JP2016209898A - Cooling control method for thick steel plate, cooling control apparatus, manufacturing method, and manufacturing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a cooling control method for a thick steel plate capable of enhancing the control precision of thick plate temperature during cooling and after cooling.SOLUTION: A cooling control method for a thick steel plate has: a step of inversely calculating a temperature at which an ebullition state changes, in a thick steel plate temperature prediction model, so that a predicted value of the surface temperature of a thick steel plate may be coincident with a measured value of the surface temperature of a thick steel plate under being cooled; a step of accumulating temperatures, obtained by inverse calculation in a database, with data of a manufacturing condition for the thick steel plate at which the temperature has been obtained; a step of extracting data, from the database, the manufacturing condition of which is similar to that of a thick steel plate to be cooled; a step of calculating a temperature, by using the extracted data and temperature, at which an ebullition state of a thick steel plate to be cooled changes; a step of estimating a cooling stop temperature of a thick steel plate to be cooled by using the calculated temperature in the previous step; and a step of deciding an operating condition of a cooling apparatus so that the estimated cooling stop temperature may be coincident with a target temperature after cooling the thick steel plate.SELECTED DRAWING: Figure 5

Description

  The present invention relates to a cooling control method for controlling the cooling mode of a thick steel plate, a method for manufacturing a thick steel plate using the same, a cooling control device for controlling the cooling mode of the thick steel plate, and a manufacturing apparatus for the thick steel plate using the same.

  In temperature control in the cooling process of thick steel plates, the steel plate temperature change due to water cooling is estimated by heat transfer calculation before cooling starts, and the amount of cooling water and the steel plate conveyance speed in the cooling device are determined so as to obtain the target cooling stop temperature To do. Therefore, the estimation accuracy of the steel plate temperature change by water cooling influences the control accuracy of the cooling stop temperature. If the actual cooling stop temperature is far from the target value, the mechanical characteristics required for the product cannot be obtained, and it is necessary to control the cooling stop temperature with high accuracy.

  When estimating the temperature change of the steel sheet in the cooling process by heat transfer calculation, the heat flux between the steel sheet surface and the cooling water has many disturbance factors, which causes deterioration in control accuracy. At the interface between the steel sheet surface and the cooling water, when the steel sheet surface temperature is high, it is in a state of film boiling where a water vapor film is formed, but when the steel sheet surface temperature decreases, the boiling state changes from transition boiling to nucleate boiling. To do. Where the boiling state changes, the change in heat flux also increases, so it is important to accurately predict the boiling state.

  However, in the vicinity of the boundary temperature at which the boiling state changes, the steel plate surface temperature (MHF (Minumum Heat Flux) point) that changes from film boiling to transition boiling changes depending on the scale thickness and surface roughness of the steel plate surface. It is a disturbance factor for steel plate temperature control.

  So far, methods have been developed to learn and correct prediction errors in models used for steel sheet cooling control. Representative methods include deriving model offset errors and deriving appropriate parameters for heat transfer models so that the measured values of the steel plate temperature before cooling and the steel plate temperature after cooling match the model calculations. And the like.

  For example, Patent Document 1 discloses a process of calculating a predicted value of a cooling stop temperature of a thick steel plate in the cooling process using a temperature prediction model of the thick steel plate, and a past record for each slab. A process of extracting past performance data of a steel plate having a manufacturing condition similar to that of the steel plate from the database in which the data is accumulated, and a predicted value of the cooling stop temperature of the steel plate from the past performance data extracted in the process Calculating a correction value of the cooling stop temperature from the estimated value of the cooling stop temperature of the thick steel plate and the error of the estimated prediction value of the cooling stop temperature of the thick steel plate And a step of determining the cooling water amount and / or the steel plate conveyance speed so that the correction value of the cooling stop temperature becomes the target value, and cooling the thick steel plate at the determined cooling water amount and / or the steel plate conveyance speed. A step, cooling control method for a steel plate is disclosed. Patent Document 2 discloses a technique for correcting a heat transfer coefficient used for heat transfer calculation by a search method based on the actual steel sheet surface temperature values on the entry side and the exit side of the cooling device. In addition, in Patent Document 3, a cooling model using a cooling water amount density as a parameter is applied to a portion where the cooling water from the header portion collides with the thick steel plate, and a portion of the thick steel plate corresponding to a position between the header portions is applied. Applying a cooling model with the flow rate of cooling water moving in the plate width direction of the steel plate as a parameter, the heat transfer coefficient is calculated, the temperature distribution in the width direction of the steel plate after passing through the water cooling zone is predicted, and the cooling of each header A cooling control method for a thick steel plate for determining a water amount distribution is disclosed.

JP 2012-81518 A JP 2004-244721 A JP 2011-167754 A

  In the techniques described in Patent Literature 1 to Patent Literature 3, the steel plate temperature before cooling and the steel plate temperature after cooling are only used, and the surface temperature of the thick steel plate measured during the cooling is not used. It is difficult to raise. In addition, in the techniques described in Patent Documents 1 to 3, the boundary temperature at which the boiling state changes is not optimized, so that there is a risk that the control accuracy of the steel sheet temperature during cooling and after cooling stops may be insufficient. It was.

  The present invention has been made in view of the above problems, and is capable of increasing the temperature control accuracy during cooling and after cooling stop, and a cooling control method, a cooling control device, and a manufacturing of the thick steel plate. It is an object to provide a method and a manufacturing apparatus.

In order to improve the temperature control accuracy during cooling and after cooling stop, the present inventors decided to use not only the temperature measurement values before and after cooling but also the temperature measurement values during cooling. Then, from the history of temperature change, a boundary temperature (CHF (Critical Heat Flux) point or MHF point, the same applies hereinafter) at which the boiling state changes is derived, and the temperature prediction result and the temperature measurement value match. The boundary temperature was corrected. The boundary temperature thus corrected was associated with the operating conditions and accumulated in a database, and the operating conditions of the cooling device were determined using past data with similar manufacturing conditions for the thick steel plate. The inventors of the present invention have found that by cooling the thick steel plate under the operating conditions thus determined, it is possible to increase the control accuracy of the cooling temperature of the thick steel plate as compared with the conventional case.
The present invention has been completed based on such findings. The present invention will be described below.

  A first aspect of the present invention is a method for controlling cooling of a thick steel plate that passes through a water cooling zone of a cooling device, and the surface temperature of the thick steel plate cooled by using the cooling device is determined using a steel plate temperature prediction model. Predicting the temperature, measuring the surface temperature of the thick steel plate cooled by using the cooling device, the surface temperature predicted in the temperature prediction process and the surface temperature measured in the temperature measurement process So that the boundary temperature at which the boiling state changes and the boundary temperature back-calculated in the back-calculation step are used in the steel sheet temperature prediction model, and the manufacturing conditions of the thick steel plate from which the boundary temperature is obtained Along with the data, the data accumulation process accumulated in the operation results database and the past production condition data similar in production condition to the steel plate to be cooled will be displayed along with the boundary temperature at which the boiling state changes. The extraction process extracted from the database, and the boiling change temperature calculation process for calculating the boundary temperature at which the boiling state of the steel plate to be cooled is changed using the data extracted at the extraction process and the boundary temperature at which the boiling state changes. And by substituting the boundary temperature calculated in the boiling change temperature calculation step and the manufacturing condition data of the thick steel plate to be cooled into the steel plate temperature prediction model, the cooling stop temperature of the thick steel plate to be cooled from now on A cooling stop temperature estimation step that estimates the cooling stop temperature, and a determination step that determines the operating conditions of the cooling device so that the cooling stop temperature estimated in the cooling stop temperature estimation step matches the target temperature of the thick steel plate at the time of cooling stop, and This is a method for controlling the cooling of a thick steel plate.

Here, in the present invention, the “production condition of the thick steel plate” includes information on the thick steel plate cooled by the cooling device (for example, the plate thickness, the plate width, the plate length, the chemical component value, etc.), the cooling water. Temperature. Further, in the present invention, “whether or not the manufacturing conditions are similar to the steel sheet to be cooled from now on” is an information vector indicating the manufacturing conditions of the past manufacturing results and information indicating the manufacturing conditions of the steel sheet to be cooled A distance function between the vectors can be defined, and the determination can be made based on the magnitude of the distance.
In the first aspect of the present invention, the change in the steel sheet surface temperature in the cooling device is measured, and this is used to change the heat transfer that becomes the core of the steel sheet temperature prediction model, that is, the boundary temperature (boiling state) where the boiling state changes. In the following, “MHF point or CHF point” may be referred to), and the calculated MHF point or CHF point is accumulated in the operation result database. Then, by performing learning control utilizing the operation results database, the MHF point and CHF point of the thick steel plate to be cooled are determined, and the target value of the cooling stop temperature of the thick steel plate is determined using the MHF point and CHF point. Determine the operating conditions of the cooling device to cool the air. By cooling under the operating conditions determined in this way, the temperature of the thick steel plate during and after cooling can be controlled with high accuracy.

  Further, in the first aspect of the present invention, by using a plurality of similarity evaluation criteria in the extraction step, a plurality of groups of data is extracted from the operation performance database, and using the extracted plurality of groups of data, In the boiling change temperature calculation step, the temperature at which the boiling state of the thick steel plate to be cooled is changed is preferably calculated. By adopting such a form, it becomes easy to calculate appropriate values as the MHF point and CHF point of the thick steel plate to be cooled from now on, so that it is easy to control the temperature of the thick steel plate during cooling and after cooling stop with high accuracy. Become.

  A second aspect of the present invention includes a step of rolling a thick steel plate, and a step of cooling the thick steel plate after the rolling step, and the cooling step includes the first aspect of the present invention. A method for producing a thick steel plate, characterized in that such a method for controlling cooling of a thick steel plate is used. In the second aspect of the present invention, since the thick steel plate cooling control method according to the first aspect of the present invention is used, the temperature of the thick steel sheet during the cooling and after the cooling stop is controlled with high accuracy. Thereby, since it becomes possible to stabilize the mechanical characteristic of a thick steel plate, the manufacturing method of a thick steel plate which can reduce an additive element and can reduce manufacturing cost can be provided.

  A third aspect of the present invention is a cooling control device that controls operating conditions of a cooling device that cools a rolled thick steel plate, and predicts the surface temperature of the thick steel plate that is cooled by using the cooling device, as a steel plate temperature prediction. Measured using a temperature prediction unit that predicts using a model, a thermometer that measures the surface temperature of a thick steel plate that is cooled using a cooling device, and a surface temperature and thermometer that are predicted by the temperature prediction unit In order to match the surface temperature, the back-calculation part for back-calculating the boundary temperature at which the boiling state changes, and the boundary temperature back-calculated in the back-calculation part are used in the steel sheet temperature prediction model. The operation results database that is stored together with the manufacturing condition data and the past manufacturing condition data that has similar manufacturing conditions to the steel plates to be cooled are extracted from the operation results database together with the boundary temperature at which the boiling state changes. A boiling change temperature calculation unit for calculating the boundary temperature at which the boiling state of the steel plate to be cooled is changed using the extraction part, the data extracted by the extraction unit and the boundary temperature at which the boiling state changes, and the boiling change Cooling to estimate the cooling stop temperature of the thick steel plate to be cooled by substituting the boundary temperature calculated by the temperature calculation unit and the manufacturing condition data of the thick steel plate to be cooled into the steel plate temperature prediction model. A stop temperature estimation unit, and a determination unit that determines the operating conditions of the cooling device so that the cooling stop temperature estimated by the cooling stop temperature estimation unit matches the target temperature of the thick steel plate at the time of cooling stop, It is a cooling control apparatus of a thick steel plate.

  The thick steel plate cooling control device according to the third aspect of the present invention can implement the thick steel plate cooling control method according to the first aspect of the present invention. Therefore, according to the third aspect of the present invention, it is possible to provide a thick steel plate cooling control apparatus capable of controlling the temperature of the thick steel plate during cooling and after cooling stop with high accuracy.

  Moreover, in the said 3rd aspect of this invention, by an extraction part using several similarity evaluation criteria, the data of several groups are extracted from the operation performance database, and the data of several groups extracted are used. The boiling change temperature calculation unit preferably calculates the temperature at which the boiling state of the thick steel plate to be cooled is changed. By adopting such a form, it becomes easy to calculate appropriate values as the MHF point and CHF point of the thick steel plate to be cooled from now on, so that it is easy to control the temperature of the thick steel plate during cooling and after cooling stop with high accuracy. Become.

  A fourth aspect of the present invention comprises a rolling mill for rolling a thick steel plate, a cooling device for cooling the thick steel plate rolled by the rolling mill, and a cooling control device for controlling the operation of the cooling device, The cooling control device is a thick steel plate manufacturing device, which is a thick steel plate cooling control device according to the third aspect of the present invention. Since the fourth aspect of the present invention includes the thick steel plate cooling control device according to the third aspect of the present invention, the temperature of the thick steel sheet during and after cooling is controlled with high accuracy. Can do. Thereby, since it becomes possible to stabilize the mechanical characteristic of a thick steel plate, the manufacturing apparatus of a thick steel plate which can reduce an additive element and can reduce manufacturing cost can be provided.

  It is possible to provide a thick steel plate cooling control method, a cooling control device, a thick steel plate manufacturing method, and a manufacturing device capable of increasing the temperature control accuracy during and after cooling.

It is a figure which shows a boiling curve. It is a figure explaining the example of arrangement | positioning of the thermometer in a cooling device. It is a figure which shows the example of the temperature measurement value by the steel plate temperature estimated with the steel plate temperature prediction model and a thermometer. It is a figure explaining a MHF point and a CHF point. FIG. 4A is a diagram for explaining a reference MHF point and a CHF point, and FIG. 4B is a diagram for explaining an MHF point and a CHF point after back calculation. It is a figure explaining the cooling control method of the thick steel plate of this invention. It is a figure explaining the outline | summary of a water cooling heat transfer model. It is a figure explaining the manufacturing method of the thick steel plate of this invention. It is a figure explaining the cooling control apparatus 40 of the thick steel plate of this invention, and the manufacturing apparatus 100 of the thick steel plate of this invention. It is a figure explaining the similar data search method for deriving a plurality of regression models. It is a figure explaining ensemble learning by a plurality of regression models. It is a figure which shows the result of the surface temperature (predicted temperature) of the steel plate estimated using the MHF point and CHF point used as a reference | standard, without calculating MHF point and CHF point back. Fig.11 (a) is a figure explaining the prediction accuracy in the position in a cooling device, and FIG.11 (b) is a figure explaining the prediction accuracy in the exit side position of a cooling device. It is a figure which shows the result of the surface temperature (predicted temperature) of the steel plate estimated using the back-calculated MHF point and CHF point. FIG. 12A is a diagram for explaining the prediction accuracy at the position in the cooling device, and FIG. 12B is a diagram for explaining the prediction accuracy at the outlet side position of the cooling device.

  The steel plate temperature in the cooling process of the thick steel plate can be predicted by performing heat transfer calculation with a steel plate temperature prediction model in which the heat flux on the surface of the steel plate in contact with the cooling water and the heat conduction inside the steel plate are formulated. . In this steel plate temperature prediction model, for example, as disclosed in Patent Document 3, the heat flux that is a boundary condition between the cooling water and the steel plate surface changes according to the amount of cooling water and the steel plate surface temperature, and the water temperature The heat transfer calculation is performed considering that the steel plate surface temperature (MHF point or CHF point) at which the boiling state changes due to the change.

  FIG. 1 shows a boiling curve. The MHF point and the CHF point shown in FIG. 1 change according to the scale of the steel sheet surface. Therefore, the MHF point and the CHF point change depending on the operating conditions in the heating process and rolling process, which are the previous processes of the cooling process.

  Therefore, in the present invention, as shown in FIG. 2, a thermometer capable of measuring the steel sheet surface temperature in the cooling device is installed in the cooling device, and the history of the steel sheet surface temperature is measured. Then, the MHF point and CHF point are set so that the measured value of the steel plate surface temperature measured in the cooling device and the measured value of the cooling stop temperature measured on the cooling device delivery side agree with the steel plate surface temperature estimated by the steel plate temperature prediction model. Calculate backwards. An example of the steel plate surface temperature estimated by the steel plate temperature prediction model and the temperature measurement value by the thermometer is shown in FIG. 3, the reference MHF point and CHF point are shown in FIG. 4A, and the post-calculation MHF point and CHF point are shown in FIG. Each is shown in b). In the present invention, by adaptively learning the back-calculated MHF point and CHF point, the accuracy of the steel sheet temperature prediction model is improved, and highly accurate cooling control is enabled.

  The adaptive learning in the present invention will be outlined. In the present invention, the result of back-calculating the MHF point and the CHF point and the manufacturing condition of the thick steel plate are linked and accumulated in the operation performance database. When determining the operating conditions of the cooling device when cooling steel plates, data on past manufacturing conditions similar to the steel plates to be cooled will be extracted from the operation results database, and the extracted data will be It is used to estimate the MHF point and CHF point of the thick steel plate to be cooled. Then, by substituting the estimated MHF point and CHF point into the steel plate temperature prediction model, the steel plate temperature at the time of cooling stop is predicted, and the predicted steel plate temperature becomes the target value of the steel plate temperature at the time of cooling stop. The operating conditions of the cooling device are determined. In the present invention, the steel plate surface temperature is measured by a thermometer installed in the cooling device when the thick steel plate is cooled under the operating conditions determined in this way, and further installed on the outlet side of the cooling device. The cooling stop temperature is measured by the thermometer. These temperature measurement results are compared with the steel plate surface temperature predicted by the steel plate temperature prediction model, and the MHF point and the CHF point at which the temperature measurement result matches the predicted value of the steel plate surface temperature are calculated backward. The back-calculated MHF point and CHF point are linked to the manufacturing conditions of the thick steel plate from which they were obtained, accumulated in the operation performance database, and operating conditions of the cooling device when cooling the subsequent thick steel plates It is used in determining

  Thus, in this invention, the change of steel plate temperature is estimated using the MHF point and CHF point estimated based on the past data with similar manufacturing conditions. Thereby, since the prediction accuracy of the steel plate temperature is improved, it becomes possible to appropriately set the operating conditions of the cooling device, and to obtain the target cooling stop temperature.

  The present invention will be described more specifically with reference to the drawings. In addition, the form shown below is an illustration of this invention and this invention is not limited to the form shown below.

1. Thick Steel Plate Cooling Control Method FIG. 5 is a diagram for explaining the thick steel plate cooling control method of the present invention. As shown in FIG. 5, the cooling control method for a thick steel plate according to the present invention includes a temperature prediction step S11, a temperature measurement step S12, a back calculation step S13, a data accumulation step S14, an extraction step S15, and a boiling change temperature. It includes a calculation step S16, a cooling stop temperature estimation step S17, a determination step S18, and a cooling step S19.

1.1. Temperature prediction step S11
The temperature prediction step S11 is a step of predicting the temperature of the thick steel plate cooled using the cooling device using the steel plate temperature prediction model.

  The temperature of the thick steel plate can be represented by a one-dimensional heat conduction equation in the plate thickness direction shown in the following formula (1).

  The boundary conditions on the upper and lower surfaces of the thick steel plate are given by the following formulas (2) and (3).

Here, T is the temperature [° C.], t is the time [s], x is the coordinate [m] in the thickness direction, c is the specific heat [J / kg · s], ρ is the density [kg / m ³ ], λ Is the thermal conductivity [W / m · ° C.], q _w is the heat flux [W / m ² ] by water cooling, q _e is the heat flux [W / m ² ] by convection, and q _r is the heat flux [W / m ² by radiation. m ² ], u is a subscript representing the upper surface, and d is a subscript representing the lower surface.

The heat flux q _w by water cooling and the heat flux q _e by convection can be written as follows using the heat transfer coefficient.

Here, T _s is the surface temperature of the steel plate [° C.], T _w is the temperature of the cooling water [° C.], _Ta is the temperature of the atmosphere [° C.], H _w is the water cooling heat transfer coefficient, and H _a is the convection. Heat transfer coefficient.

The heat flux _qr due to radiation can be written as follows using the emissivity ε and the Stefan-Boltzmann constant σ.

  Control of thick steel sheet by solving the above equation (1) online using the finite difference method under the boundary condition equations (2) to (6) reflecting the water cooling conditions in each cooling zone. The temperature for the point can be calculated.

Here, the water-cooling heat transfer coefficient _Hw is calculated by a method disclosed in Patent Document 3, for example.

  In the model shown in FIG. 6, when calculating the heat transfer coefficient of the portion of the thick steel plate sandwiched between the upper header and the lower header (hereinafter referred to as “upper and lower header”), a set of upper and lower nozzles (upper header) The region on the thick steel plate is divided into concentric cells centering on the jet direction of the nozzles arranged in the nozzle and the nozzle arranged in the lower header. The reason why the model divided into the concentric cells is used is that the cooling water ejected from the nozzle spreads concentrically on the thick steel plate. A cell divided into concentric circles can be predicted with higher accuracy as the division width becomes narrower. However, since the calculation load increases, the cell may be divided into cells having a certain width. More specifically, the maximum radius of the model (the maximum radius of the cell assumed by the model) so that the models at each nozzle partially overlap each other in consideration of the distance between the nozzles arranged in parallel. And the model may be divided into about five cells. In the example shown in FIG. 6, since the distance between the nozzles arranged in parallel is 50 mm, a model having a maximum radius of 25.7 mm was formed. Here, each cell was divided into four ring-shaped cells having a width of 5.7 mm and one circular cell having a radius of 2.9 mm at the center.

After dividing into a plurality of cells in this way, the heat transfer coefficient is calculated for each cell. When calculating the heat transfer coefficient, first, the water temperature of each cell is calculated. The water temperature rises under the influence of the temperature of the thick steel plate as it moves away from just below the nozzle. The water temperature can be easily obtained by heat transfer calculation.
Subsequently, the ratio of nucleate boiling and film boiling is determined for each cell. The boiling heat transfer phenomenon has a small heat transfer coefficient in the film boiling state and a large heat transfer coefficient in the nucleate boiling state. When the temperature of the thick steel plate is high, film boiling is the main component, but when the temperature is low, transition to nucleate boiling tends to cause a rapid increase in the heat transfer coefficient. Therefore, the heat transfer coefficient varies greatly depending on this ratio. It is known that the relationship between the CHF point and the MHF point can be obtained by experiments or the like. The temperature region between the CHF point and the MHF point is called a transition boiling region where nucleate boiling and film boiling occur simultaneously, as shown in FIG. If the surface temperature of the thick steel plate is below the CHF point, the nucleate boiling rate is 100%, and if it is above the MHF point, the film boiling rate is 100%. Therefore, if the surface temperature of the thick steel plate is in the transition boiling region, the nucleate boiling ratio (film boiling ratio) is determined according to the ratio.

Then, the heat transfer coefficient is calculated for each cell using this ratio. The calculation calculates the heat transfer coefficient H _{n in} the case of nucleate boiling and the heat transfer coefficient H _f in the case of film boiling, and calculates the heat transfer coefficient H of each cell from the ratio. More specifically, since the heat transfer coefficient in each of nucleate boiling and film boiling is calculated by the following formulas (7) and (8), the ratio of the boiling state is added to these, and the heat transfer coefficient is calculated by formula (9). A transmission rate H is calculated.

Here, Nu _n is the nucleate boiling Nusselt number, Nu _f is the film boiling Nusselt number, λ _w is the thermal conductivity of water [W / m · ° C.], L is the representative length [m], and ΔT _sat is the superheat degree [ [° C.], ΔT _sub is the subcool degree [° C.], T _s is the surface temperature [° C.] of the thick steel plate, T _w is the jet water temperature [° C.], and B is the nucleate boiling rate (0 ≦ B ≦ 1).

  Finally, an average value (average heat transfer coefficient) is calculated for the heat transfer coefficient H calculated for each cell, and this is used as the heat transfer coefficient of the thick steel plate sandwiched between the upper and lower headers. Here, the calculation of the average value may be a simple average, but it is preferable to take an average value integrated in consideration of the cell width in order to make a more accurate prediction.

  Although one nozzle model has been described above, all nozzles are calculated by the same calculation formula. If the water density of the injected cooling water is the same, the calculated value can be used, but if the water density of the cooling water is different, a separate calculation is required.

On the other hand, in the portion of the thick steel plate corresponding to the upper and lower headers adjacent to each other, the heat transfer coefficient is calculated in consideration of the cooling water flow velocity at the position in the plate width direction of the thick steel plate. For example, assuming that the x-axis is taken in the width direction with respect to the central portion of the thick steel plate, the flow rate ν _p of the cooling water at the plate width direction position x is expressed by a quadratic expression of x as shown in the following formula (10). The calculation may be performed using the model. Since ν _p is a parameter of the Reynolds number Re, Re is represented by the following formula (11). This Re is reflected in the Nusselt number, and the heat transfer coefficient H can be calculated using the equations (9) to (11).

Here, a ₁ , a ₂ , and a ₃ represent coefficients. L is the representative length [m], ρ is the cooling water density [kg / m ³ ], and μ is the viscosity coefficient [m · s / kg] of the cooling water.

  By using the above two models, the heat transfer coefficient can be calculated and the temperature of the thick steel plate can be predicted.

1.2. Temperature measurement step S12
Temperature measurement process S12 is a process of measuring the surface temperature of the thick steel plate cooled using the cooling device, using a thermometer. If the thermometer used at this process can measure the surface temperature of the thick steel plate cooled using the cooling device, the form will not be specifically limited. Moreover, the surface temperature is measured in the temperature measurement step S12, and the measurement location only needs to include at least one location in the cooling device and the outlet side of the cooling device. In addition to the outlet side of the cooling device, the cooling device You may measure the surface temperature of the thick steel plate currently cooled in two or more places.

1.3. Back calculation process S13
In the reverse calculation step S13, the boundary temperature (MHF point and the boiling point state) used in the steel plate temperature prediction model is used so that the surface temperature predicted in the temperature prediction step S11 and the surface temperature measured in the temperature measurement step S12 coincide with each other. (CHF point) is a back calculation process. When performing the reverse calculation step S13, a specific value (reference value) is once given as the MHF point and the CHF point, and the surface temperature of the steel plate is substituted by substituting the manufacturing conditions of the thick steel plate into the steel plate temperature prediction model. Is calculated. After that, by changing the MHF point and CHF point, the surface temperature of the steel sheet is recalculated, and the MHF point and CHF point are changed so that the result of the recalculation coincides with the surface temperature measured in the temperature measurement step S2. To do. In the back calculation step S13, the MHF point and the CHF point are back calculated so that the surface temperature predicted in the temperature prediction step S11 matches the surface temperature measured in the temperature measurement step S12.

1.4. Data accumulation step S14
The data accumulation step S14 accumulates the MHF point and the CHF point calculated in the reverse calculation step S13 together with the manufacturing condition data of the steel plate from which the MHF point and the CHF point were obtained (MHF point and CHF point). Are stored in the operation result database in association with the manufacturing conditions of the thick steel plate from which these are obtained. The data accumulated in the operation performance database in the data accumulation step S14 is used as appropriate when cooling the subsequent thick steel plates.

1.5. Extraction step S15
The extraction step S15 is a step of extracting past manufacturing condition data similar in manufacturing condition to the steel plate to be cooled from the operation result database together with the MHF point and the CHF point. In the extraction step S15, in order to determine the MHF point and the CHF point of the thick steel plate to be cooled from now, the past manufacturing condition data similar in manufacturing condition to the thick steel plate to be cooled from the operation performance database is obtained. It is a process of extracting together with CHF points. In the extraction step S15, whether or not the manufacturing conditions of the thick steel plate to be cooled and the past manufacturing conditions are similar depends on the information vector representing the past manufacturing conditions and the manufacturing conditions of the thick steel plate to be cooled from now on. A distance function between the information vectors to be expressed can be defined, and the determination can be made based on the magnitude of the value of the distance function. This distance function can be expressed as, for example, the following formula (12).

Here, d _i (X _i ) is the weighted Euclidean distance in the m-dimensional space, j is the number of the explanatory variable, m is the total number of explanatory variables, w _j is the weight for each explanatory variable, and X _ij is stored in the operation performance database. X _nj is an explanatory variable of the past production condition i, and is an explanatory variable of the production condition n of the thick steel plate to be cooled. In the present invention, what is used as an explanatory variable is not particularly limited, for example, plate thickness, plate width, plate length, cooling stop temperature instruction value, chemical component value, cooling water temperature, production results in the rolling process, etc. Can be used. In the extraction step S15, the similarity of the manufacturing conditions may be determined based on only one criterion or may be determined based on a plurality of different criteria. Here, “determining based on a plurality of different criteria” means, for example, emphasizing the similarity of the cooling stop temperature instruction value or emphasizing the similarity of the chemical component values of the thick steel plate from a plurality of viewpoints. Say to judge based on the set criteria. However, from the viewpoint of making it easy to calculate an appropriate value as the MHF point or CHF point of the thick steel plate to be cooled in the future, it is possible to control the temperature of the thick steel plate during cooling and after stopping cooling with high accuracy. It is preferable to extract a plurality of similar data groups constituted by past manufacturing conditions with similar manufacturing conditions by judging the similarity of the manufacturing conditions based on a plurality of different criteria.

1.6. Boiling change temperature calculation step S16
The boiling change temperature calculation step S16 is a step of calculating the MHF point and the CHF point of the thick steel plate to be cooled from now using the data extracted in the extraction step. The MHF point and the CHF point can be expressed by the following formulas (13) and (14), for example.

Here, α _j and β _j are coefficients, and α ₀ and β ₀ are constant terms. In the boiling change temperature calculation step S16, first, α _{j in the} equation (13) and β _j in the equation (14) are determined by, for example, PLS regression using the data extracted in the extraction step S15. Next, by substituting the data of the manufacturing conditions of the thick steel plate to be cooled from now into X _nj of formula (13) and formula (14), the MHF point and CHF point of the thick steel plate to be cooled can be calculated. it can.

1.7. Cooling stop temperature estimation step S17
The cooling stop temperature estimation step S17 is performed by substituting the MHF point and CHF point calculated in the boiling change temperature calculation step S16 and data on the manufacturing conditions of the thick steel plate to be cooled into the steel plate temperature prediction model. This is a step of estimating the cooling stop temperature of the thick steel plate to be cooled.

1.8. Decision process S18
The determination step S18 is a step of determining the operating condition of the cooling device so that the cooling stop temperature estimated in the cooling stop temperature estimation step S17 matches the target temperature of the thick steel plate at the time of cooling stop. Examples of the operating conditions of the cooling device determined in the determination step S18 include the amount of cooling water sprayed from the cooling device toward the thick steel plate, the conveyance speed of the thick steel plate conveyed in the cooling device, and the like. .

1.9. Cooling step S19
The cooling step S19 is a step of cooling the thick steel plate by operating the cooling device under the operating conditions determined in the determining step S18. In the cooling step S19, the thick steel plate is cooled so that the surface temperature of the thick steel plate after cooling becomes the target value of the cooling stop temperature. When the thick steel plate is being cooled in the cooling step S19, the surface temperature of the thick steel plate is measured, and the MHF point and the CHF point calculated using this measurement result are obtained when the subsequent thick steel plates are cooled. Are used as appropriate.

  As described above, in the cooling control method for a thick steel plate according to the present invention, the change in the surface temperature of the steel plate cooled by the cooling device is measured, and using this measurement result, the MHF point that becomes the core of the steel plate temperature prediction model is measured. And the CHF point is calculated backward, and the calculated MHF point and CHF point are accumulated in the operation result database together with the manufacturing conditions of the thick steel plate. Then, by performing learning control utilizing the operation result database, the MHF point and CHF point of the thick steel plate to be cooled are determined, and the cooling stop temperature predicted using the determined MHF point and CHF point is the cooling stop. The operating conditions of the cooling device are determined so that the target temperature value is reached, and the thick steel plate is cooled. By cooling the thick steel plate in this way, it becomes possible to control the temperature of the thick steel plate during the cooling and after the cooling stop with high accuracy. Therefore, according to the present invention, it is possible to provide a cooling control method for a thick steel plate capable of increasing the temperature control accuracy during cooling and after cooling stop.

2. Thick Steel Plate Manufacturing Method FIG. 7 is a diagram for explaining the thick steel plate manufacturing method of the present invention. As shown in FIG. 7, the method for manufacturing a thick steel plate according to the present invention includes a rolling step S21 for rolling the thick steel plate, and a cooling step S22 for cooling the thick steel plate after the rolling step S21. In S22, the cooling control method for the thick steel plate of the present invention is used. As described above, according to the thick steel plate cooling control method of the present invention, it is possible to increase the control accuracy of the temperature of the thick steel plate during and after cooling. By having such cooling process S22, in the manufacturing method of the thick steel plate of this invention, the temperature of the thick steel plate in the middle of cooling and after cooling stop is controlled with high precision. Thereby, since it becomes possible to stabilize the mechanical characteristic of a thick steel plate, the manufacturing method of a thick steel plate which can reduce an additive element and can reduce manufacturing cost can be provided.

3. Thick Steel Plate Cooling Control Device and Thick Steel Plate Manufacturing Device FIG. 8 is a diagram for explaining an example of a thick steel plate manufacturing device according to the present invention, including the thick steel plate cooling control device according to the present invention. As shown in FIG. 8, the thick steel plate manufacturing apparatus 100 according to the present invention includes a rolling mill 10 that rolls a thick steel plate, a cooling device 20 that cools the thick steel plate rolled by the rolling mill 10, and a cooling device 20. It has a thermometer 35 that measures the temperature of the thick steel plate before being cooled, and a cooling control device 40 that controls the operating conditions of the cooling device 20. The cooling control device 40 is cooled using the cooling device 20. Thermometers 31, 32, and 33 that measure the temperature of the thick steel plate that is being cooled, and a thermometer 34 that measures the temperature of the thick steel plate after being cooled using the cooling device 20.

  The cooling control device 40 further includes a temperature prediction unit 41 that predicts the temperature of the thick steel plate cooled using the cooling device 20 using a steel plate temperature prediction model, thermometers 31 to 34, and a temperature prediction unit 41. A back calculation unit 42 that back-calculates the MHF point and CHF point used in the steel sheet temperature prediction model so that the predicted surface temperature matches the surface temperature measured using the thermometers 31 to 34, and back calculation by the back calculation unit 42 And an operation result database 43 for storing the MHF point and CHF point together with data on the manufacturing conditions of the thick steel plate from which the MHF point and CHF point were obtained. The cooling control device 40 further extracts data on past manufacturing conditions similar to the steel plate to be cooled in the future from the operation result database 43 together with the MHF point and the CHF point, and an extraction unit 44 Using the extracted data, the MHF point, and the CHF point, the boiling change temperature calculation unit 45 that calculates the MHF point and the CHF point of the thick steel plate to be cooled from now, and the MHF point calculated by the boiling change temperature calculation unit 45 And the cooling stop temperature estimation unit 46 for estimating the cooling stop temperature of the thick steel plate to be cooled by substituting the CHF point and the manufacturing condition data of the thick steel plate to be cooled in the future into the steel plate temperature prediction model, and The cooling stop temperature of the cooling device 20 is adjusted so that the cooling stop temperature estimated by the cooling stop temperature estimation unit 46 matches the target temperature of the thick steel plate at the time of cooling stop. It has a determination unit 47 for determining the work conditions, the. The cooling device 20 is operated under the operation conditions determined by the determination unit 47. According to the cooling control device 40 configured as described above, it is possible to carry out the thick steel plate cooling control method of the present invention. Therefore, by adopting such a form, according to the present invention, there is provided a thick steel plate cooling control device 40 capable of controlling the temperature of the thick steel plate during cooling and after cooling stop with high accuracy. Can do.

  Thus, the thick steel plate manufacturing apparatus 100 of the present invention is provided with the thick steel plate cooling control device 40 capable of controlling the temperature of the thick steel plate during cooling and after cooling stop with high accuracy. Therefore, the temperature of the thick steel plate during the cooling and after the cooling stop can be controlled with high accuracy. As a result, the mechanical properties of the thick steel plate can be stabilized. Therefore, according to the present invention, the thick steel plate manufacturing apparatus 100 capable of reducing the manufacturing cost by reducing the additive elements is provided. can do.

  In the above description regarding the present invention, the temperature prediction unit 41, the reverse calculation unit 42, the extraction unit 44, the boiling change temperature calculation unit 45, the cooling stop temperature estimation unit 46, and the determination unit 47 are shown separately. The steel sheet cooling control device is not limited to this form, and may be a form in which one control unit having all these functions is provided.

  The present invention will be further described with reference to examples.

  The thick steel plate was cooled by operating the cooling device 20 under the operating conditions determined using the cooling control device 40 provided in the thick steel plate manufacturing apparatus 100 shown in FIG.

  In the prior art different from the present invention, errors and parameters of the steel sheet temperature prediction model are estimated so that the temperature measurement value by the thermometer 34 installed on the outlet side of the cooling device 20 matches the temperature prediction value, Used for cooling control after material. On the other hand, in the present invention, in addition to the thermometer 34, the thermometers 31, 32, and 33 measure the steel plate temperature during cooling in the cooling device 20. Thereby, the MHF point and the CHF point can be calculated backward so that the measured temperature value and the predicted temperature value match not only the outlet side of the cooling device 20 but also the temperature history during the cooling by the cooling device 20. It becomes possible.

The back calculation method of the MHF point (boundary between film boiling and transition boiling) and the CHF point (boundary between transition boiling and nucleate boiling) will be described below. In this example, the MHF point (T_MHF) is represented by the following formula (15) or formula (16), and the CHF point (T_CHF) is represented by the following formula (17) or formula (18). In the following formulas (15) to (18), _Tw is the temperature of the cooling water.

In the above equations (15) to (18), the parameters for calculating the reference MHF point and CHF point are a_MHF = 900, b_MHF = 200, c_MHF = 87.5, a_CHF = 500, b_CHF = 150, c_CHF = 87. .5, substituting a_MHF = α × 900, b_MHF = α × 200, c_MHF = α × 87.5, a_CHF = β × 500, b_CHF = β × 150, c_CHF = β × 87.5 By substituting the MHF point and CHF point obtained by the above, the actual value of the cooling water amount, and the actual value of the steel plate conveyance speed into the steel plate temperature prediction model, the predicted value T _cal of the steel plate temperature at the positions of the thermometers 31 to 34 _{, i} was calculated. As the thermometer 31 to 34 according to the temperature measurements _{T act, i} and _{T cal,} and the _i roughly match, i.e., so as to minimize an evaluation function J of the formula (19), the parameter alpha, beta Was calculated backwards. Here, the parameters for calculating the reference MHF point and CHF point are the results of cooling a plurality of thick steel plates, and the relationship between the derived MHF point and CHF point and the steel plate surface temperature and water temperature at that time. It is desirable to seek and decide.

Here, i = 1 means the position of the thermometer 31, i = 2 means the position of the thermometer 32, i = 3 means the position of the thermometer 33, and i = 4 means the position of the thermometer 34. W _Ti is a weight related to an error at each temperature measurement position. In the present embodiment, the weight on the cooling device outlet side is increased to w _T1 = 0.1, w _T2 = 0.2, w _T3 = 0.4, and w _T4 = 1.0.

  The back-calculated values of the MHF point and CHF point thus obtained were stored in the operation performance database together with the manufacturing conditions of the thick steel plate from which these back-calculated values were obtained, and used appropriately when predicting the temperature of the next material and subsequent materials.

  In this example, an operation performance database was used in which manufacturing conditions for 10,000 thick steel plates were stored in association with back-calculated MHF points and CHF points. Then, the manufacturing conditions of the thick steel plate to be cooled from now on, specifically, the data for which the cooling stop temperature, the rolling dimension, the chemical composition, etc. are similar to each other 100 are extracted, and the extracted 100 MHF points and CHF are extracted. The relationship between the points and the manufacturing conditions of the thick steel plate was represented by a regression model.

An example of a similar data search method for deriving a regression model is shown in FIG. In the present embodiment, the regression model is derived by extracting from the data having the smallest value of the evaluation function d _i (X _i ) represented by the above formula (12) to the data having the 100th smallest value.

In this embodiment, the concept of ensemble learning is applied in order to ensure the reliability of the MHF points and CHF points derived from the regression model.
Specifically, in the search of the similar data, seven different evaluation functions d _i (X _i ) are used to extract seven groups of data and obtain seven regression models.
Then, manufacturing conditions of the thick steel plate to be cooled are input to each of the seven regression models, and seven MHF points and CHF points of the thick steel plate to be cooled are derived.

Here, in this example, the variables (Nos. 1 to 16) shown in Table 1 were used as variables for explaining the similarity of data. The weights shown in Table 1 correspond to w _j in the above formula (12), and weights 1 to 7 in Table 1 are weights used when seven regression models are created.

  In this example, the MHF point and the CHF point derived from the seven regression models were weighted average in this way, thereby specifying the MHF point and the CHF point of the thick steel plate to be cooled. FIG. 10 shows a method for specifying the MHF point and the CHF point of the thick steel plate to be cooled. As shown in FIG. 10, an MHF point and a CHF point are derived for each extracted group. Then, by obtaining five average values excluding the maximum and minimum values of the obtained seven MHF points, the MHF points of the thick steel plate to be cooled are identified, and the maximum values of the seven CHF points obtained are obtained. Then, by obtaining five average values excluding the minimum value, the CHF point of the thick steel plate to be cooled is specified.

  FIG. 11 shows the results of the surface temperature (predicted temperature) of the steel sheet predicted using the MHF point and the CHF point as a reference, and the surface temperature of the steel sheet predicted using the MHF point and CHF point derived by the method of the present invention ( The results of (predicted temperature) are shown in FIG. 11A is a diagram for explaining the prediction accuracy at the position in the cooling device, FIG. 11B is a diagram for explaining the prediction accuracy at the outlet side position of the cooling device, and FIG. FIG. 12B is a diagram for explaining the prediction accuracy at the outlet side position of the cooling device. 11 and 12, “N” is the number of cooled steel plates, “Ave” is the average value of prediction errors, and “σ” is the standard deviation.

  As shown in FIG. 11A, when the surface temperature of the steel sheet in the cooling device was predicted by a method different from the present invention, the average value of the prediction error was 54.4 ° C. On the other hand, as shown in FIG. 12A, when the surface temperature of the steel sheet in the cooling device was predicted according to the present invention, the average value of the prediction error was 31.7 ° C. Comparing both, according to the present invention, the prediction error of the surface temperature of the steel plate in the cooling device can be reduced to about 60% of the conventional prediction error. From this result, it was found that according to the present invention, it is possible to improve the temperature control accuracy during the cooling.

  On the other hand, as shown in FIG. 11B, when the surface temperature (cooling stop temperature) of the steel sheet on the cooling device exit side is predicted by a method different from the present invention, the average value of the prediction error is 13.3 ° C. It was. On the other hand, as shown in FIG. 12B, when the cooling stop temperature was predicted according to the present invention, the average value of the prediction error was −8.8 ° C. Comparing both, according to the present invention, the prediction error of the cooling stop temperature can be reduced to about 2/3 of the conventional prediction error. From this result, it was found that according to the present invention, it is possible to improve the control accuracy of the cooling stop temperature.

DESCRIPTION OF SYMBOLS 10 ... Rolling mill 20 ... Cooling device 31, 32, 33, 34, 35 ... Thermometer 40 ... Cooling control device 41 ... Temperature prediction part 42 ... Back calculation part 43 ... Operation performance database 44 ... Extraction part 45 ... Boiling change temperature calculation part 46: Cooling stop temperature estimation unit 47 ... Determination unit

Claims (6)

A method of controlling cooling of a thick steel plate that passes through a water cooling zone of a cooling device,
Predicting the surface temperature of the thick steel plate cooled using the cooling device, a temperature prediction step,
Measuring the surface temperature of the thick steel plate being cooled using the cooling device, a temperature measuring step;
Back-calculate the temperature at which the boiling state changes in the steel sheet temperature prediction model used in the temperature prediction step so that the surface temperature predicted in the temperature prediction step matches the surface temperature measured in the temperature measurement step. , Reverse calculation process,
A data accumulating step for accumulating the temperature at which the boiling state changes in the reverse calculation step, together with data on manufacturing conditions of the thick steel plate from which the temperature was obtained, in an operation result database;
The extraction process of extracting data of manufacturing conditions similar to the steel sheet to be cooled from now, together with the temperature at which the boiling state changes, from the operation performance database,
Using the data extracted in the extraction step and the temperature at which the boiling state changes, the temperature at which the boiling state of the steel plate to be cooled is calculated, the boiling change temperature calculating step,
By substituting the temperature calculated in the boiling change temperature calculation step and the manufacturing condition data of the thick steel plate to be cooled in the steel plate temperature prediction model, the cooling stop temperature of the thick steel plate to be cooled from now on is calculated. Estimating a cooling stop temperature estimating step;
A determination step of determining operating conditions of the cooling device such that the cooling stop temperature estimated in the cooling stop temperature estimation step matches a target temperature after cooling the thick steel plate;
A method for controlling cooling of a thick steel plate. By using a plurality of similarity evaluation criteria in the extraction step, a plurality of groups of data are extracted from the operation performance database,
The method for controlling cooling of a thick steel plate according to claim 1, wherein the temperature at which the boiling state of the thick steel plate to be cooled is changed is calculated in the boiling change temperature calculating step using the extracted data of a plurality of groups. . Rolling a thick steel plate;
Cooling the thick steel plate after the rolling step,
A method for manufacturing a thick steel plate, wherein the cooling control method for a thick steel plate according to claim 1 or 2 is used in the cooling step. A cooling control device for controlling operating conditions of a cooling device for cooling a rolled steel plate,
Predicting the surface temperature of the thick steel plate cooled using the cooling device, a temperature prediction unit,
A thermometer for measuring the surface temperature of the thick steel plate being cooled using the cooling device;
Back-calculate the temperature at which the boiling state changes in the steel sheet temperature prediction model used in the temperature prediction unit so that the surface temperature predicted by the temperature prediction unit matches the surface temperature measured using the thermometer. The back calculation part,
The temperature at which the boiling state is changed in the reverse calculation unit is accumulated together with data on manufacturing conditions of the thick steel plate from which the temperature is obtained, and an operation result database,
An extraction unit that extracts data of manufacturing conditions similar to the steel sheet to be cooled from now, together with the temperature at which the boiling state changes, from the operation performance database,
Using the data extracted by the extraction unit and the temperature at which the boiling state changes, the temperature at which the boiling state of the steel plate to be cooled is calculated, the boiling change temperature calculating unit,
By substituting the temperature calculated by the boiling change temperature calculation unit and the manufacturing condition data of the thick steel plate to be cooled in the steel plate temperature prediction model, the cooling stop temperature of the thick steel plate to be cooled is estimated. A cooling stop temperature estimation unit,
A determination unit that determines operating conditions of the cooling device so that the cooling stop temperature estimated by the cooling stop temperature estimation unit matches a target temperature after cooling the thick steel plate;
A thick steel plate cooling control device. In the extraction unit, by using a plurality of similarity evaluation criteria, a plurality of groups of data are extracted from the operation performance database,
The cooling control apparatus for a thick steel plate according to claim 4, wherein the boiling change temperature calculation unit calculates the temperature at which the boiling state of the thick steel plate to be cooled is changed by using the extracted data of a plurality of groups. . A rolling mill for rolling a thick steel plate, a cooling device for cooling the thick steel plate rolled by the rolling mill, and a cooling control device for controlling the operation of the cooling device,
An apparatus for manufacturing a thick steel plate, wherein the cooling control device is the cooling control device for a thick steel plate according to claim 4 or 5.