JP5647917B2 - Control apparatus and control method - Google Patents

Description

  The present invention relates to a control apparatus that controls a hot rolling apparatus, and more particularly, to a control apparatus and a control method that include two or more measuring devices and continue operation appropriately even when one measuring device fails.

  In recent years, customer requirements for the quality of rolled products including steel are becoming stricter, and high-level quality control is required. For example, in a hot rolling apparatus, there are management items such as the thickness of the material to be rolled, the sheet width, the finishing mill exit temperature, and the winding temperature, so that the values of the management items are within a predetermined management width. In addition, it is required to control each device of the hot rolling apparatus.

  For example, the sheet width of the rolled material is about 600 mm to 2,300 mm, whereas the sheet width accuracy after finish rolling is within the control width of about 0 to 8 mm with respect to the target value. It is requested. Furthermore, this management width is a value required for the entire length in the longitudinal direction (conveying direction) of the material to be rolled with respect to the sheet width required as a product, and deviates from the management width with respect to the target value in this section. If the value is negative, it cannot be corrected.

  Therefore, in order to perform quality control with high sheet width accuracy so as not to deviate from the management width, it is necessary to perform control using width correction means such as a slab sizing press or an edger. When an edger is used, the plate thickness is also reduced using a horizontal mill. At this time, a locally deformed portion at the end in the width direction called a dog bone when using the edger, and a width direction deformation due to the horizontal mill reduction of this portion and a width expansion due to the horizontal mill occur. Further, the ease of deformation of the material to be rolled, that is, the deformation resistance, also changes due to temperature changes during rolling. For this reason, it has been very difficult to accurately calculate the amount of change in sheet width by the edger and the horizontal mill using a calculation model. Therefore, a width measuring instrument that measures the sheet width is provided at the main control points on the hot rolling mill (rolling line), and accuracy management and learning of the calculation model are performed using the sheet width measured by the width measuring instrument. Is often performed.

  Patent Document 1 discloses a width deviation between a width measured by a width measuring device disposed before a rolled material winder and a predetermined target width before the winding device, and a width measuring device disposed on the finish rolling mill outlet side. Describes a sheet width control method for correcting the finishing mill output side width target value so that the width deviation between the measured width and the predetermined hot finishing mill output side width target value becomes equal.

Japanese Patent No. 3341622

  Up to now, since the measuring instruments require maintenance, it has not been so often performed to install a plurality of measuring instruments at the same location on the rolling line, unless the measuring instruments themselves are inexpensive. However, in recent years, there have been cases where a plurality of measuring instruments are provided and these measuring instruments can be switched and used.

  For example, there are multiple measurement methods such as a width measuring device that detects the edge with a CCD camera and measures the plate width, and a width measuring device that detects a number of X-ray detectors and detects the edge from the X-ray transmission amount. There is a case where a width measuring device is provided and used by switching between them. However, the measured plate width value may differ depending on the measurement method of the width measuring instrument.

  Further, depending on the line configuration, for example, a plurality of width measuring instruments of the same measurement method may be arranged at the same or close quality control points such as the finishing mill exit side. For example, when a first width measuring device used as a normal system (main) and a second width measuring device used as an emergency system (backup) are provided and the first width measuring device fails, Some devices continue the width measurement by switching to the width measuring device.

  However, even when the first width measuring device and the second width measuring device, which are measuring devices of the same measuring method, measure the same measurement target point of the material to be rolled, the measured values may be different. . This is thought to be due to the fact that the measurement points are slightly shifted and consequently the measurement points are different, and that there are individual differences in the measuring device itself. These are called machine differences. If there is this machine difference, the measured value may be different even if the first width measuring device fails, even if the second width measuring device is switched. Thereby, since a measured value changes suddenly, it influences control of a hot rolling apparatus, and the quality of a product fluctuates as a result.

  Similarly, even when the sheet width control method described in Patent Document 1 is used, when the width measuring device is switched and used, the measured value suddenly changes when switched, thus affecting the control of the hot rolling apparatus, As a result, product quality varies.

  The present invention has been made in view of the above problems, and even when one measuring device is switched to the other measuring device, the control of the hot rolling apparatus is appropriately continued without suddenly changing the measured value. The present invention provides a control device and a control method.

  In order to achieve the above object, a first feature of the control device according to the present invention is that a first measurement unit that measures a process value in a rolling line for rolling the material to be rolled as a first process value, and the rolling line A second measurement unit that measures a process value of the same type as the first process value as a second process value, an abnormality detection unit that detects an abnormality of the first measurement unit, and the first process value The second process value, the measurement time information indicating the time when the first process value and the second process value were measured, and the abnormality detection indicating the time when the abnormality was detected by the abnormality detection unit A process information storage unit that stores the process information in association with the time point information, and learning for correcting the second process value so as to approach the first process value based on the process information A learning term calculation unit that calculates the learning term, a learning term storage unit that stores the learning term calculated by the learning term calculation unit, and the second term value stored in the learning term storage unit based on the process information A correction unit that generates a correction process value by correcting with the learned term, the first process value before the point in time when an abnormality is detected by the abnormality detection unit based on the process information, and the abnormality detection A selection unit that selects the correction process value after the point in time when an abnormality is detected by the unit, and an equipment control unit that controls the rolling line based on the process value and the correction process value selected by the selection unit; It is in having.

  In order to achieve the above object, a second feature of the control device according to the present invention is that the learning term calculation unit, based on the process information stored in the process value storage unit, A difference from the second process value is calculated, and based on the calculated difference and the learning term at the previous rolling stored in the learning term storage unit, the new learning term is obtained by exponential smoothing. And calculating and storing the calculated new learning term in the learning term storage unit.

  In order to achieve the above object, a third feature of the control device according to the present invention is that the abnormality detection unit calculates a standard deviation of the first process value, and the standard deviation exceeds a predetermined range. Alternatively, when it is zero, it is determined that an abnormality has occurred in the first measurement unit.

  In order to achieve the above object, a fourth feature of the control device according to the present invention is that the learning term calculation unit has a positional relationship with respect to the material to be rolled by the first measurement unit and the second measurement unit. Based on the difference between the first process value and the second process value measured at substantially the same location, a learning term for correcting the second process value so as to approach the first process value is calculated. There is to do.

  In order to achieve the above object, a first feature of the control method according to the present invention is that a first measurement step for measuring a process value in a rolling line for rolling the material to be rolled as a first process value, and the rolling line A second measurement step of measuring a process value of the same type as the first process value as a second process value, an abnormality detection step of detecting an abnormality of the first measurement step, and the first process value The second process value, the measurement time information indicating the time when the first process value and the second process value were measured, and the abnormality detection indicating the time when the abnormality was detected by the abnormality detection step. A process information storage step for storing the process information in the process information storage unit in association with the time point information, and the second process based on the process information. Learning term calculation step for calculating a learning term for correcting the value so as to be close to the first process value, and a learning term storage step for storing the learning term calculated by the learning term calculation step in a learning term storage unit; A correction step of generating a correction process value by correcting the second process value with the learning term stored in the learning term storage step based on the process information; and the abnormality based on the process information A selection step for selecting the first process value before the time point when the abnormality is detected by the detection step and the correction process value after the time point when the abnormality is detected by the abnormality detection step, and the selection step selects Equipment control step for controlling the rolling line based on the measured process value and the corrected process value It lies in the fact that with a.

  In order to achieve the above object, a second feature of the control method according to the present invention is that the learning term calculation step is based on the process information stored in the process value storage unit and the first process value. A difference from the second process value is calculated, and based on the calculated difference and the learning term at the previous rolling stored in the learning term storage unit, the new learning term is obtained by exponential smoothing. And calculating and storing the calculated new learning term in the learning term storage unit.

  In order to achieve the above object, a third feature of the control method according to the present invention is that the abnormality detection step calculates a standard deviation of the first process value, and the standard deviation exceeds a predetermined range. Alternatively, when it is zero, it is determined that an abnormality has occurred in the first measurement unit.

  In order to achieve the above object, a fourth feature of the control method according to the present invention is that the learning term calculation step has a positional relationship with respect to the material to be rolled with each other by the first measurement step and the second measurement step. Based on the difference between the first process value and the second process value measured at substantially the same location, a learning term for correcting the second process value so as to approach the first process value is calculated. There is to do.

  According to the control device and the control method of the present invention, even when one measuring device is switched to the other measuring device, the control of the hot rolling device can be appropriately continued without suddenly changing the measurement value. it can.

It is the block diagram which showed the structure of the hot rolling apparatus controlled by the control apparatus which concerns on one Embodiment which concerns on this invention. It is the block diagram which showed the function structure of the control apparatus which concerns on one Embodiment which concerns on this invention. It is the figure which showed an example of the process information memorize | stored in the process information storage part with which the control apparatus which concerns on one Embodiment which concerns on this invention is provided. It is the figure which showed an example of the learning term memorize | stored in the learning term memory | storage part with which the control apparatus which concerns on one Embodiment which concerns on this invention is provided. It is the flowchart which showed the process sequence of the control apparatus which concerns on one Embodiment which concerns on this invention. In the control apparatus which concerns on one Embodiment which concerns on this invention, it is the figure explaining the process information memorize | stored in the process information storage part, and the performance correction process information memorize | stored in the performance correction process information storage part. It is the figure which showed typically the hot rolling apparatus provided with the two thermometers controlled by the control apparatus which concerns on one Embodiment which concerns on this invention.

  A configuration of a control device according to an embodiment of the present invention will be described.

≪Configuration of hot rolling equipment≫
FIG. 1 is a configuration diagram showing a configuration of a hot rolling apparatus controlled by a control apparatus according to an embodiment of the present invention. The arrow in FIG. 1 has shown the conveyance direction in which the to-be-rolled material 200 rolled in a hot rolling apparatus (hot rolling line) is conveyed. In general, the material to be rolled 200 is also referred to as a slab, a bar, and a coil in the process of being rolled in a hot rolling apparatus, but here, the material to be rolled 200 is unified.

  As shown in FIG. 1, the hot rolling apparatus 100 includes a heating furnace 101, a primary descaler 103, a rough edger 105, a roughing mill 107, a roughing side plate width meter 109, and a roughing side thermometer 111. Finishing side thermometer 113, crop shear 115, secondary descaler 117, finishing rolling mill 119, finishing side plate thickness meter 121, multigauge 123, finishing side thermometer 125, flatness A total 127, a run-out laminar spray cooling unit 129, a coiler inlet side thermometer 131, a coiler inlet side plate width meter 133, and a coiler 135 are provided.

  The heating furnace 101 is a furnace for heating the material to be rolled 200.

  The primary descaler 103 removes the oxide film formed on the surface of the material to be rolled 200 by the heating of the heating furnace 101 by spraying high-pressure water from above and below the material to be rolled 200.

  The rough edger 105 performs rolling in the width direction of the material to be rolled 200 when viewed from the upper surface direction of the hot rolling line.

  The rough rolling mill 107 includes one or more stands, and performs rolling in the vertical direction of the material to be rolled 200. Further, the rough rolling mill 107 includes a reversible rolling mill because it is necessary to shorten the line length from the viewpoint of preventing temperature decrease and the like, and further, rolling by a plurality of passes (reciprocating motion in the conveying direction) is necessary. It is often composed of. The rough rolling mill 107 is equipped with a descaler for spraying high-pressure water onto the material to be rolled 200 that is a semi-finished product to remove the oxide film on the surface. Since rolling is performed at a high temperature, an oxide film is likely to be formed, and it is necessary to use an apparatus for removing such an oxide film as appropriate.

  The roughing side plate width meter 109 measures the plate width of the material 200 to be rolled, which is a semi-finished product during rolling.

  The roughing side thermometer 111 measures the surface temperature of the material 200 to be rolled, which is a semi-finished product during rolling.

  The finish entry side thermometer 113 measures the surface temperature of the material to be rolled 200 at the entrance of the finish rolling mill 119 because the distance between the rough rolling mill 107 and the finish rolling mill 119 is long.

  The crop shear 115 cuts the leading end of the material to be rolled 200.

  The secondary descaler 117 is provided at the entrance of the finish rolling mill 119 because the distance between the rough rolling mill 107 and the finish rolling mill 119 is long, and in order to improve the surface properties of the material 200 after finish rolling, The oxide film formed on the surface of the rolled material 200 is removed by spraying high-pressure water from above and below the material 200 to be rolled.

  The finish rolling mill 119 employs a tandem type in which a plurality of rolling rolls called stands are installed. By rolling in the vertical direction with a plurality of rolling rolls, a material to be rolled 200 having a target plate thickness can be obtained. it can. A spray is provided between the stands of the finish rolling mill 119 in order to suppress oxide film formation and to control the temperature.

  The finishing delivery side thickness gauge 121 measures the thickness of the material to be rolled 200 rolled by the finishing mill 119. In order to control the thickness of the material to be rolled 200 within the target range, the value of the thickness measured by the finishing delivery side thickness gauge 121 is used as a process value for feedback control of the rolling position of the finishing mill 119.

  Multi-Channel Gauge 123, which is a kind of X-ray measuring device, has a form in which X-ray detectors are arranged in the width direction of the material to be rolled 200, and the thickness distribution in the width direction can be measured. Therefore, it is a composite measuring instrument that can measure a plurality of types of process values such as plate thickness, crown, and plate width with a single unit.

  The finish delivery side thermometer 125 measures the surface temperature of the material 200 to be rolled after being rolled by the finish rolling mill 119. The temperature of the material to be rolled 200 is closely related to the formation and material of the metal structure of the product, and needs to be managed at an appropriate temperature.

  The flatness meter 127 measures the flatness of the material to be rolled 200 after being rolled by the finish rolling mill 119. Further, the flatness meter 127 includes a plurality of CCD cameras, and can measure the plate width of the material to be rolled 200. Since the flatness meter 127 has higher measurement accuracy than the multigauge 123, the plate width value measured by the flatness meter 127 is often used as the plate width value on the exit side of the finishing mill 119.

  The run-out laminar spray cooling unit 129 is a device that cools the material to be rolled 200 with cooling water in order to control the temperature of the material to be rolled 200. These may be provided with a forced cooling device in the front and rear in addition to a normal run-out table cooling device.

  The coiler entry side thermometer 131 measures the surface temperature of the material to be rolled 200 cooled by the run-out laminar spray cooling unit 129. The temperature of the material to be rolled 200 is closely related to the formation and material of the metal structure of the rolled product, and needs to be managed at an appropriate temperature.

  The coiler entry side plate width meter 133 measures the plate width of the material to be rolled 200 cooled by the run-out laminar spray cooling unit 129. In normal rolling, the material to be rolled 200 heated to the austenite region is transformed into a structure such as ferrite or pearlite in the run-out laminar spray cooling unit 129, and thus the plate width after transformation is measured. In addition, since it is about 860 ° C. on the exit side of the finish rolling mill 119 and about 600 ° C. on the inlet side of the coiler 135, a state where there is less error from room temperature due to linear expansion by measuring in a state closer to room temperature The board width can be measured with.

  The coiler 135 is wound up to convey the material to be rolled 200.

≪Functional configuration of control device 1≫
FIG. 2 is a configuration diagram illustrating a functional configuration of the control device 1 according to an embodiment of the present invention. In FIG. 2, a case where the multi-gauge 123 is the first measurement unit and the flatness meter 127 is the second measurement unit will be described as an example.

  As shown in FIG. 2, the control device 1 according to an embodiment of the present invention includes a process information storage unit 2, a learning term storage unit 3, a control unit 4, and a performance correction process information storage unit 5. The plate width value measured by the multigauge 123 is supplied, and the plate width value measured by the flatness meter 127 is supplied.

  The process information storage unit 2 includes a plate width value that is a first process value measured by the multigauge 123, a plate width value that is a second process value measured by the flatness meter 127, and these plate width values. Is stored as process information in association with measurement time point information indicating the time point at which measurement is performed and abnormality detection time point information indicating a point in time when an abnormality is detected by the abnormality detection unit 12 described later.

  FIG. 3 is a diagram illustrating an example of process information stored in the process information storage unit 2 included in the control device 1 according to an embodiment of the present invention.

  As shown in FIG. 3, at each measurement time t (reference numeral 201), the plate width value (reference numeral 202) that is the first process value measured by the multigauge 123 and the second measured by the flatness meter 127. The process width is associated with a plate width value (reference numeral 203) and an error code (abnormality detection time information) (reference numeral 204) indicating whether or not an abnormality is detected by an abnormality detection unit 12 to be described later. Is remembered as Here, when the error code (abnormality detection time information) is “1”, it indicates that an abnormality has been detected by the abnormality detection unit 12, and when it is “0”, no abnormality has been detected by the abnormality detection unit 12. Is shown.

  The learning term storage unit 3 stores the learning term calculated by the learning term calculation unit 13 described later.

  FIG. 4 is a diagram illustrating an example of learning terms stored in the learning term storage unit 3 included in the control device 1 according to an embodiment of the present invention.

  As shown in FIG. 4, learning term tables 300 to 500 are stored in the learning term storage unit 3 for each of the steel types A to C.

The learning term tables 300 to 500 store learning terms Z corresponding to the target plate width X (1)... X (m) and the target plate thickness Y (1). Yes. For example, in the learning term table 300 for steel type A, a learning term Z _A (m) 1 is stored as a learning term corresponding to the target plate width X (m) and the target plate thickness Y (1).

  When the process information does not include an error code indicating that an abnormality has been detected by the abnormality detection unit 12, the actual correction process information storage unit 5 displays the plate width value measured by the multigauge 123 as the actual correction process information. And the plate width value measured by the multi-gauge 123 before the time point when the abnormality is detected by the abnormality detection unit 12 and the error code indicating that the abnormality is detected by the abnormality detection unit 12 Then, the correction plate width value generated by the correction unit 14 to be described later after the abnormality is detected by the abnormality detection unit 12 is selected and stored as actual correction process information.

  The control unit 4 performs central control of the control device 1. The control unit 4 includes a process information storage control unit 11, an abnormality detection unit 12, a learning term calculation unit 13, a correction unit 14, a selection unit 15, and a device control unit 16.

  The process information storage control unit 11 includes a first process value, a second process value, measurement time point information indicating a time point at which the first process value and the second process value are measured, and an abnormality detection unit 12. The process information is stored in the process information storage unit 2 as process information in association with the abnormality detection time point information indicating the time point when the abnormality is detected.

  The abnormality detection unit 12 detects an abnormality in the multigauge 123 (first measurement unit). Specifically, the abnormality detection unit 12 calculates a standard deviation of a constant time or constant length process value for one piece of the material 200 to be rolled, and when the standard deviation is larger than a certain value or zero, It is determined that an abnormality has occurred in the multigauge 123 (first measurement unit).

  The learning term calculation unit 13 calculates a learning term for correcting the second process value so as to approach the first process value based on the process information.

  The correction unit 14 generates a correction plate width value (correction process value) by correcting the second process value with the learning term Z stored in the learning term storage unit 3 based on the process information.

  When the error code indicating that an abnormality has been detected by the abnormality detection unit 12 is not included based on the process information, the selection unit 15 selects the plate width value measured by the multigauge 123, and the abnormality detection unit 12 includes an error code indicating that an abnormality has been detected, the first process value before the time when the abnormality is detected by the abnormality detection unit 12, and the time when the abnormality is detected by the abnormality detection unit 12. Thereafter, a correction plate width value (correction process value) generated by the correction unit 14 described later is selected.

  The equipment control unit 16 controls the rolling line based on the setting conditions such as the target plate width, target plate thickness, and steel type supplied from the outside, the process value and the correction process value selected by the selection unit 15. Specifically, the heating furnace 101, the primary descaler 103, the rough edger 105, the roughing mill 107, the crop shear 115, the secondary descaler 117, the finish rolling mill 119, and the run-out laminar spray cooling unit 129 And the coiler 135 are controlled.

  In addition, the equipment control unit 16 performs setting calculation for making the setting of the material 200 to be rolled next highly accurate, for example, the plate width after rolling of the material 200 and the target value of the plate width if it is a plate width. In order to reduce the width deviation amount, the sheet width learning which is a setting calculation learning function is performed by comparing the sheet width actual value and the sheet width target value for each material 200 to be rolled.

  In this setting calculation learning function, a calculation value obtained by model calculation after rolling is compared with an actual measurement value to absorb a model calculation error. This setting calculation learning function is described in p.291 of “Theory and Practice of Sheet Rolling” (Japan Steel Association: 1984).

  In addition, when an abnormality is detected by the abnormality detection unit 12, the device control unit 16 replaces the plate width value that is the first process value with the plate width that is the second process value measured by the flatness meter 127. Feedback control may be performed using the value.

<< Operation of the control device 1 >>
FIG. 5 is a flowchart showing a processing procedure of the control device 1 according to an embodiment of the present invention. In FIG. 5, a case where the multi-gauge 123 is the first measurement unit and the flatness meter 127 is the second measurement unit will be described as an example.

  As shown in FIG. 5, the learning term calculation unit 13 of the control unit 4 determines whether or not the rolling of the material to be rolled 200 has been completed (step S101). Specifically, when the equipment control unit 16 sequentially supplies the material 200 to be rolled from the heating furnace 101 provided in the hot rolling apparatus 100 and is wound by the coiler 135, the material 200 to be rolled is used. It is determined that the rolling is completed.

  In step S101, when it is determined that the rolling of the material to be rolled 200 is completed (in the case of YES), the learning term calculation unit 13 reads the process information stored in the process information storage unit 2 (step S103).

Next, the learning term calculation unit 13 calculates a difference between measurement values (step S105). Specifically, the learning term calculation unit 13 is flat with the plate width value B ₁ (first process value) measured by the multigauge 123 at an arbitrary time point based on the process information read in step S103. The plate width value B ₂ (second process value) measured by the dynamometer 127 is calculated, and the difference δ is calculated using the following (Equation 1).

次に、学習項算出部１３は、学習項Ｚを算出する（ステップＳ１０７）。具体的には、学習項算出部１３は、ステップＳ１０１において圧延が完了した被圧延材２００の鋼種、目標板厚Ｘ、及び目標板厚Ｙに対応する学習項を、前回の圧延時に算出された学習項Ｚ_ｎ−１として学習項記憶部３から抽出する。そして、学習項算出部１３は、ステップＳ１０５において算出された差δと、下記の式２とを用いて、指数平滑法により前回の圧延時に算出された学習項Ｚを平滑化することにより新たな学習項Ｚ_ｎを算出する。なお、式２では、Ｚ_ｎ−１を前回の圧延時に算出された学習項とし、Ｚ_ｎを今回の圧延時に算出する学習項として表記している。

Next, the learning term calculation unit 13 calculates a learning term Z (step S107). Specifically, the learning term calculation unit 13 calculates the learning terms corresponding to the steel type, the target plate thickness X, and the target plate thickness Y of the material to be rolled 200 that has been rolled in step S101 at the time of the previous rolling. The learning term Z _n−1 is extracted from the learning term storage unit 3. And the learning term calculation part 13 uses the difference (delta) calculated in step S105, and the following formula 2, and smoothes the learning term Z calculated at the time of the last rolling by the exponential smoothing method, and is new. A learning term Z _n is calculated. In Equation 2, Z _n-1 is represented as a learning term calculated during the previous rolling, and Z _n is represented as a learning term calculated during the current rolling.

ここで、βは、学習ゲインであり０〜１．０の範囲である。この学習ゲインβは、１．０に近いほど学習速度が速くなるが異常値の影響を受けやすくなり、通常は０．３〜０．４程度にすることが多い。

Here, β is a learning gain and is in the range of 0 to 1.0. As the learning gain β is closer to 1.0, the learning speed increases, but the learning gain β tends to be affected by an abnormal value, and is usually about 0.3 to 0.4.

Then, the learning term calculation unit 13 stores the calculated learning term Z in the learning term storage unit 3 (step S109). Specifically, the learning term calculation unit 13 uses the learning term table stored in the learning term storage unit 3 and the steel type, the target plate thickness X, and the target plate thickness Y of the material 200 to be rolled that have been rolled in step S101. The learning term Z _n−1 corresponding to is overwritten with the learning term Z _n calculated in step S107.

Next, the correction unit 14 of the control unit 4 generates a correction process value (step S111). Specifically, the correction unit 14 stores the plate width value B ₂ measured by the flatness meter 127 based on the process information stored in the process information storage unit 2 using the following equation 3 as a learning term. Correction plate width value B ₂ ^COMP which is a correction process value is generated by correcting with learning term Z stored by unit 3.

次に、制御部４の選択部１５は、ステップＳ１０１において圧延完了した被圧延材２００の圧延中に、マルチゲージ１２３（第１の測定部）に異常があったか否かを判定する（ステップＳ１１３）具体的には、選択部１５は、プロセス情報記憶部２に記憶された、ステップＳ１０１において圧延完了した被圧延材２００に対応するプロセス情報に含まれるエラーコード（異常検出時点情報）に“１”を示しているデータが存在する場合、マルチゲージ１２３に異常があったと判定する。

Next, the selection unit 15 of the control unit 4 determines whether or not there is an abnormality in the multi-gauge 123 (first measurement unit) during rolling of the material to be rolled 200 that has been rolled in step S101 (step S113). Specifically, the selection unit 15 stores “1” in the error code (abnormality detection time point information) included in the process information corresponding to the material to be rolled 200 that has been rolled in step S101 and stored in the process information storage unit 2. Is present, it is determined that there is an abnormality in the multi-gauge 123.

  In step S113, when it is determined that there is no abnormality in the multigauge 123 (in the case of NO), the selection unit 15 selects the plate width value measured by the multigauge 123 (first measurement unit). Specifically, the selection unit 15 selects the sheet width value measured by the multigauge 123 from the process information corresponding to the material to be rolled 200 that has been rolled in step S101, stored in the process information storage unit 2. Select (step S115).

  On the other hand, when it is determined in step S113 that there is an abnormality in the multigauge 123 (first measurement unit) (in the case of YES), the selection unit 15 is measured by the multigauge 123 before the time when the abnormality is detected. The selected plate width value and the corrected plate width value generated by the correction unit 14 after the point in time when the abnormality is detected are selected (step S117).

  For example, in the example of FIG. 3, if the value of the error code 204 is changed from “0” to “1” when the time t201 is “00:10:10”, the selection unit 15 determines that the time t201 is The plate width value 202 measured by the multigauge 123 up to the time “00:10:00” is selected, and the correction generated by the correction unit 14 after the time t201 is “00:10:10”. Select the board width value.

  Next, the selection unit 15 determines the plate width value measured by the multigauge 123 selected in step S115 or the plate width value measured by the multigauge 123 before the time point when the abnormality selected in step S117 is detected. And the correction board width value produced | generated by the correction | amendment part 14 after the time when abnormality was detected is memorize | stored in the performance correction process information storage part 5 as performance correction process information (step S119).

  FIG. 6 illustrates the process information stored in the process information storage unit 2 and the actual correction process information stored in the actual correction process information storage unit 5 in the control device 1 according to the embodiment of the present invention. FIG.

  6 indicates a plate width value measured by the multi-gauge 123, 602 indicates a plate width value measured by the flatness meter 127, and 603 is generated by the correction unit 14. The corrected plate width value is shown.

  As shown in FIG. 6, the plate width value measured by the flatness meter 127 is corrected based on the difference between the plate width values measured by the multi-gauge 123 and the flatness meter 127 at the time point t1, which is an arbitrary time point. A learning term Z is calculated for this purpose.

  Since the plate width value measured by the multigauge 123 is a constant value at time t2, it is determined that the standard deviation is zero, and it is determined that there is an abnormality in the multigauge 123 after time t2. Therefore, the selection unit 15 of the control unit 4 includes the plate width value 601 measured by the multigauge 123 before the time t2 when the abnormality is detected by the abnormality detection unit 12, and the time t2 when the abnormality is detected by the abnormality detection unit 12. Thereafter, the correction plate width value 603 generated by the correction unit 14 is selected and stored in the result correction process information storage unit 5 as result correction process information.

  Thereby, even when an abnormality occurs in the multi-gauge 123, the measurement value can be stored without sudden change.

As mentioned above, according to the control apparatus 1 which concerns on one Embodiment which concerns on this invention, the board width value (process value) in the rolling line (hot rolling apparatus) which rolls the to-be-rolled material 200 is set to the 1st process value. Multi-gauge 123 (first measuring unit) to measure as a flatness to measure a sheet width value which is the same process value as the first process value in a rolling line (hot rolling apparatus) as a second process value 127 in total (second measurement unit), abnormality detection unit 12 for detecting an abnormality in multi-gauge 123 (first measurement unit), first process value, second process value, and first process The process information storage unit 2 stores the measurement time information indicating the time when the value and the second process value are measured and the abnormality detection time information indicating the time when the abnormality is detected by the abnormality detection unit 12 in association with the process information. And Based on the process information, a learning term calculation unit 13 that calculates a learning term for correcting the second process value so as to approach the first process value, and a learning term Z calculated by the learning term calculation unit 13 Based on the learning term storage unit 3 to be stored and the process information, the correction plate width value (correction process value) is generated by correcting the second process value with the learning term Z stored in the learning term storage unit 3. Based on the correction unit 14 and the process information, the first process value before the time when the abnormality is detected by the abnormality detection unit 12, and the correction plate width value (correction) after the time when the abnormality is detected by the abnormality detection unit 12. Process value), and a device control unit 16 that controls the rolling line based on the process value and the correction process value selected by the selection unit 15. Even when an abnormality occurs in the gauge 123 (first measurement unit), the control of the hot rolling apparatus 100 is appropriately continued without suddenly changing the measurement value. Note that the control according to the embodiment of the present invention. In the apparatus 1, the new learning term Z is calculated by the exponential smoothing method, but the present invention is not limited to this.

  For example, a layered learning method using the target plate thickness, target plate width, alloy type, or the like as a layered key, or a learning method using a neural network using the same type of measured value and difference δ as teaching data can be used.

  Moreover, in the control apparatus 1 which concerns on one Embodiment which concerns on this invention, although the multigauge 123 was made into the 1st measurement part and the flatness meter 127 was made into the 2nd measurement part, it is not restricted to this.

  For example, the flatness meter 127 may be the first measurement unit, the multi-gauge 123 may be the second measurement unit, the flatness meter 127 is the first measurement unit, and the coiler entry side plate width meter 133 is the second measurement unit. It may be a measuring unit.

  Further, the process value to be measured is not limited to the plate width, and the temperature measured by the roughing side thermometer 111, the finishing side thermometer 113, the finishing side thermometer 125, or the coiler inlet side thermometer 131 is the process value. Alternatively, the plate thickness value measured by the finishing delivery side plate thickness meter 121 may be used as the process value.

  That is, the first measuring unit and the second measuring unit may be a measuring device that is provided on the hot rolling apparatus 100 (rolling line) and measures the same type of process value.

  Moreover, in the control apparatus 1 which concerns on one Embodiment which concerns on this invention, although the difference of the board width value each measured in the multi gauge 123 and the flatness meter 127 in arbitrary time was calculated, it is not restricted to this.

  For example, the difference between the average of the plate width values measured by the multigauge 123 and the average of the plate width values measured by the flatness meter 127 may be calculated. Further, two thermometers may be provided, and the difference in temperature measured between the two thermometers at a location where the positional relationship with the material to be rolled 200 is approximately the same may be calculated.

  FIG. 7 is a diagram schematically showing a hot rolling apparatus provided with two thermometers.

  As shown in FIG. 7, a first finishing side thermometer 140 and a second finishing side thermometer 141 are provided.

  And the learning term calculation part 17 is based on each temperature measured in the location where the positional relationship with respect to the to-be-rolled material 200 is substantially the same with the 1st finishing side thermometer 140 and the 2nd finishing side thermometer 141. Calculate the difference. For example, the learning term calculation unit 17 calculates the difference based on the temperature at a point separated from the tip of the material to be rolled 200 by a predetermined distance S on the material to be rolled 200 based on the conveyance speed of the material to be rolled 200.

  Then, the learning term calculation unit 13 may calculate a learning term Z for correcting the temperature measured by the second finisher-side thermometer 141 based on the calculated difference.

Moreover, in the control apparatus 1 which concerns on one Embodiment which concerns on this invention, although the difference of the board width value each measured by the multigauge 123 and the flatness meter 127 was calculated, it is not restricted to this, It is made to calculate ratio. Also good. Specifically, the learning term calculation unit 13 calculates the following (formula 4) based on the plate width value B ₁ measured by the multigauge 123 and the plate width value B ₂ measured by the flatness meter 127. To calculate the ratio φ.

そして、学習項算出部１３は、比φを算出した場合、学習項算出部１３は、算出された比φを、下記の式５を用いて、指数平滑法により前回の圧延時に算出された学習項Ｚを平滑化することにより新たな学習項Ｚ_ｎを算出する。なお、式５では、Ｚ_ｎ−１を前回の圧延時に算出された学習項とし、Ｚ_ｎを今回の圧延時に算出する学習項として表記している。

Then, when the learning term calculation unit 13 calculates the ratio φ, the learning term calculation unit 13 uses the following equation 5 to calculate the calculated ratio φ by the exponential smoothing method during the previous rolling. A new learning term Z _n is calculated by smoothing the term Z. In Equation 5, Z _n-1 is represented as a learning term calculated during the previous rolling, and Z _n is represented as a learning term calculated during the current rolling.

さらに、制御部４の補正部１４は、プロセス情報記憶部２に記憶されたプロセス情報に基づいて、下記の式６を用いて、平坦度計１２７により測定された板幅値Ｂ_２を学習項記憶部３により記憶された学習項Ｚで補正することにより補正プロセス値である補正板幅値Ｂ_２ ^ＣＯＭＰを生成する。

Further, the correction unit 14 of the control unit 4 uses the following equation 6 based on the process information stored in the process information storage unit 2 to calculate the plate width value B ₂ measured by the flatness meter 127 as a learning term. Correction plate width value B ₂ ^COMP which is a correction process value is generated by correcting with learning term Z stored in storage unit 3.

  As described above, the control device and the control method according to the present invention can be used for the control device and the control method for controlling the hot rolling device for rolling the material to be rolled.

DESCRIPTION OF SYMBOLS 1 ... Control apparatus 2 ... Process information storage part 3 ... Learning term memory | storage part 4 ... Control part 5 ... Performance correction process information storage part 11 ... Process information storage control part 12 ... Abnormality detection part 13 ... Learning term calculation part 14 ... Correction part DESCRIPTION OF SYMBOLS 15 ... Selection part 16 ... Equipment control part 17 ... Learning term calculation part 100 ... Hot rolling apparatus 101 ... Heating furnace 103 ... Primary descaler 105 ... Rough edger 107 ... Rough rolling mill 109 ... Roughening side plate width meter 111 ... Roughening Side thermometer 113 ... Finishing input side thermometer 115 ... Crop shear 117 ... Secondary descaler 119 ... Finishing rolling mill 121 ... Finishing side thickness gauge 123 ... Multi gauge 125 ... Finishing side thermometer 127 ... Flatness meter 129 ... Runout Laminar spray cooling part 131 ... Coiler inlet side thermometer 133 ... Coiler inlet side board width meter 135 ... Coiler 140 ... First finish side thermometer 141 ... 2nd finishing delivery side thermometer

Claims (8)

A first measuring unit for measuring a process value in a rolling line for rolling the material to be rolled as a first process value;
A second measuring unit that measures a process value of the same type as the first process value in the rolling line as a second process value;
An abnormality detection unit for detecting an abnormality in the first measurement unit;
The first process value, the second process value, the measurement time point information indicating the time point when the first process value and the second process value are measured, and the abnormality detection unit detects an abnormality. A process information storage unit that stores the process information in association with the abnormality detection time point information indicating the time point,
A learning term calculation unit that calculates a learning term for correcting the second process value so as to approach the first process value based on the process information;
A learning term storage unit for storing a learning term calculated by the learning term calculation unit;
A correction unit that generates a correction process value by correcting the second process value with a learning term stored in the learning term storage unit based on the process information;
A selection for selecting the first process value before the time point when the abnormality is detected by the abnormality detection unit and the correction process value after the time point when the abnormality is detected by the abnormality detection unit based on the process information And
Based on the process value selected by the selection unit and the correction process value, an equipment control unit for controlling the rolling line;
A control device comprising: The learning term calculation unit
Based on the process information stored in the process value storage unit, a difference between the first process value and the second process value is calculated, and the calculated difference and the learning term storage unit are stored. The new learning term is calculated by exponential smoothing based on the learning term at the time of the previous rolling, and the calculated new learning term is stored in the learning term storage unit. The control device according to 1. The abnormality detection unit
A standard deviation of the first process value is calculated, and when the standard deviation exceeds a predetermined range or is zero, it is determined that an abnormality has occurred in the first measurement unit. The control device according to claim 1. The learning term calculation unit
Based on the difference between the first process value and the second process value measured by the first measurement unit and the second measurement unit at locations where the positional relationship with respect to the material to be rolled is substantially the same. 4. The control device according to claim 1, wherein a learning term for correcting the process value of 2 so as to be close to the first process value is calculated. 5. A first measurement step for measuring a process value in a rolling line for rolling the material to be rolled as a first process value;
A second measurement step of measuring a process value of the same type as the first process value in the rolling line as a second process value;
An abnormality detection step of detecting an abnormality in the first measurement step;
An abnormality is detected by the first process value, the second process value, measurement time information indicating the time when the first process value and the second process value are measured, and the abnormality detection step. A process information storage step for storing the process information in the process information storage unit in association with the abnormality detection time point information indicating the time point,
A learning term calculation step for calculating a learning term for correcting the second process value so as to approach the first process value based on the process information;
A learning term storage step of storing the learning term calculated in the learning term calculation step in a learning term storage unit;
A correction step of generating a correction process value by correcting the second process value with the learning term stored in the learning term storage step based on the process information;
Selection for selecting the first process value before the time point when the abnormality is detected by the abnormality detection step and the correction process value after the time point when the abnormality is detected by the abnormality detection step based on the process information Steps,
An equipment control step for controlling the rolling line based on the process value and the correction process value selected by the selection step;
A control method characterized by comprising: The learning term calculation step includes:
Based on the process information stored in the process value storage unit, a difference between the first process value and the second process value is calculated, and the calculated difference and the learning term storage unit are stored. The new learning term is calculated by exponential smoothing based on the learning term at the time of the previous rolling, and the calculated new learning term is stored in the learning term storage unit. 5. The control method according to 5. The abnormality detection step includes:
A standard deviation of the first process value is calculated, and when the standard deviation exceeds a predetermined range or is zero, it is determined that an abnormality has occurred in the first measurement unit. The control method according to claim 5. The learning term calculation step includes:
Based on the difference between the first process value and the second process value measured at locations where the positional relationship with respect to the material to be rolled is substantially the same by the first measurement step and the second measurement step. The control method according to claim 5, further comprising: calculating a learning term for correcting the process value of 2 so as to approach the first process value.

Images