WO2020204043A1 - 高炉の異常判定装置、高炉の異常判定方法、及び高炉の操業方法 - Google Patents
高炉の異常判定装置、高炉の異常判定方法、及び高炉の操業方法 Download PDFInfo
- Publication number
- WO2020204043A1 WO2020204043A1 PCT/JP2020/014881 JP2020014881W WO2020204043A1 WO 2020204043 A1 WO2020204043 A1 WO 2020204043A1 JP 2020014881 W JP2020014881 W JP 2020014881W WO 2020204043 A1 WO2020204043 A1 WO 2020204043A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- abnormality
- blast furnace
- index
- sensor
- ventilation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- 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
-
- 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 specially adapted for furnaces of these types
- F27B1/28—Arrangements of monitoring devices, of indicators, of alarm devices
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B5/00—Making pig-iron in the blast furnace
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/007—Controlling or regulating of the top pressure
-
- 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
- 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—Arrangement of monitoring devices; Arrangement of safety devices
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0297—Reconfiguration of monitoring system, e.g. use of virtual sensors; change monitoring method as a response to monitoring results
-
- 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—Arrangement of monitoring devices; Arrangement of safety devices
- F27D2021/0007—Monitoring the pressure
Definitions
- the present invention relates to a blast furnace abnormality determination device, a blast furnace abnormality determination method, and a blast furnace operation method.
- a ventilation index indicating a ventilation state in a furnace a ventilation resistance or the like calculated from a difference value between a furnace top pressure and a ventilation pressure has been used. Then, a threshold value for determining the deterioration of ventilation is set for each ventilation index, and when the ventilation index exceeds the threshold value, an abnormality determination as deterioration of ventilation is performed.
- Patent Document 1 describes a method of calculating an abnormality index by using a shaft pressure as an input and performing an abnormality determination by using a statistical method such as principal component analysis.
- the ratio of the Q statistic calculated based on the stability limit data and the Q statistic calculated based on the operation data of the abnormality detection target is taken.
- the abnormality index is converted into an abnormality index, and a threshold value is set for the abnormality index to determine the abnormality.
- the abnormality determination of the sensor is performed by using the output value of the peripheral sensor.
- the range of output values for regarding the sensor as normal is set widely, and as a result, the abnormality can be detected only when the degree of abnormality of the sensor becomes large, and the abnormality detection is delayed.
- the shaft pressure sensor used in the blast furnace is an important sensor for controlling ventilation, etc., but it often becomes abnormal due to clogging of blast furnace dust, etc., so early abnormality detection is required. Be done.
- the output value of the sensor suddenly changes from a normal value to a large value and becomes abnormal, it is possible to set a threshold value to determine the abnormality, but an abnormality that gradually deviates from the normal value is a threshold value. It is difficult to detect with.
- abnormality determination is performed based on the average value or dispersion value of the output values of the sensors installed in the vicinity of the sensor to be determined, but the output of the sensor group observed in the normal state is performed.
- the synchrony of the values is not considered, and there is a risk that the detection of an abnormality that can be detected early is delayed by considering the asynchronousness of the output values of a plurality of sensor groups.
- the present invention has been made in view of the above problems, and an object of the present invention is a blast furnace abnormality determination device capable of distinguishing between an operation abnormality state and a sensor abnormality state, a blast furnace abnormality determination method, and a blast furnace. Is to provide a method of operation.
- the abnormality determination device for a blast furnace was calculated by an abnormality index calculating means for calculating an abnormality index indicating an abnormality degree of a blast furnace, a ventilation index calculating means for calculating a ventilation index of a blast furnace, and the abnormality index calculating means.
- the blast furnace is provided with a determination means for determining an abnormal state of the blast furnace by using the abnormality index and the ventilation index calculated by the ventilation index calculation means.
- the abnormality index calculating means may calculate the abnormality index using the output value of the shaft pressure sensor group installed around the furnace body of the blast furnace.
- the abnormality index should be a Q statistic based on principal component analysis.
- the contribution is evaluated with respect to the Q statistic, and if the ventilation index does not exceed a predetermined threshold value and the Q statistic exceeds a predetermined threshold value, a sensor abnormality that identifies an abnormal sensor based on the contribution degree is specified. It is advisable to provide specific means.
- An abnormality sensor removing means that continues the abnormality determination by newly calculating the Q statistic by removing the signal value of the sensor identified as the sensor that becomes abnormal by the sensor abnormality identifying means from the calculation of the Q statistic. You should prepare.
- the ventilation index is calculated by the following formula (1), and in the parameter X in the formula (1), the gas generation amount in the blast furnace and the ventilation index are substantially linear even if the gas generation amount in the blast furnace changes. It is good that the numerical value is adjusted so that it can be expressed by the relationship.
- the blast furnace abnormality determination method was calculated in the abnormality index calculation step for calculating the abnormality index indicating the degree of abnormality of the blast furnace, the ventilation index calculation step for calculating the ventilation index of the blast furnace, and the abnormality index calculation step. It includes a determination step of determining an abnormal state of the blast furnace using the abnormality index and the ventilation index calculated in the ventilation index calculation step.
- the method for operating the blast furnace according to the present invention includes a step of operating the blast furnace while determining the abnormal state of the blast furnace using the abnormality determination device for the blast furnace according to the present invention.
- the blast furnace abnormality determination device According to the blast furnace abnormality determination device, the blast furnace abnormality determination method, and the blast furnace operation method according to the present invention, it is possible to distinguish between the operation abnormality state and the sensor abnormality state.
- FIG. 1 is a block diagram showing a configuration of an abnormality determination device for a blast furnace according to the first embodiment of the present invention.
- FIG. 2 is a diagram showing an example of changes over time in shaft pressure, ventilation index, and Q statistic.
- FIG. 3 is a block diagram showing a configuration of an abnormality determination device for a blast furnace sensor group according to a second embodiment of the present invention.
- FIG. 4 is a diagram showing an example of changes in shaft pressure over time.
- FIG. 5 is a diagram showing an example of time-series changes in contribution.
- FIG. 1 is a block diagram showing a configuration of an abnormality determination device for a blast furnace according to the first embodiment of the present invention.
- the blast furnace abnormality determination device 1 according to the first embodiment of the present invention is an apparatus for determining the operation of the blast furnace and the abnormality of the sensor 2 used in the blast furnace, and is a data collection device 11.
- the abnormality index calculation device 12, the abnormality index determination device 13, the ventilation index determination device 14, the abnormality determination device 15, and the display device 16 are provided as main components.
- the data collecting device 11 collects and stores the output value of the sensor 2.
- a shaft pressure sensor group can be exemplified.
- the shaft pressure sensor group is installed at a plurality of locations around the furnace body of the blast furnace in the height direction and the circumferential direction.
- the abnormality index calculation device 12 calculates the abnormality index of the blast furnace using the output value of the sensor 2 stored in the data collection device 11.
- the calculation method of the abnormality index is not limited, and any method may be used as long as it is a method of unifying a plurality of input data and using it as an abnormality index. That is, for example, the Q statistic method as described in Patent Document 1 may be used, or the method may be one index by independent component analysis or one index using a machine learning method.
- the index calculated from the operation data group used as the stability limit value is divided by the index calculated from the data to be judged to be an abnormal index, that is, the standardization process of the abnormal index such that the stability limit becomes 1 is added. May be good.
- the abnormality index determination device 13 determines whether or not the abnormality index calculated by the abnormality index calculation device 12 exceeds a predetermined threshold value determined in advance based on the stability limit data, thereby determining the presence or absence of an abnormality in the blast furnace. judge. For example, when a standardization process for an abnormality index such as 1 at the stability limit is added, the predetermined threshold value may be set to 1.
- the ventilation index determination device 14 calculates the ventilation index of the blast furnace using the output value of the sensor 2 stored in the data collection device 11, and determines whether or not the calculated ventilation index exceeds a predetermined threshold value. By doing so, it is determined whether or not there is an abnormality in the ventilation state of the blast furnace.
- the aeration index for example, the aeration index shown in the following mathematical formula (1) can be used.
- the furnace pressure value A is the output value of the pressure gauge A installed in the blast furnace
- the furnace pressure value B is the downstream side in the gas flow direction (upper side of the blast furnace) from the pressure gauge A in the blast furnace. ) Is the output value of the pressure gauge B installed.
- X is a numerical value, and is a parameter for adjusting so that the gas generation amount in the blast furnace and the ventilation index can be expressed as linearly as possible even if the gas generation amount in the blast furnace changes.
- a predetermined threshold value lower than the threshold value for determining an abnormality of the ventilation index used in normal operation.
- the abnormality determination device 15 makes a final and comprehensive abnormality determination of the blast furnace based on the abnormality determination results of the abnormality index determination device 13 and the ventilation index determination device 14.
- the display device 16 displays and outputs the determination result of the abnormality determination device 15. In particular, when both the abnormality index and the ventilation index exceed the threshold value and are determined to be abnormal, the display device 16 notifies the operator of the abnormality by displaying and outputting to that effect, and prompts the operator to take measures such as self-check.
- the display device 16 notifies the operator of the "sensor abnormality" and the sensor 2 has an abnormality. Prompt for inspection.
- a threshold value is separately provided for each unit regardless of the present invention. Therefore, when only the ventilation index exceeds the threshold value, the conventional device notifies the operator of an abnormal state.
- the blast furnace abnormality determination device 1 calculates an abnormality index indicating the degree of abnormality of the blast furnace, and a ventilation index calculation means for calculating the ventilation index of the blast furnace. Since the abnormal state of the blast furnace is determined by using the abnormality index and the ventilation index, it is possible to distinguish between the abnormal state of the operation and the abnormal state of the sensor. In addition, this makes it possible to reduce erroneous detection of an abnormal state of operation due to a sensor abnormality. Furthermore, by looking at the abnormality index at the same time as the ventilation resistance, the accuracy of determining that the blast furnace is in an abnormal state is increased. In addition, these make it possible to take appropriate measures such as wind reduction in an abnormal state at an early stage, and prevent troubles due to an abnormality and avoid a large reduction in production.
- the Q statistic was calculated as an abnormality index. Since the calculation of the Q statistic itself is based on principal component analysis and is general, it is not particularly presented here (see Non-Patent Document 1).
- the plurality of data used (input) in this embodiment are the output values of the shaft pressure sensor group of the blast furnace.
- the Q statistic and the principal component analysis that is the basis of the Q statistic will be described in a form suitable for the blast furnace process.
- Principal component analysis is a small number of variables that well reflect the characteristics of the original data group while minimizing the loss of information in the original data group for multiple (multiple dimension) data groups to be synchronized. Refers to mathematical processing that replaces (lowers the dimension) with.
- the component of the movement in which each shaft pressure is synchronized during stable operation of the blast furnace appears in the first principal component value, which has the largest dispersion in the principal component analysis.
- the Q statistic was calculated by considering only the first principal component as the principal component.
- FIG. 2A shows the time series data of the shaft pressure used for calculating the Q statistic
- FIG. 2C shows the calculated Q statistic.
- the Q statistic is small when the shaft pressure fluctuates synchronously in normal times.
- FIG. 3 is a block diagram showing a configuration of an abnormality determination device for a blast furnace sensor group according to a second embodiment of the present invention.
- the abnormality determination device 20 of the blast furnace sensor group according to the second embodiment of the present invention is an apparatus for determining an abnormality of the sensor group 3 used in the blast furnace operation, and is a data collection device 21.
- An abnormality index calculation device 22, an abnormality determination device 23, and a display device 24 are provided as main components.
- the data collecting device 21 collects and stores the output value of the sensor group 3.
- a shaft pressure sensor group can be exemplified as a sensor group for collecting output values.
- the shaft pressure sensor group is installed at a plurality of locations around the furnace body of the blast furnace in the height direction and the circumferential direction.
- the abnormality index calculation device 22 calculates an abnormality index for each sensor by executing principal component analysis (PCA) on the output value of the sensor group 3 stored in the data collection device 21. .. Specifically, the abnormality index calculation device 22 calculates the abnormality index for each sensor by using the Q statistic, which is one index of the MSPC (Multivariate Statistical Process Control) method. As shown in Non-Patent Document 1, the method for calculating the Q statistic itself is a well-known method and will not be described in detail, but the Q statistic can be calculated by the following mathematical formula (2). In the formula (2), N indicates the total number of sensors.
- the Q statistic is an index showing the degree of deviation from the correlation between variables held by the PCA model creation data, and by monitoring this index, abnormalities of each input variable can be detected.
- Each element of the Q statistic represents the contribution of the output value of each sensor to the Q statistic, and each input variable (output value of each sensor) and each input indicate which sensor's output value affected the detected abnormality. It can be obtained from the difference (contribution) of the estimated values of variables.
- the degree of contribution can be calculated by the following mathematical formula (3).
- this quantified contribution is used.
- the abnormality determination threshold value of each sensor can be made common (to the same value). By setting a predetermined threshold value for determining an abnormality in the contribution of each sensor in this way, it is possible to present a sensor that is a candidate for an abnormality.
- the loss of the amount of information of the original data group is minimized, and the characteristics of the original data group are well reflected. It means a mathematical process that replaces (lowers the dimension) with a variable of.
- shaft pressure data of a blast furnace about 30 shaft pressure sensors are installed for one blast furnace, but applying principal component analysis to this, the characteristics of the data group of 30 points are good. If it is replaced by several variables (principal component values) reflected in, the furnace can be monitored by monitoring a small number of variables generated by principal component analysis without observing all of these 30 data groups. It shows that the state inside can be estimated more easily.
- synchronization means that the behavior of variables in operation is coordinated with respect to the time transition or operation action in the process.
- the component of the synchronized movement of each shaft pressure during stable operation of the blast furnace appears in the first principal component value, which has the largest variance in the principal component analysis.
- movements other than asynchronous movements appear. If there is a sensor with a higher degree of asynchrony than other sensors, that sensor can be regarded as abnormal.
- the difference between the above-mentioned input variable and the estimated value of the first principal component of each input variable is calculated as a Q statistic, and the absolute value of each element is calculated. Anomaly judgment is performed with (contribution) as the asynchronous degree.
- the abnormality determination device 23 determines the abnormality of the sensor based on the abnormality index calculated by the abnormality index calculation device 22. Specifically, the abnormality determination device 23 determines the abnormality of each sensor by comparing the degree of contribution of the measured value of each sensor calculated by the abnormality index calculation device 22 to the Q statistic and the magnitude relationship between the predetermined threshold value. I do.
- the predetermined threshold value may be determined from the case of an abnormality in the output value of the sensor in the past.
- the display device 24 displays and outputs the determination result of the abnormality determination device 23.
- the abnormality determination result of the sensor is displayed and output to the display device 24, so that the operator can be contacted and a response such as a sensor check can be urged.
- the abnormality determination device 20 of the blast furnace sensor group performs principal component analysis on the output value of the sensor group 3, and Q statistic and contribution degree. Is calculated, and the magnitude relationship between the calculated data and the predetermined threshold value is compared to determine the abnormality of the sensor group 3. Therefore, even if the abnormality is difficult to detect based on the threshold value of the output value of the sensor group 3, it is early. It is possible for the maintenance staff to perform inspection / repair based on the detection. Further, by completing the repair of the sensor at an early stage, it becomes possible to determine the abnormal state of the operation in the state where a large number of sensors are used, and the accuracy of the abnormality determination is further increased.
- an abnormal sensor is detected by the first embodiment, and the sensor determined to be abnormal is excluded from the calculation of the Q statistic, and the second embodiment is performed. Calculate the Q statistic again depending on the form. Subsequently, the recalculated Q statistic and the aeration index can be simultaneously considered to determine the abnormal state of the blast furnace. As a result, it is possible to remove the sensor having a possibility of abnormality and continue the abnormality determination based on the Q statistic and the ventilation index based on the other sensor values, and improve the accuracy of the abnormality determination. This process can be repeated even for the sensor detected as abnormal.
- the sensor determined to be sensor abnormality can be presented to the maintenance staff for inspection and restoration.
- Example In this embodiment, first, standardization processing (mean value 0, variance 1) was performed for each data of the shaft pressure sensor group. When standardizing, time series data was used when normal operation and sensors were normal. Next, the Q statistic was calculated for the standardized data, and the difference (contribution) between each input variable and the estimated value of each input variable was obtained. Next, an abnormality was determined based on the threshold value for the contribution.
- FIG. 2 shows the shaft pressure used in the calculation.
- the example shown in FIG. 4 is an example in which the output value of one sensor slowly deviates from the movement of the output value of the other sensor and then returns.
- FIG. 5 shows the difference (contribution degree) between each input variable in the same time zone and the estimated value of the first principal component of each input variable. As shown in FIG.
- the contribution of the sensor exhibiting abnormal behavior is large. As shown in FIG. 5, 0 is the smallest contribution, and it is considered that the contribution increases as the absolute value increases.
- An abnormality was determined by setting a threshold value for this contribution. In this case, as shown in FIG. 3, ⁇ 10 was set as the threshold value for determining an abnormality, and the case where it exceeded +10 and the case where it was below -10 were regarded as abnormal.
- the threshold value for this abnormality determination was determined based on the calculation result of this method using the data in which the actual sensor abnormality occurred. By using this threshold value, it was possible to determine that one sensor was abnormal in the data used this time.
- a blast furnace abnormality determination device capable of distinguishing between an operation abnormality state and a sensor abnormality state, a blast furnace abnormality determination method, and a blast furnace operation method.
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Organic Chemistry (AREA)
- Metallurgy (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Waste-Gas Treatment And Other Accessory Devices For Furnaces (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Blast Furnaces (AREA)
Abstract
Description
図1は、本発明の第1の実施形態である高炉の異常判定装置の構成を示すブロック図である。図1に示すように、本発明の第1の実施形態である高炉の異常判定装置1は、高炉の操業及び高炉で使用されているセンサ2の異常を判定する装置であり、データ収集装置11、異常指標計算装置12、異常指標判定装置13、通気指標判定装置14、異常判定装置15、及び表示装置16を主な構成要素として備えている。
本実施例では、異常指標としてQ統計量を算出した。Q統計量の算出自体は主成分分析によるものであり、一般的なものであるため、ここでは特に提示しない(非特許文献1を参照のこと)。本実施例では使用(入力と)する複数データは、高炉のシャフト圧力センサ群の出力値とする。まず、高炉プロセスに即した形で、Q統計量及びその元となる主成分分析ついて説明する。主成分分析は、同期する複数個(複数次元)のデータ群について、元のデータ群の持つ情報量の損失をできる限り小さくしつつ、元のデータ群の持つ特徴が良く反映された少数の変数へと置換(低次元化)する数学的処理を指す。これは、高炉のシャフト圧力データの場合であれば、高炉一基に対しシャフト圧力計は約30点設置されているが、これに主成分分析を適用し30点のデータ群の特徴を良好に反映する数個の変数(主成分値)に仮に置き換えられたとすれば、これら30点のデータ群全てを観察することなく、主成分分析により生成された少数の変数を監視することにより、炉内の状態をより簡便に推定可能であることを表している。
図3は、本発明の第2の実施形態である高炉センサ群の異常判定装置の構成を示すブロック図である。図3に示すように、本発明の第2の実施形態である高炉センサ群の異常判定装置20は、高炉操業において使用されるセンサ群3の異常を判定する装置であり、データ収集装置21、異常指標計算装置22、異常判定装置23、及び表示装置24を主な構成要素として備えている。
本実施例では、まず、シャフト圧力センサ群のデータ毎に基準化処理(平均値0、分散1)を行った。基準化を行う際には、正常操業及びセンサが正常の際の時系列データを用いた。次に、基準化したデータに対してQ統計量を算出し、各入力変数と各入力変数の推定値の差(寄与度)を求めた。次に、寄与度に対して閾値によって異常判定を行った。図2に計算に用いたシャフト圧力を示す。図4に示す例は、1つのセンサの出力値が他のセンサの出力値の動きからゆっくりと乖離して、その後戻っている例である。図5に同時間帯の各入力変数と各入力変数の第1主成分の推定値との差(寄与度)を示す。図5に示すように、異常な挙動を示すセンサの寄与度が大きくなっている。なお、寄与度は図5に示すように0が一番小さく、その絶対値が大きくなるに従って寄与度が大きくなると見なす。この寄与度に対して、閾値を設けて異常判定を行った。今回のケースでは図3に示すように±10を異常判定の閾値として、+10を上回った場合及び-10を下回った場合を異常とした。この異常判定の閾値は実際のセンサ異常が発生しているデータを用いた本手法の計算結果に基づいて決定した。この閾値を用いることで今回用いたデータにおいて、途中から一つのセンサが異常であると判定できた。
2 センサ
3 センサ群
11,21 データ収集装置
12,22 異常指標計算装置
13 異常指標判定装置
14 通気指標判定装置
15,23 異常判定装置
16,24 表示装置
20 高炉センサ群の異常判定装置
Claims (8)
- 高炉の異常度を示す異常指標を算出する異常指標算出手段と、
高炉の通気指標を算出する通気指標算出手段と、
前記異常指標算出手段によって算出された異常指標と前記通気指標算出手段によって算出された通気指標とを用いて高炉の異常状態を判定する判定手段と、
を備える、高炉の異常判定装置。 - 前記異常指標算出手段は、高炉の炉体周りに設置されたシャフト圧力センサ群の出力値を用いて前記異常指標を算出する、請求項1に記載の高炉の異常判定装置。
- 前記異常指標は、主成分分析に基づくQ統計量である、請求項1又は2に記載の高炉の異常判定装置。
- 前記Q統計量について寄与度を評価し、前記通気指標が所定の閾値を超えず、且つ、前記Q統計量が所定の閾値を超える場合、異常となるセンサを前記寄与度に基づき特定するセンサ異常特定手段を備える、請求項3に記載の高炉の異常判定装置。
- 前記Q統計量の計算から前記センサ異常特定手段によって異常となるセンサとして特定されたセンサの信号値を除いて、新たにQ統計量を算出することにより、異常判定を継続する異常センサ除去手段を備える、請求項4に記載の高炉の異常判定装置。
- 高炉の異常度を示す異常指標を算出する異常指標算出ステップと、
高炉の通気指標を算出する通気指標算出ステップと、
前記異常指標算出ステップにおいて算出された異常指標と前記通気指標算出ステップにおいて算出された通気指標とを用いて高炉の異常状態を判定する判定ステップと、
を含む、高炉の異常判定方法。 - 請求項1~6のうち、いずれか1項に記載の高炉の異常判定装置を用いて高炉の異常状態を判定しながら高炉を操業するステップを含む、高炉の操業方法。
Priority Applications (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2020546515A JP6825753B1 (ja) | 2019-04-03 | 2020-03-31 | 高炉の異常判定装置、高炉の異常判定方法、及び高炉の操業方法 |
| CN202080023159.9A CN113614253A (zh) | 2019-04-03 | 2020-03-31 | 高炉异常判定装置、高炉异常判定方法以及高炉操作方法 |
| BR112021018897-5A BR112021018897B1 (pt) | 2019-04-03 | 2020-03-31 | Aparelho de determinação de falha de alto-forno, método para determinar uma falha em um alto-forno e método para operar um alto-forno |
| KR1020217029637A KR102574567B1 (ko) | 2019-04-03 | 2020-03-31 | 고로의 이상 판정 장치, 고로의 이상 판정 방법, 및 고로의 조업 방법 |
| US17/440,913 US12320590B2 (en) | 2019-04-03 | 2020-03-31 | Blast furnace fault determination apparatus, method for determining fault in blast furnace, and method for operating blast furnace |
| EP20783058.9A EP3950966B1 (en) | 2019-04-03 | 2020-03-31 | Blast furnace abnormality assessment device, blast furnace abnormality assessment method, and blast furnace operation method |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019-071016 | 2019-04-03 | ||
| JP2019-071012 | 2019-04-03 | ||
| JP2019071016 | 2019-04-03 | ||
| JP2019071012 | 2019-04-03 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020204043A1 true WO2020204043A1 (ja) | 2020-10-08 |
Family
ID=72668446
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2020/014881 Ceased WO2020204043A1 (ja) | 2019-04-03 | 2020-03-31 | 高炉の異常判定装置、高炉の異常判定方法、及び高炉の操業方法 |
Country Status (7)
| Country | Link |
|---|---|
| US (1) | US12320590B2 (ja) |
| EP (1) | EP3950966B1 (ja) |
| JP (1) | JP6825753B1 (ja) |
| KR (1) | KR102574567B1 (ja) |
| CN (1) | CN113614253A (ja) |
| TW (1) | TWI745912B (ja) |
| WO (1) | WO2020204043A1 (ja) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112961949A (zh) * | 2021-03-12 | 2021-06-15 | 鞍钢股份有限公司 | 一种快速判断高炉出现管道行程的方法 |
| KR20230098293A (ko) * | 2020-12-08 | 2023-07-03 | 제이에프이 스틸 가부시키가이샤 | 이상 판정 모델 생성 장치, 이상 판정 장치, 이상 판정 모델 생성 방법 및 이상 판정 방법 |
| JPWO2023190234A1 (ja) * | 2022-03-29 | 2023-10-05 | ||
| JP2024105129A (ja) * | 2023-01-25 | 2024-08-06 | トヨタ自動車株式会社 | 情報処理装置、情報処理方法、及び情報処理プログラム |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113614253A (zh) | 2019-04-03 | 2021-11-05 | 杰富意钢铁株式会社 | 高炉异常判定装置、高炉异常判定方法以及高炉操作方法 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017088950A (ja) * | 2015-11-09 | 2017-05-25 | Jfeスチール株式会社 | 高炉への原料装入方法 |
| JP2017128805A (ja) | 2016-01-19 | 2017-07-27 | Jfeスチール株式会社 | 高炉の操業方法 |
| JP2017190482A (ja) | 2016-04-12 | 2017-10-19 | 株式会社神戸製鋼所 | 高炉のセンサ故障検知システム及び異常状況の予測システム |
| JP2018165399A (ja) * | 2017-03-28 | 2018-10-25 | Jfeスチール株式会社 | 高炉炉況状態判定装置、高炉の操業方法、及び、高炉炉況状態判定方法 |
| JP2018204076A (ja) * | 2017-06-06 | 2018-12-27 | Jfeスチール株式会社 | 高炉吹込み還元材の燃焼率推定方法、高炉操業方法及び送風羽口 |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU992590A1 (ru) | 1981-08-17 | 1983-01-30 | Всесоюзный Научно-Исследовательский И Проектный Институт По Очистке Технологических Газов,Сточных Вод И Использованию Вторичных Энергоресурсов Предприятий Черной Металлургии | Устройство дл обнаружени неисправности охлаждаемых элементов металлургических агрегатов |
| SU1447859A1 (ru) | 1987-02-23 | 1988-12-30 | Кузнецкий металлургический комбинат им.В.И.Ленина | Способ контрол целостности элементов охлаждени доменной печи |
| JP2009054843A (ja) * | 2007-08-28 | 2009-03-12 | Omron Corp | プロセス異常検出装置および方法並びにプログラム |
| CN101886152A (zh) | 2010-06-02 | 2010-11-17 | 河北省首钢迁安钢铁有限责任公司 | 高炉炉缸三维非稳态监测和异常诊断及维护系统 |
| JP5699832B2 (ja) | 2011-07-08 | 2015-04-15 | Jfeスチール株式会社 | 高炉操業方法 |
| TWI435936B (zh) | 2011-11-10 | 2014-05-01 | China Steel Corp | 高爐管道流現象之預測方法 |
| CN103361454B (zh) | 2012-03-30 | 2015-03-11 | 鞍钢股份有限公司 | 基于数据过滤的高炉悬料判断方法 |
| EP2851437B1 (en) | 2012-05-18 | 2018-10-03 | JFE Steel Corporation | Method for loading raw material into blast furnace |
| US9799110B2 (en) | 2013-07-29 | 2017-10-24 | Jfe Steel Corporation | Abnormality detection method and blast furnace operation method |
| LU92351B1 (en) | 2014-01-09 | 2015-07-10 | Tmt Tapping Measuring Technology Sarl | Method and probe for determining the material distribution in a blast furnace |
| JP6003909B2 (ja) | 2014-01-28 | 2016-10-05 | Jfeスチール株式会社 | 高炉通気性予測装置及び高炉通気性予測方法 |
| JP6617619B2 (ja) | 2016-03-14 | 2019-12-11 | 日本製鉄株式会社 | 高炉の操業方法 |
| JP6690081B2 (ja) | 2016-07-14 | 2020-04-28 | 株式会社神戸製鋼所 | 操業状況評価システム |
| JP6686947B2 (ja) | 2017-03-23 | 2020-04-22 | Jfeスチール株式会社 | 高炉炉況状態判定装置及び高炉の操業方法 |
| CN108595380B (zh) | 2018-03-14 | 2021-07-09 | 山东科技大学 | 一种高炉异常炉况检测方法 |
| CN113614253A (zh) | 2019-04-03 | 2021-11-05 | 杰富意钢铁株式会社 | 高炉异常判定装置、高炉异常判定方法以及高炉操作方法 |
-
2020
- 2020-03-31 CN CN202080023159.9A patent/CN113614253A/zh active Pending
- 2020-03-31 JP JP2020546515A patent/JP6825753B1/ja active Active
- 2020-03-31 KR KR1020217029637A patent/KR102574567B1/ko active Active
- 2020-03-31 US US17/440,913 patent/US12320590B2/en active Active
- 2020-03-31 WO PCT/JP2020/014881 patent/WO2020204043A1/ja not_active Ceased
- 2020-03-31 EP EP20783058.9A patent/EP3950966B1/en active Active
- 2020-04-01 TW TW109111280A patent/TWI745912B/zh active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017088950A (ja) * | 2015-11-09 | 2017-05-25 | Jfeスチール株式会社 | 高炉への原料装入方法 |
| JP2017128805A (ja) | 2016-01-19 | 2017-07-27 | Jfeスチール株式会社 | 高炉の操業方法 |
| JP2017190482A (ja) | 2016-04-12 | 2017-10-19 | 株式会社神戸製鋼所 | 高炉のセンサ故障検知システム及び異常状況の予測システム |
| JP2018165399A (ja) * | 2017-03-28 | 2018-10-25 | Jfeスチール株式会社 | 高炉炉況状態判定装置、高炉の操業方法、及び、高炉炉況状態判定方法 |
| JP2018204076A (ja) * | 2017-06-06 | 2018-12-27 | Jfeスチール株式会社 | 高炉吹込み還元材の燃焼率推定方法、高炉操業方法及び送風羽口 |
Non-Patent Citations (1)
| Title |
|---|
| STATISTICAL PROCESS CONTROL USING PROCESS CHEMOMETRICS, SYSTEMS, CONTROL AND INFORMATION, vol. 48, no. 5, 2004, pages 165 - 170 |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20230098293A (ko) * | 2020-12-08 | 2023-07-03 | 제이에프이 스틸 가부시키가이샤 | 이상 판정 모델 생성 장치, 이상 판정 장치, 이상 판정 모델 생성 방법 및 이상 판정 방법 |
| KR102881149B1 (ko) | 2020-12-08 | 2025-11-04 | 제이에프이 스틸 가부시키가이샤 | 이상 판정 모델 생성 장치, 이상 판정 장치, 이상 판정 모델 생성 방법 및 이상 판정 방법 |
| CN112961949A (zh) * | 2021-03-12 | 2021-06-15 | 鞍钢股份有限公司 | 一种快速判断高炉出现管道行程的方法 |
| JPWO2023190234A1 (ja) * | 2022-03-29 | 2023-10-05 | ||
| JP7715195B2 (ja) | 2022-03-29 | 2025-07-30 | Jfeスチール株式会社 | 高炉の異常判定装置、高炉の異常判定方法、高炉の操業方法、高炉の操業システム、高炉の異常判定サーバ装置、高炉の異常判定サーバ装置のプログラム、及び表示端末装置 |
| JP2024105129A (ja) * | 2023-01-25 | 2024-08-06 | トヨタ自動車株式会社 | 情報処理装置、情報処理方法、及び情報処理プログラム |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2020204043A1 (ja) | 2021-04-30 |
| CN113614253A (zh) | 2021-11-05 |
| BR112021018897A2 (pt) | 2021-11-30 |
| US12320590B2 (en) | 2025-06-03 |
| KR20210127223A (ko) | 2021-10-21 |
| TW202043489A (zh) | 2020-12-01 |
| EP3950966B1 (en) | 2024-06-19 |
| EP3950966A1 (en) | 2022-02-09 |
| US20220163260A1 (en) | 2022-05-26 |
| KR102574567B1 (ko) | 2023-09-04 |
| TWI745912B (zh) | 2021-11-11 |
| JP6825753B1 (ja) | 2021-02-03 |
| EP3950966A4 (en) | 2022-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6825753B1 (ja) | 高炉の異常判定装置、高炉の異常判定方法、及び高炉の操業方法 | |
| JP5975883B2 (ja) | 故障している計器を検出する方法及び装置 | |
| KR102408426B1 (ko) | 설비 노화 지수를 이용한 이상 감지 방법 및 장치 | |
| JP2013073414A (ja) | プラントのセンサ診断装置およびセンサ診断方法 | |
| KR102351538B1 (ko) | 패턴 학습 및 진단 알고리즘을 이용한 자동화 설비의 고장 예측 시스템 및 그 방법 | |
| KR102303406B1 (ko) | 산업설비의 이상유무 진단장치 및 방법 | |
| JP5621967B2 (ja) | 異常データ分析システム | |
| WO2020166236A1 (ja) | 作業効率評価方法、作業効率評価装置、及びプログラム | |
| KR20180116322A (ko) | 제조 설비의 이상 진단 장치 | |
| WO2023190234A1 (ja) | 高炉の異常判定装置、高炉の異常判定方法、高炉の操業方法、高炉の操業システム、高炉の異常判定サーバ装置、高炉の異常判定サーバ装置のプログラム、及び表示端末装置 | |
| CN116878728A (zh) | 一种压力传感器故障检测分析处理系统 | |
| JP5565357B2 (ja) | 設備診断装置、設備診断方法、設備診断プログラムおよびこれを記録したコンピュータ読み取り可能な記録媒体 | |
| JP5091604B2 (ja) | 分布の評価方法、製品の製造方法、分布の評価プログラム及び分布の評価システム | |
| TW202105107A (zh) | 裝置診斷裝置、電漿處理裝置及裝置診斷方法 | |
| JP2010276339A (ja) | センサ診断方法およびセンサ診断装置 | |
| JP5948998B2 (ja) | 異常診断装置 | |
| CN113551764B (zh) | 振动分析设备及其方法 | |
| RU2794126C1 (ru) | Устройство для определения неисправности доменной печи, способ определения неисправности доменной печи и способ эксплуатации доменной печи | |
| US20220184766A1 (en) | Processing abnormality diagnostic device and processing abnormality diagnostic method of machine tool | |
| JP7467982B2 (ja) | 異常判定装置及び異常判定方法 | |
| JP7127477B2 (ja) | 学習方法、装置及びプログラム、並びに設備の異常診断方法 | |
| JP6760503B2 (ja) | 製造プロセス監視装置 | |
| JP2023145899A (ja) | 高炉の異常判定装置、異常判定方法、及び操業方法 | |
| WO2021140942A1 (ja) | 診断装置、診断方法およびプログラム | |
| JP2013164720A (ja) | プラント監視装置およびプラント監視方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| ENP | Entry into the national phase |
Ref document number: 2020546515 Country of ref document: JP Kind code of ref document: A |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20783058 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 20217029637 Country of ref document: KR Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112021018897 Country of ref document: BR |
|
| ENP | Entry into the national phase |
Ref document number: 2020783058 Country of ref document: EP Effective date: 20211103 |
|
| ENP | Entry into the national phase |
Ref document number: 112021018897 Country of ref document: BR Kind code of ref document: A2 Effective date: 20210922 |
|
| WWG | Wipo information: grant in national office |
Ref document number: 17440913 Country of ref document: US |