JP2005211963A - Method for correcting model parameter in steel production process, and method for manufacturing hot rolled steel sheet using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a parameter correcting method in which a model parameter in a steel production process is corrected to an optimum value in practicable computation time. <P>SOLUTION: The model parameter in the steel production process is corrected within a scope designated beforehand by using an optimizing method with restrictive conditions so that the result of calculations obtained with the above model is consistent with the result obtained from actual measurement data. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for correcting model parameters in a steel process and a method for producing a hot-rolled steel sheet using the method.

  In the steel process, various physical phenomena are controlled and / or monitored by describing them in a model used for control and / or monitoring (hereinafter both are referred to as “models”). In order to maintain and further improve the performance, it is important to modify model parameters (hereinafter referred to as “model parameters”) based on actual data. For example, as a model parameter used in the steel hot rolling process, in the slab heating in the heating furnace, there is a general heat absorption rate representing the absorption efficiency of radiant heat from the furnace body to the slab, and a rough rolling mill and a finishing rolling mill In the press edger control for controlling the width of the plate installed in the apparatus, there is a width adjustment efficiency by the press edger.

Moreover, the following are known as model parameters in the setup model of the finishing mill.
(1) Contact heat transfer coefficient between roll and material used in calculating steel sheet temperature and heat transfer coefficient of descaler water cooling device (2) Friction coefficient between roll and material during rolling (3) By rolling mill reduction Mill rigidity coefficient for calculating mill elongation (4) Model parameter of profile change model due to wear and heat up when calculating roll profile (5) Model parameter of width shrinkage model due to heat shrinkage, tension and crown change (6 ) Model parameters of deformation resistance model corresponding to the force required for plastic deformation of steel sheet

  Hereinafter, winding temperature control in a hot rolling factory will be described as an example. In a hot rolling facility at a hot rolling mill, a hot rolled steel sheet rolled by a finish rolling mill is cooled to a predetermined temperature in a cooling zone and wound into a roll shape for reasons such as improvement of its characteristics. It is done. The cooling zone is provided with a plurality of upper and lower cooling banks, and the hot-rolled steel sheet that has been rolled by the finish rolling mill is cooled from above and below by water cooling or air cooling. The cooled hot-rolled steel sheet is wound into a roll by a winder. Hereinafter, when the upper cooling bank and the lower cooling bank are not particularly distinguished, they are simply referred to as a cooling bank.

  In the above process after rolling by a finish rolling mill, winding temperature control is performed to control the degree of cooling by the cooling zone so that the hot rolled steel sheet reaches a predetermined temperature before being wound by the winder. Has been. In order to carry out this winding temperature control with high accuracy, it is necessary to accurately set the cooling condition of the hot-rolled steel sheet. This cooling condition can be expressed by a unit temperature, a unit time, and an amount of heat per unit area, and is generally called a heat transfer coefficient as a parameter. This heat transfer coefficient varies from one cooling bank to another due to the influence of fluctuations in the cooling capacity due to header deterioration in the cooling bank and differences in header specifications between the cooling banks.

  As a method for obtaining the heat transfer coefficient, for example, based on the actual temperature values measured on the inlet side and the outlet side of the cooling zone, by using the sequential least square method, the heat transfer coefficient of each upper cooling bank at the time of water cooling Focus on the heat transfer coefficient of the lower cooling banks, the technology for correcting the heat transfer coefficients of the upper and lower cooling banks during air cooling (see, for example, Patent Document 1), and the nonlinearity of the heat transfer coefficient. In addition, a technique is known in which the actual temperature values measured on the inlet side and the outlet side of the cooling zone, and the heat transfer coefficient configured by the neural network based on physical elements are sequentially updated by learning (for example, Patent Document 2).

These techniques are all the temperature actual value _{T inact} the inlet side of the cooling zone, the actual value from the actual value _{T OUTact} of delivery temperature the amount of temperature drop in the cooling zone △ a _{T act} (1) formula The heat transfer coefficient a _i ^{U of} each upper cooling bank during water cooling, the heat transfer coefficient a _i ^{L of} each lower cooling bank during water cooling, the heat transfer coefficient b _i ^{U of} each upper cooling bank during air cooling, and air cooling From the heat transfer coefficient b _i ^{L of} each lower cooling bank at the time, the speed pattern V _i in the cooling zone, and the actual temperature value T _{INact on the inlet} side, the calculated value ΔT _cal of the temperature drop in the cooling zone is Calculation is performed using equation (2) as a model. The heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L are corrected so that the actual value ΔT _act and the calculated value ΔT _cal are equal. The function Ω shown in the equation (2) is a known non-linear function composed of the basic equation of heat conduction described in Chapter 6 of “Theory and Practice of Sheet Rolling”, for example.

ΔT _act = T _INact −T _OUTact (1)
^{_{△ T cal = Ω (a i}} U, a i L, b i U, b i L, T INact, V i) (2)

  Further, there is also known a method of squeezing the one with the smallest error square sum at a predetermined step size between the maximum value and the minimum value of the range that can be taken by the parameter (see, for example, Patent Document 3).

Japanese Patent Publication No. 6-88060 JP-A-4-232214 JP 10-149203 A

However, the above-described technique simply corrects the actual value ΔT _act and the calculated value ΔT _cal to be equal without considering the upper and lower limit values of each heat transfer coefficient and the restrictions between the heat transfer coefficients. There is a problem that the difference in header specifications between the cooling banks is not taken into consideration. The header specifications refer to, for example, the amount of water in each cooling bank, water pressure, nozzle shape, and the like.

  Therefore, for example, when the heat transfer coefficient is three times different even if the specifications of the cooling headers of adjacent cooling banks are exactly the same, or when the heat transfer coefficient is a negative value, that is, the temperature rises due to cooling. There is a problem that can occur even if they do not match. Therefore, when the coiling temperature control is performed using such an inappropriate heat transfer coefficient, there is a problem that desired performance cannot be exhibited.

  As described above, when correcting the model parameters in the steel process, if the model parameters are not limited and corrected within the range specified in advance, there is a possibility of becoming a model parameter that cannot be physically or in specification, If the model parameters are used as they are, not only the desired performance cannot be obtained, but there is also a risk of equipment failure or damage. In addition, in the method of calculating the possible values of the parameters in a predetermined step size, the calculation time may be enormous and may not be practically used.

  Therefore, the present invention provides a parameter correction method for correcting steel process model parameters to an optimum value in a calculation time that can withstand practical use, and a hot-rolled steel sheet using the correction method, in order to eliminate such conventional problems. It aims at providing the manufacturing method of.

  The present invention will be described below. In order to facilitate understanding of the present invention, reference numerals in the accompanying drawings are appended in parentheses, but the present invention is not limited to the illustrated embodiment.

  In the parameter correction method of the present invention, the calculation results obtained by the model are obtained from the actually measured data within the range specified in advance by using the optimization method with constraints on the parameters of the steel process model. The above-mentioned problem is solved by correcting the result so as to match the obtained result.

  According to the present invention, a range designated in advance for a parameter can be expressed by, for example, an equation function or / and an inequality function, and the value of the parameter can be corrected using a known solution of the optimization problem. For example, there is a method for obtaining a parameter that minimizes the objective function while satisfying the constraint condition, using the square of the difference between the calculation result obtained from the model and the result obtained from the actual data as the objective function. The “preliminarily specified range” means a value that a parameter can take or a magnitude relationship with other parameters. Therefore, since the obtained parameter value is always within the range specified in advance, an abnormal value is not calculated, and the optimum parameter value can be obtained in a calculation time that can be practically used.

  Further, each time a steel material is manufactured in the steel process, the actual data is accumulated, and when the number of accumulated actual data reaches a predetermined number, at a predetermined timing or at an arbitrary timing The parameter may be corrected based on the accumulated result data. Thereby, the parameter can be corrected at a predetermined timing based on the accumulated performance data, and the accuracy of the correction result can be improved. “Every time a steel material is manufactured” refers to a manufacturing unit of steel material in a steel process, and varies depending on the type of steel material to be manufactured and the manufacturing environment. “When the number of actual data reaches a predesignated number” means that the actual data is accumulated every time the steel material is manufactured, so the number of production units reaches the predesignated number. .

  The steel plate (0) rolled by the finish rolling mill in the steel process is cooled in the cooling step (41, 42) provided with a plurality of cooling devices (B1U ... BNU, B1L ... BNL) in the cooling step. You may correct each heat transfer coefficient of a some cooling device based on the said performance data in which the measured value of the temperature in the at least entrance side and the exit side of the said cooling zone was included. Thereby, the parameter of the model regarding the temperature at the time of winding the steel plate which exited the finishing mill can be corrected with high accuracy. Furthermore, a plurality of water environment thermometers may be installed in the cooling zone, and the parameters may be corrected based on the actual data including the measurement values measured by the water environment thermometers. And the hot-rolled steel sheet may be manufactured by controlling the temperature of the steel sheet cooled by the cooling zone using the heat transfer coefficient corrected by the parameter correcting method.

  Heating efficiency of a heating device (131, 132, 133) in which a steel plate (12) rolled by a rough rolling mill is heated over the entire plate width, and / or between each stand of a finishing rolling mill constituted by a plurality of stands The heat transfer coefficient of the cooling device (151, 152) provided between at least one of the stands and water-cooling the heated steel sheet is set to a temperature at least on the inlet side of the heating device and on the outlet side of the finishing mill. You may correct based on the said performance data in which the measured value was included. Thereby, the parameter of the model regarding temperature control until the steel plate which exited the rough rolling mill is heated and exits the finish rolling mill can be corrected with high accuracy.

  And the temperature of the steel sheet heated by the heating device or / and the cooling by the cooling device by correcting the heating efficiency or / and the heat transfer coefficient by the parameter correction method. A hot-rolled steel sheet may be manufactured by controlling the temperature of the steel sheet.

  As described above, according to the present invention, it is possible to obtain an optimum value within a calculation time that can withstand practical use by correcting parameters of a steel process model using an optimization method with constraints. Therefore, highly accurate control and monitoring can be performed without impairing the performance of the model.

  First, as a first embodiment, a case where the present invention is applied to the above-described winding temperature control in a hot rolling factory will be described. FIG. 1 is a schematic configuration diagram when the present invention is applied to winding temperature control. Between the finishing mill final stand 1 and the winding device 5, a preceding cooling zone 41 composed of n upper and lower cooling banks B1U ... BnU, B1L ... BnL as cooling devices, and Nn upper and lower each. A rear cooling zone 42 composed of the cooling banks Bn + 1U... BNU, Bn + 1L. Hereinafter, when it is not particularly necessary to distinguish between the types of cooling banks, they are simply referred to as an upper cooling bank BU, a lower cooling bank BL, or a cooling bank B. The front cooling zone 41 and the rear cooling zone 42 are collectively referred to as a cooling zone 4.

  Thermometers 2 and 3 for detecting the temperature are respectively installed on the inlet side and the outlet side of the cooling zone 4, and a speed detector for detecting the feeding speed of the steel plate 0 is provided in the finishing mill final stand 1. 6 is installed. In addition, each cooling bank B is connected to a water volume record capturing device 7 that captures the water volume record of each cooling bank B.

  In addition, although the speed detector 6 in this form is installed in the finishing mill final stand 1, it will not be restricted to this if a speed is detected. For example, you may install in the table roller in the cooling zone 4, or the winding device 5. FIG. Moreover, although the thermometers 2 and 3 in this form are installed in the entrance side of the front | former stage cooling zone 41, and the exit side of the back | latter stage cooling zone 42, temperature can be measured like the exit side of the front stage cooling zone 41, for example. It can be anywhere. Further, if a water environment thermometer capable of measuring the temperature in the cooling zone 4 is installed in the cooling zone and the actual temperature value is used, the heat transfer coefficient can be predicted with higher accuracy.

  The steel plate 0 that has passed through the finishing mill final stand 1 is cooled in the cooling zone 4 and wound up by the winding device 5. In the meantime, the actual speed value is detected by the speed detector 6, the actual temperature values on the input and output sides are detected by the thermometers 2 and 3, and the actual water amount in the cooling zone 4 is detected by the actual water amount capturing device 7. These actual values are stored in the data storage device 8 for each rolling as all the data necessary for learning the heat transfer coefficient, together with coil rolling information such as coil material and manufacturing dimensions. The data storage device 8 stores all past data necessary for learning the heat transfer coefficient and is configured to be used by the heat transfer coefficient learning device 9.

  Next, the learning timing determination device 10 instructs the heat transfer coefficient learning device 9 to correct the heat transfer coefficient at a predetermined timing. As the correction timing, for example, when the number of data stored in the data storage device 8 reaches a predetermined number, when the operation is temporarily stopped, a roll change, maintenance date, a periodic repair, etc. There are arbitrary timings etc. during operation.

  At the timing instructed by the learning timing determination device 10, the heat transfer coefficient learning device 9 reads the current heat transfer coefficient value from the heat transfer coefficient table 11 and stores the past values stored in the data storage device 8. Using data, the heat transfer coefficient is corrected (learned) by a heat transfer coefficient correction process described later, and the corrected heat transfer coefficient is written in the heat transfer coefficient table 11. In addition, since the heat transfer coefficient used by coiling temperature control is read for every rolling from the heat transfer coefficient table 11, even if the fluctuation | variation of cooling capacity arises, the heat transfer coefficient corrected so that the fluctuation | variation might be used can be used. , Model performance degradation can be prevented.

  FIG. 2 is a flowchart showing a heat transfer coefficient correction process performed by the heat transfer coefficient learning device 9. Steps 1 to 6 are executed in order at the timing instructed by the learning timing determination device 10. Hereinafter, each step will be described.

  Since the heat transfer coefficient has a strong non-linearity, it is treated as a table value divided according to factors such as the component, size, and temperature of the steel sheet. In step 1, the value of the heat transfer coefficient of the category to be corrected is read from the heat transfer coefficient table 11.

  Next, in step 2, only data belonging to the classification of the heat transfer coefficient read in step 1 is extracted from the data storage device 8 in which data necessary for learning of the heat transfer coefficient is stored. “Data belonging to a category” is data necessary for correcting the value of the heat transfer coefficient of the category.

In step 3, the heat transfer coefficient a _i ^{U of} each upper cooling bank BU during water cooling, the heat transfer coefficient a _i ^L of the lower cooling bank BL during water cooling, and the heat transfer coefficient b _i ^{U of} each upper cooling bank BU during air cooling. , And the restrictions determined by the physical meaning of the heat transfer coefficient b _i ^L of each lower cooling bank BL during air cooling and the specifications of the cooling header of each cooling bank B are described. This constraint can be described by N _E equality constraint functions g _j as in equation (3) and N _N inequality constraint functions h _k as in equation (4).

g _j (a _i ^U , a _i ^L , b _i ^U , b _i ^L ) = 0 (j = 1, 2,..., N _E ) (3)
^{_{h k (a i U, a}} i L, b i U, b i L)> 0 (k = 1,2, ..., N N) (4)
For example, when the specifications of the cooling headers of the upper cooling banks B2U and B3U are the same, the equation constraint function g _j can be described as follows.
^{_{g 1 (a i U, a}} i L, b i U, b i L) = a 2 U -a 3 U

Also, it is understood that the heat transfer coefficient a ₂ ^U of the upper cooling bank B2U cannot be more than three times the heat transfer coefficient a ₃ ^u of the upper cooling bank B3U due to differences in specifications such as the nozzle shape of the header and the amount of water. If so, the inequality constraint function h _k is
^{_{h 1 (a i U, a}} i L, b i U, b i L) = - a 2 U + 3a 3 U

The function g _j and the function h _k depend on the specifications of the cooling header of each cooling bank B, but need to be linear functions of the respective heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L. Not a nonlinear function. Therefore, it can be applied to any cooling device.

In step 4, initial values of the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L are set. Usually, the value itself read in step 1 is used, but any value that satisfies the expressions (3) and (4) may be used and may be determined by a random number.

In step 5, the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L are actually corrected. Here, if a _i ^U , a _i ^L , b _i ^U , and b _i ^L to be obtained are expressed by a vector x, equation (5) is obtained.

The problem of obtaining the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , b _i ^L is the vector x that minimizes the expression (6) while satisfying the constraints of the expressions (7) and (8). Can be considered as an optimization problem with constraints. Here, M is the total number of data read in step 2, and ΔT _{act, m} and ΔT _{cal, m} (X) are respectively the expression (1) for the m (<= M) th data and Equation (2) is shown. For example, when the past data read in step 2 is data for the past five times, M = 5, and the first accumulated data is m = 1.

The above-mentioned optimization problem with constraints may be solved by a publicly known method such as “System Engineering Kyoritsu Publishing Shinsuke Akagi”. Further, (6) of the function f (x), may be added to the section of the unknown variables of ^{x 2.} Thereby, it is possible not to increase the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L more than necessary.

In the final step 6, the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , b _i ^L obtained in step 5 are stored in the heat transfer table 11. Since the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , b _i ^L can be corrected within the range specified in advance by the above-described heat transfer coefficient correction processing, the specification of the header between the cooling banks B The heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , b _i ^L can be corrected without departing from the difference.

In the above description, the heat transfer coefficients a _i ^U , a _i ^L , b _i ^U , and b _i ^L are corrected, but for example, correction coefficients α _i expressed by the equations (9) and (10) are used. ^U may be modified.
a _i ^U = A _i ^U (1 + α _i ^U ) (9)
a _i ^U = A _i ^U · α _i ^U (10)

  As a second embodiment, the case where the present invention is applied to heating / cooling control in a hot rolling factory will be described. FIG. 3 is a schematic configuration diagram in a case where the present invention is applied to heating / cooling control until the rough rolled material rolled by the rough rolling mill is heated and applied to the finishing mill 14.

  On the entry side of the finish rolling mill 14, three heating devices 131, 132, and 133 that can heat the rough rolled material over the entire sheet width direction are installed. In the finishing mill 14, a plurality of stands F1... F7 are provided, and cooling devices 151 and 152 for cooling the steel plate 12 are installed between the stand F1 and the sudan F2, and between the stand F3 and the stand F4, respectively. ing. Hereinafter, when it is not particularly necessary to distinguish between the heating devices 131, 132, and 133, it is simply referred to as the heating device 13, and particularly when it is not necessary to distinguish between the cooling devices 151 and 152, it is simply referred to as the cooling device 15. By using the heating efficiency of each heating device 13 and the heat transfer coefficient of each cooling device 15, the heating / cooling control disclosed in JP-A-2002-11502 is performed. The heating / cooling control method may be any method that uses the heating efficiency of the heating device and the heat transfer efficiency of the cooling device, and is not limited to the control method described in JP-A-2002-11502.

  A thermometer 161 is installed on the entry side of the first heating device 131, and a thermometer 162 is installed on the exit side of the finishing mill 14, and each heating device 13 takes in the results of the power supplied to each heating device 13. An actual power acquisition device 17 is connected, and each cooling device 15 is connected to a water amount actual acquisition device 18 that takes in an actual value of the amount of cooling water in each cooling device 15.

  The steel plate 12 as the rough rolled material that has passed through the rough rolling mill is heated by the heating device 13, rolled by the finish rolling mill 14, and cooled by the cooling device 15. Each actual value from when the steel plate 12 passes through the finishing mill 14 is detected by the thermometers 161 and 162, the input power actual acquisition device 17 and the water amount actual acquisition device 18, and the heating efficiency and heat All data necessary for learning the transfer coefficient is stored in the data storage device 19 for each rolling.

Thereafter, since it is the same as the winding temperature control which is the first form, the details are omitted, but the heating efficiency of the heating device 13 to be corrected is η _i (i = 1, 2, 3), and the cooling device 15 the heat transfer coefficient in order to determine the _a i (i = 1,2) is the actual value of the thermometer 161 for the data m and _{T inact, m,} the actual value of the thermometer 162 _{T OUTact,} and _m, to be obtained When η _i (i = 1, 2, 3) and a _i (i = 1, 2) are vectors x as shown in the following equation (11),
x = {η ₁ , η ₂ , η ₃ , a ₁ , a ₂ } (11)

It is sufficient to solve the optimization problem with a constraint condition that obtains the vector x that minimizes the equation (12) while satisfying the constraint conditions of the equations (13) and (14). The function Ω is a model for calculating the temperature difference between the inlet side temperature of the heating device 131 and the outlet side temperature of the finishing mill 14 from the vector x and the measured value T _{INact, m of the inlet} side thermometer 161. This optimization problem may be solved using a known numerical method as described above. The feature of the present invention is that the relationship between the heating efficiency η _i between the respective heating devices 13 and the relationship between the heat transfer coefficients a _i between the respective cooling devices 15 are designated in advance, and the optimum parameters are calculated within the range. Therefore, the relationship can be specified in advance as an equality function and an inequality function as in the equations (13) and (14).

Although the equations (13) and (14) differ depending on the equipment and specifications between the facilities, for example, only the heating device 133 is added after the introduction of the other heating devices 131 and 132, so that the heating efficiency η ₃ becomes the other heating efficiency η. _1, it is slightly improved over eta ₂ has been found by the manufacturer's specifications, and the cooling apparatus 152, the water pressure headers shape, that is large heat transfer coefficient is impossible twice or more the cooling device 151 Can be described as in the following formulas (15) and (16). Note that equation (16) is added to correct the parameters within a range that makes physical sense.

An embodiment when the present invention is applied to winding temperature control in a hot rolling factory will be described. Table 1 shows the results of comparing the prediction accuracy of the temperature drop amount in the cooling zone between the conventional technology and the present invention for a low carbon steel having a thickness of 2.3 to 4 mm. The column (a) shows the result of learning the heat transfer coefficient, and both methods have the same accuracy. The column (b) shows the results when the value of the heat transfer coefficient obtained in the column (a) is used as it is thereafter. The “average” and “standard deviation” are values obtained by calculating the average and standard deviation for the target data number 100 based on the difference between the actual value ΔT _act and the calculated value ΔT _cal of the temperature drop in the cooling zone. is there.

  In the prior art, the prediction accuracy is deteriorated because the abnormal heat transfer coefficient is neglected in the header (a) column of Table 1. On the other hand, the present invention is good because it does not have an abnormal heat transfer coefficient as in the prior art and can maintain the prediction accuracy.

  The present invention is not limited to the above-described form, and may be implemented in various forms. For example, in the present embodiment, the parameters in the rolling process related to the steel plate have been described, but the present invention is not limited to this, and the steel material formed by the rolling mill may be a rolled steel material such as a wire, a shape steel, or a bar steel.

  The present invention is not limited to the correction (learning) of the heat transfer coefficient in the coiling temperature control at the hot rolling factory and the correction (learning) of the heating efficiency and the heat transfer coefficient in the heating / cooling control at the hot rolling factory. The present invention can be widely applied to the field of identifying, correcting, or learning an optimum value within a range in which model parameters of a model used in the steel process are designated in advance. In that case, the model parameter of the model is set as an unknown variable, and the model (2) is replaced with a model to which the model parameter is applied, and the constraint on the model parameter or the like according to the model is described by the formulas (7) and (8).

The schematic block diagram at the time of applying this invention to the coiling temperature control in a steel process. The flowchart of the correction method of a heat transfer coefficient. The schematic block diagram at the time of applying this invention to the heating and cooling control in a steel process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 0 Steel plate 1 Final stand of finishing mill 2,161 Inlet side thermometer 3,162 Outlet side thermometer 4 Cooling zone 5 Winding device 6 Speed detector B Cooling bank 8, 19 Data storage device 131-133 Heating device 151, 152 Cooling device

Claims (7)

  By using the optimization method with constraints on the steel process model parameters, the calculation results obtained by the model are corrected to match the results obtained from the actual measured data within the specified range. A parameter correction method characterized by:   Each time the steel material is produced in the steel process, the actual data is accumulated, and when the number of accumulated actual data reaches a predetermined number, at a predetermined timing, or at an arbitrary timing, The parameter correction method according to claim 1, wherein the parameter is corrected based on the accumulated performance data.   A steel sheet rolled by a finish rolling mill in the steel process is cooled in a cooling zone provided with a plurality of cooling devices, and each heat transfer coefficient of the plurality of cooling devices in a cooling step is set at least in the cooling zone. The parameter correction method according to claim 1, wherein correction is performed based on the actual data including the measured values of the temperatures on the entry side and the exit side.   4. A plurality of water environment thermometers are installed in the cooling zone, and the parameters are corrected based on the actual data including measurement values measured by the water environment thermometers. Parameter correction method.   A hot-rolled steel sheet is manufactured by controlling the temperature of the steel sheet cooled by the cooling zone using the heat transfer coefficient corrected by the parameter correcting method according to claim 3 or 4. A method for producing a hot-rolled steel sheet.   The heating efficiency of the heating device in which the steel sheet rolled by the rough rolling mill is heated over the entire plate width, or / and provided between at least one stand among the stands of the finishing mill constituted by a plurality of stands, The heat transfer coefficient of a cooling device for water-cooling the heated steel sheet is corrected based on the actual data including at least the temperature measurement values on the entry side of the heating device and the exit side of the finishing mill. 3. The parameter correction method according to 1 or 2. The temperature of the steel sheet heated by the heating device or / and cooling by the cooling device by correcting the heating efficiency or / and the heat transfer coefficient by the parameter correcting method according to claim 6. A hot-rolled steel sheet is produced by controlling the temperature of the steel sheet to be manufactured.

Images