WO2019132476A1 - System and method for evaluating operational conditions of blast furnace - Google Patents
System and method for evaluating operational conditions of blast furnace Download PDFInfo
- Publication number
- WO2019132476A1 WO2019132476A1 PCT/KR2018/016580 KR2018016580W WO2019132476A1 WO 2019132476 A1 WO2019132476 A1 WO 2019132476A1 KR 2018016580 W KR2018016580 W KR 2018016580W WO 2019132476 A1 WO2019132476 A1 WO 2019132476A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- combustion state
- index
- air
- image data
- classification
- Prior art date
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D19/00—Arrangements of controlling devices
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B5/00—Making pig-iron in the blast furnace
- C21B5/006—Automatically controlling the process
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/16—Tuyéres
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/24—Test rods or other checking devices
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27B—FURNACES, KILNS, OVENS, OR RETORTS IN GENERAL; OPEN SINTERING OR LIKE APPARATUS
- F27B1/00—Shaft or like vertical or substantially vertical furnaces
- F27B1/10—Details, accessories, or equipment peculiar to furnaces of these types
- F27B1/28—Arrangements of monitoring devices, of indicators, of alarm devices
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
- F27D21/02—Observation or illuminating devices
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B2300/00—Process aspects
- C21B2300/04—Modeling of the process, e.g. for control purposes; CII
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
- F27D21/02—Observation or illuminating devices
- F27D2021/026—Observation or illuminating devices using a video installation
Definitions
- the present application relates to a blast furnace operation situation evaluation system and method.
- Japanese Laid-Open Patent Application No. 2015-52148 discloses a control method based on determination of the operational status of a furnace.
- an embodiment of the present invention provides a blast furnace operation situation evaluation system.
- a blast furnace operation status evaluation system includes an image acquisition unit that acquires a plurality of wind turbine image data provided in a blast furnace; An image collecting unit for collecting the image data of the wind type acquired by the image acquiring unit; A tidal burning state determination unit for classifying the air-fueled combustion states based on artificial intelligence using the air-conditioned image data; A tidal burning state index generating unit for generating an air-fueled combustion state index by using the air-fueled combustion state classification result classified by the tidal burning state determining unit; And an integrated evaluation unit for generating an integrated combustion state index based on the air-fuel ratio combustion state index.
- another embodiment of the present invention provides a method for evaluating blast furnace operation status.
- a method for evaluating blast furnace operation status comprising: collecting a plurality of blast furnace image data provided in a blast furnace; Classifying the combustion state of each airway based on artificial intelligence using the airway image data; Generating an air-fueled combustion state index by using the air-fuel ratio classified state classification result; And generating an integrated combustion state index based on the air-fueled combustion state index.
- the present invention it is possible to classify the tuyere combustion state based on the deep running based on the tuyere image data.
- the analysis result of the tuyere image data and the analysis result of the blast furnace data are additionally , And it is possible to collectively evaluate and control the blast furnace operation status.
- FIG. 1 is a configuration diagram of a blast furnace operation situation evaluation system according to an embodiment of the present invention.
- FIG. 2 is a view for explaining a concept of classifying a tuyere combustion state primarily on a deep learning basis according to an embodiment of the present invention.
- FIG. 3 and FIG. 4 are views for explaining a concept of determining a tidal combustion state classification based on a result obtained by accumulating results obtained by primary classification on a deep learning basis according to an embodiment of the present invention.
- FIG. 5 is a flowchart of a method for evaluating blast furnace operation status according to another embodiment of the present invention.
- FIG. 1 is a configuration diagram of a blast furnace operation situation evaluation system according to an embodiment of the present invention.
- a blast furnace operation status evaluation system 100 includes an image acquisition unit 110, an image collection unit 120, a tile combustion state determination unit 130, A production information collection unit 150, an integrated evaluation unit 160, and an aging control unit 170.
- the image acquiring unit 110 may acquire image data for each tuyere 11 provided in the blast furnace 10.
- the image acquiring unit 110 may include a plurality of cameras installed in the respective tuyeres 11, and acquire the respective blind image data in real time (for example, in units of ms) can do.
- the image collecting unit 120 may collect the image data of each type obtained by the image obtaining unit 110.
- the image capturing unit 120 may collect image data obtained in real time for each tuyere from a plurality of cameras included in the image capturing unit 110.
- the image collection unit 120 may map the collected image data to collection environment information including a tug number and a data acquisition time.
- the image data that has been mapped by the image collection unit 120 may be stored in a data storage (not shown) provided in the blast operation status evaluation system 100 or may be transmitted to the tile combustion status determination unit 130 in real time .
- the toughening combustion state determination unit 130 is for classifying the combustion state of each airway using the airway distinction image data received from the image collection unit 120.
- the airway state determination unit 130 includes an AI based determination unit 131, (132).
- the AI-based determination unit 131 can classify the combustion state of each air-fuel mixture based on artificial intelligence using the image data of the air-conditioned air. For example, the AI-based determination unit 131 can classify the combustion state of each air-fuel mixture based on the deep-run basis.
- the AI-based determination unit 131 may classify each air-fueled combustion state based on the CNN (Convolutional Neural Network) using the air-differentiated image data.
- CNN Convolutional Neural Network
- the AI-based determination unit 131 determines the classification of the tidal combustion state based on the result of accumulating the primary classification results of the classified combustion state classification in time series to further improve the consistency of the combustion state classification .
- FIG. 2 is a view for explaining a concept of classifying a tuyere combustion state primarily on a deep learning basis according to an embodiment of the present invention.
- the AI-based determination unit 131 performs image deep-processing (for example, based on CNN) on each of the first tugging image to the Nth tugging image data 21 to 2N obtained for each tug To obtain the first to tenth flue gas combustion state classification to Nth tuyere combustion state classification results 21 'to 2N'.
- N means the number of tongues.
- FIG. 3 and FIG. 4 are views for explaining a concept of determining a tidal combustion state classification based on a result obtained by accumulating results obtained by primary classification on a deep learning basis according to an embodiment of the present invention.
- the AI-based determination unit 131 accumulates the combustion state classification results primarily classified for each taurine in time series, that is, the first tidal combustion state classification 31-1, 31- 2, 31-3, the second tidal combustion state classifications 32-1, 32-2, 32-3 and the Nth tidal combustion state classifications 3N-1, 3N-2, 3N-3, It is possible to determine the tidal combustion state classification by tungsten, and obtain the determined tuyball combustion state classification result (31 'to 33').
- any combustion state classification is performed based on a plurality of combustion state classification results primarily classified during any time period (t-1 to t + 1) Or more, it can be decided to classify the combustion state. This makes it possible to improve the accuracy of classification of the tuyere combustion state.
- the AI-based determination unit 131 calculates the first tidal combustion state classification 41-1, 41-2, and 41-3 as a result of accumulating the combustion state classification results primarily classified for each taurine, 2, 41 - 3), the second tidal combustion state classifications 42 - 1, 42 - 2, 42 - 3 and the N th tug combustion state classifications 4 N - 1, 4 N - 2, 4 N - Based on the time-series deep running, the tuyere combustion state classification can be determined for each tuyere, and the confirmed tuyere combustion state classification results (41 'to 43') can be obtained.
- the AI-based determination unit 131 may determine whether or not an RNN (Recurrent Neural Network) or an RNN (Recurrent Neural Network) based on a plurality of combustion state classification results that are primarily classified for each time period t- Based on the Recurrent Convolutional Neural Network (RCNN), it is possible to determine the classification of each type of tidal burning condition.
- RNN Recurrent Neural Network
- RCNN Recurrent Convolutional Neural Network
- the accuracy of the classification of the tuyere combustion state is further improved by applying the image time- .
- the tuyball combustion state classification can be established by accumulating the result of primary classification for the time period set by the user, starting from the point in time when the tuyball combustion state classification is first detected .
- the time period described above is adjusted according to the elapsed time information from when the tidal combustion state classification is first detected to when the tidal combustion state classification is transferred to another state, Can be improved.
- the combustion state of the tuyeres classified by the AI-based determination unit 131 may include, for example, a normal combustion state, a poor combustion state, a non-pulverized coal injection, a unreduced melt drop (dropping light drop), a cock turning, and the like.
- the pulverized coal is judged whether or not the pulverized coal is blown, and the unreduced melted material drop (live fall drop) means whether or not the raw material to be reduced at the furnace portion is not reduced,
- the turning of the coke means judging whether or not the coke turns in the middle of the furnace.
- the image processing base determination unit 132 can diagnose the tuyere apparatus through the image processing of the image data of the tandem image, and can determine the tuyere combustion state.
- the image processing-based determination unit 132 may perform an image processing on the image data of the background image to determine whether or not the toughening equipment abnormality including the presence or absence of the tornado, presence of the tornado, clogging of the tornado, lance banding, It can be judged.
- the image processing base determination unit 132 may extract the combustion area and the combustion brightness (i.e., luminance) through image processing on the image data of the air-conditioned image.
- the image processing base determination unit 132 can determine the pulverized coal flow rate through image processing on the image data of the wind type.
- the determination by the AI-based determination unit 131 and the image processing-based determination unit 132 may be performed in parallel.
- the results of the classification of the air condition classified by the air conditioner combustion state determiner 130 and the results of the air conditioner diagnosis may be mapped and stored and managed together with the image data and the collected environment information.
- the tidal burning state index generating unit 140 may generate the air-fueled combustion state index using the air-fuel burning state classification result classified by the tidal burning state determining unit 130.
- the air-fuel ratio combustion state index generated by the air-flow combustion state index generating unit 140 may include a combustion state failure index, a pulverized coal non-spent index, a unreduced melt falling index, A combustion state level index, a pulverized coal flow rate index, a ruggedness index (Raceway) index, and the like.
- the tidal burning state index generator 140 counts the number of times that an arbitrary classification result is generated based on the classification result of each air-fuel mixture combustion state by the tidal combustion state determiner 130 every predetermined period,
- the related index can be generated by scoring according to the number of times counted for each cycle.
- the tidal combustion state index generator 140 scales the combustion state level index according to the combustion area and the combustion brightness (i.e., brightness) extracted by the tidal flame combustion state determiner 130, Can be combined to generate a combustion status level index.
- the reference information used for generating the combustion state level index can be updated according to the input signal by the manager. Accordingly, the updated reference information is reflected in real time, and the index information reflecting the blast condition can be generated.
- the tidal burning state index generator 140 may generate the tidal system abnormality index by scoring the tidal system diagnosis result determined by the tidal burning state determiner 130.
- the irregularity abnormality index may include a tug curvature index, a tug attachment index, a tug blockage index, a lance damage index, and the like.
- the operation information collecting unit 150 is for collecting, in real time, operational information generated during blast furnace operation.
- the operating information may include, for example, the temperature of the corona body, the pressure, the flow rate of the cooling water, and the like.
- the operation information collected in real time by the operation information collecting unit 150 can be mapped and stored and managed with the tidal combustion state index information generated by the tidal combustion state index generating unit 140 described above.
- the integrated evaluation unit 160 calculates the integrated value of the tug combustion state index information generated by the tug burning state index generating unit 140 based on the tug burning state index information generated for each tug and the operation information collected by the operation information collection number 150 It can be evaluated integrally.
- the integrated evaluation unit 160 may generate an integrated combustion state index by comprehensively considering the taurine combustion state index information generated for each taurine by the taurine combustion state index generating unit 140.
- the integrated combustion state index is 1: 1 matched to the tidal combustion state index generated for each tuyere such as integrated combustion state failure index, integrated pulverized coal unloading index, integrated unreduced melt drop
- the integrated combustion state index is 1: 1 matched to the tidal combustion state index generated for each tuyere such as integrated combustion state failure index, integrated pulverized coal unloading index, integrated unreduced melt drop
- the integrated evaluation unit 160 may generate the columnar balance index based on the tuyere combustion race index generated for each tuyere.
- the integrated evaluation unit 160 can generate an integrated tuff facility abnormality index based on the tuff equipment abnormality index generated for each tornado.
- the sulfur content control unit 170 Based on the tug burning state index information generated for each tuyere by the tug burning state index generating unit 140 or the integrated combustion state index generated by the integrated evaluation unit 160, the sulfur content control unit 170 performs pulverized coal injection control, N2 Purge control, and blast furnace charge control to control blast furnace burning.
- the aging control unit 170 may perform the pulverized coal injection control when the pulverized coal unloading index for any tuyere exceeds a preset reference value.
- the sulfur content control unit 170 may perform the blast furnace charge control when the drop of the live light occurs in an arbitrary tidal area and the unreduced melt drop index exceeds a preset reference value.
- the aging control unit 170 may integrally control the plurality of tuyere based on the integrated combustion state index or the circumferential balance index information.
- the sulfur content control unit 170 can control the content of the blast furnace, such as changing the charge direction by changing the charge content distribution.
- the blast operation status evaluation system 100 described above with reference to FIG. 1 includes a processing device capable of performing application of an artificial intelligence algorithm to input data and image processing and capable of calculating various indices, a control device Or a combination of devices.
- FIG. 5 is a flowchart of a method for evaluating blast furnace operation status according to another embodiment of the present invention.
- the image obtaining unit 110 and the image collecting unit 120 calculate image data for each tile provided in the blast furnace in real time (S510).
- the tuyere combustion state determination unit 130 can classify the combustion states of the respective tuyere using the image data for each airway classification (S520). Specifically, the AI-based determination unit 131 first classifies the tidal combustion state on the basis of artificial intelligence using the image data of each airway type (S521), and then classifies the tidal combustion state based on the classification result (S522). In parallel with this, the image processing based determination unit 132 may classify the combustion state of each tuyere through image processing of each tandem image data, and diagnose the tuyere apparatus (S525).
- the tidal flue combustion state index generator 140 generates a flue state combustion state index based on the result of the air-fuel ratio combustion state classification (S530), and the integrated evaluation unit 160 calculates the air-
- the operating status of the blast furnace can be evaluated integrally in the circumferential direction based on the state index (S540).
- the longevity control unit 170 can control the longevity based on the collectively evaluated operating conditions (S550).
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Blast Furnaces (AREA)
- Manufacture Of Iron (AREA)
Abstract
A system and method for evaluating the operational conditions of a blast furnace is disclosed. The system for evaluating the operational conditions of a blast furnace may comprise: an image capture unit for capturing image data according to multiple tuyeres disposed in a blast furnace; an image collection unit for collecting the image data captured according to the tuyeres by the image capture unit; a tuyere combustion state determination unit for classifying, on the basis of artificial intelligence, combustion states according to the tuyeres by using the image data according to the tuyeres; a tuyere combustion state index generation unit for generating combustion state indexes according to the tuyeres by using the result of classifying the combustion states according to the tuyeres by the tuyere combustion state determination unit; and an integrated evaluation unit for generating an integrated combustion state index on the basis of the combustion state indexes according to the tuyeres.
Description
본 출원은 고로 조업 상황 평가 시스템 및 방법에 관한 것이다.The present application relates to a blast furnace operation situation evaluation system and method.
고로 조업 상황을 평가하기 위해서 고로의 풍구 등을 통해 촬영한 영상 데이터를 분석하여 노내 상황을 판단하거나, 조업 데이터를 모니터링하여 노내 상황을 판단하고자 하는 시도가 이루어지고 있다.In order to evaluate the operation status of the blast furnace, attempts have been made to analyze the image data taken through the blast furnace and determine the situation in the furnace or to monitor the operating data to determine the situation in the furnace.
그러나, 종래에는 조업자가 단순히 영상 데이터를 통해 고로 연소성 또는 노황을 정성적으로 판단하거나, 영상 데이터의 휘도 분석에 의해 노내 상황을 판단하는데 그쳤다. Conventionally, however, the operator has simply judged the blast furnaceability or the luminescence of the blast furnace through the image data or judged the furnace condition by the luminance analysis of the image data.
이와 관련하여, 일본공개특허 제2015-52148호 (공개일: 2015.03.19.)에는, 용광로의 조업 상황 판정에 기초하는 제어 방법을 개시하고 있다.In this regard, Japanese Laid-Open Patent Application No. 2015-52148 (Laid-open date: Mar. 19, 2015) discloses a control method based on determination of the operational status of a furnace.
당해 기술분야에서는 풍구 영상 데이터를 기초로 풍구 연소성 상태를 정량적으로 평가하고, 이를 기초로 고로 조업 상황을 통합적으로 평가하기 위한 방안이 요구되고 있다.In the related art, there is a need for a method for quantitatively evaluating the tuff combustion state based on the tug-guided image data and for integrally evaluating the blast furnace operation status based on the evaluation.
상기 과제를 해결하기 위해서, 본 발명의 일 실시예는 고로 조업 상황 평가 시스템을 제공한다.In order to solve the above problems, an embodiment of the present invention provides a blast furnace operation situation evaluation system.
본 발명의 일 실시예에 따른 고로 조업 상황 평가 시스템은, 고로에 구비된 복수의 풍구별 영상 데이터를 획득하는 영상 획득부; 상기 영상 획득부에 의해 획득된 풍구별 영상 데이터를 수집하는 영상 수집부; 상기 풍구별 영상 데이터를 이용하여 인공지능 기반으로 각 풍구별 연소상태를 분류하는 풍구 연소상태 판단부; 상기 풍구 연소상태 판단부에 의해 분류된 풍구별 연소상태 분류 결과를 이용하여 풍구별 연소상태 지수를 생성하는 풍구 연소상태 지수 생성부; 및 상기 풍구별 연소상태 지수를 기초로 통합 연소상태 지수를 생성하는 통합 평가부를 포함할 수 있다.According to an embodiment of the present invention, a blast furnace operation status evaluation system includes an image acquisition unit that acquires a plurality of wind turbine image data provided in a blast furnace; An image collecting unit for collecting the image data of the wind type acquired by the image acquiring unit; A tidal burning state determination unit for classifying the air-fueled combustion states based on artificial intelligence using the air-conditioned image data; A tidal burning state index generating unit for generating an air-fueled combustion state index by using the air-fueled combustion state classification result classified by the tidal burning state determining unit; And an integrated evaluation unit for generating an integrated combustion state index based on the air-fuel ratio combustion state index.
한편, 본 발명의 다른 실시예는 고로 조업 상황 평가 방법을 제공한다.Meanwhile, another embodiment of the present invention provides a method for evaluating blast furnace operation status.
본 발명의 다른 실시예에 따른 고로 조업 상황 평가 방법은, 고로에 구비된 복수의 풍구별 영상 데이터를 수집하는 단계; 상기 풍구별 영상 데이터를 이용하여 인공지능 기반으로 각 풍구별 연소상태를 분류하는 단계; 풍구별 연소상태 분류 결과를 이용하여 풍구별 연소상태 지수를 생성하는 단계; 및 상기 풍구별 연소상태 지수를 기초로 통합 연소상태 지수를 생성하는 단계를 포함할 수 있다.According to another aspect of the present invention, there is provided a method for evaluating blast furnace operation status, comprising: collecting a plurality of blast furnace image data provided in a blast furnace; Classifying the combustion state of each airway based on artificial intelligence using the airway image data; Generating an air-fueled combustion state index by using the air-fuel ratio classified state classification result; And generating an integrated combustion state index based on the air-fueled combustion state index.
덧붙여 상기한 과제의 해결수단은, 본 발명의 특징을 모두 열거한 것이 아니다. 본 발명의 다양한 특징과 그에 따른 장점과 효과는 아래의 구체적인 실시형태를 참조하여 보다 상세하게 이해될 수 있을 것이다.In addition, the means for solving the above-mentioned problems are not all enumerating the features of the present invention. The various features of the present invention and the advantages and effects thereof will be more fully understood by reference to the following specific embodiments.
본 발명의 일 실시예에 따르면, 풍구 영상 데이터를 이용하여 딥러닝 기반으로 풍구 연소상태를 분류할 수 있으며, 풍구 연소상태 분류 결과에 더하여 풍구 영상 데이터의 분석 결과 및 고로 조업 데이터의 분석 결과를 추가적으로 활용하여, 각 풍구별 풍구 연소상태 지수를 추출하고, 고로 조업 상황을 통합적으로 평가 및 제어할 수 있다.According to one embodiment of the present invention, it is possible to classify the tuyere combustion state based on the deep running based on the tuyere image data. In addition to the tuyere combustion state classification result, the analysis result of the tuyere image data and the analysis result of the blast furnace data are additionally , And it is possible to collectively evaluate and control the blast furnace operation status.
이에 따라, 고로 연소성 및 노황을 정량적으로 평가하여 안정적인 고로 조업을 가능하게 하고, 생산성을 향상시킬 수 있다. As a result, it is possible to quantitatively evaluate blast furnaceability and sulfur content to enable stable blast furnace operation and improve productivity.
도 1은 본 발명의 일 실시예에 따른 고로 조업 상황 평가 시스템의 구성도이다.1 is a configuration diagram of a blast furnace operation situation evaluation system according to an embodiment of the present invention.
도 2는 본 발명의 일 실시예에 따라 딥러닝 기반으로 1차적으로 풍구 연소상태를 분류하는 개념을 설명하는 도면이다.FIG. 2 is a view for explaining a concept of classifying a tuyere combustion state primarily on a deep learning basis according to an embodiment of the present invention.
도 3 및 도 4는 본 발명의 일 실시예에 따라 딥러닝 기반으로 1차적으로 분류한 결과를 시계열로 축적한 결과에 기초하여 풍구 연소상태 분류를 확정하는 개념을 설명하는 도면이다.FIG. 3 and FIG. 4 are views for explaining a concept of determining a tidal combustion state classification based on a result obtained by accumulating results obtained by primary classification on a deep learning basis according to an embodiment of the present invention.
도 5는 본 발명의 다른 실시예에 따른 고로 조업 상황 평가 방법의 흐름도이다.5 is a flowchart of a method for evaluating blast furnace operation status according to another embodiment of the present invention.
이하, 첨부된 도면을 참조하여 본 발명이 속하는 기술분야에서 통상의 지식을 가진 자가 본 발명을 용이하게 실시할 수 있도록 바람직한 실시예를 상세히 설명한다. 다만, 본 발명의 바람직한 실시예를 상세하게 설명함에 있어, 관련된 공지 기능 또는 구성에 대한 구체적인 설명이 본 발명의 요지를 불필요하게 흐릴 수 있다고 판단되는 경우에는 그 상세한 설명을 생략한다. 또한, 유사한 기능 및 작용을 하는 부분에 대해서는 도면 전체에 걸쳐 동일한 부호를 사용한다.Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, in order that those skilled in the art can easily carry out the present invention. In the following detailed description of the preferred embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In the drawings, like reference numerals are used throughout the drawings.
덧붙여, 명세서 전체에서, 어떤 부분이 다른 부분과 '연결'되어 있다고 할 때, 이는 '직접적으로 연결'되어 있는 경우뿐만 아니라, 그 중간에 다른 소자를 사이에 두고 '간접적으로 연결'되어 있는 경우도 포함한다. 또한, 어떤 구성요소를 '포함'한다는 것은, 특별히 반대되는 기재가 없는 한 다른 구성요소를 제외하는 것이 아니라 다른 구성요소를 더 포함할 수 있다는 것을 의미한다.In addition, in the entire specification, when a part is referred to as being 'connected' to another part, it may be referred to as 'indirectly connected' not only with 'directly connected' . Also, to "include" an element means that it may include other elements, rather than excluding other elements, unless specifically stated otherwise.
도 1은 본 발명의 일 실시예에 따른 고로 조업 상황 평가 시스템의 구성도이다.1 is a configuration diagram of a blast furnace operation situation evaluation system according to an embodiment of the present invention.
도 1을 참조하면, 본 발명의 일 실시예에 따른 고로 조업 상황 평가 시스템(100)은 영상 획득부(110), 영상 수집부(120), 풍구 연소상태 판단부(130), 풍구 연소상태 지수 생성부(140), 조업 정보 수집부(150), 통합 평가부(160) 및 노황 제어부(170)를 포함하여 구성될 수 있다. 1, a blast furnace operation status evaluation system 100 according to an exemplary embodiment of the present invention includes an image acquisition unit 110, an image collection unit 120, a tile combustion state determination unit 130, A production information collection unit 150, an integrated evaluation unit 160, and an aging control unit 170. The production unit 140, the operation information collection unit 150,
영상 획득부(110)는 고로(10)에 구비된 각 풍구(11)별 영상 데이터를 획득할 수 있다.The image acquiring unit 110 may acquire image data for each tuyere 11 provided in the blast furnace 10.
예를 들어, 영상 획득부(110)는 각각의 풍구(11)에 설치된 복수의 카메라를 포함할 수 있으며, 각각의 카메라를 통해 각 풍구별 영상 데이터를 실시간으로(예를 들어, ms 단위) 획득할 수 있다.For example, the image acquiring unit 110 may include a plurality of cameras installed in the respective tuyeres 11, and acquire the respective blind image data in real time (for example, in units of ms) can do.
영상 수집부(120)는 영상 획득부(110)에 의해 획득된 각 풍구별 영상 데이터를 수집할 수 있다.The image collecting unit 120 may collect the image data of each type obtained by the image obtaining unit 110.
예를 들어, 영상 수집부(120)는 영상 획득부(110)에 포함된 복수의 카메라로부터 각 풍구별로 실시간으로 획득된 영상 데이터를 수집할 수 있다.For example, the image capturing unit 120 may collect image data obtained in real time for each tuyere from a plurality of cameras included in the image capturing unit 110.
또한, 영상 수집부(120)는 수집된 영상 데이터를 풍구 번호 및 데이터 획득 시간 등을 포함하는 수집 환경 정보와 매핑할 수 있다.The image collection unit 120 may map the collected image data to collection environment information including a tug number and a data acquisition time.
또한, 영상 수집부(120)에 의해 매핑이 완료된 영상 데이터는 고로 조업 상황 평가 시스템(100)에 구비된 데이터 저장소(미도시)에 저장되거나, 풍구 연소상태 판단부(130)로 실시간으로 전달될 수 있다.The image data that has been mapped by the image collection unit 120 may be stored in a data storage (not shown) provided in the blast operation status evaluation system 100 or may be transmitted to the tile combustion status determination unit 130 in real time .
풍구 연소상태 판단부(130)는 영상 수집부(120)로부터 전달받은 각 풍구별 영상 데이터를 이용하여 각 풍구별 연소상태를 분류하기 위한 것으로, AI 기반 판단부(131) 및 영상처리 기반 판단부(132)를 포함하여 구성될 수 있다.The toughening combustion state determination unit 130 is for classifying the combustion state of each airway using the airway distinction image data received from the image collection unit 120. The airway state determination unit 130 includes an AI based determination unit 131, (132).
AI 기반 판단부(131)는 풍구별 영상 데이터를 이용하여 인공지능 기반으로 각 풍구별 연소상태를 분류할 수 있다. 예를 들어, AI 기반 판단부(131)는 딥러닝 기반으로 각 풍구별 연소상태를 분류할 수 있다.The AI-based determination unit 131 can classify the combustion state of each air-fuel mixture based on artificial intelligence using the image data of the air-conditioned air. For example, the AI-based determination unit 131 can classify the combustion state of each air-fuel mixture based on the deep-run basis.
일 실시예에 따르면, AI 기반 판단부(131)는 1차적으로 풍구별 영상 데이터를 이용하여 CNN(Convolutional Neural Network) 기반으로 각 풍구별 연소상태를 분류할 수 있다.According to an exemplary embodiment, the AI-based determination unit 131 may classify each air-fueled combustion state based on the CNN (Convolutional Neural Network) using the air-differentiated image data.
필요에 따라, AI 기반 판단부(131)는 1차적으로 분류된 각 풍구별 연소상태 분류 결과를 시계열로 축적한 결과를 기초로 풍구 연소상태 분류를 확정하여, 연소상태 분류의 정합성을 더욱 향상시킬 수 있다. If necessary, the AI-based determination unit 131 determines the classification of the tidal combustion state based on the result of accumulating the primary classification results of the classified combustion state classification in time series to further improve the consistency of the combustion state classification .
AI 기반 판단부(131)에 의해 풍구 연소상태를 분류 및 확정하는 개념은 도 2 내지 도 4를 참조하여 보다 구체적으로 설명한다.The concept of classifying and establishing the tuyere combustion state by the AI-based determination unit 131 will be described in more detail with reference to FIG. 2 to FIG.
도 2는 본 발명의 일 실시예에 따라 딥러닝 기반으로 1차적으로 풍구 연소상태를 분류하는 개념을 설명하는 도면이다.FIG. 2 is a view for explaining a concept of classifying a tuyere combustion state primarily on a deep learning basis according to an embodiment of the present invention.
도 2를 참조하면, AI 기반 판단부(131)는 각 풍구별로 획득된 제1 풍구 영상 내지 제N 풍구 영상 데이터(21~2N)에 대해 각각 영상 딥러닝, 예를 들어 CNN 기반으로 연소상태를 분류하여 제1 풍구 연소상태 분류 내지 제N 풍구 연소상태 분류 결과(21'~2N')를 획득할 수 있다. 여기서, N은 풍구의 개수를 의미한다.Referring to FIG. 2, the AI-based determination unit 131 performs image deep-processing (for example, based on CNN) on each of the first tugging image to the Nth tugging image data 21 to 2N obtained for each tug To obtain the first to tenth flue gas combustion state classification to Nth tuyere combustion state classification results 21 'to 2N'. Here, N means the number of tongues.
도 3 및 도 4는 본 발명의 일 실시예에 따라 딥러닝 기반으로 1차적으로 분류한 결과를 시계열로 축적한 결과에 기초하여 풍구 연소상태 분류를 확정하는 개념을 설명하는 도면이다.FIG. 3 and FIG. 4 are views for explaining a concept of determining a tidal combustion state classification based on a result obtained by accumulating results obtained by primary classification on a deep learning basis according to an embodiment of the present invention.
우선, 도 3을 참조하면, AI 기반 판단부(131)는 각 풍구별로 1차적으로 분류된 연소상태 분류 결과를 시계열로 축적한 결과, 즉, 제1 풍구 연소상태 분류(31-1, 31-2, 31-3), 제2 풍구 연소상태 분류(32-1, 32-2, 32-3) 및 제N 풍구 연소상태 분류(3N-1, 3N-2, 3N-3)를 기초로 각 풍구별로 풍구 연소상태 분류를 확정하고, 확정된 풍구 연소상태 분류 결과(31'~33')를 획득할 수 있다.Referring to FIG. 3, the AI-based determination unit 131 accumulates the combustion state classification results primarily classified for each taurine in time series, that is, the first tidal combustion state classification 31-1, 31- 2, 31-3, the second tidal combustion state classifications 32-1, 32-2, 32-3 and the Nth tidal combustion state classifications 3N-1, 3N-2, 3N-3, It is possible to determine the tidal combustion state classification by tungsten, and obtain the determined tuyball combustion state classification result (31 'to 33').
본 실시예에서는, 풍구 연소상태 분류의 확정을 위해 임의의 시간 주기(t-1 내지 t+1) 동안 1차적으로 분류된 복수의 연소상태 분류 결과를 기초로 임의의 연소상태 분류가 기 설정된 횟수이상 발생한 경우 해당 연소상태 분류로 확정할 수 있다. 이를 통해 풍구 연소상태 분류의 정확성을 보다 향상시킬 수 있다.In this embodiment, for determining the classification of the tidal burning state, any combustion state classification is performed based on a plurality of combustion state classification results primarily classified during any time period (t-1 to t + 1) Or more, it can be decided to classify the combustion state. This makes it possible to improve the accuracy of classification of the tuyere combustion state.
다음으로, 도 4를 참조하면, AI 기반 판단부(131)는 각 풍구별로 1차적으로 분류된 연소상태 분류 결과를 시계열로 축적한 결과, 즉, 제1 풍구 연소상태 분류(41-1, 41-2, 41-3), 제2 풍구 연소상태 분류(42-1, 42-2, 42-3) 및 제N 풍구 연소상태 분류(4N-1, 4N-2, 4N-3)에 대해 영상 시계열 딥러닝을 기반으로 각 풍구별로 풍구 연소상태 분류를 확정하고, 확정된 풍구 연소상태 분류 결과(41'~43')를 획득할 수 있다.Next, referring to FIG. 4, the AI-based determination unit 131 calculates the first tidal combustion state classification 41-1, 41-2, and 41-3 as a result of accumulating the combustion state classification results primarily classified for each taurine, 2, 41 - 3), the second tidal combustion state classifications 42 - 1, 42 - 2, 42 - 3 and the N th tug combustion state classifications 4 N - 1, 4 N - 2, 4 N - Based on the time-series deep running, the tuyere combustion state classification can be determined for each tuyere, and the confirmed tuyere combustion state classification results (41 'to 43') can be obtained.
예를 들어, AI 기반 판단부(131)는 임의의 시간 주기(t-1 내지 t+1) 동안 각 풍구별로 1차적으로 분류된 복수의 연소상태 분류 결과를 이용하여 RNN(Recurrent Neural Network) 또는 RCNN(Recurrent Convolutional Neural Network) 기반으로 각 풍구별 풍구 연소상태 분류를 확정할 수 있다.For example, the AI-based determination unit 131 may determine whether or not an RNN (Recurrent Neural Network) or an RNN (Recurrent Neural Network) based on a plurality of combustion state classification results that are primarily classified for each time period t- Based on the Recurrent Convolutional Neural Network (RCNN), it is possible to determine the classification of each type of tidal burning condition.
풍구의 연소상태는 시간 흐름에 따라 연속성을 가지고 변하는 것이므로 임의의 일 시점의 상태만으로 풍구의 연소상태를 판단하기에는 정확성이 떨어질 수 있다. Since the combustion state of the tuyere changes with continuity with time, it may be less accurate to judge the state of combustion of the tuyere by only a state at an arbitrary point in time.
따라서, 본 실시예에 따르면, 시간적 흐름에 따른 풍구의 연소상태의 변화를 종합적으로 고려하여 풍구의 연소상태 분류를 확정하기 위해, 영상 시계열 딥러닝을 적용함으로써 풍구 연소상태 분류의 정확성을 더욱 향상시킬 수 있다.Therefore, according to the present embodiment, in order to determine the classification of the combustion state of the tuyere by comprehensively considering the change of the combustion state of the tuyere according to the temporal flow, the accuracy of the classification of the tuyere combustion state is further improved by applying the image time- .
한편, 도 3 및 도 4에 도시된 바와 같이, 시계열로 축적한 결과를 기초로 풍구 연소상태 분류를 확정함에 있어서, 1차적으로 분류한 결과를 축적하는 시간 주기(예를 들어, t-1 내지 t+1) 및 해당 시간 주기의 개시 시점(t-1)에 따라 분류의 정확성에 영향을 미칠 수도 있다.On the other hand, as shown in FIG. 3 and FIG. 4, in establishing the classification of the tidal combustion state based on the results accumulated in time series, a time period (for example, t- t + 1) and the start time (t-1) of the corresponding time period.
일 실시예에 따르면, 초기 수행시에는, 해당 풍구 연소상태 분류가 최초로 감지된 시점을 시작으로, 사용자에 의해 설정된 시간 주기 동안 1차적으로 분류한 결과를 축적하여 풍구 연소상태 분류를 확정할 수 있다.According to an embodiment, at the time of initial execution, the tuyball combustion state classification can be established by accumulating the result of primary classification for the time period set by the user, starting from the point in time when the tuyball combustion state classification is first detected .
또한, 풍구 연소상태 분류의 확정 결과가 누적되면, 풍구 연소상태 분류가 최초로 감지된 시점으로부터 풍구 연소상태 분류가 다른 상태로 전이하는 시점까지의 경과 시간 정보에 따라 상술한 시간 주기를 조절함으로써, 분류의 정확성을 더욱 향상시킬 수 있다.If the determination result of the tidal combustion state classification accumulates, the time period described above is adjusted according to the elapsed time information from when the tidal combustion state classification is first detected to when the tidal combustion state classification is transferred to another state, Can be improved.
AI 기반 판단부(131)에 의해 분류되는 풍구의 연소상태는 예를 들어, 연소 상태 정상, 연소 상태 불량, 미분탄 미취입, 미환원 용융물 낙하(생광낙하), 코크 선회 등을 포함할 수 있다. 여기서, 미분탄 미취입은 미분탄이 취입이 되는지 여부를 판단하는 것을 의미하고, 미환원 용융물 낙하(생광낙하)는 노 상부에서 환원이 되어야 하는 원료들이 미환원되어 낙하하는 미환원 원료 용융물이 낙하하는지 여부를 판단하는 것을 의미하며, 코크 선회는 노중부에서 코크가 선회하는지 여부를 판단하는 것을 의미한다.The combustion state of the tuyeres classified by the AI-based determination unit 131 may include, for example, a normal combustion state, a poor combustion state, a non-pulverized coal injection, a unreduced melt drop (dropping light drop), a cock turning, and the like. Here, the pulverized coal is judged whether or not the pulverized coal is blown, and the unreduced melted material drop (live fall drop) means whether or not the raw material to be reduced at the furnace portion is not reduced, And the turning of the coke means judging whether or not the coke turns in the middle of the furnace.
영상처리 기반 판단부(132)는 풍구별 영상 데이터에 대한 영상처리를 통해 풍구 설비를 진단함과 함께, 풍구 연소 상태를 판단할 수 있다.The image processing base determination unit 132 can diagnose the tuyere apparatus through the image processing of the image data of the tandem image, and can determine the tuyere combustion state.
일 실시예에 따르면, 영상처리 기반 판단부(132)는 풍구별 영상 데이터에 대한 영상처리를 통해 풍구의 곡손 유무, 풍구 부착물 유무, 풍구 막힘, 랜스 밴딩 또는 버닝 등을 포함하는 풍구 설비 이상상태를 판단할 수 있다.According to one embodiment, the image processing-based determination unit 132 may perform an image processing on the image data of the background image to determine whether or not the toughening equipment abnormality including the presence or absence of the tornado, presence of the tornado, clogging of the tornado, lance banding, It can be judged.
또한, 영상처리 기반 판단부(132)는 풍구별 영상 데이터에 대한 영상처리를 통해 연소 면적 및 연소 밝기(즉, 휘도)를 추출할 수 있다.In addition, the image processing base determination unit 132 may extract the combustion area and the combustion brightness (i.e., luminance) through image processing on the image data of the air-conditioned image.
또한, 영상처리 기반 판단부(132)는 연소 상태 정상인 경우에 있어서는 풍구별 영상 데이터에 대한 영상처리를 통해 미분탄 유량을 판별할 수 있다.In addition, when the combustion state is normal, the image processing base determination unit 132 can determine the pulverized coal flow rate through image processing on the image data of the wind type.
영상처리 기반 판단부(132)에서 풍구별 영상 데이터에 대한 영상처리를 위해 통상의 기술자에게 공지된 다양한 영상처리 기법이 적용될 수 있으며, 이에 대한 구체적인 설명은 생략한다.Various image processing techniques known to those skilled in the art may be applied to the image processing based on the image data by the image processing base determination unit 132, and a detailed description thereof will be omitted.
상술한 AI 기반 판단부(131) 및 영상처리 기반 판단부(132)에 의한 판단은 병렬적으로 수행될 수 있다.The determination by the AI-based determination unit 131 and the image processing-based determination unit 132 may be performed in parallel.
풍구 연소상태 판단부(130)에 의해 분류된 각 풍구별 연소상태 분류 결과 및 풍구 설비 진단 결과는 각 풍구별 영상 데이터 및 수집 환경 정보와 함께 매핑되어 저장 및 관리될 수 있다. The results of the classification of the air condition classified by the air conditioner combustion state determiner 130 and the results of the air conditioner diagnosis may be mapped and stored and managed together with the image data and the collected environment information.
풍구 연소상태 지수 생성부(140)는 풍구 연소상태 판단부(130)에 의해 분류된 풍구별 연소상태 분류 결과를 이용하여 풍구별 연소상태 지수를 생성할 수 있다.The tidal burning state index generating unit 140 may generate the air-fueled combustion state index using the air-fuel burning state classification result classified by the tidal burning state determining unit 130.
일 실시예에 따르면, 풍구 연소상태 지수 생성부(140)에 의해 생성되는 풍구별 연소상태 지수는, 연소 상태 불량 지수, 미분탄 미취입 지수, 미환원 용융물 낙하(생광낙하) 지수, 코크 선회 지수, 연소 상태 수준 지수, 미분탄 유량 지수, 풍구 연소대(Raceway) 지수 등을 포함할 수 있다. According to one embodiment, the air-fuel ratio combustion state index generated by the air-flow combustion state index generating unit 140 may include a combustion state failure index, a pulverized coal non-spent index, a unreduced melt falling index, A combustion state level index, a pulverized coal flow rate index, a ruggedness index (Raceway) index, and the like.
예를 들어, 풍구 연소상태 지수 생성부(140)는 기 설정된 주기마다 풍구 연소상태 판단부(130)에 의한 각 풍구별 연소상태 분류 결과를 기초로 임의의 분류 결과가 발생한 횟수를 카운트 하고, 해당 주기별로 카운트된 횟수에 따라 점수화하여 관련 지수를 생성할 수 있다. For example, the tidal burning state index generator 140 counts the number of times that an arbitrary classification result is generated based on the classification result of each air-fuel mixture combustion state by the tidal combustion state determiner 130 every predetermined period, The related index can be generated by scoring according to the number of times counted for each cycle.
또한, 풍구 연소상태 지수 생성부(140)는 연소 상태 수준 지수에 대해서는 풍구 연소상태 판단부(130)에 의해 추출된 연소 면적 및 연소 밝기(즉, 휘도)에 따라 점수화하고, 기설정된 주기동안 계산된 점수를 종합하여 연소 상태 수준 지수를 생성할 수 있다. 여기서, 연소 상태 수준 지수를 생성하기 위해 사용되는 기준 정보는 관리자에 의한 입력 신호에 따라 갱신될 수 있다. 이에 따라, 갱신된 기준 정보가 실시간으로 반영되어 고로 상황이 반영된 지수 정보를 생성할 수 있다. The tidal combustion state index generator 140 scales the combustion state level index according to the combustion area and the combustion brightness (i.e., brightness) extracted by the tidal flame combustion state determiner 130, Can be combined to generate a combustion status level index. Here, the reference information used for generating the combustion state level index can be updated according to the input signal by the manager. Accordingly, the updated reference information is reflected in real time, and the index information reflecting the blast condition can be generated.
또한, 풍구 연소상태 지수 생성부(140)는 풍구 연소상태 판단부(130)에 의해 판단된 풍구 설비 진단 결과를 점수화하여 풍구 설비 이상 지수를 생성할 수 있다. 여기서, 풍구 설비 이상 지수는 풍구 곡손 지수, 풍구 부착물 지수, 풍구 막힘 지수, 랜스 손상 지수 등을 포함할 수 있다.The tidal burning state index generator 140 may generate the tidal system abnormality index by scoring the tidal system diagnosis result determined by the tidal burning state determiner 130. Here, the irregularity abnormality index may include a tug curvature index, a tug attachment index, a tug blockage index, a lance damage index, and the like.
조업 정보 수집부(150)는 고로 조업 중에 생성되는 조업 정보를 실시간으로 수집하기 위한 것이다. 여기서, 조업 정보는 예를 들어 고로 노체 온도, 압력, 냉각수 유량 등을 포함할 수 있다.The operation information collecting unit 150 is for collecting, in real time, operational information generated during blast furnace operation. Here, the operating information may include, for example, the temperature of the corona body, the pressure, the flow rate of the cooling water, and the like.
조업 정보 수집부(150)에 의해 실시간으로 수집되는 조업 정보는 상술한 풍구 연소상태 지수 생성부(140)에 의해 생성된 풍구 연소상태 지수 정보들과 매핑되어 저장 및 관리될 수 있다.The operation information collected in real time by the operation information collecting unit 150 can be mapped and stored and managed with the tidal combustion state index information generated by the tidal combustion state index generating unit 140 described above.
통합 평가부(160)는 풍구 연소상태 지수 생성부(140)에 의해 각 풍구별로 생성된 풍구 연소상태 지수 정보 및 조업 정보 수집수(150)에 의해 수집된 조업 정보들을 기초로 고로의 원주 방향으로 통합적으로 평가할 수 있다.The integrated evaluation unit 160 calculates the integrated value of the tug combustion state index information generated by the tug burning state index generating unit 140 based on the tug burning state index information generated for each tug and the operation information collected by the operation information collection number 150 It can be evaluated integrally.
일 실시예에 따르면, 통합 평가부(160)는 풍구 연소상태 지수 생성부(140)에 의해 각 풍구별로 생성된 풍구 연소상태 지수 정보들을 종합적으로 고려하여 통합 연소상태 지수를 생성할 수 있다. 예를 들어, 이러한 통합 연소상태 지수는 통합 연소 상태 불량 지수, 통합 미분탄 미취입 지수, 통합 미환원 용융물 낙하(생광낙하) 지수 등과 같이 각 풍구별로 생성된 풍구 연소상태 지수에 1:1로 매칭되는 통합 연소상태 지수를 포함할 수 있다.According to an embodiment, the integrated evaluation unit 160 may generate an integrated combustion state index by comprehensively considering the taurine combustion state index information generated for each taurine by the taurine combustion state index generating unit 140. [ For example, the integrated combustion state index is 1: 1 matched to the tidal combustion state index generated for each tuyere such as integrated combustion state failure index, integrated pulverized coal unloading index, integrated unreduced melt drop And may include an integrated combustion state index.
또한, 통합 평가부(160)는 각 풍구별로 생성된 풍구 연소대(raceway) 지수를 기초로 원주 밸런스 지수를 생성할 수 있다. Also, the integrated evaluation unit 160 may generate the columnar balance index based on the tuyere combustion race index generated for each tuyere.
또한, 통합 평가부(160)는 각 풍구별로 생성된 풍구 설비 이상 지수를 기초로 통합 풍구 설비 이상 지수를 생성할 수 있다. Also, the integrated evaluation unit 160 can generate an integrated tuff facility abnormality index based on the tuff equipment abnormality index generated for each tornado.
노황 제어부(170)는 풍구 연소상태 지수 생성부(140)에 의해 각 풍구별로 생성된 풍구 연소상태 지수 정보 또는 통합 평가부(160)에 의해 생성된 통합 연소상태 지수를 기초로 미분탄 취입 제어, N2 퍼지 제어, 고로 장입물 제어 중 적어도 하나를 실시하여 고로 노황을 제어할 수 있다.Based on the tug burning state index information generated for each tuyere by the tug burning state index generating unit 140 or the integrated combustion state index generated by the integrated evaluation unit 160, the sulfur content control unit 170 performs pulverized coal injection control, N2 Purge control, and blast furnace charge control to control blast furnace burning.
일 실시예에 따르면, 노황 제어부(170)는 임의의 풍구에 대한 미분탄 미취입 지수가 기 설정된 기준값을 초과한 경우 미분탄 취입 제어를 수행할 수 있다.According to one embodiment, the aging control unit 170 may perform the pulverized coal injection control when the pulverized coal unloading index for any tuyere exceeds a preset reference value.
또한, 노황 제어부(170)는 임의의 풍구 영역에서 생광낙하가 발생하여 미환원 용융물 낙하 지수가 기 설정된 기준값을 초과한 경우 고로 장입물 제어를 수행할 수 있다.In addition, the sulfur content control unit 170 may perform the blast furnace charge control when the drop of the live light occurs in an arbitrary tidal area and the unreduced melt drop index exceeds a preset reference value.
다른 실시예에 따르면, 노황 제어부(170)는 통합 연소상태 지수 또는 원주 밸런스 지수 정보를 기초로 복수의 풍구에 대해서 통합적으로 제어를 실시할 수도 있다. According to another embodiment, the aging control unit 170 may integrally control the plurality of tuyere based on the integrated combustion state index or the circumferential balance index information.
예를 들어, 노황 제어부(170)는, 임의의 한 방향으로만 생광낙하가 발생하는 경우, 장입물 분포를 변경하여 장입물이 떨어지는 방향을 변경하는 등과 같이 고로 장입물을 제어할 수 있다. For example, when the live fall falls only in a certain direction, the sulfur content control unit 170 can control the content of the blast furnace, such as changing the charge direction by changing the charge content distribution.
도 1을 참조하여 상술한 고로 조업 상황 평가 시스템(100)은 입력 데이터에 대한 인공지능 알고리즘의 적용 및 영상처리를 수행하고, 각종 지수의 연산이 가능한 프로세싱 장치와, 고로 제어를 수행할 수 있는 제어 장치의 결합에 의해 구현될 수 있다.The blast operation status evaluation system 100 described above with reference to FIG. 1 includes a processing device capable of performing application of an artificial intelligence algorithm to input data and image processing and capable of calculating various indices, a control device Or a combination of devices.
도 5는 본 발명의 다른 실시예에 따른 고로 조업 상황 평가 방법의 흐름도이다.5 is a flowchart of a method for evaluating blast furnace operation status according to another embodiment of the present invention.
도 5를 참조하면, 본 발명의 다른 실시예에 따른 고로 조업 상황 평가 방법에 따르면, 영상 획득부(110) 및 영상 수집부(120)에 의해, 고로에 구비된 각 풍구별로 영상 데이터를 실시간으로 수집할 수 있다(S510).Referring to FIG. 5, according to another embodiment of the present invention, the image obtaining unit 110 and the image collecting unit 120 calculate image data for each tile provided in the blast furnace in real time (S510).
이후, 풍구 연소상태 판단부(130)에 의해, 각 풍구별 영상 데이터를 이용하여 각 풍구의 연소상태를 분류할 수 있다(S520). 구체적으로, AI 기반 판단부(131)에 의해, 각 풍구별 영상 데이터를 이용하여 인공지능 기반으로 1차적으로 풍구 연소상태를 분류한 후(S521), 분류 결과를 기초로 풍구 연소상태 분류를 확정할 수 있다(S522). 또한, 이와 병렬적으로, 영상처리 기반 판단부(132)에 의해, 각 풍구별 영상 데이터에 대한 영상처리를 통해 각 풍구의 연소상태를 분류함과 더불어 풍구 설비를 진단할 수 있다(S525).Thereafter, the tuyere combustion state determination unit 130 can classify the combustion states of the respective tuyere using the image data for each airway classification (S520). Specifically, the AI-based determination unit 131 first classifies the tidal combustion state on the basis of artificial intelligence using the image data of each airway type (S521), and then classifies the tidal combustion state based on the classification result (S522). In parallel with this, the image processing based determination unit 132 may classify the combustion state of each tuyere through image processing of each tandem image data, and diagnose the tuyere apparatus (S525).
이후, 풍구 연소상태 지수 생성부(140)에 의해, 각 풍구별 연소상태 분류 결과를 기초로 풍구별 연소상태 지수를 생성하고(S530), 통합 평가부(160)에 의해, 생성된 풍구별 연소상태 지수를 기초로 고로의 조업 상황을 원주 방향으로 통합적으로 평가할 수 있다(S540).Thereafter, the tidal flue combustion state index generator 140 generates a flue state combustion state index based on the result of the air-fuel ratio combustion state classification (S530), and the integrated evaluation unit 160 calculates the air- The operating status of the blast furnace can be evaluated integrally in the circumferential direction based on the state index (S540).
이후, 노황 제어부(170)에 의해, 통합적으로 평가된 조업 상황을 기초로 노황을 제어할 수 있다(S550). Thereafter, the longevity control unit 170 can control the longevity based on the collectively evaluated operating conditions (S550).
도 5를 참조하여 상술한 각 단계의 구체적인 수행 방법은 도 1 내지 도 4를 참조하여 상술한 바와 동일하므로, 이에 대한 중복적인 설명은 생략한다.5 is the same as that described above with reference to FIG. 1 to FIG. 4, so that a detailed description thereof will be omitted.
본 발명은 전술한 실시예 및 첨부된 도면에 의해 한정되는 것이 아니다. 본 발명이 속하는 기술분야에서 통상의 지식을 가진 자에게 있어, 본 발명의 기술적 사상을 벗어나지 않는 범위 내에서 본 발명에 따른 구성요소를 치환, 변형 및 변경할 수 있다는 것이 명백할 것이다.The present invention is not limited to the above-described embodiments and the accompanying drawings. It will be apparent to those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (12)
- 고로에 구비된 복수의 풍구별 영상 데이터를 획득하는 영상 획득부;An image acquiring unit for acquiring a plurality of pieces of image data of the blind furnace;상기 영상 획득부에 의해 획득된 풍구별 영상 데이터를 수집하는 영상 수집부;An image collecting unit for collecting the image data of the wind type acquired by the image acquiring unit;상기 풍구별 영상 데이터를 이용하여 인공지능 기반으로 각 풍구별 연소상태를 분류하는 풍구 연소상태 판단부;A tidal burning state determination unit for classifying the air-fueled combustion states based on artificial intelligence using the air-conditioned image data;상기 풍구 연소상태 판단부에 의해 분류된 풍구별 연소상태 분류 결과를 이용하여 풍구별 연소상태 지수를 생성하는 풍구 연소상태 지수 생성부; 및A tidal burning state index generating unit for generating an air-fueled combustion state index by using the air-fueled combustion state classification result classified by the tidal burning state determining unit; And상기 풍구별 연소상태 지수를 기초로 통합 연소상태 지수를 생성하는 통합 평가부를 포함하는 고로 조업 상황 평가 시스템.And an integrated evaluation unit for generating an integrated combustion state index based on the air-fuel ratio combustion state index.
- 제 1 항에 있어서, 상기 풍구 연소상태 판단부는,The method according to claim 1,상기 풍구별 영상 데이터를 이용하여 딥러닝 기반으로 각 풍구별 연소상태를 분류하는 AI 기반 판단부를 포함하는 고로 조업 상황 평가 시스템.And an AI-based determination unit for classifying the combustion state of each air-fuel mixture based on the deep-run based on the air-terminated image data.
- 제 2 항에 있어서,3. The method of claim 2,상기 AI 기반 판단부는 딥러닝 기반으로 분류된 풍구별 연소상태 분류 결과를 기 설정된 시간 주기동안 시계열로 축적한 결과를 기초로 풍구별 연소상태 분류를 확정하는 고로 조업 상황 평가 시스템.Wherein the AI-based determination unit determines the classification of the blast state according to the result of accumulating the blast combustion state classification result classified by the deep running based on time series for a predetermined time period.
- 제 3 항에 있어서,The method of claim 3,상기 AI 기반 판단부는 상기 시간 주기동안 임의의 연소상태 분류가 기 설정된 횟수 이상 발생하면 연소상태 분류로 확정하는 고로 조업 상황 평가 시스템.Wherein the AI-based determination unit determines the combustion state classification when an arbitrary combustion state classification occurs more than a predetermined number of times during the time period.
- 제 3 항에 있어서,The method of claim 3,상기 AI 기반 판단부는 상기 시간 주기동안 축적된 풍구별 연소상태 분류 결과에 대해 시계열 딥러닝 기반으로 연소상태 분류를 확정하는 고로 조업 상황 평가 시스템.Wherein the AI-based determination unit determines the combustion state classification based on the time-series deep-learning based on the air-classified combustion state accumulation result accumulated during the time period.
- 제 3 항에 있어서,The method of claim 3,상기 풍구 연소상태 분류가 최초로 감지된 시점으로부터 상기 풍구 연소상태 분류가 다른 상태로 전이하는 시점까지는 경과 시간 정보에 따라 상기 시간 주기를 조절하는 고로 조업 상황 평가 시스템.Wherein the time period is adjusted according to the elapsed time information from the time when the tidal combustion state classification is first detected to the time when the tidal combustion state classification is changed to another state.
- 제 2 항에 있어서, 상기 풍구 연소상태 판단부는,3. The apparatus according to claim 2, wherein the tidal burning-상기 풍구별 영상 데이터에 대한 영상처리를 통해 풍구 설비를 진단하고 풍구 연소 상태를 판단하는 영상처리 기반 판단부를 더 포함하는 고로 조업 상황 평가 시스템.Further comprising an image processing-based determination unit for diagnosing the tuyere apparatus through image processing of the image data for the tandem image and determining the tuyere combustion state.
- 제 1 항에 있어서,The method according to claim 1,상기 연소상태 분류는 연소 상태 정상, 연소 상태 불량, 미분탄 미취입, 미환원 용융물 낙하 및 코크 선회를 포함하는 고로 조업 상황 평가 시스템. Wherein the combustion state classification includes a normal combustion state, a poor combustion state, no pulverized coal injection, a non-reduced melt drop, and a cock turning.
- 제 1 항에 있어서,The method according to claim 1,상기 풍구별 연소상태 지수 또는 상기 통합 연소상태 지수를 기초로 미분탄 취입 제어, N2 퍼지 제어, 고로 장입물 제어 중 적어도 하나를 실시하는 노황 제어부를 더 포함하는 고로 조업 상황 평가 시스템.Further comprising an aging control section for performing at least one of pulverized coal intake control, N2 purge control, and blast furnace charge control based on the air-fuel ratio combustion state index or the integrated combustion state index.
- 제 1 항에 있어서,The method according to claim 1,상기 풍구별 연소상태 지수는 연소 상태 불량 지수, 미분탄 미취입 지수, 미환원 용융물 낙하 지수, 코크 선회 지수, 연소 상태 수준 지수, 미분탄 유량 지수 및 풍구 연소대 지수 중 적어도 하나를 포함하는 고로 조업 상황 평가 시스템.The blast state index according to the present invention includes at least one of a blast condition index, a blast furnace unfilled index, a unreduced melt drop index, a coke turning index, a combustion state level index, a pulverized coal flow index, system.
- 고로에 구비된 복수의 풍구별 영상 데이터를 수집하는 단계;Collecting a plurality of pieces of image data for each of the blast furnaces;상기 풍구별 영상 데이터를 이용하여 인공지능 기반으로 각 풍구별 연소상태를 분류하는 단계;Classifying the combustion state of each airway based on artificial intelligence using the airway image data;풍구별 연소상태 분류 결과를 이용하여 풍구별 연소상태 지수를 생성하는 단계; 및Generating an air-fueled combustion state index by using the air-fuel ratio classified state classification result; And상기 풍구별 연소상태 지수를 기초로 통합 연소상태 지수를 생성하는 단계를 포함하는 고로 조업 상황 평가 방법.And generating an integrated combustion state index based on the air-fuel ratio combustion state index.
- 제 11 항에 있어서, 상기 풍구별 연소상태를 분류하는 단계는,12. The method of claim 11, wherein classifying the air-상기 풍구별 영상 데이터를 이용하여 딥러닝 기반으로 각 풍구별 연소상태를 분류하는 단계; 및Classifying each air-fueled combustion state on the basis of a deep learning using the air-terminated image data; And상기 딥러닝 기반으로 분류된 풍구별 연소상태 분류 결과를 기 설정된 시간 주기동안 시계열로 축적한 결과를 기초로 풍구별 연소상태 분류를 확정하는 단계를 포함하는 고로 조업 상황 평가 방법.And determining the classification of the blast state according to the result of accumulating the blast combustion state classification result classified by the deep running basis in time series for a predetermined time period.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201880084408.8A CN111527217B (en) | 2017-12-26 | 2018-12-24 | System and method for evaluating operating state of blast furnace |
EP18893418.6A EP3733876B1 (en) | 2017-12-26 | 2018-12-24 | System and method for evaluating operational conditions of blast furnace |
JP2020535559A JP7064598B2 (en) | 2017-12-26 | 2018-12-24 | Blast furnace operation status evaluation system and method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020170180170A KR102075223B1 (en) | 2017-12-26 | 2017-12-26 | System and method of evaluating furnace operation state |
KR10-2017-0180170 | 2017-12-26 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019132476A1 true WO2019132476A1 (en) | 2019-07-04 |
Family
ID=67067906
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2018/016580 WO2019132476A1 (en) | 2017-12-26 | 2018-12-24 | System and method for evaluating operational conditions of blast furnace |
Country Status (5)
Country | Link |
---|---|
EP (1) | EP3733876B1 (en) |
JP (1) | JP7064598B2 (en) |
KR (1) | KR102075223B1 (en) |
CN (1) | CN111527217B (en) |
WO (1) | WO2019132476A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177364A (en) * | 2021-05-21 | 2021-07-27 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere raceway |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021182041A1 (en) * | 2020-03-12 | 2021-09-16 | Jfeスチール株式会社 | Abnormality determination method for blast furnace, training method for stabilization period model, operation method for blast furnace, and abnormality determination device for blast furnace |
CN112094974A (en) * | 2020-08-21 | 2020-12-18 | 广东韶钢松山股份有限公司 | Blast furnace taphole spray control method, system and computer readable storage medium thereof |
CN112501368B (en) * | 2020-11-17 | 2022-07-08 | 中冶南方工程技术有限公司 | Blast furnace smelting method and computer equipment |
JP7380604B2 (en) * | 2021-01-12 | 2023-11-15 | Jfeスチール株式会社 | Learning model generation method, learning model generation device, blast furnace control guidance method, and hot metal manufacturing method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR950018489A (en) * | 1993-12-28 | 1995-07-22 | 조말수 | Blast Furnace Chart Temperature Prediction and Action Control Amount Guide System Using Artificial Neural Network |
JPH07305105A (en) * | 1994-05-02 | 1995-11-21 | Nippon Steel Corp | Method for evaluating raceway condition of blast furnace |
JPH07116527B2 (en) * | 1990-04-20 | 1995-12-13 | 新日本製鐵株式会社 | Sintering process controller |
KR19980050007A (en) * | 1996-12-20 | 1998-09-15 | 김종진 | Blast furnace combustion monitoring device and method |
KR19990080341A (en) * | 1998-04-15 | 1999-11-05 | 이해규 | Combustion diagnosis device of boiler burner |
JP2015048508A (en) * | 2013-09-02 | 2015-03-16 | 新日鐵住金株式会社 | Method and device for observing state of blast furnace tuyere |
JP2015052148A (en) | 2013-09-06 | 2015-03-19 | 新日鐵住金株式会社 | Method for controlling blast furnace by determining operational situation thereof |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN2589484Y (en) * | 2002-12-12 | 2003-12-03 | 武汉钢铁(集团)公司 | Blast furnace tuyere combustion status monitoring device |
CN2913380Y (en) * | 2006-06-22 | 2007-06-20 | 重庆大学 | On-line monitoring device for blast furnace tuyere operating mode |
KR101747591B1 (en) * | 2013-06-19 | 2017-06-14 | 제이에프이 스틸 가부시키가이샤 | Method for detecting abnormality at blast furnace and method for operating blast furnace |
JP5644910B1 (en) * | 2013-07-29 | 2014-12-24 | Jfeスチール株式会社 | Abnormality detection method and blast furnace operation method |
EP3029160B1 (en) * | 2013-07-29 | 2018-04-25 | JFE Steel Corporation | Abnormality detection method and blast-furnace operation method |
JP6350159B2 (en) * | 2014-09-17 | 2018-07-04 | 新日鐵住金株式会社 | Blast furnace tuyere state observation method and blast furnace tuyere state observation apparatus |
CN106191350B (en) * | 2016-08-30 | 2018-04-17 | 武汉钢铁有限公司 | Bottom house air port working condition appraisal procedure based on fixed point radar |
-
2017
- 2017-12-26 KR KR1020170180170A patent/KR102075223B1/en active IP Right Grant
-
2018
- 2018-12-24 WO PCT/KR2018/016580 patent/WO2019132476A1/en unknown
- 2018-12-24 EP EP18893418.6A patent/EP3733876B1/en active Active
- 2018-12-24 CN CN201880084408.8A patent/CN111527217B/en active Active
- 2018-12-24 JP JP2020535559A patent/JP7064598B2/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07116527B2 (en) * | 1990-04-20 | 1995-12-13 | 新日本製鐵株式会社 | Sintering process controller |
KR950018489A (en) * | 1993-12-28 | 1995-07-22 | 조말수 | Blast Furnace Chart Temperature Prediction and Action Control Amount Guide System Using Artificial Neural Network |
JPH07305105A (en) * | 1994-05-02 | 1995-11-21 | Nippon Steel Corp | Method for evaluating raceway condition of blast furnace |
KR19980050007A (en) * | 1996-12-20 | 1998-09-15 | 김종진 | Blast furnace combustion monitoring device and method |
KR19990080341A (en) * | 1998-04-15 | 1999-11-05 | 이해규 | Combustion diagnosis device of boiler burner |
JP2015048508A (en) * | 2013-09-02 | 2015-03-16 | 新日鐵住金株式会社 | Method and device for observing state of blast furnace tuyere |
JP2015052148A (en) | 2013-09-06 | 2015-03-19 | 新日鐵住金株式会社 | Method for controlling blast furnace by determining operational situation thereof |
Non-Patent Citations (1)
Title |
---|
See also references of EP3733876A4 |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177364A (en) * | 2021-05-21 | 2021-07-27 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere raceway |
WO2022242109A1 (en) * | 2021-05-21 | 2022-11-24 | 东北大学 | Soft measurement modeling method for temperature of tuyere raceway of blast furnace |
CN113177364B (en) * | 2021-05-21 | 2023-07-14 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere convolution zone |
Also Published As
Publication number | Publication date |
---|---|
JP2021509440A (en) | 2021-03-25 |
EP3733876B1 (en) | 2023-06-07 |
KR102075223B1 (en) | 2020-02-07 |
EP3733876A4 (en) | 2021-01-20 |
CN111527217B (en) | 2022-08-16 |
KR20190078319A (en) | 2019-07-04 |
JP7064598B2 (en) | 2022-05-10 |
EP3733876A1 (en) | 2020-11-04 |
CN111527217A (en) | 2020-08-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019132476A1 (en) | System and method for evaluating operational conditions of blast furnace | |
WO2020141683A1 (en) | Image-based traffic information providing system and method thereof | |
WO2015126029A1 (en) | Power entry point energy measuring apparatus and energy measurement information labeling system using same | |
WO2013051828A2 (en) | Battery management system and battery management method | |
WO2019117337A1 (en) | Device and method for classifying appliances by using electric power data analysis | |
WO2014061904A1 (en) | System and method for managing electric vehicle | |
CN105308191A (en) | Method for detecting abnormality in blast furnace, and method for operating blast furnace | |
WO2020004749A1 (en) | Apparatus and method for allowing equipment to learn by using video file | |
WO2021194008A1 (en) | Cloud service method and device for detecting and managing abnormal symptom of sow | |
WO2022114654A1 (en) | Process monitoring system and method | |
WO2022149890A1 (en) | Battery diagnosis device, battery system, and battery diagnosis method | |
WO2023128669A1 (en) | Partial discharge monitoring system and partial discharge monitoring method | |
WO2014030833A1 (en) | Method and device for predicting number of suicides using social information | |
CN112633049A (en) | System for monitoring working state of blast furnace tuyere | |
WO2019117493A1 (en) | Blast control device of blast furnace and method therefor | |
WO2019098484A1 (en) | Blast control device for blast furnace and method therefor | |
WO2021125662A1 (en) | Atmospheric condition provision system using atmospheric pollution source measurement module | |
WO2023158068A1 (en) | Learning system and method for improving object detection rate | |
WO2013191457A1 (en) | Client monitoring system | |
WO2019103208A1 (en) | Device for analyzing multiple distributed image data | |
CN206710860U (en) | Common control equipment | |
WO2021261905A1 (en) | Apparatus and method for image analysis-based work operation recognition and production measurement | |
CN115002448A (en) | Video image quality diagnosis method and system applied to security monitoring | |
WO2019112207A1 (en) | Apparatus and method for controlling amount of pulverized coal injection | |
WO2014193057A1 (en) | Method for determining problematic equipment using defect map of faulty samples and apparatus for same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18893418 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2020535559 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2018893418 Country of ref document: EP Effective date: 20200727 |