TW202328468A - Method and system for controlling thickness of galvanized film of steel strip - Google Patents
Method and system for controlling thickness of galvanized film of steel strip Download PDFInfo
- Publication number
- TW202328468A TW202328468A TW111100839A TW111100839A TW202328468A TW 202328468 A TW202328468 A TW 202328468A TW 111100839 A TW111100839 A TW 111100839A TW 111100839 A TW111100839 A TW 111100839A TW 202328468 A TW202328468 A TW 202328468A
- Authority
- TW
- Taiwan
- Prior art keywords
- steel strip
- thickness
- air knife
- zinc layer
- center position
- Prior art date
Links
Images
Landscapes
- Coating With Molten Metal (AREA)
Abstract
Description
本發明是關於一種控制方法及其系統,且特別是關於一種鋼帶鍍鋅膜厚控制方法及其系統。The present invention relates to a control method and its system, and in particular to a steel strip galvanized film thickness control method and its system.
熱浸鍍鋅鋼板是目前使用最廣泛也是最有效的鋼鐵防蝕方法,而鍍鋅鋼帶的耐蝕性主要取決於鋼帶的鍍鋅層的厚度。在傳統生產方法中,當鋼板完成熱浸鍍鋅後,待鋅液凝固後會透過量測裝置以X射線(X-ray)量測鍍鋅層的厚度,再依據前後表面之鍍鋅層的厚度差異來以人工方式調整鋼帶中心線位置,直到前後表面鋅層厚度差異最小。然而,這樣的鋼帶中心線調整方式,無法即時調整鋼帶中心位置,易造成部分鋼帶的鋅層厚度不符合目標,另外,當產線有異況時,需等待後端的鋅層厚度量測裝置的量測結果才能進行相對處置,無法即時預知,再者,以人工方式進行調整會導致判斷標準並不一致,且變異較大,無法建立調整標準。Hot-dip galvanized steel sheet is currently the most widely used and most effective steel corrosion protection method, and the corrosion resistance of galvanized steel strip mainly depends on the thickness of the galvanized layer of the steel strip. In the traditional production method, after the hot-dip galvanizing of the steel plate is completed, the thickness of the galvanized layer will be measured by X-ray (X-ray) through the measuring device after the zinc liquid is solidified, and then according to the thickness of the galvanized layer on the front and rear surfaces Manually adjust the position of the centerline of the steel strip according to the difference in thickness until the difference in the thickness of the zinc layer on the front and rear surfaces is the smallest. However, this method of adjusting the center line of the steel strip cannot immediately adjust the center position of the steel strip, which may easily cause the thickness of the zinc layer of some steel strips to fail to meet the target. In addition, when there is an abnormality in the production line, it is necessary to wait for the thickness of the zinc layer at the back end Only the measurement results of the measuring device can be relatively dealt with, which cannot be predicted in real time. Furthermore, manual adjustments will lead to inconsistent judgment standards and large variations, making it impossible to establish adjustment standards.
另外,一般在鍍鋅產線為維持鋅槽設備穩定,約運作3個禮拜後,需有1個禮拜之設備定修時段。定修期間會將所有鋅槽內相關之設備,如沉浸輥、校正輥、穩定輥等皆移除更換維修,造成設備重新安裝時,參數可能會有偏移誤差,如中心位置參數並未校正,會造成鋼帶前後表面鋅層厚度偏移。而中心位置參數的校正方式,同樣是如同上述以後端的鋅層厚度量測裝置的量測結果來以人工方式調整,除了會有如同上述所討論的問題之外,且鋼帶前後表面鋅層厚度不一致的影響因素並不只有中心線位置,還有其他因素亦會影響,人力很難準確判斷並修正製程參數。In addition, in order to maintain the stability of the zinc tank equipment in the galvanizing production line, after about 3 weeks of operation, there is a period of 1 week for equipment maintenance. During the maintenance period, all related equipment in the zinc tank, such as immersion rollers, correction rollers, stabilization rollers, etc. will be removed for replacement and maintenance. As a result, when the equipment is reinstalled, the parameters may have offset errors, such as the center position parameters have not been corrected , will cause the thickness of the zinc layer on the front and rear surfaces of the steel strip to deviate. The correction method of the center position parameter is also manually adjusted by the measurement results of the rear-end zinc layer thickness measuring device as mentioned above. In addition to the problems discussed above, the thickness of the zinc layer on the front and rear surfaces of the steel strip The inconsistency is not limited to the position of the center line, but other factors also affect it. It is difficult for manpower to accurately judge and correct the process parameters.
又如鋅槽內各轉輥軸承,生產過程中會持續磨損造成各轉輥位置偏移,亦會造成生產過程中,製程參數變異性加大。Another example is the roller bearings in the zinc tank. During the production process, the continuous wear will cause the position of each roller to shift, which will also increase the variability of process parameters during the production process.
另一重要設備為氣刀,氣刀在鍍鋅過程中會有鋅液黏附、堵塞出氣口造成氣流不均,因此生產過程中需以刮刀清理氣刀唇,然而刮刀清理之方式隨著長時間累積會造成氣刀唇退縮狀況,造成實際氣刀與鋼帶距離變遠,鋅層厚度變異變大。然而,目前氣刀唇退縮的問題僅能在停機維修時以人工量測來進行修正,無法即時預知。Another important piece of equipment is the air knife. During the galvanizing process, zinc liquid will adhere to the air knife and block the air outlet, causing uneven air flow. Therefore, it is necessary to use a scraper to clean the lip of the air knife during the production process. Accumulation will cause the lip of the air knife to shrink, resulting in a longer distance between the actual air knife and the steel strip, and greater variation in the thickness of the zinc layer. However, at present, the problem of retraction of the air knife lip can only be corrected by manual measurement when the machine is shut down for maintenance, and cannot be predicted immediately.
本發明之目的在於提出一種鋼帶鍍鋅膜厚控制方法,適用於鍍鋅鋼帶產線,上述鍍鋅鋼帶產線包括鋅液槽及氣刀設備,上述鍍鋅鋼帶產線將鋼帶通過鋅液槽以達到熱浸鍍鋅並透過氣刀設備吹出氣體以刮除鋼帶上多餘的鋅液。上述鋼帶鍍鋅膜厚控制方法包括:量測鋼帶的前表面鋅層厚度與後表面鋅層厚度;根據前表面鋅層厚度與後表面鋅層厚度來判斷氣刀設備是否有氣刀唇退縮;及當判斷氣刀設備有氣刀唇退縮時,校正氣刀設備相對於鋼帶之位置。The object of the present invention is to propose a method for controlling the galvanized film thickness of a steel strip, which is suitable for a galvanized steel strip production line. The above-mentioned galvanized steel strip production line includes a zinc liquid tank and an air knife device. The strip passes through the zinc bath to achieve hot-dip galvanizing and blows air through the air knife device to scrape off excess zinc on the strip. The method for controlling the galvanized film thickness of the steel strip includes: measuring the thickness of the zinc layer on the front surface and the thickness of the zinc layer on the rear surface of the steel strip; judging whether the air knife device has an air knife lip Retraction; and when it is judged that the air knife device has a retraction of the air knife lip, correct the position of the air knife device relative to the steel belt.
在一些實施例中,上述鋼帶鍍鋅膜厚控制方法更包括:當前表面鋅層厚度或後表面鋅層厚度大於目標鋅層厚度時,判斷氣刀設備有氣刀唇退縮。In some embodiments, the method for controlling the galvanized film thickness of the steel strip further includes: when the thickness of the zinc layer on the front surface or the thickness of the zinc layer on the rear surface is greater than the target zinc layer thickness, judging that the air knife device has an air knife lip retraction.
在一些實施例中,上述鋼帶鍍鋅膜厚控制方法更包括:在鋼帶通過氣刀設備之前,將鋼帶的多個製程參數傳送至鋼帶中心位置計算裝置以計算鋼帶的中心位置;及當判斷氣刀設備有氣刀唇退縮時,根據中心位置、前表面鋅層厚度、後表面鋅層厚度、目標鋅層厚度來移動鋅液槽中的校正輥以校正氣刀設備相對於鋼帶之位置。In some embodiments, the method for controlling the galvanized film thickness of the steel strip further includes: before the steel strip passes through the air knife equipment, transmitting a plurality of process parameters of the steel strip to the calculation device for the center position of the steel strip to calculate the center position of the steel strip ; And when judging that the air knife equipment has the air knife lip retraction, move the correcting roller in the zinc liquid tank to correct the air knife equipment relative to The position of the steel belt.
在一些實施例中,上述鋼帶鍍鋅膜厚控制方法更包括:當前表面鋅層厚度與後表面鋅層厚度之間的差值不為零時,判斷中心位置有偏離;及當判斷中心位置有偏離時,根據中心位置、前表面鋅層厚度與後表面鋅層厚度來移動校正輥以校正中心位置。In some embodiments, the method for controlling the galvanized film thickness of the steel strip further includes: when the difference between the thickness of the zinc layer on the current surface and the thickness of the zinc layer on the rear surface is not zero, judging that the center position deviates; and when judging the center position When there is a deviation, move the correction roller to correct the center position according to the center position, the thickness of the zinc layer on the front surface and the thickness of the zinc layer on the rear surface.
在一些實施例中,上述鋼帶中心位置計算裝置依據預測模型進行預測以計算中心位置,上述預測模型之學習過程包括數據收集步驟、數據分析步驟及模型訓練步驟。數據收集步驟是取得製程參數以及與產線相關的產線製程參數。數據分析步驟是將產線速度對時間作積分運算以計算出鋼帶目前位置。模型訓練步驟是係透過分類、分群和迴歸之人工智慧學習方式以建立預測模型。In some embodiments, the said device for calculating the central position of the steel strip performs prediction according to the forecasting model to calculate the central position, and the learning process of the said forecasting model includes a data collection step, a data analysis step and a model training step. The data collection step is to obtain process parameters and production line process parameters related to the production line. The data analysis step is to integrate the production line speed with time to calculate the current position of the steel strip. The model training step is to establish a predictive model through artificial intelligence learning methods of classification, grouping and regression.
本發明之目的在於另提出一種鋼帶鍍鋅膜厚控制系統,適用於鍍鋅鋼帶產線,上述鍍鋅鋼帶產線包括鋅液槽及氣刀設備,上述鍍鋅鋼帶產線將鋼帶通過鋅液槽以達到熱浸鍍鋅並透過氣刀設備吹出氣體以刮除鋼帶上多餘的鋅液。上述鋼帶鍍鋅膜厚控制系統包括:鋅層厚度量測裝置與校正裝置。鋅層厚度量測裝置用以量測鋼帶的前表面鋅層厚度與後表面鋅層厚度。校正裝置用以:根據前表面鋅層厚度與後表面鋅層厚度來判斷氣刀設備是否有氣刀唇退縮;及當校正裝置判斷氣刀設備有氣刀唇退縮時,校正裝置校正氣刀設備相對於鋼帶之位置。The purpose of the present invention is to propose another steel strip galvanized film thickness control system, which is applicable to the galvanized steel strip production line. The above-mentioned galvanized steel strip production line includes zinc liquid tank and air knife equipment. The above-mentioned galvanized steel strip production line will The steel strip passes through the zinc bath to achieve hot-dip galvanizing and blows air through the air knife device to scrape off the excess zinc liquid on the steel strip. The above-mentioned galvanized film thickness control system for the steel strip includes: a zinc layer thickness measuring device and a correcting device. The zinc layer thickness measuring device is used for measuring the thickness of the zinc layer on the front surface and the thickness of the zinc layer on the rear surface of the steel strip. The calibration device is used to: judge whether the air knife device has the air knife lip retraction according to the thickness of the front surface zinc layer and the rear surface zinc layer thickness; and when the calibration device judges that the air knife device has the air knife lip retraction, the calibration device corrects the air knife device Relative to the position of the steel belt.
在一些實施例中,其中當前表面鋅層厚度或後表面鋅層厚度大於目標鋅層厚度時,上述校正裝置判斷氣刀設備有氣刀唇退縮。In some embodiments, when the thickness of the zinc layer on the front surface or the thickness of the zinc layer on the rear surface is greater than the target thickness of the zinc layer, the correction device determines that the air knife device has an air knife lip retraction.
在一些實施例中,上述鋼帶鍍鋅膜厚控制系統更包括鋼帶中心位置計算裝置用以:在鋼帶通過氣刀設備之前,將鋼帶的多個製程參數傳送至鋼帶中心位置計算裝置以計算鋼帶的中心位置。其中校正裝置更用以:當校正裝置判斷氣刀設備有氣刀唇退縮時,校正裝置根據中心位置、前表面鋅層厚度、後表面鋅層厚度、目標鋅層厚度來移動鋅液槽中的校正輥以校正氣刀設備相對於鋼帶之位置。In some embodiments, the above-mentioned steel strip galvanized film thickness control system further includes a steel strip center position calculation device for: before the steel strip passes through the air knife equipment, a plurality of process parameters of the steel strip are transmitted to the steel strip center position calculation device to calculate the center position of the steel strip. Among them, the correction device is more used: when the correction device judges that the air knife device has the retraction of the air knife lip, the correction device moves the zinc liquid tank according to the center position, the thickness of the zinc layer on the front surface, the thickness of the zinc layer on the rear surface, and the thickness of the target zinc layer. Calibration rollers are used to correct the position of the air knife device relative to the steel belt.
在一些實施例中,上述校正裝置更用以:當前表面鋅層厚度與後表面鋅層厚度之間的差值不為零時,校正裝置判斷中心位置有偏離;及當校正裝置判斷中心位置有偏離時,校正裝置根據中心位置、前表面鋅層厚度與後表面鋅層厚度來移動校正輥以校正中心位置。In some embodiments, the correction device is further used: when the difference between the thickness of the zinc layer on the front surface and the thickness of the zinc layer on the rear surface is not zero, the correction device judges that the center position deviates; and when the correction device judges that the center position has a deviation When it deviates, the correction device moves the correction roller according to the center position, the thickness of the zinc layer on the front surface and the thickness of the zinc layer on the rear surface to correct the center position.
在一些實施例中,上述鋼帶中心位置計算裝置依據預測模型進行預測以計算中心位置,上述預測模型之學習過程包括數據收集步驟、數據分析步驟及模型訓練步驟。數據收集步驟是取得製程參數以及與產線相關的產線製程參數。數據分析步驟是將產線速度對時間作積分運算以計算出鋼帶目前位置。模型訓練步驟是係透過分類、分群和迴歸之人工智慧學習方式以建立預測模型。In some embodiments, the said device for calculating the central position of the steel strip performs prediction according to the forecasting model to calculate the central position, and the learning process of the said forecasting model includes a data collection step, a data analysis step and a model training step. The data collection step is to obtain process parameters and production line process parameters related to the production line. The data analysis step is to integrate the production line speed with time to calculate the current position of the steel strip. The model training step is to establish a predictive model through artificial intelligence learning methods of classification, grouping and regression.
為讓本發明的上述特徵和優點能更明顯易懂,下文特舉實施例,並配合所附圖式作詳細說明如下。In order to make the above-mentioned features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail together with the accompanying drawings.
以下仔細討論本發明的實施例。然而,可以理解的是,實施例提供許多可應用的概念,其可實施於各式各樣的特定內容中。所討論、揭示之實施例僅供說明,並非用以限定本發明之範圍。關於本文中所使用之『第一』、『第二』、…等,並非特別指次序或順位的意思,其僅為了區別以相同技術用語描述的元件或操作。Embodiments of the invention are discussed in detail below. It should be appreciated, however, that the embodiments provide many applicable concepts that can be implemented in a wide variety of specific contexts. The discussed and disclosed embodiments are for illustration only, and are not intended to limit the scope of the present invention. The terms “first”, “second”, etc. used herein do not specifically refer to a sequence or sequence, but are only used to distinguish elements or operations described with the same technical terms.
圖1係根據本發明的實施例之鍍鋅鋼帶產線的多個設備之相對位置的配置圖。圖2係根據本發明的實施例之鍍鋅鋼帶產線的多個設備之另一視角的相對位置的配置圖。圖3係根據本發明的實施例之鍍鋅鋼帶產線的結構示意圖。Fig. 1 is a configuration diagram of the relative positions of a plurality of devices in a galvanized steel strip production line according to an embodiment of the present invention. Fig. 2 is a configuration diagram of relative positions of a plurality of devices of a galvanized steel strip production line according to another viewing angle of an embodiment of the present invention. Fig. 3 is a structural schematic diagram of a galvanized steel strip production line according to an embodiment of the present invention.
鍍鋅鋼帶產線包括氣刀設備110及鋅液槽120。首先,將鋼帶100通過熔融鋅的鋅液槽120以達到熱浸鍍鋅,此時,鋼帶100的表面會附著有熔融後的鋅液,當鋼帶100穿出鋅液槽120後,會透過氣刀設備110吹出氣體,由氣體來刮除鋼帶100上多餘的鋅液,使鋼帶100的表面上所附著的鋅液鍍層厚度達到預設的厚度。The galvanized steel strip production line includes an
在本發明的實施例中,氣刀設備110包括第一氣刀及第二氣刀分別設置於鋼帶100之相異兩側。在本發明的實施例中,氣刀設備110之可調整的氣刀設備參數值包括:氣刀壓力、第一氣刀與鋼帶之間的距離值、第二氣刀與鋼帶之間的距離值,可調整上述氣刀設備參數值使鋼帶100的表面上所附著的鋅液鍍層厚度達到預設的厚度。In an embodiment of the present invention, the
在本發明的實施例中,鋼帶鍍鋅膜厚控制系統包括鋅層厚度量測裝置200。在鋼帶100通過氣刀設備110之後,鋼帶鍍鋅膜厚控制系統透過鋅層厚度量測裝置200來量測鋼帶100的前表面的前表面鋅層厚度以及鋼帶100的後表面的後表面鋅層厚度。在本發明的實施例中,鋅層厚度量測裝置200可例如為塗層測厚儀(coating thickness gauge),可例如利用X射線(X-ray)來量測鍍鋅層的厚度。In an embodiment of the present invention, the steel strip galvanized film thickness control system includes a zinc layer
在本發明的實施例中,鋼帶鍍鋅膜厚控制系統還包括彼此通訊連接的鋼帶中心位置計算裝置210與校正裝置220。鋼帶中心位置計算裝置210用以計算鋼帶100的中心位置。如圖2與圖3所示,在鋅液槽120中包括鋅槽輥121、校正輥122及穩定輥123,校正輥122通訊連接校正裝置220。在本發明的實施例中,當鋼帶100的中心位置產生偏離時,校正裝置220會透過校正輥122的移動以調整鋼帶100的位置。In the embodiment of the present invention, the steel strip galvanized film thickness control system further includes a steel strip center
鋼帶中心位置計算裝置210用以:在鋼帶100通過氣刀設備110之前,將鋼帶100的多個製程參數傳送至鋼帶中心位置計算裝置210以計算鋼帶100的中心位置。The strip center
校正裝置220用以根據鋅層厚度量測裝置200所量測得的前表面鋅層厚度與後表面鋅層厚度來判斷氣刀設備110是否有氣刀唇退縮。當校正裝置220判斷氣刀設備110有氣刀唇退縮時,校正裝置220會透過校正輥122的移動以校正氣刀設備110相對於鋼帶100之位置。The
關於鋼帶中心位置計算裝置210,在鋼帶100通過氣刀設備110之前,將鋼帶100的多個製程參數傳送至鋼帶中心位置計算裝置210以計算鋼帶100的中心位置。具體而言,鋼帶中心位置計算裝置210依據鋼帶100的多個製程參數,使用預測模型進行預測,以計算鋼帶100的中心位置。Regarding the strip center
在本發明的實施例中,鋼帶中心位置計算裝置210所使用的預測模型之學習過程包括數據收集步驟、數據分析步驟及模型訓練步驟。In the embodiment of the present invention, the learning process of the prediction model used by the steel strip center
其中,數據收集步驟可以依據現場設備情況取得與產線相關的多個產線製程參數,例如:產線速度、鋼帶種類、鋼帶尺寸(如寬度、長度)、設備狀態(如輥徑、輥位置等)、氣刀設備110的氣刀開度、壓力、距離等,此外,還可以參考鋼帶的相關參數(即,鋼帶100的多個製程參數),例如:訂單編號(用以判別鋼帶的使用對象,使用對象不同製程中也需要做調整)、母鋼捲編號(串連資料用以判斷鋼捲的來源),以及上游相關製程參數(如退火溫度或退火時間等)等都有可能會影響到鋅層厚度的產線製程參數、鋼帶100的製程參數、材料資訊等任何有可能會影響鋅層厚度之資訊或參數都可以抓取,端看使用者對於鍍鋅膜厚的預測模型的需求,以及鋅層厚度量測裝置200所量測之前表面鋅層厚度與後表面鋅層厚度。其中,預測模型的建立還是要依據實務上的需求,例如:要求鋼帶表面的品質,就必須參考相關的軋延條件(例如:壓力、軋輥輥徑等),將這些相關數據放入建立預測模型的參數,之後再透過相對應的前表面鋅層厚度與後表面鋅層厚度的量測,以作為預測模型訓練的相對依據。Among them, the data collection step can obtain multiple production line process parameters related to the production line according to the conditions of the field equipment, such as: production line speed, steel strip type, steel strip size (such as width, length), equipment status (such as roll diameter, roll position, etc.), air knife opening, pressure, distance, etc. Identify the object of use of the steel strip, which needs to be adjusted in different processes), the serial number of the parent steel coil (connecting data to determine the source of the steel coil), and upstream related process parameters (such as annealing temperature or annealing time, etc.), etc. Any information or parameters that may affect the thickness of the zinc layer, such as the production line process parameters that may affect the thickness of the zinc layer, the process parameters of the
其中,關於數據分析步驟,必須將上列的收集好的各項參數數據經過數據處理後才可以進行預測模型訓練,由於鋼帶100表面層的厚度資料通常會與製程參數有距離延遲的問題,因此必需要根據資料延遲部分做修正,資料主要是時間序列,但產線的速度隨時都可能有變化產生,無法準確估算出所需的時間,因此,主是以鋼帶位置作為依據,將產線速度對時間作積分運算,以計算出鋼帶目前位置,算式如下:
其中,V
鋼帶為產線速度,Local為鋼帶目前位置。
Among them, with regard to the data analysis step, the above-mentioned collected parameter data must be processed before the prediction model training can be carried out. Because the thickness data of the surface layer of the
除了校正鋅層厚度量測延遲以外,篩選影響參數較少的數據也是提高預測模型準確度的方法之一,例如:可依據鋅層寬度變異量進行數據篩選(例如將變異量較大者濾除),寬度、方向、鋅層厚度的變異量較小,表示鋼帶100本身的形變越少,比較平整,平整的鋼帶對於判斷中心線位置較容易,如果鋼帶本身不平整,比較難去定義鋼帶100的中心位置。In addition to correcting the delay in zinc layer thickness measurement, screening data with fewer influencing parameters is also one of the methods to improve the accuracy of the prediction model. For example, data screening can be performed based on the variation of zinc layer width (for example, filtering out those with large variation ), the variation in width, direction, and thickness of the zinc layer is small, which means that the
最後進行模型訓練步驟,經過資料的收集整理過後,就可以利用所收集的資料藉由迴歸模型進行預測模型的訓練,透過分類、分群和迴歸之人工智慧學習方式(例如機器學習、深度學習)以建立預測模型。鋼帶中心位置計算裝置210的預測模型是預測一特定數值,屬於迴歸方式。Finally, the model training step is carried out. After the data is collected and sorted, the collected data can be used to train the prediction model through the regression model. Through the artificial intelligence learning methods of classification, grouping and regression (such as machine learning, deep learning) to Build predictive models. The prediction model of the steel strip center
在本發明的實施例中,透過支持向量機(support vector machine,SVM)的監督式的學習方法,用統計風險最小化的原則來預估一個分類的超平面(hyperplane),其概念為找到一個超平面可以將兩類不同之數據區隔開來。而支持向量迴歸(support vector regression,SVR)是SVM的變形,假設資料表示為
,其中
表示輸入的製程參數,
表示所對應的迴歸值,鋅層厚度令
,如果
,則吾人可知這樣的
能夠從
,準確地預測
,這個
即是SVR要找的平面,計算上可由所收集的產線製程參數和相對應的鋼帶100之前表面鋅層厚度與後表面鋅層厚度,加上一定的誤差範圍找出
,即可訓練出產線製程參數和前表面鋅層厚度與後表面鋅層厚度相對應的關係式,再由鋼帶100的中心位置和前表面鋅層厚度與後表面鋅層厚度差異訓練預測模型。
In the embodiment of the present invention, through the supervised learning method of support vector machine (SVM), the principle of statistical risk minimization is used to estimate a classified hyperplane (hyperplane), the concept of which is to find a A hyperplane can separate two different types of data. The support vector regression (SVR) is a deformation of SVM, assuming that the data is expressed as ,in Indicates the input process parameters, Indicates the corresponding regression value, the thickness of the zinc layer is ,if , then we know that such can from , accurately predicting ,this It is the plane that SVR is looking for. In terms of calculation, it can be found from the collected production line process parameters and the corresponding thickness of the zinc layer on the front surface of the
而模型訓練步驟的另一種訓練方式是梯度提升迴歸(gradient boosting regression,GBR),其主要方式是透過許多的弱分類器分類,最後將所有的弱分類器分類結果集合成答案,弱分類是一種簡單的分類器,可先依不同的條件區分答案,最後集合所有弱分類器,投票選出最佳之答案,而GBR的預測模型中,假設資料為 ,先以一簡單的公式(可以是線性函數或多項式等)計算出 ,目標是讓 ,若不為0就繼續將 帶入第二個弱分類器計算求得 ,直到 ,最後 總合即為所需之模型,最後將預測模型加上參數輸入方式的程式即可建立預測模型。 Another training method in the model training step is gradient boosting regression (GBR). The main method is to classify through many weak classifiers, and finally collect the classification results of all weak classifiers into an answer. Weak classification is a kind of A simple classifier can first distinguish answers according to different conditions, and finally gather all weak classifiers to vote for the best answer. In GBR's prediction model, the data is assumed to be , first calculate it with a simple formula (it can be a linear function or a polynomial, etc.) , the goal is to have , if it is not 0, continue to Bring it into the second weak classifier to calculate ,until ,at last The total is the required model, and finally the forecast model can be established by adding the program of the parameter input method.
綜合上述的方法,只要有製程參數、產線相關製程參數和相對應的前表面鋅層厚度與後表面鋅層厚度等資訊,可由不同方式訓練得到預測模型。Combining the above methods, as long as there are process parameters, production line-related process parameters and corresponding information such as the thickness of the front surface zinc layer and the thickness of the rear surface zinc layer, the prediction model can be trained in different ways.
圖4係根據本發明的實施例之氣刀設備有無氣刀唇退縮之鋼帶的中心位置與表面鋅層厚度差異的對照示意圖。如圖4所示,氣刀設備110包括第一氣刀110
前,設置於鋼帶100的前表面的該側;氣刀設備110包括第二氣刀110
後,設置於鋼帶100的後表面的該側。如圖4所示,鋼帶100的前表面的前表面鋅層厚度以CW
前表示,鋼帶100的後表面的後表面鋅層厚度以CW
後表示。如圖4所示,第一氣刀110
前與鋼帶100之間的距離值以d
前表示,第二氣刀110
後與鋼帶100之間的距離值以d
後表示。
Fig. 4 is a schematic diagram of the difference between the center position and the surface zinc layer thickness of the steel strip with or without the retracted air knife lip of the air knife equipment according to the embodiment of the present invention. As shown in Figure 4, before the
如圖4所示,當氣刀設備110的第二氣刀110
後有氣刀唇退縮會造成氣刀設備110的第二氣刀110
後與鋼帶100之間的距離值d
後變遠,從而導致後表面鋅層厚度CW
後未能相等於目標鋅層厚度而是大於目標鋅層厚度。在本發明的實施例中,目標鋅層厚度為鍍鋅鋼帶產線之預設鋅層厚度或者是利用歷史數據所得出的鍍鋅鋼帶產線之目標鋅層厚度。
As shown in Figure 4, when the second air knife 110 of the
換言之,如圖4所示,前表面鋅層厚度CW
前等於目標鋅層厚度,但後表面鋅層厚度CW
後大於目標鋅層厚度時,校正裝置220可據以判斷氣刀設備110的第二氣刀110
後有氣刀唇退縮。接著,校正裝置220根據鋼帶100的中心位置、前表面鋅層厚度CW
前、後表面鋅層厚度 CW
後、目標鋅層厚度來校正氣刀設備110之第二氣刀 110
後相對於鋼帶100之位置,以校正第二氣刀110
後之氣刀唇退縮的問題。在本發明的實施例中,校正氣刀設備100相對於鋼帶100之位置的目標是要減少氣刀設備110的第二氣刀110
後與鋼帶100之間的距離值d
後,從而使得後表面鋅層厚度CW
後相等於目標鋅層厚度,具體方法可為(1)移動氣刀設備110的第二氣刀110
後以減少d
後;和/或(2)移動鋅液槽120中的校正輥122。
In other words, as shown in FIG. 4 , the front surface zinc layer thickness CW is equal to the target zinc layer thickness, but when the rear surface zinc layer thickness CW is greater than the target zinc layer thickness, the
具體而言,當前表面鋅層厚度CW
前大於目標鋅層厚度時,校正裝置220判斷氣刀設備110的第一氣刀 110
前有氣刀唇退縮;當後表面鋅層厚度CW
後大於目標鋅層厚度時,校正裝置220判斷氣刀設備110的第二氣刀110
後有氣刀唇退縮。
Specifically, when the thickness CW of the current surface zinc layer is greater than the target zinc layer thickness, the
具體而言,當校正裝置220判斷氣刀設備110有氣刀唇退縮時,校正裝置220根據鋼帶的中心位置、前表面鋅層厚度CW
前、後表面鋅層厚度CW
後、目標鋅層厚度來移動氣刀設備110或者是移動鋅液槽120中的校正輥122,以校正氣刀設備110相對於鋼帶100之位置。
Specifically, when the
圖5係根據本發明的實施例之鋼帶的中心位置有無偏離之鋼帶的中心位置與表面鋅層厚度差異的對照示意圖。具體而言,如圖5所示,鋼帶100的中心位置即為相對於氣刀設備110之鋼帶100的相對位置。如圖5所示,氣刀設備110包括第一氣刀110
前,設置於鋼帶100的前表面的該側;氣刀設備110包括第二氣刀110
後,設置於鋼帶100的後表面的該側。
Fig. 5 is a comparative schematic diagram of the difference between the central position of the steel strip and the thickness of the surface zinc layer according to the embodiment of the present invention, whether the central position of the steel strip deviates or not. Specifically, as shown in FIG. 5 , the central position of the
如圖5所示,當鋼帶100的中心位置無偏離時,鋼帶100的前表面的前表面鋅層厚度以CW
前會相等於鋼帶100的後表面的後表面鋅層厚度以CW
後。換言之,當前表面鋅層厚度CW
前與後表面鋅層厚度CW
後之間的差值為零時,校正裝置220判斷鋼帶100的中心位置無偏離。
As shown in Figure 5, when the central position of the
如圖5所示,當偏移鋼帶100'的中心位置有偏離時,偏移鋼帶100'的前表面的前表面鋅層厚度以CW
前'不相等於偏移鋼帶100'的後表面的後表面鋅層厚度以CW
後'。換言之,當前表面鋅層厚度CW
前'與後表面鋅層厚度 CW
後'之間的差值不為零時,校正裝置220可判斷出偏移鋼帶100'的中心位置有偏離。當校正裝置220判斷偏移鋼帶100'的中心位置有偏離時,校正裝置220根據偏移鋼帶100'的中心位置、前表面鋅層厚度CW
前'與後表面鋅層厚度CW
後'來移動鋅液槽120中的校正輥122以校正偏移鋼帶100'的中心位置。而上述之校正輥122的移動的目標是要使得前表面鋅層厚度與後表面鋅層厚度之間的差值為零,從而使得鋼帶100的中心位置無偏離。
As shown in Figure 5, when the center position of the offset steel strip 100' deviates, the front surface zinc layer thickness of the front surface of the offset steel strip 100' is not equal to the rear of the offset steel strip 100' in terms of CW front' The rear surface zinc layer thickness of the surface is given in CW post' . In other words, when the difference between the front surface zinc layer thickness CWbefore ' and the rear surface zinc layer thickness CWafter' is not zero, the
具體而言,當校正裝置220判斷鋼帶的中心位置有偏離時,校正裝置220根據鋼帶的中心位置、前表面鋅層厚度、後表面鋅層厚度來移動鋅液槽120中的校正輥122以校正鋼帶100的中心位置。Specifically, when the
圖6係根據本發明的實施例之鋼帶鍍鋅膜厚控制方法的流程圖。於步驟S1,將鋼帶100通過鋅液槽120以達到熱浸鍍鋅。於步驟S2,透過氣刀設備110吹出氣體以刮除鋼帶100上多餘的鋅液。於步驟S3,鋅層厚度量測裝置200量測鋼帶100的前表面鋅層厚度與後表面鋅層厚度。於步驟S4,校正裝置220根據前表面鋅層厚度與後表面鋅層厚度來判斷氣刀設備110是否有氣刀唇退縮。於步驟S5,當校正裝置220判斷氣刀設備110有氣刀唇退縮時,校正裝置220校正氣刀設備110相對於鋼帶100之位置。Fig. 6 is a flow chart of a method for controlling the thickness of a galvanized steel strip according to an embodiment of the present invention. In step S1, the
圖7係根據本發明的實施例之鍍鋅產線的製造方法的示意流程圖。於步驟U1,獲取目前鍍鋅產線的各種製程參數。於步驟U2,驗證設備狀態以判斷是否有誤差,具體方式為將目前鍍鋅產線的各種製程參數與歷史的產線的各種製程參數進行比對。若於步驟U2判斷無誤差則進入步驟U4,若於步驟U2判斷有誤差則進入步驟U3,更新製程參數,具體方式為以歷史的產線的製程參數來進行更新。於步驟U4,使用預測模型進行預測,以計算鋼帶的中心位置。於步驟U5,進行製程參數優化計算,具體方式為使用步驟U4計算出之鋼帶的中心位置來優化製程參數。於步驟U6,設定製程參數,具體方式為使用步驟U5優化後的製程參數來重設製程參數。於步驟U7,鋅層厚度回饋判斷鋅層厚度是否有誤差,若於步驟U7判斷無誤差則進行產線生產,若於步驟U7判斷有誤差則回到步驟U2以進行修正。FIG. 7 is a schematic flowchart of a manufacturing method of a galvanizing production line according to an embodiment of the present invention. In step U1, various process parameters of the current galvanizing production line are obtained. In step U2, verify the state of the equipment to determine whether there is an error, the specific way is to compare various process parameters of the current galvanizing production line with various process parameters of the historical production line. If it is judged in step U2 that there is no error, then go to step U4, and if it is judged in step U2 that there is an error, then go to step U3 to update the process parameters. The specific method is to update the process parameters of the historical production line. In step U4, use the prediction model to perform prediction to calculate the center position of the steel strip. In step U5, the optimization calculation of the process parameters is carried out. The specific method is to use the center position of the steel strip calculated in the step U4 to optimize the process parameters. In step U6, the process parameters are set, specifically by using the optimized process parameters in step U5 to reset the process parameters. In step U7, the zinc layer thickness feedback judges whether there is an error in the thickness of the zinc layer. If it is judged in step U7 that there is no error, the production line production will be carried out. If it is judged in step U7 that there is an error, it will return to step U2 for correction.
綜合上述,本發明透過人工智慧學習方式和歷史數據提供一種透過機器學習與深度學習,配合鋼帶鍍鋅膜厚控制系統的鋅層厚度量測裝置與校正裝置以及產線製程參數,而建立用以計算鋼帶的中心位置之預測模型並比對量測得到的前表面鋅層厚度、後表面鋅層厚度,可透過預測模型自動判斷目前鍍鋅產線重要設備之狀態,自動修正設備之偏移與誤差量,穩定鍍鋅產線生產,同時根據訂單規格,自動設定各製程參數,達到精準、快速、穩定的生產控制,最後可根據產線上之鋅層厚度量測裝置回饋鋅層厚度與預測模型預測值比對,自動修正預測值與實際量測之誤差,同時可即時提供現場人員監控鋼帶製程,減少因鋼帶偏移或氣刀設備之氣刀唇退縮造成鋼帶前後表面鍍鋅厚度不同的問題,並透過鋼帶鍍鋅膜厚控制系統實現可自動判斷與生產之鍍鋅鋼帶智慧製造系統,達到智能化生產及全自動化鋼帶中心位置調控之目的。To sum up the above, the present invention provides a zinc layer thickness measurement device and correction device and production line process parameters through machine learning and deep learning through machine learning and deep learning, and a steel strip galvanized film thickness control system, as well as production line process parameters. Using the prediction model for calculating the center position of the steel strip and comparing the measured front surface zinc layer thickness and rear surface zinc layer thickness, the prediction model can be used to automatically judge the status of important equipment in the current galvanizing production line, and automatically correct the deviation of the equipment To stabilize the production of the galvanizing production line, and automatically set the process parameters according to the order specifications to achieve accurate, fast and stable production control. Finally, the thickness of the zinc layer and the thickness of the zinc layer can be reported according to the zinc layer thickness measuring device on the production line. Comparing the prediction value of the prediction model, automatically correcting the error between the prediction value and the actual measurement, and at the same time providing on-site personnel to monitor the steel strip process in real time, reducing the plating on the front and rear surfaces of the steel strip caused by the deviation of the steel strip or the retraction of the air knife lip of the air knife equipment The zinc thickness is different, and through the steel strip galvanized film thickness control system, a galvanized steel strip intelligent manufacturing system that can automatically judge and produce is realized, achieving the purpose of intelligent production and fully automatic steel strip center position control.
以上概述了數個實施例的特徵,因此熟習此技藝者可以更了解本發明的態樣。熟習此技藝者應了解到,其可輕易地把本發明當作基礎來設計或修改其他的製程與結構,藉此實現和在此所介紹的這些實施例相同的目標及/或達到相同的優點。熟習此技藝者也應可明白,這些等效的建構並未脫離本發明的精神與範圍,並且他們可以在不脫離本發明精神與範圍的前提下做各種的改變、替換與變動。The features of several embodiments are outlined above, so those skilled in the art can better understand aspects of the present invention. Those skilled in the art should appreciate that they can easily use the present invention as a basis to design or modify other processes and structures, thereby achieving the same goals and/or achieving the same advantages as the embodiments described herein . Those skilled in the art should also understand that these equivalent constructions do not depart from the spirit and scope of the present invention, and that they can make various changes, substitutions and alterations without departing from the spirit and scope of the present invention.
100:鋼帶 100':偏移鋼帶 110:氣刀設備 110 前:第一氣刀 110 後:第二氣刀 120:鋅液槽 121:鋅槽輥 122:校正輥 123:穩定輥 200:鋅層厚度量測裝置 210:鋼帶中心位置計算裝置 220:校正裝置 CW 前,CW 前':前表面鋅層厚度 CW 後,CW 後':後表面鋅層厚度 d 前,d 後:距離值 S1~S5,U1~U7:步驟 100: steel belt 100': offset steel belt 110: air knife equipment 110 before : first air knife 110 after : second air knife 120: zinc liquid tank 121: zinc tank roll 122: correction roll 123: stabilizing roll 200: Zinc layer thickness measurement device 210: steel strip center position calculation device 220: correction device CW front , CW front' : front surface zinc layer thickness CW rear , CW rear' : rear surface zinc layer thickness d front , d rear : distance value S1~S5, U1~U7: steps
從以下結合所附圖式所做的詳細描述,可對本發明之態樣有更佳的了解。需注意的是,根據業界的標準實務,各特徵並未依比例繪示。事實上,為了使討論更為清楚,各特徵的尺寸都可任意地增加或減少。 [圖1]係根據本發明的實施例之鍍鋅鋼帶產線的多個設備之相對位置的配置圖。 [圖2]係根據本發明的實施例之鍍鋅鋼帶產線的多個設備之另一視角的相對位置的配置圖。 [圖3]係根據本發明的實施例之鍍鋅鋼帶產線的結構示意圖。 [圖4]係根據本發明的實施例之氣刀設備有無氣刀唇退縮之鋼帶的中心位置與表面鋅層厚度差異的對照示意圖。 [圖5]係根據本發明的實施例之鋼帶的中心位置有無偏離之鋼帶的中心位置與表面鋅層厚度差異的對照示意圖。 [圖6]係根據本發明的實施例之鋼帶鍍鋅膜厚控制方法的流程圖。 [圖7]係根據本發明的實施例之鍍鋅產線的製造方法的示意流程圖。 A better understanding of aspects of the present invention can be obtained from the following detailed description in conjunction with the accompanying drawings. It is to be noted that, in accordance with the standard practice in the industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or decreased for clarity of discussion. [ Fig. 1 ] is an arrangement diagram of relative positions of a plurality of devices of a galvanized steel strip production line according to an embodiment of the present invention. [ FIG. 2 ] is an arrangement diagram of relative positions of a plurality of devices of a galvanized steel strip production line according to an embodiment of the present invention from another viewing angle. [ Fig. 3 ] is a structural schematic diagram of a galvanized steel strip production line according to an embodiment of the present invention. [ Fig. 4 ] is a comparison schematic diagram of the difference between the central position of the steel strip and the thickness of the surface zinc layer with or without the retraction of the air knife lip of the air knife equipment according to the embodiment of the present invention. [ Fig. 5 ] is a schematic diagram showing the difference between the central position of the steel strip and the thickness of the surface zinc layer according to the embodiment of the present invention. [ Fig. 6 ] is a flowchart of a method for controlling the thickness of a galvanized steel strip according to an embodiment of the present invention. [ Fig. 7 ] is a schematic flowchart of a manufacturing method of a galvanizing line according to an embodiment of the present invention.
S1~S5:步驟 S1~S5: steps
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111100839A TW202328468A (en) | 2022-01-07 | 2022-01-07 | Method and system for controlling thickness of galvanized film of steel strip |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111100839A TW202328468A (en) | 2022-01-07 | 2022-01-07 | Method and system for controlling thickness of galvanized film of steel strip |
Publications (1)
Publication Number | Publication Date |
---|---|
TW202328468A true TW202328468A (en) | 2023-07-16 |
Family
ID=88147428
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW111100839A TW202328468A (en) | 2022-01-07 | 2022-01-07 | Method and system for controlling thickness of galvanized film of steel strip |
Country Status (1)
Country | Link |
---|---|
TW (1) | TW202328468A (en) |
-
2022
- 2022-01-07 TW TW111100839A patent/TW202328468A/en unknown
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113039305B (en) | Plating amount control device and plating amount control method | |
WO2023130666A1 (en) | Strip steel plate convexity prediction method based on data-driving and mechanism model fusion | |
CN107379494B (en) | A kind of the film thickness monitoring system and bolt method for self-locating self-positioning based on bolt | |
CN110064667B (en) | Laminar cooling method for steel plate | |
CN105316613A (en) | Zinc coating thickness control method and system based on time-varying delay deviation correction technology | |
TW202328468A (en) | Method and system for controlling thickness of galvanized film of steel strip | |
CN106637026B (en) | A kind of galvanizing process air pressure method for real-time optimization control and system | |
CN110184447A (en) | A kind of normalizing pickling unit intelligence spray cooling control method | |
CN112805399B (en) | Method for controlling coating weight uniformity in industrial galvanization production line | |
CN110076215B (en) | Uncoiler and deviation correction control system and method thereof | |
TWI725861B (en) | The utility model relates to a virtual measuring device for the central position of a steel strip, and a measuring method thereof | |
CN105568197A (en) | Method for accurately controlling weight of continuously-hot-galvanized zinc layer | |
CN107322905A (en) | A kind of film thickness monitoring method self-positioning based on bolt | |
CN114634294B (en) | Automatic substrate glass thickness adjusting method and system | |
JP6958488B2 (en) | Plating adhesion control method, hot-dip plating steel strip manufacturing method, plating adhesion control device and plating adhesion control program | |
CN109593951B (en) | Furnace temperature-based hot dip product dezincification defect dynamic control method | |
KR100815815B1 (en) | Method for controlling coating weight on strip in continuous galvanizing process | |
CN115700412A (en) | Control method of hot-dip galvanized strip steel coating based on deep learning | |
US7751923B2 (en) | Method for control of the thickness of extruded film | |
KR20010018280A (en) | Coating weight predictive control in continuous galvanizing line | |
CN114266451A (en) | Precision evaluation method for roll bending equipment of hot rolling mill | |
CN116881613B (en) | Setting method of high-order item plate-shaped target curve based on flatness | |
TWI695089B (en) | System and method for hot-dip galvanization | |
JP2020139175A (en) | Time-series event prediction method, plating deposition amount control method, method of producing hot-dip coated steel strip, time-series event prediction device, plating deposition amount control device, and plating deposition amount control program | |
JP5954043B2 (en) | Method for judging quality of continuously cast slab and method for continuously casting steel |