JP2022024340A - Steel strip material prediction method, material control method, production method and method for creating material prediction model - Google Patents

Abstract

To provide a steel strip material prediction method where the mechanical properties of a steel strip in the outlet side of a continuous annealing facility are predicted at high precise, a method for creating a material prediction model, a steel strip material control method where variation in the mechanical properties of a steel strip in the outlet side of a continuous annealing facility is reduced, and a production method therefor.SOLUTION: A steel strip material prediction method comprises: an input data acquisition step of, in a continuous annealing facility where a production process including a steel strip annealing stage and a reheating stage is practiced, acquiring one or more parameters selected from operation parameters in a continuous annealing facility and transformation ratio information measured using a transformation ratio meter 20 in at least one of the annealing stage and the reheating stage as input data; and a step of predicting the mechanical properties of the steel strip in the downstream side of the reheating stage using a material prediction model learnt by mechanical learning with information regarding the mechanical properties of the steel strip in the downstream side of the reheating stage as output data.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a steel strip material prediction method, a material control method, a manufacturing method, and a material prediction model generation method. The present disclosure relates to a material prediction method, a material control method, a manufacturing method, and a method for generating a material prediction model of a steel strip in a continuous annealing facility for manufacturing a cold-rolled steel sheet or a hot-dip galvanized steel sheet.

In the continuous annealing equipment for manufacturing cold-rolled steel sheets or hot-dip galvanized steel sheets, materials are manufactured using phase transformation by heat-treating the steel strips after hot rolling and cold rolling.

The continuous annealing equipment (CAL, continuous annealing line) for producing cold-rolled steel sheets has a pre-tropical, heating zone, soaking zone and cooling zone for performing the annealing process, and then an overaging zone for performing the reheating process. Is common. A continuous hot-dip galvanizing facility (CGL, hot-dip galvanizing line) for producing hot-dip galvanized steel sheets is equipped with pre-tropical, heating, soaking and cooling zones for performing annealing steps, and steel strips cooled to a predetermined temperature. It is generally provided with a zinc plating tank (zinc pot) for immersing the steel, and an alloying zone, a tropical zone, and a cooling zone for performing a reheating step after adjusting the amount of zinc. In this way, the annealing step and the reheating step are executed in any of the continuous annealing facilities, and the mechanical properties (yield point elongation, yield stress, tensile strength, elongation, etc.) of the steel strip are adjusted on the downstream side thereof. The annealing rolling process is performed. Further, on the downstream side of the temper rolling process, an inspection step of the steel strip is executed to inspect the surface defects of the steel strip and the like. In the inspection process, test pieces are collected from the steel strip and inspected for mechanical properties offline.

Various methods have been conventionally proposed for controlling the mechanical properties of steel strips in such continuous annealing equipment.

For example, Patent Document 1 discloses a material prediction method using a neural network for tensile strength. In Patent Document 1, as the material influencing factors of the steel strip, the content of C, Si, Mn which is the component composition of the steel strip, the product size (plate thickness), the annealing conditions (annealing temperature, quenching start temperature, quenching water temperature, tempering). Aspects including temperature) as a variable are described. Further, in order to reduce the variation in the material, a method of controlling the material so that the material is within the target range by using a control factor selected from the material influencing factors is disclosed.

Further, Patent Document 2 discloses an apparatus and a method for measuring a transformation rate of a steel strip in a continuous annealing facility. In Patent Document 2, by measuring the transformation rate of the steel strip especially on the upstream side of the reheating step, the control of the temperature rising process by the induction heating equipment in the reheating step is stabilized, and the material of the steel strip to be the product is stabilized. It is disclosed that it can be converted.

Japanese Unexamined Patent Publication No. 2010-106314 Japanese Unexamined Patent Publication No. 2019-7907

However, in the method described in Patent Document 1, the above-mentioned material influence factor is disclosed as a factor affecting the material of the steel strip, but it includes information on the internal structure of the steel strip in the continuous annealing equipment. Not done. In particular, in the manufacture of high-strength steel sheets, microstructure control utilizing phase transformation is actively performed, and information on phase transformation in the manufacturing process is not used, so the accuracy of predicting the mechanical properties of steel strips is sufficient. I can't say.

On the other hand, the method described in Patent Document 2 measures the transformation rate of a steel strip in a continuous annealing facility and identifies the austenite fraction of the steel strip, thereby stabilizing the heating characteristics in induction heating. Is. However, there is no description of how to reduce the variation in the mechanical properties of the steel strip by measuring the transformation rate of the steel strip.

The object of the present disclosure made to solve the above problems is used in a steel strip material prediction method for predicting the mechanical properties of a steel strip on the exit side of a continuous annealing facility with high accuracy, and a material prediction method thereof. The purpose is to provide a method for generating a material prediction model. Another object of the present disclosure is to provide a material control method and a manufacturing method of a steel strip in which the material prediction method is used to reduce the variation in the mechanical properties of the steel strip on the exit side of the continuous annealing equipment. be.

The method for predicting the material of the steel strip according to the embodiment of the present disclosure is as follows.
A method for predicting the material of a steel strip in which the mechanical properties of the steel strip on the downstream side of the reheating step are predicted in a continuous annealing facility that executes a manufacturing process including an annealing step and a reheating step of the steel strip.
As input data, an input data acquisition step for acquiring one or more parameters selected from the operation parameters of the continuous annealing facility and transformation rate information measured in at least one of the annealing step and the reheating step.
Using a material prediction model learned by machine learning, which uses information on the mechanical properties of the steel strip on the downstream side of the reheating process as output data, the mechanical properties of the steel strip on the downstream side of the reheating process can be determined. Includes predicting.

The method for controlling the material of the steel strip according to the embodiment of the present disclosure is as follows.
Using the above method for predicting the material of the steel strip, the mechanical properties of the steel strip on the downstream side of the reheating step are predicted, and if the predicted mechanical properties are outside the preset target range, the mechanical properties are mechanical. Includes resetting one or more operating parameters selected from the operating parameters of the continuous annealing facility so that the properties fall within the target range.

The method for manufacturing a steel strip according to an embodiment of the present disclosure is as follows.
A method for manufacturing a steel strip in a continuous annealing facility that executes a manufacturing process including an annealing step and a reheating step of the steel strip.
To acquire one or more parameters selected from the operation parameters of the continuous annealing equipment and transformation rate information measured in at least one of the annealing step and the reheating step as input data.
Using a material prediction model learned by machine learning, which uses information on the mechanical properties of the steel strip on the downstream side of the reheating process as output data, the mechanical properties of the steel strip on the downstream side of the reheating process can be determined. Predicting and
If the predicted mechanical properties are outside the preset target range, reset one or more operating parameters selected from the operating parameters of the continuous annealing equipment so that the mechanical properties fall within the target range. And, including.

The method for generating the steel strip material prediction model according to the embodiment of the present disclosure is as follows.
A method for generating a steel strip material prediction model for predicting the mechanical properties of a steel strip downstream of the reheating step in a continuous annealing facility that executes a manufacturing process including a steel strip annealing step and a reheating step. There,
At least one or more operation record data selected from the operation record data of the continuous annealing facility and the record data of the transformation rate information measured in at least one of the annealing step and the reheating step are acquired as input record data. To do and
Acquiring a plurality of learning data using information on the mechanical properties of the steel strip on the downstream side of the reheating process based on the input actual data as output actual data.
It includes generating a steel strip material prediction model by machine learning using a plurality of acquired training data.

According to the present disclosure, a method for predicting the material of a steel strip that predicts the mechanical properties of the steel strip on the exit side of a continuous annealing facility with high accuracy, and a method for generating a material prediction model used in the material prediction method are provided. be able to. According to the present disclosure, it is possible to provide a material control method and a manufacturing method of a steel strip that reduce variations in mechanical properties of the steel strip on the exit side of a continuous annealing facility by using the material prediction method.

FIG. 1 is a diagram showing a continuous annealing line for manufacturing a cold-rolled steel sheet as a continuous annealing facility. FIG. 2 is a diagram showing a hot-dip plating line for manufacturing a galvanized steel sheet as a continuous annealing facility. FIG. 3 is a diagram showing an example of heat history in a continuous annealing line for manufacturing a cold-rolled steel sheet. FIG. 4 is a diagram showing an example of thermal history in a hot-dip plating line for manufacturing a galvanized steel sheet. FIG. 5 is a diagram showing a method of generating a material prediction model. FIG. 6 is a diagram showing a material control method.

The method for predicting the material of the steel strip according to the present embodiment is manufactured by heat-treating a steel sheet whose thickness has been reduced to a predetermined plate thickness through a hot rolling step, a pickling step, and a cold rolling step by a continuous annealing facility. Predict the mechanical properties of cold-rolled or hot-rolled zinc-plated steel sheets on the exit side of continuous annealing equipment. At least after the hot rolling step, the thin steel sheet is wound into a coil and then heat-treated. Therefore, in the present embodiment, the thin steel sheet is referred to as a "steel strip".

<Continuous annealing equipment>
In the present embodiment, the continuous annealing equipment is intended for heat treatment equipment that executes a manufacturing process including an annealing step and a reheating step of a steel strip. It can be applied to both a continuous annealing line (CAL) for producing a cold-rolled steel sheet and a hot-dip plating line (CGL) for producing a galvanized steel sheet. FIG. 1 is a diagram showing a continuous annealing line for manufacturing a cold-rolled steel sheet. FIG. 2 is a diagram showing a configuration of a hot-dip plating line for manufacturing a galvanized steel sheet. Hereinafter, the present disclosure will be specifically described with reference to the drawings.

<Continuous annealing line for manufacturing cold-rolled steel sheets>
FIG. 1 is a schematic view showing an example of a continuous annealing facility for manufacturing a cold-rolled steel sheet according to the present embodiment. The arrow in FIG. 1 indicates the direction of travel of the line. The continuous annealing equipment is roughly divided into entrance side equipment, furnace body part and exit side equipment. The entry side equipment includes a payoff reel 1, a welding machine 2, an electrolytic cleaning device 3, and an entry side looper 4. The furnace body part is composed of an annealing part and a reheating part. The annealed portion may have a heating zone 6, an average tropic 7 and a cooling zone 8, and may have a pre-tropical 5 on the upstream side of the heating zone 6. The annealing step in the present embodiment is a heat treatment step performed in the annealed portion. On the other hand, the reheating section has a reheating zone 9, a super-aging zone 10, and a final cooling zone 11, and an induction heating device is arranged in the reheating section. The reheating step in the present embodiment is a heat treatment step performed in the reheating section. The outgoing side equipment includes an outgoing side looper 12, a tempering rolling equipment 13, an inspection equipment 14, and a tension reel 15. The inspection equipment 14 has a sampling equipment for collecting a sample material for measuring mechanical properties offline such as a tensile test from a steel strip.

The annealing step is a step of raising the temperature of the steel strip from around room temperature, holding it at a predetermined temperature, and then lowering the temperature to around room temperature. In the continuous annealing equipment shown in FIG. 1, the annealing step is executed by the heating zone 6, the soaking zone 7, and the cooling zone 8. The reheating step is a step of performing an overaging treatment of the steel strip that has passed through the cooling zone 8. In the continuous annealing equipment shown in FIG. 1, the reheating step is executed by the reheating zone 9, the aging zone 10, and the final cooling zone 11.

The heating zone 6 is a facility for raising the temperature of the steel strip, and heats the steel strip to a preset temperature in the range of about 600 to 900 ° C. depending on the steel type. In the heating zone 6, a direct flame or radiant combustion burner is used. Since these heating devices have a large heating capacity and a relatively quick response, it is easy to change the temperature rise history when the heat cycle is changed. The soothing tropic 7 is a facility that keeps the steel strip at a predetermined temperature, and is a facility with a heating capacity that supplements the heat dissipated from the furnace body.

The cooling zone 8 is a facility for cooling the steel strip to a predetermined temperature, and gas jet cooling, roll cooling, water cooling, or the like is used as the cooling means. Gas jet cooling is a cooling means for blowing gas from a nozzle onto the surface of a steel strip. Roll cooling is a cooling means for cooling a steel strip by contacting it with a water-cooled roll. Water cooling is a cooling means for cooling by immersing a steel strip in a water cooling tank installed on the downstream side of the soothing tropics 7. Since the cooling rate of the steel strip by these cooling devices is different, the cooling zone 8 is divided into a plurality of cooling zones 8A such as the first cooling zone 8A and the second cooling zone 8B, and different cooling means can be combined or the same type of cooling means can be used. The cooling conditions of the steel strip may be changed to control the heat history during cooling of the steel strip.

The reheating zone 9 is arranged on the downstream side of the cooling zone 8, and after cooling the steel strip to a predetermined temperature in the cooling zone 8, it is reheated to a temperature of about 300 to 400 ° C. using an induction heating device. The overage zone 10 is a facility that performs an overage treatment for holding the reheated steel strip for a predetermined time. The final cooling zone 11 is a facility for finally cooling the overaged steel strip to near room temperature. Similar to the cooling zone 8, the final cooling zone 11 may be divided into a plurality of sections such as the first final cooling zone 11A and the second final cooling zone 11B, and the thermal history during cooling of the steel strip may be controlled.

Further, in the heating zone 6, the uniform tropical zone 7, the cooling zone 8, and the reheating zone 9, the overaging zone 10, and the final cooling zone 11 constituting the reheating step, the surfaces of the steel strips are located at a plurality of positions. A thermometer that measures the temperature is installed. In particular, in the cooling zone 8 where the temperature change of the steel strip is large, thermometers are installed on the entry side and the exit side of the cooling zone 8, and the cooling rate of the cooling zone 8 is measured by measuring the surface temperature of the steel strip at the position. The actual value is calculated. Further, as a thermometer, a radiation thermometer that continuously measures the surface temperature of the central portion of the plate width of the steel strip is used. However, the thermometer is not limited to the radiation thermometer, and a profile radiation thermometer that measures the temperature distribution in the plate width direction may be used. In addition, an in-core thermometer that measures not only the surface temperature of the steel strip but also the ambient temperature in the furnace in each zone of the annealing process and the reheating process is installed. The measured surface temperature and atmospheric temperature of the steel strip are output to the process computer that controls the continuous annealing equipment and controls the operation.

FIG. 3 is a graph showing the thermal history of the steel strip in the annealing step and the reheating step. The horizontal axis shows time and the vertical axis shows steel strip temperature. The steel strip temperature is, for example, the surface temperature of the steel strip. The annealing step is performed by the heating zone 6, the soaking zone 7 and the cooling zone 8, and then the reheating step is performed by the reheating zone 9, the overaging zone 10 and the final cooling zone 11. In order to prevent material variation due to the position of the steel strip in the longitudinal direction, the transport speed of the steel strip during the annealing process is kept constant. However, when steel strips having different plate thicknesses, plate widths, steel types, etc. are welded, the line speed may change before and after the welded portion. Therefore, the shape of the thermal history graph may vary depending on the measurement position of the steel strip.

<Hot-dip plating line for manufacturing hot-dip galvanized steel sheets>
FIG. 2 is a schematic view showing an example of a continuous annealing facility for manufacturing a hot-dip galvanized steel sheet according to the present embodiment. The arrow in FIG. 2 indicates the traveling direction of the steel strip. The continuous annealing equipment in this embodiment is roughly classified into an entrance side equipment, a furnace body part, and an exit side equipment. The entry-side equipment includes a payoff reel 1, a welding machine 2, an electrolytic cleaning device 3, and an entry-side looper 4, as in FIG. 1. The furnace body portion is composed of an annealed portion, a plating portion and a reheating portion. The annealed portion may have a heating zone 6, an average tropic 7 and a cooling zone 8, and may have a pre-tropical 5 on the upstream side of the heating zone 6. The annealing step in the present embodiment is a heat treatment step performed in the annealed portion. Further, in a hot-dip galvanizing line for manufacturing hot-dip galvanized steel sheets (hereinafter referred to as "hot-dip galvanizing line"), a steel strip having a plated portion on the downstream side of the annealed portion and cooled to a predetermined temperature in the cooling zone 8. Is immersed in the zinc plating tank 16, and the amount of zinc grain is adjusted by the wiping device 21. The reheating section on the downstream side has an alloying zone 17, a tropical zone 18, and a final cooling zone 11, and an induction heating device is arranged in the alloying zone 17. The reheating step in the present embodiment is a heat treatment step performed in the reheating section. The outgoing side equipment includes an outgoing side looper 12, a tempering rolling equipment 13, an inspection equipment 14, and a tension reel 15. The inspection facility 14 has a sampling facility for sampling a sample material for measuring mechanical properties offline such as a tensile test from a steel strip.

The annealing step is a step of raising the temperature of the steel strip from around room temperature, holding it at a predetermined temperature, and then lowering the temperature of the steel strip to a temperature suitable for galvanizing.

The heating zone 6 is a facility for raising the temperature of the steel strip, and heats the steel strip to a preset temperature in the range of about 700 to 850 ° C. depending on the steel type. In the heating zone 6, a direct flame or radiant combustion burner is used. The soothing tropic 7 is a facility that keeps the steel strip at a predetermined temperature, and is a facility with a heating capacity that supplements the heat dissipated from the furnace body. The cooling zone 8 is a facility for cooling to about 480 ° C. as a temperature suitable for performing zinc plating, and gas jet cooling is generally used as the cooling means. By dividing the cooling zone 8 into a plurality of cooling zones 8A, a second cooling zone 8B, and the like and changing the cooling conditions of the cooling means, the thermal history at the time of cooling the steel strip may be controlled.

The plating portion is composed of a snout 19, a zinc plating tank 16, and a wiping device 21 connected to the outlet of the cooling zone 8. The snout 19 is a member having a rectangular cross section that divides the space through which the steel strip passes, and a mixed gas containing hydrogen, nitrogen, and water vapor is supplied to the inside until the steel strip is immersed in the galvanizing tank 16. Atmospheric gas is adjusted. The galvanizing tank 16 has a sink roll inside, passes through the snout 19 and is immersed in the galvanizing tank 16 downward, and the steel strip having molten zinc adhered to the surface is pulled up above the plating bath. Equipment. Further, the wiping device 21 blows wiping gas from nozzles arranged on both sides of the steel strip to scrape off excess molten zinc adhering to the surface of the steel strip, and the amount of molten zinc adhered (also referred to as a grain amount). ) Is a facility to adjust.

A reheating zone 9 (called an alloyed zone 17 in the hot-dip galvanizing line) constituting the reheating portion is arranged above (downstream side) the wiping device 21 constituting the plating portion. Normally, the temperature of the steel strip that has passed through the wiping device 21 drops to about 430 ° C. Therefore, the steel strip is heated to a temperature at which the Zn—Fe alloying reaction proceeds in the alloying strip 17. The temperature at which the temperature rises in the alloying zone 17 corresponds to the target alloying temperature and varies depending on the alloy component of the steel strip, the Al concentration in the plating bath, and the like. However, normally, the temperature of the steel strip is raised to about 500 ° C. in the alloyed strip 17. After that, in the tropics 18, the temperature of the steel strip is maintained in order to secure the time required for the alloying reaction to proceed. A final cooling zone 11 is provided on the downstream side of the tropical 18 and is a facility for finally cooling the alloyed steel strip to near room temperature. Similar to the cooling zone 8, the final cooling zone 11 may be divided into a plurality of sections such as the first final cooling zone 11A and the second final cooling zone 11B, and the thermal history during cooling of the steel strip may be controlled.

Even in the hot-dip galvanizing line, steel is located at multiple positions in the heating zone 6, the soaking zone 7, the cooling zone 8, the alloying zone 17, the tropical zone 18, and the final cooling zone 11 that constitute the reheating process. A thermometer is installed to measure the surface temperature of the band. In addition, an in-core thermometer that measures not only the surface temperature of the steel strip but also the ambient temperature in the furnace in each zone of the annealing process and the reheating process is installed. The measured surface temperature and atmospheric temperature of the steel strip are output to the process computer that controls the hot-dip galvanizing line, which is a continuous annealing facility, and controls the operation.

FIG. 4 is a graph showing the thermal history of a steel strip including an annealing step and a reheating step of a continuous annealing facility for manufacturing a hot-dip galvanized steel sheet. The horizontal axis and the vertical axis are the same as those in FIG. The annealing step is performed by the pre-tropical zone 5, the heating zone 6, the leveling zone 7, and the cooling zone 8 (first cooling zone 8A and second cooling zone 8B), and then passing through the plating section to form the alloying zone 17. , The tropics 18 and the final cooling zone 11 (first final cooling zone 11A and second final cooling zone 11B) perform the reheating step. The transport speed of the steel strip during the annealing process is kept constant in order to prevent material variation due to the position of the steel strip in the longitudinal direction. However, when steel strips having different plate thicknesses, plate widths, steel types, etc. are welded, the line speed may change before and after the welded portion.

<Transformation rate meter>
The transformation rate meter 20 according to the present embodiment is a measuring instrument that measures the ratio of the austenite phase (γ phase) to the whole as the internal structure of the steel strip in the heat treatment step. In a continuous annealing facility, the structure of a steel strip is often controlled by using a phase transformation from a two-phase state of a specific austenite phase (γ phase) and a ferrite phase (α phase). As the transformation rate meter 20, a transformation rate meter 20 using X-ray diffraction can be used. Since the crystal structures of the γ phase and the α phase are different, when X-rays are applied, diffraction peaks are generated from each at a unique angle. The transformation rate (γ rate) can be quantified from this diffraction peak intensity. For example, a product called X-CAP manufactured by SMS can be used. Further, as a magnetic detector, that is, a device for measuring the magnetic transformation rate of a steel strip, a magnetic transformation rate measuring device composed of a drive coil for generating a magnetic field and a detection coil for measuring the magnetic field transmitted through the steel strip is used. Alternatively, a method of measuring the austenite phase ratio may be used. Specifically, the apparatus described in Patent Document 2 can be used.

In the present embodiment, the transformation rate meter 20 for measuring the ratio of such austenite phase is installed at a position on the line where at least one of the annealing step and the reheating step of the continuous annealing equipment is executed. In other words, in the annealing step, the reheating step, or the annealing step and the reheating step, the transformation rate is measured by using the transformation rate meter 20. As the installation location of the transformation rate meter 20, for example, the inlet of the tropics 7 where the annealing process is executed, the outlet of the tropics 7, the inlet of the cooling zone 8, and the reheating zone 9 (alloying zone) where the reheating process is executed. It is preferable to install it at the entrance or the exit of 17). Further, the transformation rate meter 20 may be installed at any one place, but it is preferable that the transformation rate meter 20 is installed at a plurality of different positions. This is because the material prediction accuracy is improved by obtaining a plurality of transformation rate information.

<Information on the mechanical properties of steel strips>
The mechanical properties of the steel strip according to the present embodiment include the yield point elongation, yield stress, tensile strength, elongation, Rankford value of the steel strip, or the value in any direction in the plane of the steel strip. , Refers to the material characteristics of steel strips. Specifically, the characteristic values obtained from the material tests such as the offline tensile test, bending test, and deep drawing test by collecting the test piece from the sample material of the steel strip collected in the inspection process of the continuous annealing equipment. be. In the present embodiment, it is particularly preferable to use the tensile strength among these characteristic values. This is because the demand for high-strength steel sheets using continuous annealing equipment has increased in recent years, and it is highly necessary to control the strength of steel strips. Further, the variation in the austenite phase in the annealing step has a great influence on the tensile strength of the steel strip.

Information on the mechanical properties uses characteristic values obtained from test pieces taken from either the tip and / or tail of the strip that is heat treated by a continuous annealing facility. When the tensile strength is used as information on the mechanical properties, specifically, the JIS No. 5 test piece is collected from the test piece collected in the direction parallel to the direction orthogonal to the longitudinal direction (C direction) of the steel strip. However, by conducting a tensile test in accordance with the provisions of JIS Z2241, information on tensile strength can be obtained.

The information on the mechanical properties of the steel strip thus obtained is the manufacturing information to the process computer together with the identification number (coil number) of the steel strip from which the test piece was sampled and, if necessary, the information on the sampling position. Etc. are sent to the host computer.

<Method of generating material prediction model>
FIG. 5 shows a method of generating a material prediction model of a steel strip according to the present embodiment.

The operation performance data of the continuous annealing facility, the performance data of the transformation rate information of the steel strip measured by the transformation rate meter 20, and the performance data of the information on the mechanical properties of the steel strip are accumulated in the database. The details of the operation performance data of the continuous annealing equipment will be described later, but the performance data selected from the operation performance data possessed by the process computer that controls the operation of the continuous annealing equipment is sent to the database of the material prediction model generation unit. Further, the actual data of the transformation rate information of the steel strip is the transformation rate information obtained from the transformation rate meter 20, and when the transformation rate information is stored in the process computer, it is sent from the process computer to the database. However, if the transformation rate information is not stored in the process computer, it is sent directly to the database of the material prediction model generation unit.

The transformation rate information is sent to the database together with incidental information that can be associated with the operation record data of the continuous annealing equipment, such as the coil number of the steel strip. Furthermore, the actual data of the information on the mechanical properties of the steel strip is the information obtained by the offline test and is accumulated in the host computer. This information is also sent to the database together with incidental information that can be associated with the operation record data of the continuous annealing equipment such as the coil number of the steel strip. Then, the operation record data of the continuous quenching facility, the track record data of the transformation rate information of the steel strip measured by the transformation rate meter 20, and the track record data of the information on the mechanical properties of the steel strip are associated with each other by a coil number or the like. It is stored in the database as a set of data sets. At this time, the data set stored in the database acquires one data set for each steel strip. However, if the actual data of the information on the mechanical properties of the steel strip is obtained at multiple locations such as the tip and tail end of the steel strip, the tip and tail end of the steel strip may be multiple. A plurality of data sets may be acquired for one steel strip by using the operation record data of the continuous quenching facility and the track record data of the transformation rate information of the steel strip acquired at the location.

Further, it is preferable that the database has one or more parameters selected from the attribute parameters of the steel strip regarding the composition of the steel strip. The actual data of the attribute parameters of the steel strip related to the composition of the steel strip are stored in the process computer or the host computer together with the coil number as the actual values in the steelmaking process, and the data set can be configured by appropriately sending to the database. .. This is because the material prediction model can be widely applied to steel strips having different component compositions by adding the attribute parameters of the steel strips related to the composition of the steel strips to the input.

The number of database data sets used to generate the material prediction model of the present embodiment is preferably 200, more preferably 1000 or more.

In the present embodiment, using the database thus created, at least one or more operation record data selected from the operation record data of the continuous annealing equipment, and a transformation rate meter in at least one of the annealing step and the reheating step. Using the actual data of the transformation rate information measured by 20 and the input actual data as the input actual data, a material prediction model of the steel strip by machine learning using the input actual data is generated.

As the machine learning method, a known learning method may be applied. The type of machine learning model is not limited as long as the accuracy of predicting the mechanical properties of the steel strip is sufficient for practical use. For example, a known machine learning method such as a neural network may be used. Examples of other methods include decision tree learning, random forest, support vector regression, and Gaussian process. Further, an ensemble model in which a plurality of models are combined may be used. In addition, the material prediction model may be updated as appropriate using the latest learning data. This is because it can respond to changes in long-term operating conditions of continuous annealing equipment.

<Operational parameters of continuous annealing equipment>
As the operation parameters in the continuous annealing equipment, any operation parameters other than the transformation rate information measured by the transformation rate meter 20 and which affect the mechanical properties of the steel strip can be used.

Using the example of the thermal history of the steel strip in the annealing step and the reheating step shown in FIG. 3, the following operating parameters of the annealing step can be used. For example, as the operation parameter of the heating zone 6, the time and the amount of temperature rise through which the steel strip passes through the heating zone 6 may be used, or the average temperature rising rate calculated from these values may be used.

As the operating parameters of the tropics 7, the soaking temperature, which is the average temperature of the steel strip in the tropics 7, and the soaking time, which is the time for passing through the tropics 7, may be used. As the operating parameters of the cooling zone 8 divided into a plurality of sections, the time and the amount of temperature decrease for the steel strip to pass through the first cooling zone 8A may be used, or the average cooling rate calculated from these values may be used. good. Further, as the operation parameter of the cooling zone 8, the time and the amount of temperature decrease for the steel strip to pass through the second cooling zone 8B may be used, or the average cooling rate calculated from these values may be used. However, when the temperature of the steel strip is cooled to room temperature in the middle of the second cooling zone 8B, it is preferable to use a substantial cooling rate before reaching room temperature. The substantial cooling rate is calculated from, for example, the temperature difference measured by two or more radiation thermometers installed in the second cooling zone 8B. After the measurement, these values are output to the process computer from the heating zone 6, the tropics 7, and the cooling zone 8.

Further, in the reheating step, the passing time and the amount of temperature rise of the radiation thermometers arranged on the entry side and the exit side of the induction heating device installed in the reheating zone 9 (alloying zone 17) may be used. Alternatively, the average heating rate calculated from these values may be used. As the operating parameters of the tropics 18, the average temperature of the steel strip at the tropics 18 and the time to pass through the tropics 18 may be used. As the operation parameter of the final cooling zone 11, the time and the amount of temperature decrease for the steel strip to pass through the final cooling zone 11 may be used, or the average cooling rate calculated from these values may be used. These values are output to the process computer after measurement.

At least one of the operation parameters of the annealing process and the reheating process as described above is used as teacher data (learning data). However, the operation parameters are not limited to these, and the control output values of the heating devices in the heating zone 6 and the reheating zone 9 (alloying zone 17) may be used as the operation parameters. Further, the control output value of the cooling device in the cooling zone 8 and the final cooling zone 11 may be used as an operation parameter. This is because these operating parameters are used to control the temperature history of the steel strip in the annealing process and the reheating process.

The operating parameters in the annealing and reheating steps as described above are due to the mechanical properties of the strip, such as recovery, recrystallization, grain growth, precipitation, phase transformation, etc. in the internal structure of the strip in continuous annealing equipment. This is because it is determined through the process of, and the factors of temperature and time as the thermal history of the steel strip have a complicated influence.

Further, the operating parameters of the annealing step and the reheating step are not limited to the above parameters, but are the line speed of the steel strip in the uniform tropical zone 7 and the tropical zone 18, the average cooling rate in the cooling zone 8 and the final cooling zone 11. The injection pressure of a cooling device such as gas injection may be used. This is because these are also factors that affect the thermal history of the steel strip.

On the other hand, as the operating parameters of the continuous annealing equipment other than the annealing step and the reheating step, the set value or the measured value of the elongation rate in the temper rolling mill may be used. This is because the temper rolling mill is the final step for adjusting the mechanical properties of the steel strip, and in particular, the elongation ratio affects the yield stress and the tensile strength.

Further, in the continuous annealing equipment for producing a hot-dip galvanized steel sheet, the operation parameters in the plated portion may be used. For example, the operating parameters may include the surface temperature of the steel strip when immersed in the zinc plating tank 16, the bath temperature of the plating bath, the wiping gas temperature in the wiping device 21, the injection pressure, and the like. This is because the temperature of the steel strip is changed by being immersed in the zinc plating tank 16, and the material of the tropical zone 18 may be changed depending on the plating thickness.

As the operation parameters in the continuous annealing equipment in the present embodiment, as the operation data, one set of operation parameters per steel strip is acquired as learning data. This is because the information on the mechanical properties, which is the output data of the material prediction model, is basically collected in units of steel strips. In that case, the above temperature data and the like are data continuously collected in the longitudinal direction of the steel strip, but a representative value is calculated for one steel strip and this is operated in the continuous annealing facility. It is a parameter. For example, data collected at a preset distance from the tip or tail of the strip may be used. Further, data obtained by averaging the measured values in the longitudinal direction of the steel strip may be used.

<Transformation rate information>
In the present embodiment, the transformation rate meter 20 for measuring the ratio of the austenite phase is installed in at least one of the annealing steps or the reheating steps of the continuous annealing equipment, and the measurement result by the transformation rate meter 20 is used as the transformation rate information. It is used as one of the training data of the above material prediction model.

The data obtained by the transformation rate meter 20 is continuous data obtained for each sampling period in the longitudinal direction of the steel strip as the ratio data of the austenite phase of the steel strip, but for one steel strip. A representative value may be calculated and used as actual data of transformation rate information. At this time, the measurement result of the transformation rate measured at the position generally corresponding to the position where the actual data of the information on the mechanical properties of the steel strip, which is the output of the material prediction model, is acquired is used as the actual data of the transformation rate information. Is preferable. In a continuous quenching facility, the transformation rate of the steel strip may fluctuate in the longitudinal direction, and the correlation between the transformation rate and the mechanical properties of the steel strip is relatively high. By associating with the actual data collection position of, it is possible to predict the material with higher accuracy.

In this embodiment, the reason for using the transformation rate information measured by the transformation rate meter 20 in addition to the operation parameters of the continuous annealing equipment is as follows. The operating parameters of the continuous annealing equipment affect the mechanical properties of the steel strip through processes such as recovery, recrystallization, grain growth, precipitation and phase transformation in the internal structure of the steel strip. However, such changes in the internal structure are not determined only by the operating parameters of the continuous annealing equipment, but are also affected by the machining history in the hot rolling process and the cold rolling process, which are the preceding processes. For example, the take-up temperature in the hot rolling process affects the size (distribution) and amount of precipitates as the internal structure of the hot-rolled steel sheet, and affects the grain growth and transformation behavior in the heat treatment process. In addition, the rolling reduction in the cold rolling process affects the recrystallization, grain growth and transformation behavior of the annealing process through the strain state accumulated in the internal structure of the cold-rolled steel sheet. Therefore, as the training data of the material prediction model, it is not possible to consider the influence of the operation parameters of the previous process than the annealing process on the mechanical properties after the heat treatment of the steel strip only by the operation parameters of the continuous annealing equipment. In addition, there was a problem that the material prediction accuracy was low.

On the other hand, by using the transformation rate information measured by the transformation rate meter 20 in the heating step or the reheating step as learning data, the operation in the hot rolling process and the cold rolling process, which are the pre-processes of the annealing process, is performed. The effect of parameters on the mechanical properties of the steel strip after heat treatment can be considered as indirect information in the process in the continuous rolling equipment. This enables highly accurate prediction as a material prediction model. On the other hand, a method of adding the operation parameters in the hot rolling process and the cold rolling process, which are the pre-processes of the annealing process, to the learning data of the material prediction model is also conceivable. However, when accumulating data sets in a database, it is necessary to collect manufacturing performance information from multiple independent manufacturing lines through a computer network and construct a database in which a large amount of corresponding data is associated with each other. Processing by a computer is required. Therefore, it is preferable to use the transformation rate information for learning data rather than directly collecting the operation parameters of the pre-process of the annealing process.

From the above, in the present embodiment, the transformation rate meter 20 for measuring the ratio of the austenite phase is installed in at least one of the annealing step or the reheating step of the continuous annealing equipment, and the measurement result by the transformation rate meter 20 is obtained as the transformation rate. As information, it is used as one of the training data of the above material prediction model.

<Attribute parameters related to the composition of steel strips>
In the present embodiment, it is preferable that the input data of the material prediction model further has one or more parameters selected from the attribute parameters of the steel strip regarding the component composition of the steel strip. This makes it possible to generate a material prediction model that predicts the mechanical properties of steel strips with various component compositions as cold-rolled steel sheets or hot-dip galvanized steel sheets manufactured by continuous annealing equipment, and the applicable range of the material prediction model. Is because it expands.

As the attribute parameter relating to the component composition of the steel strip, the content of C, Si, Mn, P, S can be used as the chemical component contained in the steel strip. Further, the attribute parameters relating to the component composition of the steel strip may include the contents of Cu, Ni, Cr, Al, Nb, Ti, V, N and B. However, it is not necessary to use all of these component compositions as attribute parameters related to the component composition of the steel strip. The composition may be appropriately selected according to the type of steel strip to be manufactured in the continuous annealing equipment. Each component composition affects the mechanical properties of the steel strip as follows.

C is an element that most contributes to the mechanical properties of the steel strip, and is effective in increasing the strength and hardenability. However, if it is too high, ductility and toughness will decrease.

If Si is 0.5% or less, it dissolves in ferrite and increases its strength without impairing its ductility and toughness.

Mn is a substitution type and is solid-dissolved to strengthen ferrite and increase the strength of pearlite in order to make it denser. Compared to C, the strength is higher, but the decrease in elongation is small.

P is an element that easily segregates and affects toughness.

Like P, S is an element that easily segregates and affects toughness.

Cu affects the strength by the precipitation effect.

Ni is an element that is effective for both strength and toughness.

Cr increases hardenability.

Al fixes N to improve strain aging and improve toughness.

Addition of 0.02 to 0.05% of Nb suppresses ferrite formation during A3 transformation, refines crystal grains, and contributes to improvement of toughness.

Ti becomes carbide or nitride with a small amount of addition, contributes to the miniaturization of crystal grains, and improves toughness.

V becomes carbide or nitride with a small amount of addition, contributes to the miniaturization of crystal grains, and improves toughness.

N is a solid solution in an intrusive solid solution like C. Excess N can cause strain aging. However, an appropriate amount of N produces Al, Nb, Ti and a nitride, which makes the crystal grains finer and improves the toughness.

B is added in a trace amount of 0.003% or less to increase hardenability.

Here, in these component compositions, the distribution in the longitudinal direction of the steel strip is generally constant, and one attribute parameter can be acquired as actual data for one steel strip.

Further, as the learning data of the material prediction model of the present embodiment, other than the attribute parameters of the steel strip regarding the component composition of the steel strip may be used. As the attribute parameter of the steel strip, the attribute parameter related to the dimension of the steel strip such as the plate thickness, the plate width, and the length of the steel strip may be used. This is because they affect the heat transfer behavior in the continuous annealing equipment, and thus affect the mechanical properties of the steel strip even at the same furnace atmosphere temperature.

<Material control method for steel strips>
FIG. 6 shows a material control method (manufacturing method) for steel strips using the material prediction method as described above.

The material control method in the present embodiment differs depending on the installation position of the transformation rate meter 20 installed in at least one of the annealing step and the reheating step of the continuous annealing equipment. Specifically, when a plurality of transformation rate meters 20 are installed as transformation rate information used for inputting the material prediction model generated as described above, the transformation rate meter installed on the most downstream side thereof. The band on the upstream side of 20 and the band on the downstream side thereof are separated. The band from the entry side of the continuous annealing equipment to the transformation rate meter 20 installed on the most downstream side is called the "material identification band". Further, the band on the downstream side of the transformation rate meter 20 installed on the most downstream side is referred to as a "material control band". Then, when the tip of the steel strip targeted for material prediction reaches the position of the transformation rate meter 20 installed on the most downstream side and the transformation rate information of the steel strip is acquired, the control flow shown in FIG. Is started.

At this point, for the steel strip targeted for material prediction, the operation record data of the continuous annealing equipment obtained in the material identification zone of the continuous annealing equipment and the transformation rate information measured by the transformation rate meter 20 are available. It is the input data of the material prediction model. The step of acquiring this input data corresponds to the input data acquisition step of the present disclosure. Further, the operation performance data of the continuous annealing equipment in the material control band at that time or the set value of the operation conditions of the continuous annealing equipment may be used as the input data of the material prediction model. Using the data obtained in this way as an input, the mechanical properties of the steel strip are predicted using the material prediction model.

In the present embodiment, the target range of the mechanical properties of the steel strip is set in the higher-level computer, and the predicted mechanical properties and the target range are compared. Here, the target range of the mechanical properties is set so as to make the variation smaller than before, based on the past operation results regarding the mechanical properties of the steel strip. For example, for a steel strip having a tensile strength of 980 MPa, the target range of mechanical properties can be set as 985 to 1050 MPa.

At this time, in the operating condition setting unit of the continuous annealing equipment, the target range of the mechanical properties preset as described above is compared with the prediction result of the mechanical properties, and the predicted mechanical properties are the target range. If the above conditions are satisfied, the operating conditions of the continuous annealing equipment are determined with the initial settings and sent to the control unit of the continuous annealing equipment. On the other hand, if the predicted mechanical properties deviate from the target range, the operating conditions in the material control zone are reset.

Specifically, among the transformation rate meters 20 installed on the most downstream side in the continuous annealing equipment (however, among the transformation rate meters 20 that give transformation rate information used for inputting the material prediction model, the most. When the downstream side) is installed at the entrance of the uniform tropical 7 in the annealing process, the material identification zone is from the entrance side of the continuous annealing equipment to the entrance of the normal tropical 7 and is downstream from the entrance of the uniform tropical 7. The side is the material control band. At this time, when the tip of the steel strip reaches the entrance of the solitary tropics 7 and the transformation rate information is acquired by the transformation rate meter 20, the material control flow shown in FIG. 6 is started. At this time, in the material control zone, the operating conditions that can be used to control the material include the cooling condition in the cooling zone 8, the reheating condition in the reheating zone 9 (alloyed zone 17), and the maintenance in the tropical zone 18. The operating conditions selected from the heat temperature, the heat retention time, the cooling rate in the final cooling zone 11, and the like can be reset. Further, the operating conditions to be reset are not necessarily limited to those used as the input of the material prediction model. For example, the elongation rate of the temper rolling process may not be used as an input of the material prediction model, or may be used as an operating condition for resetting in the material control band.

Here, when the transformation rate meter 20 on the most downstream side is installed at the outlet of the reheating zone 9 (alloyed zone 17), the material control zone is substantially limited to the band after the final cooling zone 11. Will be done. That is, the operating conditions to be reset in the continuous annealing equipment are limited to the cooling rate of the final cooling zone 11 and the elongation rate in the temper rolling process.

Therefore, the position of the most downstream transformation rate meter 20 used as the input of the material prediction model may be appropriately determined depending on the balance between the degree of freedom of the operating conditions for resetting and the prediction accuracy by the material prediction model. That is, by lengthening the material identification band, the prediction accuracy of the material is improved, but the degree of freedom of the operating conditions that can be reset in the material control band is reduced. On the other hand, if the material identification band is shortened, the prediction accuracy of the material is lowered, but the degree of freedom of the operating conditions that can be reset in the material control band is increased.

Further, the material prediction model generation unit shown in FIG. 5, the material prediction unit shown in FIG. 6, and the operation condition setting unit of the continuous annealing equipment are realized by a higher-level computer and a computer capable of communicating with various devices constituting the continuous annealing equipment. You can do it. This computer may be a process computer as an example. The configuration of the computer is not particularly limited, and includes, for example, a memory (storage device), a CPU (processing device), a hard disk drive (HDD), a communication control unit for connecting to a network, a display device, and an input device. May be. Here, various processes of the material prediction model generation unit, the material prediction unit, and the operation condition setting unit of the continuous annealing equipment may be executed by the CPU. The database of FIG. 5 may be realized by a hard disk drive. Further, the material prediction model may be stored in the memory. Further, the collection of learning data may be realized by the communication control unit.

Hereinafter, the present disclosure will be specifically described with reference to Examples.

In the hot-dip galvanized equipment as shown in FIG. 2, 200 coils of 980 MPa class hot-dip galvanized steel sheets (target tensile strength is 1020 MPa) were manufactured. The input actual data is the actual data of the attribute information of the steel strip charged in the hot-dip galvanizing equipment and the operation actual data of the operation parameters in the hot-dip galvanizing equipment. A plurality of learning data were acquired using the tensile strength of the steel strip on the outlet side of the hot-dip galvanizing facility based on the input actual data as the output actual data. Here, the output actual data based on the input actual data is the actual data (the tensile strength of the obtained steel strip) output corresponding to the case where the input actual data is used as the setting condition of the operation or the like. A material prediction model learned by machine learning using a plurality of acquired learning data was generated by the method shown in FIG.

In generating the material prediction model, the contents of C, Si, and Mn were used as the attribute parameters of the steel strip regarding the component composition of the steel strip as an input. Further, as the operation record data of the continuous annealing equipment, the steel strip temperature in the uniform tropical 7 and the transport speed when the tip of the steel strip passes through the uniform tropical 7 are input. Further, in this embodiment, an online transformation rate meter 20 is installed at two locations, an inlet and an outlet of the uniform tropical 7 of the hot-dip galvanizing line shown in FIG. 2, and the transformation measured by these transformation rate meters 20. The actual data of the rate information was used as the input actual data. Further, in this embodiment, as other inputs, a material prediction model was generated using the set values of the plate thickness and the plate width of the steel strip and the actual values of the elongation rate in the temper rolling process.

Here, the mechanical properties of the steel strip acquired as learning data are the tensile strength obtained from the tensile test using the sample of the tip of the steel strip collected at the outside inspection table of the hot-dip galvanizing facility. ..

The material prediction model thus generated was applied to the material prediction unit in the material control shown in FIG. 6, and 100 coils of 980 MPa class hot-dip galvanized steel sheet (target tensile strength of 1020 MPa) were manufactured.

At this time, using the above material prediction model, the tensile strength of the steel strip on the outlet side of the hot-dip galvanizing facility is predicted as output data, and the predicted tensile strength is within a preset allowable range (in this case, 985). The operation parameters in the hot-dip galvanizing facility were reset so as to be within (set to ~ 1050 MPa). At this time, since the transformation rate meter 20 on the most downstream side is installed at the outlet of the uniform tropical 7 in the annealing process, the material identification zone is from the entrance side of the continuous annealing equipment to the exit of the uniform tropical 7 and the uniform tropical 7 is used. The material control band is located downstream from the exit of. The flow shown in FIG. 6 is started after the tip of the steel strip reaches the exit of the flat zone 7. In the material control zone, the heat retention temperature and heat retention time in the tropical 18 were reset as the operating conditions used for controlling the material. Then, the tensile strength obtained by the tensile test of these steel strips was collected. As a result, 95% of the steel strips were within the allowable range of tensile strength (985 to 1050 MPa).

On the other hand, as a comparative example, a similar experiment was performed by the method described in Patent Document 1. As a result, the steel strip within the allowable range of tensile strength was 65%.

As described above, by applying the material prediction method according to the present disclosure, the mechanical properties of the steel strip on the exit side of the continuous annealing equipment can be predicted with high accuracy, and the variation in the mechanical properties of the steel strip can be predicted. It can be reduced.

1 Payoff reel 2 Welder 3 Electrolytic purifier 4 Enter looper 5 Pre-tropical zone 6 Heating zone 7 Average tropical zone 8 Cooling zone 8A 1st cooling zone 8B 2nd cooling zone 9 Reheating zone 10 Overaging zone 11 Final cooling zone 11A 1st final cooling zone 11B 2nd final cooling zone 12 Outside looper 13 Conditioning rolling equipment 14 Inspection equipment 15 Tension reel 16 Zinc plating tank 17 Alloyed zone 18 Conservation 19 Snout 20 Transformation rate meter 21 Wiping device

Claims (6)

A method for predicting the material of a steel strip in which the mechanical properties of the steel strip on the downstream side of the reheating step are predicted in a continuous annealing facility that executes a manufacturing process including an annealing step and a reheating step of the steel strip.
As input data, an input data acquisition step for acquiring one or more parameters selected from the operation parameters of the continuous annealing facility and transformation rate information measured in at least one of the annealing step and the reheating step.
Using a material prediction model learned by machine learning, which uses information on the mechanical properties of the steel strip on the downstream side of the reheating process as output data, the mechanical properties of the steel strip on the downstream side of the reheating process can be determined. Prediction and how to predict the material of steel strips, including. The material prediction method for a steel strip according to claim 1, wherein the input data acquisition step further acquires one or more parameters selected from the attribute parameters of the steel strip regarding the component composition of the steel strip as the input data. Using the steel strip material prediction method according to claim 1 or 2, the mechanical properties of the steel strip on the downstream side of the reheating step are predicted, and the predicted mechanical properties are outside the preset target range. In some cases, a method of controlling the material of a steel strip, comprising resetting one or more operating parameters selected from the operating parameters of the continuous annealing facility so that the mechanical properties fall within the target range. A method for manufacturing a steel strip in a continuous annealing facility that executes a manufacturing process including an annealing step and a reheating step of the steel strip.
To acquire one or more parameters selected from the operation parameters of the continuous annealing equipment and transformation rate information measured in at least one of the annealing step and the reheating step as input data.
Using a material prediction model learned by machine learning, which uses information on the mechanical properties of the steel strip on the downstream side of the reheating process as output data, the mechanical properties of the steel strip on the downstream side of the reheating process can be determined. Predicting and
If the predicted mechanical properties are outside the preset target range, reset one or more operating parameters selected from the operating parameters of the continuous annealing equipment so that the mechanical properties fall within the target range. And, including, how to make steel strips. A method for generating a steel strip material prediction model for predicting the mechanical properties of a steel strip downstream of the reheating step in a continuous annealing facility that executes a manufacturing process including a steel strip annealing step and a reheating step. There,
At least one or more operation record data selected from the operation record data of the continuous annealing facility and the record data of the transformation rate information measured in at least one of the annealing step and the reheating step are acquired as input record data. To do and
Acquiring a plurality of learning data using information on the mechanical properties of the steel strip on the downstream side of the reheating process based on the input actual data as output actual data.
A method for generating a steel strip material prediction model, including generating a steel strip material prediction model by machine learning using a plurality of acquired learning data. The method for generating a material prediction model for a steel strip according to claim 5, wherein a method selected from a neural network, decision tree learning, random forest, and support vector regression is used as the machine learning.

Images