Temperature control device for hot rolling line

Abstract

A temperature control device for controlling the temperature of a rolled material in a hot rolling line comprises a rolling data acquisition unit, a model description unit, and an identification unit. The rolling data acquisition unit acquires rolling data including a temperature actual value of the rolled material. The model description unit describes a temperature model for calculating the temperature of the rolled material, and describes, as a function of the temperature of the rolled material, a water-cooling heat transfer rate, which is a parameter of the temperature model, with a curve or a polygonal line passing through the minimum heat flux point and the critical heat flux point. The identification unit identifies the water-cooling heat transfer rate by repeatedly deforming the curve or the polygonal line so that the difference between the temperature calculation value calculated using the temperature model and the temperature actual value is reduced.

Description

Temperature control device for hot rolling lines

[0001] The present disclosure relates to a temperature control device for controlling the temperature of a rolled material in a hot rolling line, and more particularly to a technique for identifying a water-cooling heat transfer coefficient, which is a parameter of a temperature model used to predict and calculate the temperature of the rolled material.

[0002] Known methods of controlling the temperature of a rolled material in a hot rolling line include finish entry temperature control, finish exit temperature control, and coiling temperature control. Finish entry temperature control uses a water injection device provided between a roughing mill and a finishing mill to control the temperature at the entry side of the finishing mill (FET: Finisher Enter Temperature). Finish exit temperature control uses a water injection device (ISC: Inter Stand Coolant) provided between stands of the finishing mill to control the temperature at the delivery side of the finishing mill (FDT: Finisher Delivery Temperature). Coiling temperature control uses a water injection device provided on a run-out table to control the coiling temperature (CT: Coiling Temperature), which is the temperature at the entry side of the coiler.

[0003] In these temperature control processes, the temperature of the rolled material is typically predicted and calculated using the following temperature model. Specifically, when a material (rolled material) is heated to a high temperature, heat is lost through heat transfer from the material surface, such as water cooling, air cooling, and radiation. At the same time, heat is transferred from the rolled material to the rolling rolls, heat is input through friction between the rolled material and the rolling rolls, heat is input due to deformation caused by processing, and heat is generated due to phase transformation (heat balance). Heat conduction within the thick rolled material must also be taken into account. Furthermore, the physical properties of the rolled material, such as its specific heat and thermal conductivity, are also factors that determine the final temperature. These phenomena and parameter values ​​have been largely clarified through physical formulations and laboratory measurements.

[0004] On the other hand, the water injection device used in winding temperature control is more complex than other water injection devices, and there are cases where 100 to 300 water injection headers are arranged above and below the run-out table. Even if the individual heat balance for each water injection header is described and calculated using a model or parameters, the final winding temperature often does not match the actually measured temperature. Note that winding temperature control has a wider temperature range to control than other temperature control methods. Therefore, the following explanation will be given using winding temperature control as an example.

[0005] Patent Document 1 listed below discloses a temperature control device. In this device, the values ​​of multiple correction terms in a temperature model are changed so that the coiling temperature approaches an actual value, and the changed values ​​are stored in a learning table as learned values. When predicting the temperature of a new rolled material, the learned values ​​are extracted from the learning table and reflected in the temperature model.

[0006] Furthermore, Patent Document 2 listed below discloses a method for predicting the temperature of a metal sheet in hot rolling. This method finds that as the temperature of the cooling water decreases, the curve of the heat transfer coefficient (corresponding to the "heat transfer coefficient" described later) shifts to the higher temperature side, and the transition boiling onset temperature (corresponding to the "minimum heat flux (MHF) point" described later) shifts in an increasing direction. Based on this finding, regression of the actual value against the predicted value of the coiling temperature is performed manually.

[0007] International Publication No. 2014 / 006681 Japanese Patent Application Laid-Open No. 2004-331992

[0008] In both Patent Documents 1 and 2, it is recognized that the heat transfer due to water cooling has the greatest effect on the coiling temperature, and an attempt is made to learn the water-cooling heat transfer coefficient using actual temperature values.

[0009] However, the run-out table measures the temperature of the rolled material at only two or three locations, which means that sufficient actual temperature values ​​cannot be obtained. Furthermore, the time allocated to learning calculations in the control computer is short, making it difficult to sufficiently correct multiple parameters within the temperature model through short-term learning. For this reason, it is difficult to accurately learn the water-cooling heat transfer coefficient for each water injection header of the water injection device. As a result, the temperature model obtained through learning cannot be said to be suitable for actual cooling. For these reasons, when sufficient actual temperature values ​​cannot be obtained, it is desirable to accurately identify the water-cooling heat transfer coefficient of the temperature model before learning.

[0010] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a temperature control device for a hot rolling line that can accurately identify the water-cooling heat transfer coefficient of a temperature model and obtain a temperature model suitable for actual cooling even when the learning of the temperature model does not function sufficiently, such as when sufficient actual temperature values ​​cannot be obtained.

[0011] A first aspect of the present disclosure relates to a temperature control device for controlling the temperature of a rolled material in a hot rolling line. The temperature control device includes a rolling data acquisition unit, a model description unit, and an identification unit. The rolling data acquisition unit acquires rolling data including actual temperature values ​​of the rolled material. The model description unit describes a temperature model for calculating the temperature of the rolled material, and describes the water-cooling heat transfer coefficient, which is a parameter of the temperature model, as a curve or polygonal line passing through a minimum heat flux point and a critical heat flux point as a function of the temperature of the rolled material. The identification unit identifies the water-cooling heat transfer coefficient by repeatedly deforming the curve or polygonal line so as to reduce the difference between the calculated temperature value calculated using the temperature model and the actual temperature value.

[0012] The second aspect has the same features as the first aspect, but further includes the following: the hot rolling line includes a water injection device that injects cooling water, and the identification unit is configured to deform the curve or polygonal line by moving the temperature of the minimum heat flux point as a function of the cooling water temperature while keeping the critical flux point fixed.

[0013] The third aspect has the following characteristics in addition to the second aspect: the identification unit is configured to set a central cell using the temperature of the minimum heat flux point before identification in a direction in which the temperature of the minimum heat flux point is searched for as a range in which the relationship between the minimum heat flux point and the water-cooling heat transfer coefficient gain is searched for, and to set, around the central cell, a plurality of peripheral cells having different coolant temperatures or temperatures of the minimum heat flux points and different water-cooling heat transfer coefficient gains from those of the central cell, calculate an evaluation value that is the difference between a calculated temperature value and an actual temperature value for each cell, move the central cell in a direction in which the evaluation value becomes smaller, and repeat the calculation of the evaluation value and the movement of the central cell.

[0014] The fourth aspect has the same features as the second aspect, but further includes the following: the water injection device has a plurality of types of water injection headers, and the identification unit is configured to identify the water-cooling heat transfer coefficient for each type of water injection header.

[0015] The fifth aspect has the following feature in addition to the fourth aspect: the identification unit is configured to select a reference water injection header from a plurality of types of water injection headers, identify a water-cooling heat transfer coefficient of the reference water injection header, and identify a water-cooling heat transfer coefficient of a type of water injection header different from the reference water injection header by a multiple of the water-cooling heat transfer coefficient of the reference water injection header.

[0016] A sixth aspect has the following features in addition to any one of the first to fifth aspects: the rolling data includes rolling conditions, and the rolling conditions include a steel type and a size of the rolled material; the temperature control device further includes a verification unit configured to verify whether the identified water-cooling heat transfer coefficient is valid by using rolling data having rolling conditions similar to those used by the identification unit for identification; and the model description unit is configured to apply the identified water-cooling heat transfer coefficient to the temperature model when the verification unit verifies that the identified water-cooling heat transfer coefficient is valid.

[0017] The seventh aspect has the following features in addition to the sixth aspect. The temperature control device is composed of a process control computer that executes setting calculations for a control object in a rolling process of a hot rolling line. The identification unit is configured to perform identification online when a first number of rolling data pieces having similar rolling conditions are acquired, and to perform identification offline when a second number of rolling data pieces greater than the first number are acquired. The first number is set so as not to interfere with the setting calculations.

[0018] In the first aspect, the water-cooling heat transfer coefficient, which is a parameter of the temperature model, is described by a curve or polygonal line passing through the minimum heat flux point and the critical heat flux point, and the water-cooling heat transfer coefficient is identified by repeatedly deforming the described curve or polygonal line. Therefore, even when the temperature model learning does not function sufficiently, such as when sufficient actual temperature values ​​are not obtained, the water-cooling heat transfer coefficient of the temperature model can be accurately identified, and by applying the identified water-cooling heat transfer coefficient, a temperature model suitable for actual cooling can be obtained.

[0019] According to the second aspect, the curve or polygonal line can be modified by moving the temperature of the minimum heat flux point as a function of the coolant temperature while keeping the critical flux point fixed.

[0020] According to the third aspect, the water-cooling heat transfer coefficient can be identified by repeatedly calculating the evaluation values ​​of the central cell and the peripheral cells and moving the central cell.

[0021] According to the fourth aspect, even when the water injection device has multiple types of water injection headers, the water-cooling heat transfer coefficient is identified for each type of water injection header, thereby preventing a decrease in the accuracy of the temperature model. In this case, as in the fifth aspect, by identifying the water-cooling heat transfer coefficient of a type of water injection header different from the basic water injection header as a multiple of the water-cooling heat transfer coefficient of the basic water injection header, an increase in the calculation load can be suppressed.

[0022] According to the sixth aspect, the water-cooling heat transfer coefficient verified as appropriate by the verification unit is applied to the temperature model, so that a temperature model suitable for actual cooling can be obtained more reliably.

[0023] According to the seventh aspect, the water-cooling heat transfer coefficient can be identified without interfering with the setting calculation by the process control computer.

[0024] 1 is a schematic diagram showing the configuration of a hot rolling line to which a temperature control device according to an embodiment is applied; FIG. 2 is a schematic diagram showing a temperature control device according to an embodiment; FIG. 3 is a diagram showing an example of a water-cooling heat transfer coefficient; FIG. 4 is a diagram showing changes in the cooling heat transfer coefficient depending on the water temperature and classification of identification targets; FIG. 5 is a diagram showing an example of an identification method when the water-cooling heat transfer coefficient is described by a broken line; FIG. 6 is a diagram showing an example of an identification calculation for the water-cooling heat transfer coefficient; and FIG. 7 is a flowchart for explaining the flow of an identification method and a verification method.

[0025] Hereinafter, a temperature control device for a hot rolling line according to an embodiment of the present disclosure will be described with reference to the drawings. Note that elements common to the various drawings will be assigned the same reference numerals and redundant description will be omitted.

[0026] Fig. 1 is a schematic diagram showing a configuration of a hot rolling line to which a temperature control device according to an embodiment is applied, and Fig. 2 is a schematic diagram showing a temperature control device according to an embodiment.

[0027] The hot rolling line 1 is equipped with, as main rolling equipment, a heating furnace 11, a roughing mill 12, a finishing mill 13, a run-out table 14, a water injection device (also referred to as a "cooling device") 15, and a winder 16.

[0028] The heating furnace 11 heats the rolled material Mr, which is a metallic material, to a predetermined temperature. The roughing mill 12 has one or more stands R1 (one in the example shown in FIG. 1 ) and reverse-rolls the rolled material (slab) Mr extracted from the heating furnace 11. The finishing mill 13 is a tandem rolling mill having multiple stands F1 to Fn (e.g., n = 6 or 7) that rolls the rolled material Mr in one direction. The run-out table 14 has multiple table rolls (not shown) arranged in parallel along the rolling direction of the rolled material Mr and transports the rolled material (strip) Mr that has exited the final stand Fn to the winder 16. The length of the run-out table 14 is, for example, 50 to 100 meters. A water injection device 15 is attached to the run-out table 14. The water injection device 15 has multiple water injection headers W1 to Wn (e.g., n = 100 to 300) arranged above and below the run-out table 14. Although not shown, each water injection header Wj (j = 1 to n) has, for example, a water injection nozzle and a water injection valve (plug), and is configured to inject water into the rolled material Mr from the water injection nozzle by opening the water injection valve. The winder 16 winds the rolled material Mr cooled by the water injection device 15 into a coil. For example, the rolled material (slab) Mr extracted from the heating furnace 11 has a thickness of 200 mm to 250 mm, a width of 600 mm to 2000 mm, and a length of 6 m to 10 m. On the other hand, the rolled material (strip) exiting the finishing rolling mill 13 has a thickness of 1.2 mm to 25.0 mm, a width of 600 mm to 2000 mm, and a length of approximately 100 m to 2 km.

[0029] A plurality of sensors are arranged in the hot rolling line 1 to measure the state of the rolled material Mr. The plurality of sensors includes a plurality of thermometers 21 to 26 that measure the temperature of the rolled material Mr. The thermometer 21 is a roughing mill entry thermometer RET arranged on the entry side of the roughing mill 12. The thermometer 22 is a roughing mill delivery thermometer (RDT) arranged on the exit side of the roughing mill 12. The thermometer 23 is a finishing mill entry thermometer (FET) arranged on the entry side of the finishing mill 13. The thermometer 24 is a finishing mill delivery thermometer (FDT) arranged on the exit side of the finishing mill 13. The thermometer 25 is an intermediate thermometer (MT) arranged in the middle of the run-out table 14. The thermometer 26 is a coiling thermometer CT arranged on the entry side of the coiler 16. In the hot rolling line 1, other sensors (not shown) such as a thickness gauge, a width gauge, and a load cell for measuring the rolling load of each stand are arranged.

[0030] The hot rolling line 1 is operated by a control system using a computer. The computer includes a higher-level computer (also referred to as a host computer) 3 and a process control computer 4, which are connected to each other via a network. An interface screen 5, which is an operation screen for an operator, is connected to the process control computer 4 via the network. The operator can input control conditions and the like on the interface screen 5.

[0031] The process control computer 4 performs setting calculations and controls of control targets in a series of rolling processes based on rolling information (e.g., steel type, product thickness, etc.) input from the host computer 3. The process control computer 4 includes a temperature control device 40, which will be described later.

[0032] The temperature control device 40 includes a rolling data acquisition unit 41 , a temperature model description unit 42 , an identification unit 43 , a verification unit 44 , a water injection control unit 45 , and a recording unit 46 .

[0033] The rolling data acquisition unit 41 acquires rolling data from the hot rolling line 1. The rolling data includes, for example, rolling information such as the steel type and size (thickness, width) of the rolled material Mr, as well as the amount of water injected from each cooling header W1 to Wn into each segment, the actual temperature value of each segment, and the conveying speed of the rolled material Mr when the rolled material Mr is virtually divided into multiple segments having a predetermined length (e.g., 1 m to 10 m) from the leading end to the trailing end. The rolling data acquisition unit 41 records the acquired rolling data in the recording unit 46 in a predetermined format. The rolling data acquisition unit 41 may reduce noise by filtering the rolling data. The rolling data acquisition unit 41 may acquire rolling data for the entire length of the rolled material Mr, i.e., for all segments, or may be configured to acquire rolling data for a specific segment. The rolling data acquisition unit 41 sends the rolling data to the temperature model description unit 42, the identification unit 43, and the verification unit 44, respectively.

[0034] The temperature model description unit 42 obtains rolling data from the rolling data acquisition unit 41, obtains identification results (water-cooling heat transfer coefficient) from the identification unit 43, and obtains verification results from the verification unit 44. The temperature model description unit 42 has internally a temperature model including the water-cooling heat transfer coefficient and parameters. Based on the verification results obtained from the verification unit 44, the temperature model description unit 42 applies (reflects) the water-cooling heat transfer coefficient obtained from the identification unit 43 to the temperature model.

[0035] The identification unit 43 identifies the water-cooling heat transfer coefficient by a method described later. The identification unit 43 sends the identification result (water-cooling heat transfer coefficient) to the verification unit 44 and the temperature model description unit 42.

[0036] The verification unit 44 verifies whether the water-cooling heat transfer coefficient identified by the identification unit 43 is appropriate. As will be described in detail later, the verification unit 44 verifies that the water-cooling heat transfer coefficient is appropriate if the prediction (calculation) accuracy of the rolled material temperature when the identified water-cooling heat transfer coefficient is applied to the temperature model is improved compared to the accuracy when the current water-cooling heat transfer coefficient is applied to the temperature model. The verification unit 44 sends the verification result to the temperature model description unit 42. If the accuracy is improved, the water-cooling heat transfer coefficient identified by the identification unit 43 is applied (adopted) to the temperature model in the temperature model description unit 42.

[0037] The water injection control unit 45 controls the injection of water from each of the water injection headers W1 to Wn of the water injection device 15 based on the calculated temperature value of the rolled material Mr calculated using the temperature model. The water injection control unit 5 functions as a rolled material temperature control unit.

[0038] The recording unit 46 records various information such as rolling data, parameters such as temperature models and water-cooling heat transfer coefficients, calculation results and calculation processes related to identification, verification results and verification processes, etc. The recording unit 46 also functions as a log recording unit, and is configured to record information obtained from the rolling data acquisition unit 41, temperature model description unit 42, identification unit 43, and verification unit 44 as a log, and to output the recorded information upon request.

[0039] Water-cooled heat transfer coefficient h ｗ Before explaining the identification of the water-cooled heat transfer coefficient h ｗ An example of a temperature model including the above as a parameter will be described below. In the following, the heat balance related to the rolling of the rolled material Mr will not be described, that is, the heat generated by the rolling, the heat generated by friction with the rolling rolls, and the heat extraction to the rolling rolls will not be described, but the heat balance not related to the rolling, such as water cooling, air cooling, and radiation, will be described.

[0040] The temperature model calculation here can use the well-known finite difference method (FDM). The rolled material Mr has multiple nodes in the thickness direction, and in the temperature model calculation, not only is the heat transfer at the nodes on the surface calculated, but also the heat conduction between the nodes. In addition to the thickness direction of the rolled material Mr, nodes may also be provided in the width direction, and temperature calculations in two dimensions may be performed. The heat flow Q due to the heat conduction and heat transfer at the node number i is ｉ If the sum of [W] is found, the temperature ΔTi of the node is expressed by the following equation (1).

[0041] where ρ is the density [kg / mm ３ ] and C ｐ is the specific heat [J / kg / ℃], and V ｉ is the volume of the node [mm ３ ], and Δt is time [s].

[0042] heat flow Qｉ includes heat transfer on the top and bottom surfaces of the rolled material Mr, and heat conduction inside the rolled material Mr. Heat transfer refers to the transfer of heat between the rolled material Mr and a different substance, such as water, while heat conduction refers to the transfer of heat within the same substance. Heat conduction is expressed by the following Fourier equation (2):

[0043] where k is the thermal conductivity [W / mm / K] and A is the area of ​​the contacting nodes [mm ２ ], and d is the distance between nodes [mm].

[0044] Heat flow Q due to radiation from the surface of the rolled material Mr ｒａｄ is expressed by the following formula (3).

[0045] where ε is the emissivity [-] and σ is the Stephan-Boltzmann constant [W / mm ２ / K ４ ] and T ｓｕｒｆ is the surface temperature of the rolled material Mr, and T ｓｕｒｒｏｕｎｄ is the temperature around the rolled material Mr.

[0046] Heat flow Q due to heat transfer by water cooling (hereinafter also referred to as "water cooling heat transfer") ｗ is expressed by the following equation (4): Heat flow Q due to heat transfer by air-cooling convection (hereinafter also referred to as "air-cooling heat transfer") ａ is expressed by the following equation (5).

[0047] Here, h ｗ is the water cooling heat transfer coefficient [W / mm ２ / K], and A ｗ is the area through which heat is transferred, and T ｗ is the ambient water temperature. ａ is the air-cooled heat transfer coefficient [W / mm ２ / K], and A ａ is the area through which heat is transferred, and T ａ is the ambient air temperature.

[0048] In addition to the above-mentioned heat conduction and heat transfer (water-cooled heat transfer and air-cooled heat transfer), it is also possible to describe the heat transfer due to the phase transformation of the rolled material Mr. When the rolled material Mr is cooled from a high temperature, for example, if the rolled material Mr undergoes a phase transformation at around 700°C, heat is generated around that temperature, causing a decrease in cooling efficiency. This can be described as an increase in specific heat.

[0049] Of the above-mentioned heat transfer factors, water cooling, air cooling, and radiation, the water cooling heat transfer has the greatest influence. Hereinafter, the water cooling heat transfer Q described by the above formula (4) will be ｗ This article explains:

[0050] The heat flow due to water cooling heat transfer is expressed as the water cooling heat transfer coefficient h ｗ , area A through which heat is transferred ｗ , the surface temperature of the rolled material T ｓｕｒｆ and water temperature T ｗ The water cooling heat transfer coefficient h ｗ represents the efficiency of heat transfer. ｗ A large value means that the material is easily cooled when cooled.

[0051] Figure 3 shows the water-cooled heat transfer coefficient h ｗ 3 is a diagram showing an example of the water-cooling heat transfer coefficient h ｗAlthough the temperature is depicted as a curve, it can also be approximated by a broken line (see FIG. 5 ) described later. In FIG. 3 , point (a) on the highest temperature side is the cooling start point, which corresponds to the finishing mill exit temperature FDT in the case of coiling temperature (CT) control. Point (b) is the minimum heat flux (MHF) point, where film boiling transitions to transition boiling. Point (c) is the critical heat flux (CHF) point, where the heat flux is maximized. Nucleate boiling occurs below the CHF point (c). Point (d) is the boiling start point, which is the boiling end point in the case of cooling. Film boiling is a state in which steam is generated between the surface of an object and a refrigerant such as water, resulting in poor cooling efficiency. For example, when a drop of water is placed on a hot frying pan, a vapor film forms between the metal and the drop, causing the drop to move quickly across the pan. This state corresponds to film boiling. Transition boiling is a state in which the refrigerant boils violently and removes a lot of heat from the object. In the example of the frying pan, this is the state in which water droplets jump up and down vigorously on the surface of the frying pan. Nucleate boiling is a state in which the temperature of the object drops and the refrigerant comes into close contact with the object, which is known as a wet state.

[0052] Here, the cooling water temperature T ｗ Consider the effect of water temperature T ｗ It is clear that when the water temperature T ｗ If the surface temperature T ｓｕｒｆ The difference between (T ｓｕｒｆ -T ｗ ) becomes larger, and the heat flow Q ｗ However, the surface temperature T ｓｕｒｆ is several hundred degrees Celsius, while the water temperature T ｗ The water temperature T ｗ If the temperature changes by just a few degrees, the above difference (T ｓｕｒｆ -T ｗ ) does not change significantly. ｗ When the temperature T ｗ The main factor affecting the water temperature is ｗ When the temperature T ｗIf the MHF point (b) is higher, the MHF point (b) moves to the lower temperature side. If the MHF point (b) moves to the higher temperature side, the temperature at which film boiling, which has poor cooling efficiency, transition boiling, which has good cooling efficiency, becomes higher, and cooling becomes easier overall.

[0053] Next, referring to FIGS. 4 to 7, the water-cooling heat transfer coefficient h ｗ The identification method and the verification method executed by the verification unit 44 will be described. ｗ Water temperature T ｗ 5 is a diagram showing the change due to the water-cooling heat transfer coefficient h ｗ 6 is a diagram showing an example of an identification method when the water-cooling heat transfer coefficient h ｗ 7 is a flowchart illustrating the flow of the identification method and the verification method.

[0054] As described above, the MHF point (b) is the water temperature T ｗ That is, the water temperature T ｗ It is thought that the MHF point (b) moves. On the other hand, the CHF point (c) is near 200°C and is thought to be unaffected by the water temperature. Therefore, the CHF point (c) can be fixed.

[0055] As shown in FIG. 5, the coordinates of points (a) to (d) in FIG. 3 are respectively (T ａ , h ａ ), (T ｂ , h ｂ ), (T ｃ , h ｃ ), (T ｄ , h ｄ ) In this case, the line Ldc between points (d) and (c) on the low temperature side is expressed by the following formula (6). The line Lcb between points (c) and (b) is expressed by the following formula (7). The line Lba between points (b) and (a) is expressed by the following formula (8).

[0056] Here, as shown in FIG. 5, the CHF point (c) is fixed, and the MHF point (b) is located on the line Lba between the points (b) and (a).＊ ) (T ｂ ＊ , h ｂ ＊ In this case, h ｂ ＊ The value of is calculated by the following equation (9).

[0057] At this time, point (c)-(b ＊ ) Line Lcb ＊ is expressed by the following equation (10).

[0058] In this way, the water temperature T ｗ When changes, T ｂ T ｂ ＊ By changing the temperature to ｗ The MHF point (b ＊ ), while the CHF point (c) does not change.

[0059] Here, the water cooling heat transfer coefficient h ｗ When identifying the water cooling heat transfer coefficient h ｗ If it were possible to describe this with an arbitrary curve or polygonal line, the calculation load would be high. Therefore, as shown in Figure 4, the temperature is divided into two regions (divisions) from the MHF point (b), namely, the low temperature side and the high temperature side, and the shape of the curve or polygonal line is not changed in each temperature region, and the water cooling heat transfer coefficient h ｗ (hereinafter referred to as the "water-cooling heat transfer coefficient gain") can be changed. This reduces the calculation load on the process control computer 4.

[0060] Water-cooled heat transfer coefficient h ｗ An optimization method can be used to identify the water-cooling heat transfer coefficient h. The optimization method is a method for setting an objective function and minimizing or maximizing its evaluation value. The objective function is a mathematical expression that describes the relationship between the variable to be minimized or maximized and one or more variables that describe that variable. ｗWhen identifying the temperature, in order to minimize the difference between the actually measured temperature value and the calculated temperature value, the calculated coiling temperature CT is described, for example, by the temperature model shown in the above equations (1) to (5), as expressed by the following equation (11), and the water-cooled heat transfer coefficient h ｗ Regarding this, the MHF point (b) shown in FIG. 4 is moved and the water-cooled heat transfer coefficient gains on the low-temperature side and high-temperature side are changed.

[0061] Calculated coiling temperature (CT) = f (water-cooled heat transfer coefficient h ｗ , temperature model, each parameter)...(11)

[0062] As the evaluation value, the error may be defined as "the difference between the actual value of the winding temperature CT and the calculated value of the winding temperature," and the evaluation value may be expressed as a mean absolute error (MAE) or a mean squared error (MSE). The MAE is obtained by calculating the absolute value of each error in all the data to be evaluated and averaging the calculated absolute values. The MSE is obtained by squaring each error in all the data to be evaluated and averaging the squared values.

[0063] Various algorithms for optimization methods have been published, and examples that can be used include line search methods such as bisection, steepest descent, Newton's method, quasi-Newton's method, and genetic algorithms (GA). Even though they are optimization methods, they do not always yield a globally optimal solution, and may fall into a local solution. For example, bisection has a low computational load, but it is difficult to obtain a globally optimal solution. GA can easily obtain a globally optimal solution, but it has a high computational load.

[0064] In this embodiment, the water-cooling heat transfer coefficient h ｗ The timing for performing online identification and verification (described later) of is when a first number (N1) of rolling data for similar steel types and sizes has been collected. The initial value of N1 can be set to a value that does not cause any problems in the setting calculations by the process control computer 4, and can be set to, for example, a value of 10 or less (Step S1). ｗThe timing for performing offline identification and verification is when a second number (N2) of rolling data have been collected. The initial value of N2 can be set to, for example, a value greater than 10, i.e., a value greater than N1. In this way, by setting the number N1 of rolling data used for identification and the timing for performing identification in consideration of the calculation load on the process control computer 4, the margin for applicability of the optimization method can be increased. "Online" refers to a state in which the calculation load on the process control computer 4 is an issue, i.e., a state in which the process control computer 4 is performing setting calculations and control. "Offline" refers to a state in which the calculation load on the process control computer 4 is not an issue, i.e., a state in which the process control computer 4 is not performing setting calculations and control.

[0065] As shown in FIG. 6, first, the water-cooled heat transfer coefficient h ｗ In this example, in order to simplify the identification, the water temperature T ｗ The number of search ranges m1 for the MHF point (b) based on the above formula is set to 3, and the number of search ranges m2 for large and small water-cooling heat transfer coefficient gains is set to 3 (step S1). Note that the size of the initial search range Ri, i.e., the values ​​of m1 and m2, are not limited to these numerical values ​​and can be set taking into consideration the calculation load of the process control computer 4. The temperature of the MHF point and the water-cooling heat transfer coefficient gain, which are representative values ​​of each cell (each stage) divided by the number of search ranges m1 and m2, are fixed (step S1). For example, the temperature of the MHF point (b) can be set to 550, 600, or 650°C, and the water-cooling heat transfer coefficient gain can be set to 1.1, 1.0, or 0.9. The water temperature T ｗ The relationship between the water temperature T and the MHF point can be determined by a prior analysis. However, when the number of rolled materials Mr used as actual data (rolling data) is small, ｗ Since the change in m may be small, it may not be possible to set a plurality of search ranges m1. Note that in step S1, the search range expansion count L, which will be described later, is also set.

[0066] Depending on the hot rolling line 1, the water injection headers W1 to Wn may include multiple types of water injection headers. For example, there may be a water injection header for strong cooling with a relatively large amount of water injection, a water injection header for normal cooling, and a water injection header for fine cooling with a relatively small amount of water injection. In addition, the amount of water injection may differ between the water injection headers on the upper side and the lower side of the rolled material Mr. In such cases, the water cooling heat transfer coefficient h ｗ For example, as shown in FIG. 6, the water injection header for normal cooling is identified by moving the center cell (the magnitude of the water cooling heat transfer coefficient gain, the temperature movement of the MHF point), which will be described later, and for other types of water injection headers, the ratio of the cooling effect to that of the water injection header for normal cooling is changed as the water cooling heat transfer coefficient gain. For example, it is assumed that the water injection header for strong cooling has a cooling effect 1.5 times that of the water injection header for normal cooling. In this case, after identifying the water injection header for normal cooling, the water cooling heat transfer coefficient h of the water injection header for strong cooling is ｗ is the water cooling heat transfer coefficient h ｗ This allows identification using a multiple, such as 1.5 times or 1.6 times of the original value. This reduces the calculation load on the process control computer 4.

[0067] Returning to Figure 6, the water-cooled heat transfer coefficient h before identification ｗ Using the above equation (11), the coiling temperature CT of the cell at the center of the search range Ri (hereinafter referred to as the "center cell") is calculated, and an evaluation value (for example, MAE) of the center cell is calculated (step S3). For the center cell, the temperature at the MHF point (b) before identification is used. The center cell is set as the starting point for calculating the evaluation value. Next, the coiling temperatures CT of eight cells (hereinafter referred to as the "peripheral cells") set around the center cell are calculated, and an evaluation value for each peripheral cell is calculated (step S4). The peripheral cells and the center cell are related by the cooling water temperature T ｗ Alternatively, the temperature at the MHF point (b) and the water-cooled heat transfer coefficient gain are different.

[0068] Here, the process control computer 4 that calculates the evaluation value executes various controls related to rolling. In order to reduce the calculation load on the process control computer 4, the number of data items is set to N1 in the online calculation of the evaluation value. On the other hand, if the number of data items is N1 or less, there are cases where a sufficient search cannot be performed by the optimization described above, that is, the search range cannot be sufficiently expanded. Therefore, in addition to selecting the optimization method and specifying N1 and m1 and m2, it is preferable to set in advance the number of times L the search range is expanded in the initial setting of the above step S1. At this time, the calculation is naturally performed within the range of the upper and lower limit values ​​set for each parameter.

[0069] If the search range expansion count L is not reached or if there is an improvement in the evaluation value in the search range, the search range is expanded (step S6). In the example shown in FIG. 6, the evaluation values ​​of nine cells in the search range Ri are calculated, and since there is an improvement in the evaluation value, the search range Ri is expanded. The search range Ri is expanded toward the search range Ri+1, as indicated by the arrow in the figure, so that the cell with the smallest evaluation value (=8) in the search range Ri becomes the center cell. For the search range Ri+1, the evaluation value of each cell is calculated in the same way as for the search range Ri. Here, since the evaluation values ​​of four cells, including the center cell of the search range Ri+1, have already been calculated in the search range Ri, new evaluation values ​​are calculated for the remaining five cells. Once the evaluation values ​​of the five cells are calculated, the search range Ri+1 is expanded in the direction indicated by the virtual arrow so that the cell with the smallest evaluation value (=7) among the nine cells in the search range Ri+1 becomes the center cell. By repeating the expansion of the search range and the calculation of the evaluation value in this manner, an optimal solution for the evaluation value can be obtained. When the optimal solution for the evaluation value is obtained, the identification result and the evaluation value are recorded in the recording unit 46 (step S7). ｗ Includes:

[0070] Depending on the optimization calculation method used, the calculation for identification may involve the water cooling heat transfer coefficient h ｗis changed by trial and error to search for an optimal solution, which may result in an enormous computational load. For this reason, the identification data used in the optimization method can be limited to a limited number of data extracted from the entire length of the rolled material Mr. When the identification calculation is completed, the time required for the identification calculation is also recorded in step S7. If the maximum or average value of the time required for multiple identification runs falls below a predetermined time, N1, m1, and m2 set in step S1 can be increased. Furthermore, if a better optimization method that is more likely to obtain a global solution is available, even if it takes longer to calculate, it can be changed to that method.

[0071] Assume that a sufficient number N2 (N2>N1) of rolling data for rolled materials Mr of similar steel types and sizes has been obtained. When N1 rolling data have been obtained, online identification (as well as verification, which will be described later) is performed as described above. However, when N2 rolling data are to be identified, if online identification (and verification) takes too much calculation time, it is preferable to perform offline identification and verification. This is possible by recording the calculation load in step S7 above each time the above-mentioned identification calculation is performed.

[0072] Next, verification of the identification results will be described. Because the calculation time required for verification is shorter than that for identification, the verification data imported in step S8 can use total length data of an already rolled rolled material Mr of a steel type and size similar to that of the rolled material Mr used for identification, other than the rolled material Mr used for identification. The verification data includes an already recorded actual value of the coiling temperature CT and an already recorded calculated value of the coiling temperature CT. Therefore, the difference between these actual values ​​and the calculated value is calculated as an evaluation value b (before) (step S8).

[0073] Next, the identified results are applied to the rolling data for verification for which the evaluation value b was calculated, the coiling temperatures CT are calculated over the entire lengths of all rolled materials to be verified, and the difference between the coiling temperatures CT and the actual values ​​is calculated as an evaluation value a (after) (step S9). If the evaluation value a is improved compared to the evaluation value b (Yes in step S10), it is determined that the identification results are good, and the identification results are adopted for rolling subsequent materials of similar steel types and sizes (step S11). As a result, the identified water-cooling heat transfer coefficient h ｗ is applied to the temperature model and is used for the control of the water injection device 15 by the water injection control unit 45. That is, the identified water-cooling heat transfer coefficient h ｗ The amount of water injected from each of the water injection headers W1 to Wn of the water injection device 15 is controlled so that the calculated value of the coiling temperature CT calculated using the temperature model to which the above is applied becomes the actual value of the coiling temperature CT. Finally, the verification result, the evaluation value a, etc. are recorded in the recording unit 46 (step S12).

[0074] While the above description refers to similar steel types and sizes, whether or not they are similar can be determined, for example, by the following criteria. Specifically, the steel type refers to the steel type classifications used by companies that own the hot rolling line 1, or the steel type classifications established by standards such as JIS. Furthermore, even if steel is roughly classified into low carbon steel, medium carbon steel, high carbon steel, alloy steel, special steel, etc., cooling specifications may be determined separately, and it may become necessary to add cooling patterns and the like to the steel type classifications. The size refers to the plate thickness classification, plate width classification, etc., used in the setting calculations by the process control computer 4.

[0075] The specific structure of the process control computer 4 is not limited, and may be as follows, for example. FIG. 8 is a diagram showing an example of the hardware configuration of the process control computer 4. The functions of the process control computer 4, including the temperature control device 40, can be realized by the processing circuit shown in FIG. 8. This processing circuit may be dedicated hardware 4a. This processing circuit may include a processor 4b and a memory 4c. This processing circuit may be partially formed as dedicated hardware 4a and further include a processor 4b and a memory 4c. In the example of FIG. 8, part of the processing circuit is formed as dedicated hardware 4a, and the processing circuit also includes a processor 4b and a memory 4c. At least a portion of the processing circuit may be at least one dedicated hardware 4a. In this case, the processing circuit may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. The processing circuit may include at least one processor 4b and at least one memory 4c. In this case, each function of the process control computer 4 is realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in memory 4c. The processor 4b reads and executes the programs stored in memory 4c to realize the functions of each unit of the temperature control device 40. The processor 4b is also called a CPU (Central Processing Unit), processing device, arithmetic unit, microprocessor, microcomputer, or DSP. The memory 4c may be, for example, a non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, or EEPROM. The memory 4c may also serve as the recording unit 46. In this way, the processing circuit can realize each function of the temperature control device 40 by using hardware, software, firmware, or a combination of these.

[0076] As described above, according to the present disclosure, the water-cooling heat transfer coefficient h ｗis described by a curve or polygonal line passing through the minimum heat flux point (b) and the critical heat flux point (c), and the water-cooled heat transfer coefficient h is obtained by repeatedly deforming the described curve or polygonal line. ｗ Therefore, even if the learning of the temperature model does not function well, for example, if sufficient actual temperature values ​​are not obtained, that is, even if sufficient accuracy is not obtained through learning, the water-cooling heat transfer coefficient h ｗ The identified water-cooling heat transfer coefficient h ｗ By applying this to the temperature model, a temperature model suitable for actual cooling can be obtained.

[0077] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above embodiments and can be implemented in various modifications without departing from the spirit of the present disclosure. When the numbers, quantities, amounts, ranges, etc. of each element are mentioned in the above embodiments, the present invention is not limited to the mentioned numbers unless otherwise specified or clearly specified in principle. Furthermore, the structures, etc. described in the above embodiments are not necessarily essential to the present invention unless otherwise specified or clearly specified in principle.

[0078] In the above embodiment, coiling temperature control has been described as an example, but the present disclosure can also be applied to temperature control on the delivery side of a roughing rolling mill, temperature control on the entry side of a finishing rolling mill, temperature control on the delivery side of a finishing rolling mill, and temperature control in a plate rolling line.

[0079] 1...hot rolling line, 11...heating furnace, 12...roughing mill, R1...stand, 13...finishing mill, F1 to Fn...stand, 14...run-out table, 15...water injection device, W1 to Wn, Wj...water injection header, 21 to 26...thermometer, 3...host computer, 4...process control computer, processing circuit, 4a...dedicated hardware, 4b...processor, 4c...memory, 40...temperature control device, 41...rolling data acquisition unit, 42...temperature model description unit, 43...identification unit, 44...verification unit, 45...rolled material temperature control unit, 46...recording unit, 5...interface screen, Mr...rolled material, h ｗ…Water cooling heat transfer rate, a…Cooling start point, b…Minimum Heat Flow Rate (MHF), c…Clear Heat Flow Rate (CHF), d…Boiling start point

Claims

1. A temperature control device for controlling the temperature of rolled material in a hot rolling line, A rolling data acquisition unit acquires rolling data including the actual temperature values ​​of the rolled material, A model description unit describes a temperature model for calculating the temperature of the rolled material, and describes the water-cooled heat transfer coefficient, which is a parameter of the temperature model, as a function of the temperature of the rolled material, using a curve or piecewise linear representation passing through the minimum heat flux point and the critical heat flux point. An identification unit identifies the water cooling heat transfer coefficient by repeatedly deforming the curve or line so that the difference between the calculated temperature value calculated using the temperature model and the actual temperature value becomes small. Equipped with, The hot rolling line is equipped with a water injection device for injecting cooling water. The identification unit is configured to deform the curve or the piecewise linear figure by moving the temperature of the minimum heat flux point as a function of the cooling water temperature while keeping the critical heat flux point fixed. The identification unit is In the search for the relationship between the minimum heat flux point and the water cooling heat transfer coefficient gain, a central cell is set using the temperature of the minimum heat flux point before identification, and multiple peripheral cells are set around the central cell, each having a different cooling water temperature, minimum heat flux point temperature, and water cooling heat transfer coefficient gain from the central cell. For each cell, an evaluation value is calculated, which is the difference between the calculated temperature value and the actual temperature value. Moving the central cell in a direction that reduces the aforementioned evaluation value, The calculation of the evaluation value and the movement of the central cell are repeated, A temperature control device for a hot rolling line configured to perform the following:

2. In the temperature control device for a hot rolling line according to Claim 1, The water injection device has multiple types of water injection headers, The identification unit is configured to identify the water cooling heat transfer coefficient for each type of water injection header, and is a temperature control device for a hot rolling line.

3. In the temperature control device for a hot rolling line according to claim 2, The identification unit is Selecting a standard water injection header from the aforementioned multiple types of water injection headers, Identifying the water cooling heat transfer coefficient of the aforementioned reference water injection header, The water cooling heat transfer coefficient of a water injection header of a different type from the aforementioned reference water injection header is identified as a multiple of the water cooling heat transfer coefficient of the aforementioned reference water injection header, A temperature control device for a hot rolling line configured to perform the following:

4. A temperature control device for controlling the temperature of a rolled material in a hot rolling line, A rolling data acquisition unit acquires rolling data including the actual temperature values ​​of the rolled material, A model description unit describes a temperature model for calculating the temperature of the rolled material, and describes the water-cooled heat transfer coefficient, which is a parameter of the temperature model, as a function of the temperature of the rolled material, using a curve or piecewise linear representation passing through the minimum heat flux point and the critical heat flux point. An identification unit identifies the water cooling heat transfer coefficient by repeatedly deforming the curve or line so that the difference between the calculated temperature value calculated using the temperature model and the actual temperature value becomes small. Equipped with, The aforementioned rolling data includes rolling conditions, and the aforementioned rolling conditions include the steel type and size of the rolled material. The verification unit is further configured to verify whether the identified water-cooled heat transfer coefficient is valid, using rolling data with similar rolling conditions to the rolling data used for identification by the identification unit. The model description unit is configured to apply the identified water-cooled heat transfer coefficient to the temperature model when it has been verified as valid by the verification unit, and is a temperature control device for a hot rolling line.

5. A temperature control device for a hot rolling line according to claim 4, The temperature control device consists of a process control computer that performs calculations to set the control target in the rolling process of the hot rolling line, The identification unit is configured to perform the identification online when a first number of rolling data with similar rolling conditions are acquired, and to perform the identification offline when a second number of rolling data, which is greater than the first number, is acquired. The first number is a temperature control device for a hot rolling line, which is set so as not to interfere with the setting calculation.