Piercing load prediction method, piercing mill control method, piercing load prediction model generation method, and pipe manufacturing equipment

Abstract

To provide a method that predicts a load itself before start of rolling.SOLUTION: There is provided a piercing load prediction method, which predicts a piercing load, executed by an information processor in a piercing and rolling process. The method predicts the piercing load by using a piercing load prediction model that is generated by machine learning including one or more parameters selected from attribute information of the billet, one or more parameters selected from operation parameters in the heating furnace, and one or more parameters selected operation parameters in the piercing and rolling process as input data, and the piercing load as output data.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to a piercing load prediction method for predicting the load applied to a piercing roll, which is a tool used for piercing, in a piercing process for drilling a hole in a cylindrical billet in a seamless pipe manufacturing process; a piercing rolling mill control method; a piercing load prediction model generation method; and pipe manufacturing equipment. [Background technology]

[0002] In the seamless pipe manufacturing process, for example, a cylindrical billet, which is the raw material, is pierced. In this example, the billet is heated to over 1200°C in a heating furnace, and then subjected to tilted rotary piercing in a piercing mill to produce a pierced hollow. In piercing rolling, the billet is pierced by tilted rotary piercing rolls using upper and lower rolls and plugs placed between the rolls. Shoes may be placed on the left and right sides to stabilize the piercing. During processing in a piercing mill, if the load on the piercing rolls is high, there is a risk of damage to the equipment and the risk of seizure between the piercing rolls and the material.

[0003] Conventionally, as technologies related to the load in piercing-rolling, there have been proposed a technology for detecting defects in piercing-rolling based on the actual value of the rolling load, and a technology for changing the shoe position during rolling based on the actual value of the shoe load so as to reduce uneven wall thickness of the pipe after piercing (for example, Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-96265 [Patent Document 2] Japanese Patent Application Laid-Open No. 2008-161900 Summary of the Invention [Problem to be solved by the invention]

[0005] However, no technology has been proposed for predicting the load itself before rolling begins, and there has also been no technology for predicting the risk of equipment damage or the risk of seizure between the piercing roll and the material.

[0006] In view of the above circumstances, an object of the present disclosure is to provide a method for predicting the load itself before the start of rolling. [Means for solving the problem]

[0007] (1) A drilling load prediction method according to an embodiment of the present disclosure includes: A piercing load prediction method for predicting a piercing load in a piercing-rolling process including a heating furnace for heating a billet and a piercing-rolling mill for piercing, the method being executed by an information processing device, comprising: The piercing load is predicted using a piercing load prediction model generated by machine learning, which includes as input data one or more parameters selected from attribute information of the billet, one or more parameters selected from operational parameters in the heating furnace, and one or more parameters selected from operational parameters in the piercing-rolling process, and which outputs the piercing load.

[0008] (2) A drilling load prediction method according to one embodiment of the present disclosure is the drilling load prediction method according to (1), The attribute information of the billet includes a chemical composition of the billet, which includes at least one of a C content and a Si content.

[0009] (3) A drilling load prediction method according to one embodiment of the present disclosure is the drilling load prediction method according to (1) or (2), The operation parameters in the piercing-rolling process include the plug advance amount.

[0010] (4) A control method according to an embodiment of the present disclosure includes: using the piercing load prediction method according to any one of (1) to (3) to predict a piercing load before the billet is charged into the piercing-rolling mill, using attribute information of the billet, actual values ​​of operational parameters in the heating furnace, and set values ​​of operational parameters in the piercing-rolling process; The operation parameters in the piercing and rolling process are reset so that the predicted piercing load falls within the target load range.

[0011] (5) A method for generating a drilling load prediction model according to an embodiment of the present disclosure includes: A method for generating a piercing load prediction model for predicting a piercing load in a piercing-rolling process including a heating furnace for heating a billet and a piercing-rolling mill for piercing, comprising: acquiring a plurality of learning data sets, each of which has as input actual data one or more pieces of actual data selected from the attribute information of the billet, one or more pieces of operational actual data selected from the operational actual data of the heating furnace, and one or more pieces of operational actual data selected from the operational actual data of the piercing-rolling process, and has as output actual data an actual piercing load; A drilling load prediction model is generated by machine learning using the acquired multiple learning data.

[0012] (6) A method for generating a drilling load prediction model according to one embodiment of the present disclosure is a method for generating a drilling load prediction model according to (5), The drilling load prediction model is generated using machine learning selected from neural networks, decision tree learning, random forests, and support vector regression.

[0013] (7) A pipe manufacturing facility according to an embodiment of the present disclosure includes: A pipe manufacturing facility having a heating furnace for heating a billet and a piercing-rolling mill for piercing, a piercing load prediction unit that predicts a piercing load in the piercing rolling mill using a piercing load prediction model, The drilling load prediction model is a model learned by machine learning, The input data includes one or more parameters selected from attribute information of the billet, one or more parameters selected from operation parameters in the heating furnace, and one or more parameters selected from operation parameters in a piercing-rolling process, The learning model is characterized in that the piercing load of the piercing-rolling mill is used as output data. [Effects of the Invention]

[0014] According to an embodiment of the present disclosure, it is possible to provide a drilling load prediction method capable of predicting a drilling load with high accuracy, and a method for generating a drilling load prediction model used in the drilling load prediction method. Furthermore, according to an embodiment of the present disclosure, it is possible to use the drilling load prediction method to reduce the risk of equipment damage and the risk of seizure between a drilling roll and a material. [Brief explanation of the drawings]

[0015] [Figure 1] 10A and 10B are diagrams for explaining the process of drilling according to an embodiment of the present disclosure. [Figure 2] FIG. 1 is a diagram illustrating a configuration of an information processing device according to an embodiment of the present disclosure. [Figure 3] FIG. 10 is a diagram illustrating an operation for generating a drilling load prediction model by an information processing device according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a diagram for explaining parameters in a piercing-rolling process according to an embodiment of the present disclosure. [Figure 5] FIG. 4 is a diagram for explaining parameters in a piercing-rolling process according to an embodiment of the present disclosure. [Figure 6] FIG. 10 is a conceptual diagram of a process for generating a drilling load prediction model according to an embodiment of the present disclosure. [Figure 7] FIG. 10 is a diagram illustrating an operation related to drilling load prediction by an information processing device according to an embodiment of the present disclosure. [Figure 8] FIG. 1 is a conceptual diagram of a prediction process using a drilling load prediction model according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0016] An embodiment of the present disclosure will be described below. An overview of the piercing process that is the subject of prediction in the embodiment of the present disclosure is shown in Fig. 1. As shown in Fig. 1, a cylindrical billet 1 is pierced. In this example, the billet 1 is heated to 1200°C or higher in a heating furnace 2, and then subjected to tilt rotary piercing in a piercing mill 3 to produce a pierced hollow 10.

[0017] First, an overview of this embodiment will be described, and details will be provided later. A piercing load prediction method according to this embodiment is executed by an information processing device 20. The information processing device 20 predicts a piercing load in a piercing-rolling process including a heating furnace 2 that heats a billet 1 and a piercing mill 3 that performs piercing. The information processing device 20 generates a piercing load prediction model by machine learning, which includes, as input data, one or more parameters selected from attribute information of the billet 1, one or more parameters selected from operation parameters of the heating furnace 2, and one or more parameters selected from operation parameters of the piercing-rolling process, and which outputs the piercing load. The information processing device 20 also predicts the piercing load using the generated piercing load prediction model.

[0018] As described above, according to the present embodiment, it is possible to provide a method for generating a piercing load prediction model used in a piercing load prediction method, and a piercing load prediction method capable of predicting the piercing load with high accuracy. Furthermore, according to one embodiment of the present disclosure, it is possible to reduce the risk of equipment damage and the risk of seizure between the piercing roll and the material by using the piercing load prediction method.

[0019] (Configuration of information processing device) Next, a detailed description will be given of each component of the information processing device 20. As shown in Fig. 2, the information processing device 20 includes a control unit 21, a storage unit 22, an input unit 23, an output unit 24, and a communication unit 25.

[0020] The control unit 21 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC). The control unit 21 executes processes related to the operation of the information processing device 20 while controlling each unit of the information processing device 20.

[0021] The storage unit 22 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read only memory (ROM). The RAM is, for example, a static random access memory (SRAM) or a dynamic random access memory (DRAM). The ROM is, for example, an electrically erasable programmable read only memory (EEPROM). The storage unit 22 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 22 stores data used in the operation of the information processing device 20 and data obtained by the operation of the information processing device 20.

[0022] The input unit 23 includes at least one input interface. The input interface is, for example, a physical key, a capacitance key, a pointing device, or a touch screen integrated with a display. The input interface may also be, for example, a sound sensor that accepts voice input, or a camera that accepts gesture input. The input unit 23 accepts an operation to input data used for the operation of the information processing device 20. The input unit 23 may be connected to the information processing device 20 as an external input device instead of being provided in the information processing device 20. Any connection method may be used, for example, a Universal Serial Bus (USB), a High-Definition Multimedia Interface (HDMI) (registered trademark), or Bluetooth (registered trademark).

[0023] The output unit 24 includes at least one output interface. The output interface is, for example, a display that outputs information as a video, or a speaker that outputs information as a sound. The display is, for example, an LCD (liquid crystal display) or an organic EL (electro luminescence) display. The output unit 24 displays and outputs data obtained by the operation of the information processing device 20. The output unit 24 may be connected to the information processing device 20 as an external output device instead of being provided in the information processing device 20. Any connection method may be used, for example, USB, HDMI (registered trademark), or Bluetooth (registered trademark).

[0024] The communication unit 25 includes at least one external communication interface. The communication interface may be either a wired communication interface or a wireless communication interface. In the case of wired communication, the communication interface is, for example, a LAN (Local Area Network) interface or a USB (Universal Serial Bus). In the case of wireless communication, the communication interface is, for example, an interface compatible with mobile communication standards such as LTE (Long Term Evolution), 4G (4th generation), or 5G (5th generation), or an interface compatible with short-range wireless communication such as Bluetooth (registered trademark). The communication unit 25 receives data used in the operation of the information processing device 20 and transmits data obtained by the operation of the information processing device 20.

[0025] The functions of the information processing device 20 are realized by executing a program according to this embodiment on a processor corresponding to the information processing device 20. That is, the functions of the information processing device 20 are realized by software. The program causes a computer to execute the operations of the information processing device 20, thereby causing the computer to function as the information processing device 20. That is, the computer functions as the information processing device 20 by executing the operations of the information processing device 20 in accordance with the program.

[0026] In this embodiment, the program can be recorded on a computer-readable recording medium. The computer-readable recording medium includes non-transitory computer-readable media, such as a magnetic recording device, an optical disc, a magneto-optical recording medium, or a semiconductor memory. The program can be distributed, for example, by selling, transferring, or lending a portable recording medium, such as a DVD (digital versatile disc) or a CD-ROM (compact disc read only memory), on which the program is recorded. The program can also be distributed by storing the program in the storage of an external server and transmitting the program from the external server to another computer. The program can also be provided as a program product.

[0027] Some or all of the functions of the information processing device 20 may be implemented by a dedicated circuit equivalent to the control unit 21. In other words, some or all of the functions of the information processing device 20 may be implemented by hardware.

[0028] (Operation of information processing device) The operation of the information processing device 20 according to this embodiment will be described with reference to Fig. 3. First, the operation of the information processing device 20 for generating a drilling load prediction model will be described.

[0029] Step S1: The control unit 21 of the information processing device 20 acquires training data. The training data is a data set of actual values ​​of input data and actual values ​​of output data (actual drilling load). Any method can be used for the training data acquisition process. For example, the control unit 21 may acquire the training data from a higher-level computer or the like via the communication unit 25.

[0030] As described above, the piercing load prediction model is a learning model that includes, as input data, one or more parameters selected from the attribute information of the billet 1, one or more parameters selected from the operation parameters in the heating furnace 2, and one or more parameters selected from the operation parameters in the piercing-rolling process, and outputs the piercing load.

[0031] The billet 1 attribute information may include the billet diameter and billet length of the billet charged into the heating furnace 2. The billet attribute information may also include the billet's chemical composition. The contents of elemental elements such as C, Si, Mn, Ti, and Cr can be used as the billet's chemical composition. The billet attribute information preferably includes at least one of deformation resistance, C content, and Si content. This is because carbon and silicon contained in steel affect the billet's high-temperature deformation resistance and also affect the formation and composition of oxides on the surface inside the heating furnace 2. This affects temperature changes due to processing heat during rolling and scale properties. In particular, Si contained in the billet segregates on the billet surface during heating and reacts with oxygen inside the heating furnace 2 to form oxides, significantly affecting surface properties. Changes in surface properties affect the contact conditions between the billet and the rolls, which in turn affects the piercing load. The billet's chemical composition may be determined using preset or measured values ​​from the steelmaking process.

[0032] The operational parameters for the heating furnace 2 can be various parameters that are used when the billet for which the piercing load is to be predicted is inside the heating furnace 2. For example, various parameters that are expected to affect the temperature distribution inside the billet extracted from the heating furnace 2 and the state of oxides on the surface, such as the residence time in a specific heating furnace zone of the heating furnace 2, the atmospheric temperature in the final heating furnace zone of the heating furnace 2, the gas composition of the combustion gas atmosphere inside the heating furnace 2, and the surface temperature of the billet before it is charged into the heating furnace 2, can be used.

[0033] In piercing and rolling, as shown in FIG. 4, for example, a billet 1 is subjected to tilt-rotation piercing and rolling by upper and lower rolls 4 and a plug 5 arranged between the rolls 4. The number of rolls 4 is not limited to two. For example, the number of rolls 4 may be three or more. Shoes 6 may be arranged on the left and right to stabilize the piercing. Operation parameters in the piercing and rolling process include the roll gap d shown in FIGS. 4 and 5. r , shoe gap d s、 Plug outer diameter Dp Plug length L p Plug shape, plug advance amount L pa , the inclination angle θ of roll 4, and the roll outer diameter D R ·Roll shape such as roll inlet side angle α and roll outlet side angle β, shoe outer diameter D s Shoe caliber diameter D sc Shoe width W s The roll rotation speed, shoe rotation speed, etc. can also be used. The plug advance amount L pa It is desirable to include the plug advance amount L pa It was found that the amount of plug advance L pa This is because the deformation behavior of the workpiece changes, and the contact area between the roll and the workpiece changes, resulting in a change in the roll load. r , roll rotation speed, shoe gap d s , shoe rotation speed, plug advance amount L pa The tilt angle θ of the roll 4 may be a set value or an actually measured value.

[0034] The actual piercing load may be a load measured by a load cell 7 or the like installed in a piercing mill as shown in FIG. 4. The maximum value, average value, etc. of the load during piercing may be used, but it is preferable to use the maximum value in order to use a predicted value in order to avoid the risk of equipment damage and the risk of seizure between the piercing roll and the material. When measured values ​​are available for multiple rolls, it is preferable to use the maximum value, average value, etc. among them. Again, it is preferable to use the maximum value in order to use a predicted value in order to avoid the risk of equipment damage and the risk of seizure between the piercing roll and the material.

[0035] Step S2: The control unit 21 generates a drilling load prediction model based on the training data. A known learning method may be applied as the machine learning method. For example, a known machine learning method such as a neural network may be used as the machine learning method. Other examples of the machine learning method include decision tree learning, random forest, and support vector regression. Furthermore, the drilling load prediction model may be updated as appropriate using the latest training data.

[0036] Fig. 6 shows a conceptual diagram of the process for generating a piercing load prediction model. As shown in Fig. 6, a piercing load prediction model generation unit 211 collects billet attribute information record data, operation record data in the heating furnace 2, piercing-rolling operation record data, and piercing load record data from a host computer such as a process computer. The host computer collects these record data as normal operation record information. The collected data is stored in a database 212. In this way, multiple data sets of input data and output data are collected and saved in the database 212. It is preferable that the number of data stored in the database 212 be at least 80 or more, preferably 300 or more, and more preferably 500 or more.

[0037] In this embodiment, the database 212 created in this manner is used to acquire a plurality of learning data, where input actual data include at least one or more pieces of actual data selected from the attribute information of the billet, one or more pieces of operation actual data selected from the operation actual data of the heating furnace 2, and one or more pieces of operation actual data selected from the piercing-rolling operation parameters, and the actual piercing load using the input actual data is used as output actual data. The machine learning unit 213 generates a piercing load prediction model using the learning data. In other words, the piercing load prediction model is generated by machine learning using the acquired plurality of learning data. The functions of the piercing load prediction model generation unit 211 and the machine learning unit 213 are realized by the control unit 21. The database 212 is stored in the storage unit 22.

[0038] Next, with reference to Fig. 7, an operation related to the piercing load prediction by the information processing device 20 will be described. In summary, a piercing load prediction is executed using the piercing load prediction model generated in advance as described above. The piercing load prediction process is preferably executed 3 to 7 seconds after the billet to be predicted is extracted from the heating furnace 2 and before piercing and rolling is performed. This is because, if the piercing load is predicted to be excessive, an operator can change the settings to the roll gap, etc.

[0039] Step S11: The control unit 21 of the information processing device 20 acquires input data. Any method can be used for the input data acquisition process. For example, the control unit 21 may acquire the input data from a higher-level computer or the like via the communication unit 25. The input data includes one or more parameters selected from the attribute information of the billet 1, one or more parameters selected from the operation parameters of the heating furnace 2, and one or more parameters selected from the operation parameters of the piercing and rolling process.

[0040] Step S12: The control unit 21 predicts the drilling load based on the drilling load prediction model. Specifically, the control unit 21 inputs the input data acquired in step S11 into the drilling load prediction model and acquires output data to predict the drilling load.

[0041] Step S13: The control unit 21 outputs the drilling load predicted by the drilling load prediction model. Any method can be used to output the information. For example, the control unit 21 may display the information on the output unit 24.

[0042] A conceptual diagram of the prediction process using the piercing load prediction model is shown in Fig. 8. As shown in Fig. 8, attribute information of the billet and preset setting values ​​for the piercing-rolling operation are sent to the piercing load prediction unit 214 as information from a host computer. After the billet 1 to be predicted is extracted from the heating furnace 2, actual data on the operation parameters in the heating furnace 2 is sent to the piercing load prediction unit 214. Furthermore, using the generated piercing load prediction model, the piercing load prediction unit 214 outputs a predicted piercing load value.

[0043] Furthermore, as shown in FIG. 8 , the settings of the piercing-rolling mill can be reset using the piercing load predicted by the piercing load prediction model. In other words, the piercing-rolling control method according to this embodiment can reset the operation parameters in the piercing-rolling process so that the predicted piercing load falls within the target load range. An example of this method will be described below. The piercing-rolling operation condition setting unit 215 sets upper and lower limit values ​​of the piercing load to predetermined values ​​in advance. The upper limit value of the piercing load is set taking into consideration the load-bearing capacity of the equipment, the possibility of seizure between the piercing roll and the workpiece, etc. The lower limit value of the piercing load is set to a value necessary for stable piercing, taking into consideration the possibility that the pressure from the piercing roll on the workpiece will be too low, resulting in insufficient force to propel the workpiece.

[0044] The piercing-rolling operational condition setting unit 215 compares the upper and lower limit values ​​of the piercing load thus set in advance with the piercing load value predicted using the piercing load prediction model. If the predicted load does not exceed the upper limit value of the piercing load or fall below the lower limit value, piercing and rolling is performed with the initial settings. On the other hand, if the predicted load exceeds the upper limit value of the piercing load, the piercing and rolling operational condition setting unit 215 resets the set values ​​of the piercing and rolling conditions. Specifically, for example, the piercing and rolling operational condition resetting unit 216 increases the roll gap by about 1 to 10%. This is because a smaller reduction amount can reduce the piercing load. On the other hand, if the predicted load falls below the lower limit value, the piercing and rolling operational condition resetting unit 216 resets the set values ​​of the piercing and rolling conditions. Specifically, for example, the roll gap can be reduced by about 1 to 10%. This is because an increase in the reduction amount can increase the piercing load.

[0045] The reset piercing-rolling conditions may be used again as input data for the piercing load prediction model to calculate a predicted value of the predicted load, and after confirming whether the predicted load falls within the target load range, the set values ​​of the piercing-rolling conditions may be determined. By repeatedly performing such piercing load judgment, appropriate piercing-rolling conditions can be set even if the target piercing load range is set narrow, thereby enabling operation with improved equipment protection and piercing-rolling stability. The functions of the piercing load prediction unit 214, the piercing-rolling operational condition setting unit 215, and the piercing-rolling operational condition resetting unit 216 are realized by the control unit 21.

[0046] As described above, according to the present embodiment, it is possible to provide a method for generating a piercing load prediction model used in a piercing load prediction method, and a piercing load prediction method capable of predicting the piercing load with high accuracy. Furthermore, according to one embodiment of the present disclosure, it is possible to reduce the risk of equipment damage and the risk of seizure between the piercing roll and the material by using the piercing load prediction method.

[0047] (Example) An example will be described in which the present disclosure is applied to a piercing-rolling mill in a seamless pipe production line. Piercing load prediction according to the present disclosure was performed on a billet having an outer diameter of 200 to 300 mm and a length of 2500 mm. The piercing-rolling mill used was a rolling mill with two piercing rolls.

[0048] The information on the piercing load during piercing rolling, which is necessary as output performance data, was obtained by measuring the load applied to each piercing roll during piercing rolling using a load cell, and the larger of the measured values ​​for the two rolls was used as the maximum value during piercing.

[0049] The input data for the prediction model were the roll gap, plug advance, plug diameter, plug length, carbon content of the billet, the atmospheric temperature in the final heating furnace zone, and the time spent in the furnace. The carbon content was measured during the steelmaking process. The atmospheric temperature in the final heating furnace zone was measured using a thermometer installed on the furnace wall.

[0050] A neural network was used as the machine learning method, with two intermediate layers. A sigmoid function was used as the activation function. 1,000 pieces of operational performance data were prepared as above, with 800 pieces used as data for model creation (learning), and the remaining 200 pieces were used to verify prediction accuracy. The model prediction accuracy had an average error of 5.2 tons and a standard deviation of 7.1 tons. Furthermore, when the operator determined that the load prediction value for one of the 200 pieces was excessive and reset the roll gap, no problems such as equipment damage or seizure between the roll and the workpiece occurred.

[0051] Conventional operating methods do not provide a method for predicting the piercing load. Furthermore, operational parameters are not reset based on the piercing load prediction. Under the same conditions as in the above example, 200 piercings were performed without predicting the piercing load or resetting the operational parameters. An excessive piercing load occurred in one of the 200 piercings, resulting in seizure of the roll and the workpiece. If the piercing load had been predicted using this example, it may have been possible to prevent the occurrence of the excessive piercing load by changing the piercing-rolling operational parameters. As described above, by applying the prediction method according to one embodiment of the present disclosure, it is possible to reduce the risks of equipment damage, seizure of the roll and the workpiece, and the like.

[0052] Although the present disclosure has been described based on the drawings and examples, it should be noted that those skilled in the art may make various modifications and alterations based on the present disclosure. Therefore, it should be noted that these modifications and alterations are included in the scope of the present disclosure. For example, the functions included in each component or step can be rearranged so as not to be logically inconsistent, and multiple components or steps can be combined or divided into one. [Explanation of symbols]

[0053] 1 billet 2 Furnace 3 Piercing and rolling mill 4 rolls 5 Plug 6. Shoe 7 load cells 10 Hollow 20 Information processing equipment 21 Control Unit 22 Memory section 23 Input section 24 Output section 25 Communications Department 211 Drilling load prediction model generation unit 212 databases 213 Machine Learning Department 214 Drilling Load Prediction Section 215 Piercing and Rolling Operation Condition Setting Department d r Roll Gap d s Shoe Gap D p Plug Outer Diameter L p Plug Length L pa Plug advance amount θ Tilt angle D R Roll outer diameter α Roll entry side angle β Roll exit side angle D s Shoe outer diameter D sc Shoe caliber diameter W s Shoe Width

Claims

1. A piercing load prediction method for predicting a piercing load in a piercing-rolling process including a heating furnace for heating a billet and a piercing-rolling mill for piercing, the method being executed by an information processing device, comprising: A piercing load prediction method for predicting a piercing load using a piercing load prediction model generated by machine learning, the piercing load prediction model including, as input data, one or more parameters selected from attribute information of the billet, one or more parameters selected from operational parameters of the heating furnace, and one or more parameters selected from operational parameters of the piercing-rolling process, and the piercing load being output data.

2. The piercing load prediction method according to claim 1 , wherein the billet attribute information includes a chemical composition of the billet including at least one of a C content and a Si content.

3. The piercing load prediction method according to claim 1 , wherein the operational parameters in the piercing-rolling process include a plug advance amount.

4. a piercing load prediction method according to any one of claims 1 to 3, wherein a piercing load is predicted before the billet is charged into the piercing-rolling mill using attribute information of the billet, actual values ​​of operation parameters in the heating furnace, and set values ​​of operation parameters in the piercing-rolling process; A piercing-rolling mill control method that resets operation parameters in a piercing-rolling process so that a predicted piercing load falls within a target load range.

5. A method for generating a piercing load prediction model for predicting a piercing load in a piercing-rolling process including a heating furnace for heating a billet and a piercing-rolling mill for piercing, comprising: acquiring a plurality of learning data sets, each of which uses one or more pieces of performance data selected from the attribute information of the billet, one or more pieces of operation performance data selected from the operation performance data of the heating furnace, and one or more pieces of operation performance data selected from the operation performance data of the piercing-rolling process as input performance data, and which uses a performance of a piercing load as output performance data; A method for generating a drilling load prediction model, which generates a drilling load prediction model by machine learning using the acquired multiple pieces of learning data.

6. A method for generating a drilling load prediction model as described in claim 5, characterized in that machine learning selected from neural networks, decision tree learning, random forests, and support vector regression is used as the machine learning for generating the drilling load prediction model.

7. A pipe manufacturing facility having a heating furnace for heating a billet and a piercing-rolling mill for piercing, a piercing load prediction unit that predicts a piercing load in the piercing rolling mill using a piercing load prediction model, The drilling load prediction model is a model learned by machine learning, The input data includes one or more parameters selected from attribute information of the billet, one or more parameters selected from operation parameters in the heating furnace, and one or more parameters selected from operation parameters in a piercing-rolling process, A tube manufacturing facility characterized in that the learning model uses the piercing load of the piercing-rolling mill as output data.

Images