furnace operation abnormality estimation device having the above. (8) Obtaining measurements of multiple indicators during operation of the blast furnace; Calculating a first estimate of a plurality of indices using predetermined measurement values ​​from among the acquired measurement values; calculating a first degree of agreement for each index, the first degree of agreement indicating whether or not the acquired measurement value and the calculated first estimated value match for each index; An abnormality factor of the blast furnace is estimated based on the first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace. A blast furnace operation abnormality estimation program that causes a computer to execute the processing. (9) Blast furnaces; A plurality of sensors for detecting a plurality of indicators indicating an operating state of the blast furnace; A blast furnace operation abnormality estimation device that detects an abnormality in a blast furnace, A measurement value acquisition unit that acquires measurement values ​​of a plurality of indexes during operation of the blast furnace; an estimate calculation unit that calculates first estimates of a plurality of indexes using predetermined measurement values ​​from among the acquired measurement values; a coincidence calculation unit that calculates a first coincidence for each index, the first coincidence indicating whether or not the acquired measurement value and the calculated first estimated value coincide with each index; an abnormality factor estimation unit that estimates an abnormality factor of the blast furnace based on a first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace; A blast furnace operation abnormality prediction system. Effect of the Invention

[0008] In one embodiment, the blast furnace operation abnormality estimation method can estimate an abnormality factor without relying on the absolute value of an index. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a blast furnace. [Diagram 2] FIG. 1 is a diagram showing a blast furnace operation anomaly estimation system according to a first embodiment for evaluating the operational status of a blast furnace. [Diagram 3]FIG. 3 is a block diagram of the blast furnace operation abnormality estimation device shown in FIG. 2. [Figure 4] 4 is a diagram showing an example of abnormality cause information shown in FIG. 3. [Diagram 5] 3 is a flowchart of an abnormality factor information generation process executed by the blast furnace operation abnormality estimation device shown in FIG. 2. [Figure 6] 3 is a flowchart of a blast furnace operation abnormality estimation process executed by the blast furnace operation abnormality estimation device shown in FIG. 2. [Figure 7] FIG. 4 is a diagram showing a blast furnace operation anomaly estimation system according to a second embodiment for evaluating the operational status of a blast furnace. [Figure 8] 8 is a flowchart of a blast furnace operation abnormality estimation process executed by the blast furnace operation abnormality estimation device shown in FIG. 7. [Figure 9] FIG. 13 shows the changes over time in the measured and estimated values ​​of the iron production rate, top gas composition, and blast pressure in a blast furnace from which abnormality cause information was generated, where (a) shows the changes over time in the measured and estimated values ​​of the iron production rate, (b) shows the changes over time in the measured and estimated values ​​of the top gas composition, and (c) shows the changes over time in the measured and estimated values ​​of the blast pressure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, a blast furnace operation abnormality estimation method, a blast furnace operation abnormality estimation device, a blast furnace operation abnormality estimation program, and a blast furnace operation abnormality estimation system according to the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments.

[0011] (Outline of blast furnace operation abnormality estimation method according to embodiment) The blast furnace operation abnormality estimation method according to the embodiment estimates the abnormality factor of the blast furnace based on a first degree of agreement, which is the degree of agreement between the measured values ​​of a plurality of indexes and a first estimated value, and abnormality factor information indicating the relationship between the first degree of agreement calculated for each index and the abnormality factor of the blast furnace and the index of the blast furnace, thereby estimating the abnormality factor of the blast furnace without relying on the absolute value of the index. The abnormality factor information is created based on the abnormality factor and the second degree of agreement indicating whether or not a second estimated value of the index in a stable state of the blast furnace coincides with a third estimated value of the index in the case where an abnormality factor of the blast furnace occurs. (Configuration and Function of Blast Furnace Operation Anomaly Estimation System According to the First Embodiment) FIG. 1 is a diagram illustrating a schematic configuration of a blast furnace.

[0012] The blast furnace 101 is a type of vertical furnace having a cylindrical tokkuri shape, and functions as a reaction device in which raw materials 121 supplied from the furnace top 111 react with reducing gas generated by hot air 122 supplied from a tuyere 112 provided below the blast furnace 101.

[0013] The raw materials 121 supplied from the furnace top 111 mainly include iron oxides 123 such as iron ore and sintered ore (hereinafter collectively referred to as "iron ore"), coke 124, and auxiliary raw materials 125 such as limestone. The iron ore contained in the iron oxides 123 is the iron source of pig iron produced by a reaction in the blast furnace 101, and the coke 124 not only functions as a reducing agent for the iron ore and a heat source for melting the raw materials 121, but also plays a role in maintaining the ventilation inside the blast furnace 101.

[0014] Inside the blast furnace 101, an ore layer 131 made of iron ore and auxiliary materials 125 and a coke layer 132 made of coke 124 are alternately stacked. These raw materials move downward in the blast furnace 101 while maintaining the stacked state.

[0015] In addition, hot air 121 and pulverized coal that functions as a complementary reducing agent for the coke 124 are supplied from the tuyere 112, and the pulverized coal and the coke 124 are gasified by the hot air to generate high-temperature reducing gas consisting of carbon monoxide, hydrogen, and the like. This high-temperature reducing gas becomes an ascending air current moving through the furnace and is blown up to the furnace top 111. The iron ore in the furnace is indirectly reduced by this reducing gas, and furthermore, changes from solid to liquid due to the heat of the reducing gas. The liquid iron is further directly reduced by the carbon of the coke while dripping through the coke layer 132, and becomes molten iron 135 containing about 5% carbon. In addition, components other than the iron contained in the iron ore become slag 134.

[0016] The cohesive zone 133 is a portion in which solid coke exists in slits between iron components in a semi-molten state, and it is mainly in this cohesive zone that the phase change of the iron components described above occurs.

[0017] In this way, in the blast furnace 101, which is a reaction device, solids, liquids, and gases coexist and reactions proceed dynamically. In addition, the reactions proceeding in the blast furnace 101 include endothermic reactions and exothermic reactions, and the operation of manufacturing pig iron is carried out while adjusting the heat balance and mass balance between these two opposing types of reactions.

[0018] FIG. 2 is a diagram showing an example of a schematic configuration of a blast furnace operation abnormality estimation system 100 according to an embodiment for evaluating the operational status of a blast furnace 101. As shown in FIG.

[0019] The blast furnace operation abnormality estimation system 100 includes a blast furnace 101, various sensors 102 installed in the vicinity of the blast furnace 101 and each detecting an operational variable indicating the operational status of the blast furnace 101, and a blast furnace operation abnormality estimation device 1.

[0020] The various sensors 102 measure various data related to the blast furnace 101, including operational variables that indicate the operating status of the blast furnace 101. Examples of the sensors 102 include sensors that measure the blast volume, blast temperature, blast moisture, and blast pressure of hot air supplied to the tuyere of the blast furnace 101, pressure sensors that measure the pressure of gas in the furnace wall and top, and sensors that measure the feed water pressure, feed and drain flow rate, feed water temperature, and drain temperature of cooling water. Examples of the sensors 102 include an analyzer that measures the carbon monoxide (CO) and carbon dioxide (CO2) contents in the gas in the furnace top and calculates the gas reduction efficiency ηCO from the measured carbon monoxide and carbon dioxide contents, and a sensor that measures the operating conditions used when calculating the pig iron tapping rate using a blast furnace mathematical model.

[0021] The various sensors 102 are not limited to these, and other examples include a weighing sensor that weighs various raw materials charged into the blast furnace 101, a measuring device that measures the height of the charged raw materials, a thermometer that measures the temperature distribution of each part of the furnace wall, hearth, etc. of the blast furnace 101, and various analytical devices that identify the amount of Si in the pig iron, the amount of S in the pig iron, the slag composition, etc. The various data measured by the various sensors 102 are output to the blast furnace operation abnormality estimation device 1 as measured values, and are used in the blast furnace operation abnormality estimation device 1 for blast furnace operation abnormality estimation processing, etc.

[0022] (Configuration and Function of Blast Furnace Operation Abnormality Estimation Device According to the First Embodiment) FIG. 3 is a diagram showing an example of a schematic configuration of the blast furnace operation abnormality estimation device 1. As shown in FIG.

[0023] The blast furnace operation abnormality estimation device 1 has a communication unit 11, a storage unit 12, an operation unit 13, a display unit 14, and a processing unit 15, and executes various processes by referring to data stored in the storage unit 12 based on a program previously stored in the storage unit 12. The blast furnace operation abnormality estimation device 1 also executes various processes in response to instructions input by an operator via the operation unit 13, and outputs the execution results to the display unit 14. The blast furnace operation abnormality estimation device 1 is, in one example, an electronic calculator also called a computer, and in another example, a monitoring and control device that monitors and controls the blast furnace 101.

[0024] The communication unit 11 has a communication interface circuit for connecting the blast furnace operation abnormality estimation device 1 to a network (not shown). The communication unit 11 supplies data received from the sensor 102 and other devices (not shown), such as a higher-level control device, via the network to the processing unit 15. The communication unit 11 also transmits data supplied from the processing unit 15 to other devices via the network.

[0025] The storage unit 12 has at least one of, for example, a semiconductor memory, a magnetic disk device, and an optical disk device. The storage unit 12 stores an operating system program, a driver program, an application program, data, and the like used in the processing in the processing unit 15. For example, the storage unit 12 stores, as the driver program, an input device driver program for controlling the operation unit 13, an output device driver program for controlling the display unit 14, and the like. The storage unit 12 also stores, as the application program, an abnormality factor information generation program for causing the processing unit 15 to execute an abnormality factor information generation process for generating abnormality factor information 120. The storage unit 12 also stores, as the application program, a blast furnace operation abnormality estimation program for causing the processing unit 15 to execute a blast furnace operation abnormality estimation process for detecting an abnormality in the blast furnace 101. The storage unit 12 also stores, as the application program, an abnormality factor information 120 used in the blast furnace operation abnormality estimation process, and a blast furnace mathematical model for calculating estimated values ​​of a plurality of indexes indicating the operation state of the blast furnace 101, and the like. The blast furnace mathematical model stored in the memory unit 12 is, for example, a mathematical model described in "Development of a Blast Furnace Mathematical Model" (by Hiroki Nishioka et al., Nippon Steel & Sumitomo Metal Technical Report No. 410, 2018). Furthermore, the memory unit 12 may temporarily store temporary data related to a predetermined process. The blast furnace mathematical model stored in the memory unit 12 can identify abnormal factors other than furnace heat indicators, such as changes in permeability of the blast furnace 101.

[0026] FIG. 4 is a diagram showing an example of the abnormality cause information 120 stored in the storage unit 12. As shown in FIG.

[0027] The abnormality factor information 120 has an "identifier" column 121, an "indicator indicating the operational state of the blast furnace" column 122, and an "abnormality factor" column 123, and indicates the relationship between the abnormality factor of the blast furnace 101 and the indicator of the blast furnace 101. In the "identifier column" 121, the 3-bit binary information indicated in the "indicator indicating the operational state of the blast furnace" column 122 and a number identifying the abnormality factor stored in association with the binary information indicated in the "indicator indicating the operational state of the blast furnace" column 122 are entered as identifiers. In the abnormality factor information 120, the "identifier" column 121 indicates three identifiers, "1" to "3".

[0028] The "indicator indicating the operation state of the blast furnace" column 122 shows binary information of "0" or "1" corresponding to each of the amount of iron produced, the furnace top gas composition, and the blast pressure, which are examples of indicators indicating the operation state of the blast furnace 101, as 3-bit data. The binary information "0" and "1" indicate, for each indicator, whether the second estimate of the indicator in the stable state of the blast furnace 101 matches the third estimate of the indicator in the event of an abnormality factor of the blast furnace 101. The second estimate is calculated using a first operation information group consisting of a plurality of operation information in the stable state of the blast furnace and a blast furnace mathematical model. The third estimate is calculated using a second operation information group consisting of a plurality of operation information in the stable state of the blast furnace 101 and disturbance information indicating that a disturbance has occurred for one predetermined operation information among the plurality of operation information in the stable state of the blast furnace 101, and a blast furnace mathematical model. The plurality of pieces of operation information included in the second operation information group other than the disturbance information are identical to the plurality of pieces of operation information included in the first operation information group.

[0029] The first and second operation information groups are composed of groups of information input to the blast furnace mathematical model, and include, for example, the particle size of the coke. The disturbance information included in the second operation information group indicates that a disturbance has occurred. For example, when the data indicating the particle size of the coke included in the first operation information group is 50 mm, the disturbance information indicating the particle size of the coke is 46 mm. When the index of the second estimate value and the third estimate value match, the binary information is "0", and when the index of the second estimate value and the third estimate value do not match, the binary information is "1". For example, when the binary information corresponding to the amount of pig iron tapped is "0", it indicates that the estimated values ​​of the amount of pig iron tapped between the second estimate value and the third estimate value match, and when the binary information corresponding to the amount of pig iron tapped is "1", it indicates that the estimated values ​​of the amount of pig iron tapped between the second estimate value and the third estimate value do not match.

[0030] The "abnormality cause" column 123 stores abnormality causes corresponding to each binary information shown in the "indicator showing the operating state of the blast furnace" column 122. In the "abnormality cause" column 123, a blank is stored in association with the binary information "000" which is shown in the order of the iron production rate, the top gas composition, and the blast pressure. In addition, a "particle size of coke" which indicates that the particle size of the coke inserted into the blast furnace 101 has changed is stored in association with the binary information "001". In addition, multiple items such as "particle size of ore, iron content in ore" which indicate that the particle size of the ore or the iron content in the ore has changed are stored in association with the binary information "111".

[0031] The operation unit 13 may be any device capable of operating the blast furnace operation abnormality estimation device 1, such as a keyboard, a touch pad, or the like. An operator can input characters, numbers, and the like via the operation unit 13. When the operation unit 13 is operated by the operator, it generates a signal corresponding to the operation. The generated signal is then supplied to the processing unit 15 as an instruction from the operator.

[0032] The display unit 14 may be any device capable of displaying video, images, characters, etc., such as a liquid crystal display, an organic EL (Electro-Luminescence) display, etc. The display unit 14 displays video corresponding to video data supplied from the processing unit 15, images corresponding to image data, characters corresponding to character data, etc.

[0033] The processing unit 15 has one or more processors and their peripheral circuits. The processing unit 15 centrally controls the overall operation of the blast furnace operation abnormality estimation device 1, and is, for example, a CPU (Central Processing Unit). The processing unit 15 controls the operation of the communication unit 11, the display unit 14, etc. so that various processes of the blast furnace operation abnormality estimation device 1 are executed in an appropriate procedure according to the programs stored in the storage unit 12, the operation of the operation unit 13, etc. The processing unit 15 executes processing based on the programs (operating system programs, driver programs, application programs, etc.) stored in the storage unit 12. The processing unit 15 can also execute multiple programs (application programs, etc.) in parallel.

[0034] The processing unit 15 has an abnormality factor information generating unit 20 and a blast furnace operation abnormality estimation processing unit 30. The abnormality factor information generating unit 20 has a second estimated value calculation unit 21, a third estimated value calculation unit 22, a degree of agreement determination unit 23, and an abnormality factor information generating unit 24. The blast furnace operation abnormality estimation processing unit 30 has a measurement value acquisition unit 31, a first operation information group acquisition unit 32, a first estimated value calculation unit 33, a degree of agreement calculation unit 34, an abnormality factor estimation unit 35, and an abnormality factor signal output unit 36. Each of these units is a functional module realized by a program executed by a processor included in the processing unit 15. Alternatively, each of these units may be implemented in the blast furnace operation abnormality estimation device 1 as firmware.

[0035] (Processing for generating abnormality factor information by the blast furnace operation abnormality estimation device according to the first embodiment) Fig. 5 is a flowchart of an abnormality factor information generation process executed by the blast furnace operation abnormality estimation device 1. The abnormality factor information generation process shown in Fig. 5 is executed mainly by the processing unit 15 in cooperation with each element of the blast furnace operation abnormality estimation device 1 based on an abnormality factor information generation program stored in advance in the storage unit 12. The abnormality factor information generation process is executed as appropriate when the operation state of the blast furnace 101 changes due to repairs, wind shutdown, etc.

[0036] First, the second estimate calculation unit 21 acquires a first operation information group used when calculating the second estimate using a blast furnace mathematical model from the storage unit 12 (S101). The multiple pieces of operation information included in the first operation information group include operation information used when calculating the tapping rate, top gas composition, and blast pressure, which are examples of multiple indexes in the stable state of the blast furnace 101, using the blast furnace mathematical model.

[0037] Next, the second estimate calculation unit 21 calculates second estimates of the tapping rate, the top gas composition, and the blast pressure using the first operation information group acquired through processing in S101 and the blast furnace mathematical model (S102). The second estimates are estimates of indexes indicating the operation state of the blast furnace 101 when the operation of the blast furnace 101 is in a stable state. The second estimate calculation unit 21 inputs the first operation information group acquired through processing in S101 into the blast furnace mathematical model stored in the memory unit 12 to calculate second estimates, and stores the estimated second estimates in the memory unit 12.

[0038] Next, the third estimate value calculation unit 22 acquires from the storage unit 12 a second operation information group to be used when calculating the third estimate value using the blast furnace mathematical model (S103). The second operation information group is a group of information in which each of the operation information groups included in the first operation information group is replaced with disturbance information, and the same number of sets as the number of operation information are stored in the storage unit 12.

[0039] Next, the third estimate calculation unit 22 calculates third estimates of the iron tapping rate, the top gas composition, and the blast pressure using the second operation information group acquired in the process of S103 and the blast furnace mathematical model (S104). The third estimates are estimates of indexes indicating the operation state of the blast furnace 101 when a disturbance occurs. The third estimate calculation unit 22 inputs the second operation information group acquired in the process of S103 into the blast furnace mathematical model stored in the storage unit 12 to calculate a third estimate, and stores the estimated third estimate in the storage unit 12 in association with the second operation information group in which the disturbance information has been changed.

[0040] Next, the third estimate calculation unit 22 determines whether or not the third estimate has been calculated for all the second operation information groups stored in the storage unit 12 (S105). When the third estimate calculation unit 22 determines that the third estimate has not been calculated for the second operation information groups stored in the storage unit 12 (S105-NO), the process returns to S103. Thereafter, the processes of S103 to S105 are repeated until the third estimate calculation unit 22 determines that the third estimate has been estimated for the second operation information groups stored in the storage unit 12 (S105-YES).

[0041] When it is determined that the third estimated value has been calculated for all the second operation information groups stored in the storage unit 12 (S105-YES), the agreement determination unit 23 determines the agreement between the third estimated value estimated in the process of S104 and the second estimated value estimated in the process of S102 (S106). First, the agreement determination unit 23 acquires a threshold value for each of the multiple indicators. Next, the agreement determination unit 23 determines whether the difference between the second estimated value and the third estimated value for each of the tapping rate, the top gas composition, and the blast pressure is equal to or greater than the acquired threshold value. When the difference between the second estimated value and the third estimated value is less than the acquired threshold value, the agreement determination unit 23 determines that the second estimated value and the third estimated value are identical. The coincidence determination unit 23 determines that the second estimated value and the third estimated value differ from each other when the difference between the second estimated value and the third estimated value is equal to or greater than the acquired threshold value.

[0042] When the agreement determination unit 23 determines that the second estimated value and the third estimated value are identical, it associates the binary information "0" with the corresponding index. When the agreement determination unit 23 determines that the second estimated value and the third estimated value are not identical, it associates the binary information "1" with the corresponding index. When the agreement determination unit 23 associates all of the tapping rate, the top gas composition, and the blast pressure with the binary information "0" or "1", it stores the agreement information in which the binary information is associated with each of the multiple indexes in the memory unit 12 in association with the input value replaced with the disturbance input value.

[0043] Next, the matching degree determination unit 23 determines whether or not the matching degrees have been determined for all the third estimated values ​​estimated in the process of S104 (S107). When the matching degree determination unit 23 determines that the matching degrees have not been determined for all the third estimated values ​​estimated in the process of S104 (S107-NO), the process returns to S106. Thereafter, the processes of S106 to S107 are repeated until the matching degree determination unit 23 determines that the matching degrees have been determined for all the third estimated values ​​estimated in the process of S104 (S107-YES).

[0044] Then, the abnormality factor information generating unit 24 generates the abnormality factor information 120 (S108). The abnormality factor information generating unit 24 acquires the abnormality factors corresponding to the operation information associated with each piece of the coincidence information generated in the process of S106, and generates the abnormality factor information 120 by associating the acquired abnormality factors with each piece of the coincidence information generated in the process of S106. The abnormality factor information generating unit 24 stores the generated abnormality factor information 120 in the storage unit 12. The correspondence relationship between the operation information and the abnormality factors used when generating the abnormality factor information 120 in the process of S108 may be stored in the storage unit 12, or may be acquired via a graphical user interface displayed on the display unit 14.

[0045] (Blast furnace operation abnormality estimation process by the blast furnace operation abnormality estimation device according to the first embodiment) Fig. 6 is a flowchart of a blast furnace operation abnormality estimation process executed by the blast furnace operation abnormality estimation device 1. The blast furnace operation abnormality estimation process shown in Fig. 6 is executed mainly by the processing unit 15 in cooperation with each element of the blast furnace operation abnormality estimation device 1 based on a blast furnace operation abnormality estimation program stored in advance in the storage unit 12. The blast furnace operation abnormality estimation process shown in Fig. 6 is executed at a predetermined execution cycle, for example, every hour.

[0046] First, the measurement acquisition unit 31 acquires measurement values ​​of a plurality of indexes during operation of the blast furnace 101 via the plurality of sensors 102 (S201), and stores the acquired measurement values ​​in the memory unit 12. The measurement values ​​acquired by the measurement acquisition unit 31 are the measurement values ​​of three indexes, namely, the iron tapping rate, the top gas composition, and the blast pressure, which are shown in the “Indicators indicating the operational state of the blast furnace” column 122 of the abnormality cause information 120.

[0047] Next, the first operation information group acquisition unit 32 acquires a first operation information group used when estimating the first estimated values ​​of the three indexes, the tapping rate, the top gas composition, and the blast pressure, the measured values ​​of which are acquired in the process of S201, using a blast furnace mathematical model (S202). The first operation information group acquired by the first operation information group acquisition unit 32 is an information group corresponding to the first operation information group acquired in the process of S101, and is a plurality of pieces of operation information acquired at the time closest to the time at which the abnormality factor is estimated among the plurality of pieces of operation information stored in the storage unit 12. The first operation information group stored in the storage unit 12 is updated at a predetermined cycle, such as every hour, every 8 hours, or every 24 hours, by an operator who monitors and controls the blast furnace 101.

[0048] Next, the first estimate calculation unit 33 calculates the first estimates of the iron tapping rate, the top gas composition, and the blast pressure by a blast furnace mathematical model using the first group of operation information acquired in the process of S202 (S203). The first estimate calculation unit 33 stores the estimated first estimates of the iron tapping rate, the top gas composition, and the blast pressure in the memory unit 12.

[0049] Next, the agreement extraction unit 34 extracts the agreement between the measurement value acquired in the process of S201 and the first estimated value calculated in the process of S203 (S204). First, the agreement extraction unit 34 acquires a threshold value for each of the pig iron production rate, the top gas composition, and the blast pressure. Next, the agreement extraction unit 34 judges whether the difference between the measurement value of the index and the first estimated value is equal to or greater than the threshold value. When the difference between the measurement value of the index and the first estimated value is less than the threshold value, the agreement extraction unit 34 judges that the measurement value of the index and the first estimated value match, and stores the result in the storage unit 12 in association with binary information "0". When the difference between the measurement value of the index and the first estimated value is equal to or greater than the threshold value, the agreement extraction unit 34 judges that the measurement value of the index and the first estimated value differ, and stores the result in the storage unit 12 in association with binary information "1".

[0050] Specifically, when the difference between the measured value of the pig iron production rate and the first estimated value is less than a threshold value, the agreement extraction unit 34 determines that the measured value of the pig iron production rate and the first estimated value match, and associates the indicator "piston production rate" with binary information "0". When the difference between the measured value of the pig iron production rate and the first estimated value is equal to or greater than a threshold value, the agreement extraction unit 34 determines that the measured value of the pig iron production rate and the first estimated value differ, and associates the indicator "piston production rate" with binary information "1". When the difference between the measured value of the top gas composition and the first estimated value is less than a threshold value, the agreement extraction unit 34 determines that the measured value of the top gas composition and the first estimated value match, and associates the indicator "top gas composition" with binary information "0". The agreement extraction unit 34 determines that the measured value of the top gas composition and the first estimated value are different when the difference between the measured value and the first estimated value of the top gas composition is equal to or greater than a threshold value, and associates the index "top gas composition" with binary information "1". The agreement extraction unit 34 determines that the measured value of the blast pressure and the first estimated value are different when the difference between the measured value and the first estimated value of the blast pressure is less than a threshold value, and associates the index "blast pressure" with binary information "0". The agreement extraction unit 34 determines that the measured value of the blast pressure and the first estimated value are different when the difference between the measured value and the first estimated value of the blast pressure is equal to or greater than a threshold value, and associates the index "blast pressure" with binary information "1". The agreement extraction unit 34 extracts 3-bit binary information, which is indicated in the order of the iron tapping rate, the top gas composition, and the blast pressure, from the agreement as the agreement between the measured value and the first estimated value.

[0051] Next, the abnormality factor estimation unit 35 refers to the abnormality factor information 120 and estimates the abnormality factor of the blast furnace 101 based on the degree of agreement extracted in the process of S204 (S205). First, the abnormality factor estimation unit 35 determines which of the 3-bit binary information shown in the abnormality factor information 120 corresponds to the 3-bit binary information indicated in the order of the tapping rate, the top gas composition, and the blast pressure, which are the degrees of agreement extracted in the process of S204. Then, the abnormality factor information 120 estimates the abnormality factor associated with the binary information determined to match in the abnormality factor information 120 as the abnormality factor of the blast furnace 101. When the 3-bit binary information extracted in the process of S204 is "000", the abnormality factor estimation unit 35 does not estimate the abnormality factor, determining that the abnormality factor has not been found. Furthermore, when the 3-bit binary information extracted in the process of S204 is any one of “100” to “111”, the fault factor estimation unit 35 estimates the fault factor associated with the corresponding binary information in the fault factor information 120 as the fault factor.

[0052] Then, the abnormality factor signal output unit 36 ​​outputs an abnormality factor signal indicating the abnormality factor estimated in the process of S205 (S206).

[0053] (Action and effect of the blast furnace operation abnormality estimation device according to the first embodiment) The blast furnace operation abnormality estimation device 1 can estimate the abnormality factor of the blast furnace 101 without relying on the absolute value of the index indicating the operating status of the blast furnace 101 by referring to abnormality factor information 120 indicating the relationship between the abnormality factor of the blast furnace 101 and the index of the blast furnace 101.

[0054] The abnormality factor information 120 is created based on the abnormality factor and a second degree of agreement indicating whether or not the second estimated value of the index in the stable state of the blast furnace 101 coincides with the third estimated value of the index in the event of an abnormality factor of the blast furnace 101. The blast furnace operation abnormality estimation device 1 estimates the abnormality factor by comparing the degree of agreement between the second estimated value of the index in the stable state of the blast furnace 101 and the third estimated value of the index in the event of an abnormality factor of the blast furnace 101, and the degree of agreement between the measured value acquired during the operation of the blast furnace 101 and the estimated value calculated using the measured value, and therefore can estimate the abnormality factor without depending on the absolute value of the index indicating the operation state of the blast furnace 101.

[0055] In addition, the abnormality factor information 120 stores consistency information, which is binary information "0" or "1", indicating whether the second estimated value of the index when the blast furnace 101 is in a stable state and the third estimated value of the index when an abnormality factor of the blast furnace 101 occurs coincide with each other. Therefore, the blast furnace operation abnormality estimation device 1 can easily compare the consistency between the measured value and the estimated value of the index indicating the operating state of the blast furnace 101.

[0056] In addition, the blast furnace operation abnormality estimation device 1 compares the difference between the measurement values ​​acquired during the operation of the blast furnace 101 and the estimated values ​​calculated using the measurement values ​​with a threshold value corresponding to the index to determine whether the measurement values ​​and the estimated values ​​match, thereby making it possible to determine with high accuracy whether the measurement values ​​and the estimated values ​​match.

[0057] Furthermore, the blast furnace operation abnormality estimation device 1 calculates an estimate of an index indicating the operation state of the blast furnace 101 using a plurality of pieces of operation information acquired at the time closest to the time when the abnormality factor is estimated, among the plurality of pieces of operation information, so that the calculated estimate can be as close as possible to the operation state of the blast furnace 101 excluding the abnormality factor. The blast furnace operation abnormality estimation device 1 can estimate the abnormality factor with high accuracy by using an estimate of a state close to the operation state of the blast furnace 101.

[0058] (Configuration and Function of Blast Furnace Operation Abnormality Estimation Device According to Second Embodiment) FIG. 7 is a diagram showing an example of a schematic configuration of a blast furnace operation abnormality estimation device 2 according to the second embodiment. As with the blast furnace operation abnormality estimation device 1 according to the first embodiment, the blast furnace operation abnormality estimation device 2 acquires a measurement value detected by a sensor 102 shown in FIG. 2, and estimates an abnormality factor of an operation abnormality in the blast furnace 101 shown in FIG. 1 by referring to abnormality factor information stored in a storage unit 12. In addition, the blast furnace operation abnormality estimation device 2 calculates a calculated value of the material balance of the blast furnace 101, and determines whether or not the measurement value detected by the sensor 102 includes an abnormal value based on the calculated calculated value of the material balance.

[0059] As shown in FIG. 7, the blast furnace abnormality detection processing unit 40 according to the second embodiment has a material balance calculation unit 41, a measurement value abnormality determination unit 42, and a measurement value abnormality signal output unit 43. The configurations and functions of the components of the blast furnace operation abnormality estimation device 2 other than the material balance calculation unit 41, the measurement value abnormality determination unit 42, and the measurement value abnormality signal output unit 43 are the same as the configurations and functions of the components of the blast furnace operation abnormality estimation device 1 of the first embodiment. Therefore, detailed descriptions of the configurations and functions of the components of the blast furnace operation abnormality estimation device 2 other than the material balance calculation unit 41, the measurement value abnormality determination unit 42, and the measurement value abnormality signal output unit 43 will be omitted by assigning the same reference numerals.

[0060] (Processing for generating abnormality factor information by the blast furnace operation abnormality estimation device according to the second embodiment) Fig. 8 is a flowchart of a blast furnace operation abnormality estimation process executed by the blast furnace operation abnormality estimation device 2. The blast furnace operation abnormality estimation process shown in Fig. 8 is executed mainly by the processing unit 16 in cooperation with each element of the blast furnace operation abnormality estimation device 2 based on a blast furnace operation abnormality estimation program stored in advance in the storage unit 12. The blast furnace operation abnormality estimation process shown in Fig. 8 is executed at a predetermined execution cycle, for example, every hour. The processes of S301 and S302 are the same as the processes of S201 and S202, and therefore detailed description thereof will be omitted here.

[0061] Following the process of S302, a calculated value of the material balance of the blast furnace 101 is calculated (S303). The material balance calculation unit 41 extracts the measured values ​​of the indicators used to calculate the calculated value of the material balance from the measured values ​​acquired in the process of S301 over a predetermined period, which is, for example, 12 hours. The material balance calculation unit 41 calculates an average value for each of the measured values ​​of the indicators extracted over the predetermined period. Next, the material balance calculation unit 41 calculates the calculated value of the material balance using the calculated average value of the measured values ​​of the indicators. The material balance calculation unit 41 calculates, for example, the calculated value of the oxygen balance of the blast furnace 101 as the calculated value of the material balance. The method of calculating the calculated value of the material balance by the material balance calculation unit 41 is not particularly limited, and for example, known methods described in JP 2012-172222 A and JP 2022-14169 A can be used. The target element of the material balance is not particularly limited and may be, for example, oxygen, carbon, hydrogen, etc. In this specification, the target element of the material balance will be described as oxygen.

[0062] Next, the measurement value abnormality determination unit 42 acquires an actual value of the material balance of the blast furnace 101 from the measurement values ​​acquired in the process of S301 (S304). The measurement value abnormality determination unit 42 acquires an actual value of the oxygen balance of the blast furnace 101 from the amount of oxygen in the pulverized coal and hot air 122 supplied from the tuyere 112, the amount of oxygen in the raw materials 121 charged from the furnace top 111, and the amount of oxygen in the molten iron, slag, and furnace top gas, etc.

[0063] Next, the measurement value abnormality determination unit 42 determines whether or not the measurement value of the index acquired in the process of S301 includes an abnormal value based on the calculated value of the material balance calculated in the process of S303 and the actual value of the material balance acquired in the process of S304 (S305). When the difference between the calculated value of the material balance calculated in the process of S303 and the actual value of the material balance acquired in the process of S304 is within a predetermined range, the measurement value abnormality determination unit 42 determines that the measurement value of the index acquired in the process of S301 does not include an abnormal value (S305-NO). Also, when the difference between the calculated value of the material balance calculated in the process of S303 and the actual value of the material balance acquired in the process of S304 is outside the predetermined range, the measurement value abnormality determination unit 42 determines that the measurement value of the index acquired in the process of S301 includes an abnormal value (S305-YES).

[0064] The range used for the judgment in the process of S305 is determined based on the difference between the actual value and the calculated value of the oxygen balance over a predetermined period, which is, for example, one month. First, the measurement value abnormality judgment unit 42 acquires the actual value of the oxygen balance over the predetermined period, and stores the acquired actual value of the oxygen balance in the storage unit 12. Next, the measurement value abnormality judgment unit 42 calculates the calculated value of the oxygen balance over the predetermined period in the same manner as the process of S303, and stores the calculated value of the oxygen balance in the storage unit 12. Next, the measurement value abnormality judgment unit 42 calculates the average value and variance σ of the difference between the actual value of the oxygen balance and the calculated value of the oxygen balance. Then, the measurement value abnormality judgment unit 42 determines the range used for the judgment to be within ±3σ from the average value of the difference between the actual value of the oxygen balance and the calculated value of the oxygen balance. Note that the range used for the judgment in the process of S305 may be determined by other methods.

[0065] When the measurement value abnormality determination unit 42 determines that the measurement value of the index includes an abnormal value (S305-YES), the measurement value abnormality signal output unit 43 outputs a measurement value abnormality signal indicating that the measurement value of the index acquired in the process of S301 includes an abnormal value (S306). Then, the blast furnace operation abnormality estimation process ends.

[0066] When the measurement value abnormality determination unit 42 determines that the measurement value of the index does not include an abnormal value (S305-NO), the process proceeds to S306. The processes of S307 to S310 are similar to the processes of S203 to S206, and therefore detailed description thereof will be omitted here.

[0067] (Action and effect of the blast furnace operation abnormality estimation device according to the second embodiment) The blast furnace operation abnormality estimation device 2 determines whether the measurement value of the index acquired based on the calculated value of the material balance of the blast furnace 101 includes an abnormal value, and therefore can distinguish the abnormality of the measurement value caused by the abnormality and failure of the sensor 102 and a bug of the program used in the process of acquiring the measurement value of the index from the abnormality of the blast furnace 101, and estimate the abnormality factor of the blast furnace 101. When the sensor 102 is normal and outputs a normal measurement value, the calculated value of the material balance of the blast furnace 101 is an appropriate value, but when the sensor 102 outputs an abnormal measurement value, the calculated value of the material balance of the blast furnace 101 is an abnormal value. The blast furnace operation abnormality estimation device 2 can detect the abnormality of the measurement value of the index by detecting that the calculated value of the material balance of the blast furnace 101 is an abnormal value. The blast furnace operation abnormality estimation device 2 can detect the abnormality of the measurement value of the index by detecting the abnormality of the measurement value of the index, and can determine the operation abnormality of the blast furnace 101 by separating it from the abnormality of the measurement value of the index.

[0068] In addition, the blast furnace operation abnormality estimation device 2 calculates the calculated value of the material balance of the blast furnace 101 using the average value of the measured values ​​of the indexes extracted over a predetermined period, so that the calculated value of the material balance of the blast furnace 101 can be calculated without being affected by short-term fluctuations in the measured values ​​of the indexes. In addition, the blast furnace operation abnormality estimation device 2 uses the average value of the measured values ​​of the indexes over a period of about 12 hours, so that the calculated value of the material balance of the blast furnace 101 can be calculated using the average value of the measured values ​​over a period longer than the time constant of the blast furnace 101, which is usually about several hours. The blast furnace operation abnormality estimation device 2 calculates the calculated value of the material balance of the blast furnace 101 using the average value of the measured values ​​over a period longer than the time constant of the blast furnace 101, so that the influence of the time constant of the blast furnace 101 can be suppressed.

[0069] (Modification of the blast furnace operation abnormality estimation device according to the embodiment) In the blast furnace operation abnormality estimation devices 1 and 2, three indices, i.e., iron production rate, furnace top gas composition, and blast pressure, are given as indices indicating the operation state of the blast furnace 101, but in the blast furnace operation abnormality estimation device according to the embodiment, two or four or more indices may be used. For example, in the blast furnace operation abnormality estimation device according to the embodiment, the coke ratio, pulverized coal ratio, furnace top pressure, furnace top pressure, furnace top temperature, and molten iron temperature may be used as indices indicating the operation state of the blast furnace 101. In the blast furnace operation abnormality estimation device according to the embodiment, by increasing the number of indices indicating the operation state of the blast furnace 101, the abnormality factor can be estimated with higher accuracy.

[0070] Furthermore, in the blast furnace operation abnormality estimation devices 1 and 2, the abnormality factor information 120 has coincidence information in which the difference between the second estimated value and the third estimated value is binarized with "0" and "1", but in the blast furnace operation abnormality estimation device according to the embodiment, the abnormality factor information may have multi-valued indexes. Furthermore, in the blast furnace operation abnormality estimation device according to the embodiment, the abnormality factor information may have coincidence information indicated by an index indicating a correlation between the estimated value of an index other than the index calculated by the blast furnace mathematical model and the abnormality factor.

[0071] In addition, the blast furnace operation abnormality estimation devices 1 and 2 calculate an estimate of an index indicating the operation status of the blast furnace 101 using multiple pieces of operation information acquired at the time closest to the time at which the abnormality cause is estimated, but the blast furnace operation abnormality estimation device of the embodiment may calculate an estimate using an input value other than the latest value in cases such as when the fluctuation in the input value is small.

[0072] Furthermore, although the blast furnace operation abnormality estimation devices 1 and 2 execute the abnormality factor information generation process, the blast furnace operation abnormality estimation device according to the embodiment may not execute the abnormality factor information generation process, and another electronic computer may execute the abnormality factor information generation process. Furthermore, although the blast furnace operation abnormality estimation device 1 executes the abnormality factor information generation process as a process separate from the blast furnace operation abnormality estimation process, the blast furnace operation abnormality estimation device according to the embodiment may execute the abnormality factor information generation process as a process integrated with the blast furnace operation abnormality estimation process.

[0073] In addition, since the blast furnace operation abnormality estimation devices 1 and 2 use a blast furnace mathematical model that can identify indicators other than the furnace heat indicator, the heat balance of the upper part of the blast furnace 101 can be calculated in addition to the heat balance of the lower part of the blast furnace 101.

[0074] Furthermore, the blast furnace operation abnormality estimation device 2 determines whether or not the acquired measured value of the index includes an abnormal value based on the difference between the calculated value of the material balance and the actual measured value. However, the blast furnace operation abnormality estimation device according to the embodiment may determine whether or not the measured value of the index includes an abnormal value based only on the calculated value of the material balance.

[0075] For example, the blast furnace operation abnormality estimation device according to the embodiment may calculate the calculated value of the material balance at every predetermined cycle, and determine whether or not the measured value of the index includes an abnormal value based on the difference between the calculated value of the material balance calculated last time and the calculated value of the material balance calculated this time. The blast furnace operation abnormality estimation device according to the embodiment determines that the measured value of the index includes an abnormal value when the difference between the calculated value of the material balance calculated last time and the calculated value of the material balance calculated this time is equal to or greater than a predetermined threshold. In addition, the blast furnace operation abnormality estimation device according to the embodiment determines that the measured value of the index does not include an abnormal value when the difference between the calculated value of the material balance calculated last time and the calculated value of the material balance calculated this time is less than a predetermined threshold. Note that the threshold value or the like used when determining whether or not the measured value of the acquired index includes an abnormal value may be set in advance. EXAMPLES

[0076] For a given blast furnace, a blast furnace mathematical model was used to generate abnormality factor information when the grain size of the coke charged in the blast furnace was changed, with the indicators indicating the operating state of the blast furnace being the iron tapping rate, top gas composition, and blast pressure. When the data indicating the grain size of the coke was changed to disturbance data, the changes in the iron tapping rate and top gas composition were less than a predetermined threshold, and the changes in the blast pressure were equal to a predetermined threshold. The 3-bit binary information indicated in this order by the iron tapping rate, top gas composition, and blast pressure when the data indicating the grain size of the coke was changed to disturbance data was "001."

[0077] FIG. 9 is a diagram showing the time-dependent changes in the measured and estimated values ​​of the iron production rate, top gas composition, and blast pressure in a blast furnace for which abnormality factor information was generated. FIG. 9(a) shows the time-dependent changes in the measured and estimated values ​​of the iron production rate, FIG. 9(b) shows the time-dependent changes in the measured and estimated values ​​of the top gas composition, and FIG. 9(c) shows the time-dependent changes in the measured and estimated values ​​of the blast pressure. In FIG. 9(a) to (c), the horizontal axis indicates time, in FIG. 9(a) the vertical axis indicates the iron production rate, in FIG. 9(b) the vertical axis indicates the top gas composition, and in FIG. 9(c) the vertical axis indicates the blast pressure. In FIG. 9(a) to (c), the solid lines indicate the measured values, and the dashed lines indicate the estimated values.

[0078] In Figure 9(a)-(c), the particle size of the charged coke changed at time "15". The threshold for judging the difference between the measured and estimated values ​​of the iron production rate was set to 100 [t / d], the threshold for judging the difference between the measured and estimated values ​​of the top gas composition was set to 0.005 [-], and the threshold for judging the difference between the measured and estimated values ​​of the blast pressure was set to 4 [kPa].

[0079] As shown in Figure 9(a), for the iron production rate, the difference between the measured and estimated values ​​does not exceed the threshold value of 100 [t / d] after time "4650", so it is determined that the measured and estimated values ​​match, and the binary information is "0". As shown in Figure 9(b), for the top gas composition, the difference between the measured and estimated values ​​does not exceed the threshold value of 0.005 [-] after time "15", so it is determined that the measured and estimated values ​​match, and the binary information is "0".

[0080] As shown in Fig. 9(c), the difference between the measured and estimated blast pressure exceeds the threshold value of 4 kPa after time "15", so it is determined that the measured and estimated values ​​do not match, and the binary information is "1". Since the binary information of the tapping rate and top gas composition is "0", and the binary information of the blast pressure is "1", the 3-bit binary information shown in this order of the tapping rate, top gas composition, and blast pressure is "001".

[0081] Since the 3-bit binary information in the abnormality cause information and the 3-bit binary information estimated from the measured values ​​and estimated values ​​are both "001", it has been shown that the blast furnace operation abnormality estimation method of the embodiment can estimate the abnormality cause when the particle size of the coke changes. [Explanation of symbols]

[0082] 1, 2 Blast furnace operation abnormality estimation device 15, 16 Processing section 20. Anomaly cause information generation processing unit 21 Second estimate calculation unit 22 Third Estimate Calculation Unit 23 Matching degree determination section 24. Anomaly Cause Information Generation Unit 30, 40 Blast furnace operation abnormality estimation processing unit 31 Measurement value acquisition section 32 First operation information group acquisition unit 33 First estimated value calculation unit 34 Matching degree calculation unit 35 Abnormality Cause Identification Department 36 Abnormality factor signal output section 41 Material balance calculation section 42 Measurement value abnormality judgment unit 43 Measurement value abnormality signal output section 100 Blast Furnace Operation Abnormality Prediction System 101 Blast Furnace 102 Sensors

Claims

1. Obtaining measurements of multiple indicators during blast furnace operation, Calculating first estimates of a plurality of indices using predetermined measured values ​​from among the acquired measured values; calculating a first degree of agreement for each of the indices, the first degree of agreement indicating whether or not the acquired measurement value and the calculated first estimated value match for each of the indices; An abnormality factor of the blast furnace is estimated based on the first degree of agreement calculated for each of the indicators and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the indicator of the blast furnace. A method for estimating abnormalities in blast furnace operation.

2. The abnormality cause information of the blast furnace is The abnormality factor is created based on a second agreement indicating whether or not a second estimated value of the index in a stable state of the blast furnace coincides with a third estimated value of the index in a case where an abnormality factor of the blast furnace occurs. The method for estimating an abnormality in blast furnace operation according to claim 1.

3. The first estimated value is calculated using a first operational information group consisting of a plurality of operational information acquired during operation of the blast furnace and a blast furnace mathematical model; The second estimate is A first operation information group consisting of a plurality of operation information in a stable state of the blast furnace and a blast furnace mathematical model are used to calculate the blast furnace operation information; The third estimate is The blast furnace operation abnormality estimating method according to claim 2, characterized in that the blast furnace operation abnormality estimating method is calculated using a blast furnace mathematical model and a second operation information group consisting of the plurality of operation information when the blast furnace is in a stable state and disturbance information indicating that a predetermined piece of operation information among the plurality of operation information when the blast furnace is in a stable state.

4. The first degree of agreement is binary information indicating whether or not the first estimated value and the measured value agree with each other, The second degree of agreement is The blast furnace operation abnormality estimating method according to claim 3, characterized in that the information is binary information indicating whether the second estimated value and the third estimated value match each other or not.

5. The first estimate is The blast furnace operation abnormality estimating method according to claim 4, characterized in that the estimated value is calculated using a plurality of pieces of acquired operation information that are acquired at a time closest to the time at which the abnormality factor is estimated and the blast furnace mathematical model.

6. Calculating the material balance of the blast furnace from the obtained measured values; The blast furnace operation abnormality estimating method according to any one of claims 1 to 5, characterized in that it is determined whether the acquired measured value includes an abnormal value based on the calculated value of the material balance.

7. A measurement value acquisition unit that acquires measurement values ​​of a plurality of indexes during operation of the blast furnace; a first estimate calculation unit that calculates first estimates of a plurality of indexes using predetermined measurement values ​​among the acquired measurement values; a coincidence calculation unit that calculates, for each of the indices, a first coincidence indicating whether or not the acquired measurement value and the calculated first estimated value coincide with each of the indices; An abnormality factor estimation unit that estimates an abnormality factor of the blast furnace based on the first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace; A blast furnace operation abnormality estimation device having the above.

8. Obtaining measurements of multiple indicators during blast furnace operation, Calculating first estimates of a plurality of indices using predetermined measured values ​​from among the acquired measured values; calculating a first degree of agreement for each of the indices, the first degree of agreement indicating whether or not the acquired measurement value and the calculated first estimated value match for each of the indices; An abnormality factor of the blast furnace is estimated based on the first degree of agreement calculated for each of the indicators and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the indicator of the blast furnace. A blast furnace operation abnormality estimation program that causes a computer to execute the processing.

9. Blast furnaces and A plurality of sensors for detecting a plurality of indicators indicating an operating state of the blast furnace; A blast furnace operation abnormality estimation device that estimates an abnormality in the blast furnace, A measurement value acquisition unit that acquires measurement values ​​of a plurality of indexes during operation of the blast furnace; a first estimate calculation unit that calculates first estimates of a plurality of indexes using predetermined measurement values ​​among the acquired measurement values; a coincidence calculation unit that calculates, for each of the indices, a first coincidence indicating whether or not the acquired measurement value and the calculated first estimated value coincide with each of the indices; An abnormality factor estimation unit that estimates an abnormality factor of the blast furnace based on the first degree of agreement calculated for each index and abnormality factor information indicating a relationship between the abnormality factor of the blast furnace and the index of the blast furnace; A blast furnace operation abnormality prediction system.

Images