JP7272331B2 - Method for changing running plate thickness and method for manufacturing steel plate - Google Patents

Description

TECHNICAL FIELD The present invention relates to a tension variation prediction method, a tension variation prediction model generation method, a running strip thickness changing method, and a steel plate manufacturing method.

In continuous cold rolling, a plurality of materials to be rolled having different conditions such as material and plate thickness are welded and continuously rolled by a cold tandem rolling mill. In such continuous cold rolling, if the material to be rolled under different conditions is used, the reduction position (roll gap) in each rolling stand is different. Then, the strip thickness is changed by changing the pressing position of the work rolls in the rolling stand to change the strip thickness.

Normally, before the joining point reaches the entry side of the tandem rolling mill, the set values of the roll gap and roll speed corresponding to each pass schedule between the preceding material and the succeeding material (actual values are used for the preceding material) ) is calculated. The pass schedule for the preceding material is also referred to as pass schedule A, and the pass schedule for the following material is also referred to as pass schedule B. Then, in accordance with the timing at which the joint point passes each rolling stand, the amount of change (difference amount) in the roll gap and roll speed is output as command values for the respective roll gap controllers and roll peripheral speed controllers.

At that time, if the rolling conditions such as base plate thickness, finished plate thickness, deformation resistance (material), etc. are different between the preceding and succeeding materials, the inter-stand tension will fluctuate greatly when the running strip thickness is changed. occur. If the amount of tension fluctuation is too large, sheet breakage, narrowing, and the like will occur, resulting in major rolling troubles. Furthermore, before and after the joining point, if there are errors in the estimation of rolling operation parameters such as plate thickness, deformation resistance, and coefficient of friction, the tension fluctuation will increase.

In response to this, there has been proposed a technique for suppressing the tension fluctuation when the joining point between the preceding material and the succeeding material passes in the rolling stand of the tandem rolling mill.
For example, in Patent Document 1, instead of controlling toward the target value of the roll gap (the set value of the roll gap corresponding to the pass schedule B), only the direction in which the roll gap operates at the start of changing the strip thickness between runs is set, and the roll gap operation is stopped when the tension on the entry side after passing the joining point reaches the target tension. Patent Document 1 discloses that this can reduce troubles caused by tension fluctuations.

Patent Literature 2 discloses a method for suppressing changes in mass flow by keeping the rolling reduction ratio before and after changing the strip thickness constant, and for reducing tension fluctuations by reducing the amount of change in the peripheral speed of the roll. .
In Patent Document 3, an approximate curve of the friction coefficient and the rolling load is generated based on the rolling operation record, and the roll gap and the roll peripheral speed are set using the friction coefficient and the rolling load predicted from the approximate curve. is disclosed.

In Patent Document 4, as a means for predicting tension fluctuation when the strip thickness is changed, a tension fluctuation value predicted by a tension fluctuation learning means using a neural network that learns tension fluctuations when the strip thickness is actually changed as teacher data. Based on this, the value of the tension fluctuation is determined, and the time for changing the thickness between runs (hereinafter referred to as "running variation time") for changing the set values of the roll gap and roll peripheral speed from pass schedule A to pass schedule B is reset. By doing so, a means for suppressing tension fluctuation is disclosed. In addition, it is disclosed that input to the neural network includes rolling time, roll reduction change amount when rolling length is changed, rolling position control response, main engine speed control response, steel type, plate thickness, plate width, etc. .

JP 2007-61851 A JP-A-8-206712 JP 2015-36154 A JP-A-10-249423

However, in the method disclosed in Patent Document 1, the change of the roll gap from pass schedule A to pass schedule B is performed by feedback control using the actual value of the inter-stand tension. Therefore, there is a problem that the transition time from pass schedule A to pass schedule B is lengthened, and the plate thickness off-gauge increases. In particular, before and after the joining point of the preceding material and subsequent material corresponds to the unsteady part of hot rolling, which is the previous process, there are large fluctuations in plate thickness, deformation resistance, etc., and these disturbances become effective. It does not serve as an effective means of restraining tension.

The method disclosed in Patent Literature 2 suppresses tension fluctuations by imposing a constraint that the reduction rate is kept constant. However, when the base plate thickness and finished plate thickness of the preceding material and the succeeding material are greatly different, it is often difficult to keep the rolling reduction constant. The method described in Patent Document 2 is effective when the total rolling reduction from the entry side to the exit side of the finishing mill does not change significantly, as in hot rolling, but as in cold tandem rolling, It is difficult to apply when the rolling conditions of the preceding material and the succeeding material change greatly.

The method disclosed in Patent Document 3 predicts the rolling load and friction coefficient when the joint passes through the rolling stand based on the actual operation data, and the roll gap and roll speed corresponding to the pass schedule B are estimated. Setting calculation can be performed with high accuracy. However, since the tension fluctuation when passing through the joining point depends on the change in rolling conditions before and after the joining point, such as the thickness difference and deformation resistance difference between the preceding and following material, Even if the set values of the roll gap and roll speed corresponding to the pass schedule are made highly accurate, the tension fluctuation cannot be suppressed.

Patent Document 4 discloses a learning model of tension fluctuations by a neural network as means for predicting tension fluctuations. However, similar to the above, the tension fluctuation at the time of passing the joining point depends on the change in rolling conditions before and after the joining point, such as the thickness difference and the deformation resistance difference between the preceding and succeeding materials. Since the prediction model does not necessarily take these effects into consideration, there is a problem that the prediction accuracy of tension fluctuation is low.

Therefore, the present invention has been made with a focus on the above problems. A tension fluctuation prediction method, a tension fluctuation prediction model generation method, a running plate thickness changing method, and a steel plate manufacturing method that can accurately predict tension fluctuations between rolling stands and suppress tension fluctuations even if there is intended to propose.

According to one aspect of the present invention, there is provided a method for predicting tension fluctuations in a change in thickness between runs of a tandem rolling mill that continuously rolls a material to be rolled in which a preceding material and a succeeding material having different rolling conditions are joined, A rolling operation parameter of the preceding material and a rolling operation change amount parameter representing a difference value between the rolling operation parameters of the preceding material and the succeeding material are input data, and the joining point between the preceding material and the succeeding material is Predicting the amount of tension fluctuation when the joint point passes through the rolling stand using a tension fluctuation prediction model learned by machine learning, with tension fluctuation information when passing through the rolling stand as output data. Tension fluctuation. A prediction method is provided.

According to one aspect of the present invention, there is provided a method for generating a tension fluctuation prediction model in changing the thickness between runs of a tandem rolling mill that continuously rolls a material to be rolled in which a preceding material and a succeeding material having different rolling conditions are joined. The actual rolling operation data and the rolling operation change amount actual data of the preceding material are used as input actual data, and the tension fluctuation information when the joining point between the preceding material and the succeeding material passes through the rolling stand is output actual data. A method for generating a tension variation prediction model is provided, which acquires a plurality of learning data and generates a tension variation prediction model by machine learning using the plurality of learning data.

According to one aspect of the present invention, there is provided a method for suppressing tension fluctuation in changing the thickness between runs of a tandem rolling mill that continuously rolls a material to be rolled in which a preceding material and a succeeding material having different rolling conditions are joined, When determining the rolling operation parameters of the tandem rolling mill, before the joint point of the preceding material and the succeeding material reaches the tandem rolling mill, the tension fluctuation prediction method described above is used to determine whether the joint point is rolled. a tension variation prediction step of predicting a tension variation when passing through a stand; a tension variation determination step of determining whether the predicted tension variation is equal to or greater than a predetermined value; a resetting step of resetting the plate thickness of the intermediate stand of the succeeding material so that the value of the tension variation evaluation function represented by the formula (4) becomes smaller when the tension variation amount is equal to or greater than a predetermined value; A method for changing running strip thickness is provided.

ｆ：張力変動評価関数、α：重み付け係数、Ｎ：スタンド数、ｉ：スタンド番号、Δｔ：張力変動量
本発明の一態様によれば、上記の走間板厚変更方法を用いて鋼板を製造するステップを含む、鋼板の製造方法が提供される。

f: tension variation evaluation function, α: weighting factor, N: number of stands, i: stand number, Δt: amount of tension variation According to one aspect of the present invention, a steel plate is manufactured using the above-described method for changing thickness between runs. A method of manufacturing a steel sheet is provided, comprising the step of:

According to one aspect of the present invention, even if there is a difference in mother plate thickness or finished plate thickness or a difference in deformation resistance (material difference) between the preceding material and the following material, tension fluctuations between stands can be suppressed. Provided are a tension variation prediction method, a tension variation prediction model generation method, a running strip thickness changing method, and a steel plate manufacturing method that can accurately predict and suppress tension variation.

It is a mimetic diagram showing cold rolling equipment in one embodiment of the present invention. 4 is a flow chart showing a method for estimating a tension fluctuation amount and a method for determining rolling operation conditions according to one embodiment of the present invention. 4 is a graph showing tension variation in a tandem rolling mill. It is explanatory drawing which shows the production|generation method of a tension fluctuation|variation prediction model. 4 is a flow chart showing a roll gap changing method according to an embodiment of the present invention;

An embodiment of the present invention will be described below with appropriate reference to the drawings. Note that the drawings are schematic. Therefore, it should be noted that the relationship, ratio, etc. between the thickness and the planar dimensions are different from the actual ones, and the drawings include portions where the relationship and ratio of the dimensions are different from each other. Further, the embodiments shown below are examples of devices and methods for embodying the technical idea of the present invention. etc. are not specified in the following embodiments.

<Configuration of cold tandem rolling mill>
FIG. 1 is a schematic configuration diagram showing an example of a continuous cold rolling facility 1 according to one aspect of the present invention. In addition, in FIG. 1, other devices attached to the equipment (for example, devices such as a rewinder, a welder and a looper on the entry side, and a cutter and a winder on the delivery side) are omitted from illustration. .
As shown in FIG. 1, the continuous cold rolling facility 1 manages the continuous cold rolling facility 1 including a tandem rolling mill 2, a rolling control controller (PLC) 3 that controls the tandem rolling mill 2, and a rolling control controller. A control computer (process computer) 4 is provided.

The tandem rolling mill 2 is a continuous cold tandem rolling mill having a first rolling stand 2A to a fifth rolling stand 2E in order from the entry side in the sheet threading direction.
Each rolling stand of the first rolling stand 2A to the fifth rolling stand 2E has a work roll 21, a roll speed controller 22 which is an electric motor for changing the roll speed of the work roll 21, and a roll gap between the upper and lower work rolls 21. and a roll-down control device 23 for changing the are installed respectively. A rolling load detector 24 consisting of a load cell is provided under each rolling stand under the backup roll. Furthermore, each rolling stand is provided with a roll-down position detector 25 for detecting the roll-down position. Furthermore, a tension meter 26 for detecting the tension of the rolled material is provided between each rolling stand. Furthermore, plate thickness gauges 27 for detecting the plate thickness of the rolled material are provided on the delivery side of the first rolling stand 2A and the delivery side of the fifth rolling stand 2E.

In this embodiment, the control computer 4 sets a rolling schedule, which will be described later, and generates differential command values for roll gaps and roll speeds when the joining point passes each rolling stand. The rolling control controller 3 performs processing for controlling the roll speed control device 22 and the roll reduction control device 23 of each of the first rolling stand 2A to the fifth rolling stand 2E based on each value acquired from the control computer 4. to run.

The control computer 4 calculates the pass schedule, the rolling load of each rolling stand, the predicted value of the advance rate, and the set values of the roll gap and roll speed, according to the information such as the base material size and the product target size given from the host computer. At the same time, the change amount (difference command value) of the roll gap and the roll speed when passing the joint point is calculated. Then, the control computer 4 sets these calculated values in the low-order rolling controller 3 . The rolling control controller 3 continuously collects rolling data such as the rolling load detected by the rolling load detector 24 and the tension from the tension meter 26 and outputs it to the control computer 4 . Further, the rolling control controller 3 performs tracking of the joint point when the running strip thickness is changed, and outputs a predetermined control output at the timing when the joint point passes the rolling stand.

<Method of changing thickness between runs>
In the method for changing the running strip thickness according to the present embodiment, in rolling the material to be rolled in which the preceding material and the succeeding material are joined together in the tandem rolling mill 2, the joining point between the preceding material and the succeeding material is set in advance. The tension fluctuation when passing is estimated, and the rolling operation parameters such as roll gap and roll speed are determined so that the estimated tension fluctuation is equal to or less than a predetermined allowable value. Then, when rolling the material to be rolled, the thickness is changed based on the determined rolling operation parameters.

[Method for estimating tension fluctuation amount and method for determining rolling operation conditions]
A method of estimating the tension fluctuation amount and a method of determining the rolling operation conditions will be described with reference to FIG. First, the control computer 4 creates a pass schedule (thickness schedule) for preceding and succeeding materials according to information such as the dimensions of the base material coil for cold rolling and the target dimensions of the product after cold rolling given from the host computer. is set (setting step, S100). The pass schedule is the plate thickness after rolling of each rolling stand of the preceding material and the succeeding material. The plate thickness of the intermediate stand initially given can be arbitrarily set, and a plate thickness value set empirically can be set.

Next, the control computer 4 uses the rolling load model to calculate the predicted values of the rolling load and advance rate of each rolling stand (S102). At this time, the predicted value is calculated by using the set pass schedule for the rolling load model.
In the calculation using the rolling load model, the rolling load and the advance rate are calculated from the rolling conditions such as the thickness of the base material, the thickness of the delivery side of each rolling stand, the deformation resistance, the coefficient of friction, and the diameter of the work rolls. Here, the rolling load is calculated from the formula (1) using a generally used approximation formula of two-dimensional rolling theory.

Here, P: rolling load (kN), H: entry side plate thickness (mm), h: delivery side plate thickness (mm), b: plate width (mm), _km : average deformation resistance of rolled material (MPa), R': Flat roll radius (mm), Q _P : Rolling force function, μ: Friction coefficient. The flattened roll radius R′ is obtained by applying Hitchcock's roll flattening formula and from a simultaneous solution with load calculation based on two-dimensional rolling theory.
Further, the advance rate _fs defined as the ratio of the roll speed to the strip speed on the delivery side of the rolling mill is also calculated from the two-dimensional rolling theory.

After step S102, the control computer 4 calculates the set values of the roll gap and roll speed (rolling control parameter setting step, S104). In step S104, the set values of the roll gap and roll speed are calculated from the predicted values of the rolling load and advance rate corresponding to the set pass schedule. The set value S of the roll gap can be obtained from the relational expression (2) using the mill constant K.

As a result, the setting value of the roll gap corresponding to each pass schedule is calculated assuming that the preceding material and the succeeding material are in a steady state.
On the other hand, for the roll speed, the predicted value of the advance rate _fs calculated by the above method is used to balance the roll speed of each rolling stand so that the equation (3) holds for the pass schedule in the steady state. Then, based on the roll speed setting value of the rolling stand (for example, the final rolling stand) that serves as the speed reference, the roll speed setting values of the other rolling stands are calculated.

Here, i: stand number, V _R ⁽ⁱ⁾ : roll speed of i-th rolling stand (mm/min), h ⁽ⁱ⁾ : delivery side plate thickness of i-th rolling stand (mm), f _s ⁽ⁱ⁾ : It is the advance rate (-) of the i-th rolling stand.
After step S104, the control computer 4 calculates a difference command value (amount of change) between the roll gap and the roll speed as a control output when the joint point passes through the rolling stand (first difference command value generation step, S106). In step S106, assuming that the preceding material and the succeeding material are in a steady state, the amount of change from the preceding material to the succeeding material is calculated from the calculated set values (absolute values) of the roll gap and roll speed. A differential command value representing is generated. That is, a differential command value for the roll gap and a differential command value for the roll speed are generated. These differential command values are transmitted to the rolling control controller 3 through a tension fluctuation determination step (which may be reset in some cases) to be described later, and stored as differential command data in the running strip thickness change program.

After step S106, the control computer 4 predicts the tension fluctuation amount using the calculation result in step S106 (tension fluctuation amount prediction step, S108). The tension variation amount in step S108 is estimated using a tension variation prediction model generated by offline processing.
Here, the tension variation in the tandem rolling mill 2 will be described with reference to FIGS. 3(A) to 3(D). Graphs of FIGS. 3A to 3D show the tension generated between each rolling stand. FIG. 3A shows the tension between the first rolling stand 2A and the second rolling stand 2B. (B) is between the second rolling stand 2B and the third rolling stand 2C, FIG. 3(C) is between the third rolling stand 2C and the fourth rolling stand 2D, and FIG. 3(D) is the fourth rolling stand. 2D and 2E respectively show the change in tension over time between 2D and the fifth rolling stand 2E. Further, in FIG. 3, _ts is the change start time that is the timing at which the roll gap change is started in the third rolling stand 2C, and _te is the change end time that is the timing at which the roll gap change is completed in the third rolling stand 2C. indicate the time. Further, in FIG. 3, the solid line graph shows the measurement results when the thickness of the running strip is changed normally.

In the present embodiment, tension fluctuation information is defined as tension fluctuation information that occurs during the time (running change time) during which the roll gap is changed when the joining point passes through the rolling stand. For example, in the example of FIG. 3, the tension fluctuation information about the third rolling stand 2C is information about the tension fluctuation occurring between the change start time _ts and the change end time _te of the roll gap of the third rolling stand 2C. . Numerical data such as the maximum value, the minimum value, and the difference between the maximum and minimum values of the tension can be used as the tension variation information from the change start time _ts to the change end time _te . However, the tension fluctuation information may be defined by graphical information representing the time-series tension values in a two-dimensional graph.

As shown by the solid line in FIG. 3, in the normal running strip thickness change, the tension fluctuation when the joint point passes the rolling stand (the third rolling stand in the example of FIG. 3) is the rolling stand that changes the roll gap A larger variation occurs on the entry side (tension variation shown in FIG. 3(B)) than on the delivery side (tension variation shown in FIG. 3(C)). Therefore, it is preferable to define the tension fluctuation information by the fluctuation information of the entry-side tension of the rolling stand where the roll gap is changed. That is, it is preferable to define the tension fluctuation information of the i-th rolling stand by the entry-side tension fluctuation information of the i-th rolling stand between the change start time and change end time of the roll gap of the i-th roll stand. At that time, as the amount of tension, the tension per unit area may be used, or the total tension may be used. Furthermore, when the tension fluctuation information is expressed numerically, the tension fluctuation information may be defined as a rate of change with respect to the tension set value in the steady state of the preceding material. In this case, it is preferable to keep the tension fluctuation rate within the range of -5% to 10% when changing the thickness of the running plate.

The travel change time from the preceding material to the succeeding material is determined based on the roll gap change amount and roll gap change speed of each rolling stand. The change speed of the roll gap can be set as a fixed value, for example, 0.94 mm/sec under hydraulic pressure and 0.47 mm/sec under electric pressure. At this time, the travel time of each rolling stand may be calculated by dividing the absolute value of the roll gap change amount by the roll gap change speed (|roll gap change amount|÷roll gap change speed). Since it is desirable to unify the travel variation time for all rolling stands, the longest travel variation time among the travel variation times of each rolling stand is standardized.

The tension fluctuation prediction model is a prediction model using machine learning that predicts tension fluctuations under predetermined rolling conditions based on rolling operation performance data of preceding and succeeding materials and time-series data of tension fluctuations, which are actual data. is. Specifically, the tension fluctuation prediction model uses the rolling operation performance data of the preceding material and the rolling operation change amount performance data as input data, and outputs tension fluctuation information as output data. generated by the method.

(Generation method of tension fluctuation prediction model)
As shown in FIG. 4, the tension fluctuation prediction model is generated using the rolling operation performance data of the preceding material, the rolling operation performance data of the following material, and the time series data of the tension fluctuation. It is generated using an amount calculator 51 , a tension variation information calculator 52 , a database 53 and a model generator 54 . The change amount calculation unit 51, the tension variation information calculation unit 52, the database 53, and the model generation unit 54 are configured by computer programs, etc. may be configured. The rolling operation performance data of the preceding material, the rolling operation performance data of the succeeding material, and the time-series data of the tension fluctuation are performance data acquired in past operations.

In this embodiment, actual data related to at least one rolling condition of strip thickness, strip width, deformation resistance, and rolling load is used as the rolling operation performance data of the preceding material. Further, the rolling operation result data of the succeeding material is also the same as the rolling operation result data of the preceding material. The time-series data of tension fluctuation is data of tension when the joining point between the preceding material and the succeeding material passes through the rolling stands, and is measured by a tension meter installed between the rolling stands. Further, as exemplified above, the tension fluctuation information is obtained from the time-series data of the tension fluctuation.

In addition, at least one operation parameter of the thickness, width, deformation resistance and rolling load of the preceding material and the succeeding material as the rolling operation change amount actual data representing the difference value of the rolling operation parameter between the preceding material and the succeeding material The amount of difference from the leading material to the trailing material before and after the joining point between them is used. That is, at least one of the thickness difference amount, the width difference amount, the deformation resistance difference amount, and the rolling load difference amount from the preceding material to the succeeding material is used as the difference amount. The rolling operation change amount result data is calculated as the difference between the preceding material and the following material with respect to these operation parameters, and the fluctuation amount calculation unit 51 calculates the rolling operation result data of the preceding material and the following material. Calculated from operation performance data.

The tension variation information calculation unit 52 obtains tension variation information from time-series data of tension variation.
The database 53 stores the rolling operation data of the preceding material, the actual rolling operation change amount data calculated by the fluctuation amount calculation unit 51, and the tension fluctuation information obtained by the tension fluctuation information calculation unit 52.
The model generation unit 54 generates a tension fluctuation prediction model from the rolling operation data of the preceding material, the rolling operation change amount result data, and the tension fluctuation information in the operation results of a plurality of times stored in the database 53. At this time, the model generating unit 54 uses the rolling operation data and rolling operation change amount actual data of the preceding material as input actual data, and the tension variation information as output actual data, and generates a tension variation prediction model using machine learning.

A known learning method can be applied as a machine learning method. For example, as machine learning, a known machine learning method such as a neural network can be used. As other methods, decision tree learning, random forest, support vector regression, Gaussian process, k-nearest neighbor method, etc. can be used. Here, the tension variation prediction model may be updated as appropriate using the latest learning data. For example, the model may be updated to a new model by re-learning every month or every year. This is because the more data stored in the database 53, the more accurately the tension fluctuation prediction becomes possible.

It should be noted that the tension variation prediction model based on the above-described operation performance data can be generated for each rolling stand. Further, the tension fluctuation prediction model may be generated by limiting to the rolling stand where the tension fluctuation is large and troubles such as breakage are likely to occur. Additionally, the stand number may be added to the input data to generate a tension variation prediction model as a single machine learning model that predicts tension variation at any rolling stand.

Such a tension fluctuation prediction model is characterized by using both the operation parameter of the preceding material and the difference amount of the operation parameter in changing the running strip thickness as input data. As a result, for the combination of the preceding material and the succeeding material, not only the attribute information of the preceding material and the succeeding material is used, but also the order thereof is taken into consideration. That is, the preceding material _A and the succeeding material _B specified by the rolling conditions CA and CB are continuously rolled in the order of A→B and rolled in the order of B→A. The prediction model reflects the difference in tension fluctuation behavior between
Further, other rolling operation performance data can include friction coefficient, advance rate, estimated value of tension, set value of tension, etc. as parameters related to rolling conditions. Moreover, roll diameter, roll roughness, mill rigidity, and the like can be used as parameters representing equipment conditions of each rolling stand.

Then, in step S108, the control computer 4 uses the rolling operation conditions of the preceding material calculated in step S104 and the difference command value calculated in step S106 in the generated tension fluctuation prediction model to obtain the tension fluctuation information Ask for After that, the tension fluctuation amount Δt _i is predicted from the obtained tension fluctuation information. For the tension fluctuation amount Δt _i , for example, the maximum value of the predicted tension, the difference between the maximum value and the minimum value, or the like can be used.

After step S108, the control computer 4 determines whether or not the tension variation amount is equal to or greater than a predetermined value (tension variation determination step, S110). At this time, the predetermined value is set as a value that predicts the occurrence of rolling troubles such as strip breakage and narrowing based on the used tension fluctuation amount Δt _i and the actual results of the tandem rolling mill 2, for example.
If it is determined in step S110 that the tension fluctuation amount is equal to or greater than the predetermined value, the control computer 4 resets the plate thickness of the intermediate stand of the succeeding material (resetting step, S112). In step S112, as the tension variation evaluation function f, the thickness of the intermediate stand of the succeeding material is reset so that the value represented by the formula (4) becomes smaller.

Here, α: weighting coefficient, N: number of stands.
The weighting coefficient α weights each rolling stand. For example, by increasing the weight of a rolling stand that tends to cause operational troubles such as plate breakage due to tension fluctuation, the tension fluctuation in that rolling stand can be effectively suppressed.
Further, the plate thickness of the intermediate stand includes the plate thickness (base material thickness) on the entry side of the first rolling stand 2A, and the plate thickness ( It refers to the delivery side plate thickness of the first rolling stand 2A to the (N-1)th rolling stand (the fourth stand 2D in the case of this embodiment) excluding the product thickness). For example, in the example shown in Table 1, in the plate thickness schedule of the trailing material, as the plate thickness of the intermediate stand, the delivery side plate thicknesses of the first stand 2A to the fourth stand 2D are changed from h _b1 to h _b4 to h _b1 ' h _b4 '. In the pass schedule for subsequent materials, the thickness of the base plate is already decided, and the finished plate thickness cannot be changed as the product thickness. It can be set arbitrarily as long as it does not exceed the thickness of the delivery side.

At this time, if the change in tension on the entry side due to passage of the joint point is used as the tension change information, the tension on the entry side of the first stand 2A is controlled to be substantially constant by tension control using a bridle roll or the like. , (5) may be used to calculate the tension variation evaluation function f. In the calculation shown in equation (5), the tension variation from when the joint reaches the second stand until it passes the final stand is used.

As a method of reducing the tension fluctuation evaluation function f, a plurality of plate thicknesses of the intermediate stand of the succeeding material are assumed, and among those assumed values, the value of the tension fluctuation evaluation function f is smaller than the initial value. It is sufficient to determine the intermediate stand board thickness of the row material. Alternatively, the permissible value of the tension variation evaluation function f may be set in advance, and among the assumed thicknesses of the intermediate stands of the succeeding materials, those having the value of the tension variation evaluation function f equal to or less than the permissible value may be selected. . Tension fluctuations do not necessarily have to be zero, and tension fluctuations are allowed within a range that does not cause operational troubles such as plate breakage. Furthermore, the plate thickness of the intermediate stand of the succeeding material is used as a variable, and the changeable range of them is used as a constraint condition. may
However, in some cases, the thickness of the intermediate stand cannot be set so that the value of the tension fluctuation evaluation function f becomes smaller than the allowable value. Tension fluctuations may be suppressed by performing a known two-stage change in strip thickness between runs, in which an intermediate pass schedule is set.

After step S112, the control computer 4 outputs the roll gap and roll speed as control outputs when the joining point passes through the rolling stand based on the plate thickness of the intermediate stand reset in step S112, that is, the pass schedule. is calculated (second differential command value generation step, S114). In the calculation in step S114, the path schedule reset in step S112 is used to calculate the difference command value by performing the same processing as in steps S102 to S106.

If it is determined in step S110 that the tension fluctuation amount is less than the predetermined value, or after step S114, the control computer 4 calculates the set values (pass schedule) of the roll gap and roll speed in the preceding material, and finally calculates The difference command value, which is the amount of change in the roll gap and roll speed, is set as a value used for operation (S116). The set values of the roll gap and roll speed for the preceding material and the difference command value are output to the rolling control controller 3 . The rolling control controller 3 stores the obtained setting values of the roll gap and roll speed in the preceding material and the difference command value in a storage unit or the like provided therein.
By performing the processes of steps S100 to S116, the difference command values of the roll gap and the roll speed are determined.

(How to change the roll gap)
In the present embodiment, the roll gap is changed by the rolling control controller 3 based on the roll gap difference command value set as described above. In this embodiment, the roll speed is also changed based on the roll speed difference command value in accordance with the change of the roll gap. FIG. 5 shows a method of changing the roll gap.
The processing flow of FIG. 5 is started by starting rolling of the material to be rolled. At this time, the rolling of the preceding material is performed by the rolling control controller 3 controlling the roll speed control device 22 and the reduction control device 23 so as to achieve the set pass schedule.

In changing the roll gap, first, the rolling controller 3 starts tracking the joint points of the material to be rolled, and acquires joint point information as needed (tracking step, S200). The joint point tracking is started from a position where the joint point is set upstream of the first stand 2A of the tandem rolling mill by a predetermined distance. After passing through the first stand 2A, the plate speedometer on the entry side of the tandem rolling mill 2 (measurement of the peripheral speed of the bridle roll on the entry side) is used to determine the distance between the stands from the conveying distance and plate thickness of the material on the entry side of the first stand 2A. A method of calculating the conveying distance of the material in , or a method of calculating the material speed using the predicted value of the advance rate and the current value of the roll peripheral speed of each stand and converting it into the conveying distance can be applied.

Next, the rolling controller 3 determines whether or not the joint point has passed the stand based on the joint point tracking information (S202).
If it is determined in step S200 that the joint point has not passed through the stand, the process of step S202 is repeated.
On the other hand, when it is determined in step S200 that the joining point has passed through the rolling stand, the rolling control controller 3 outputs a difference command value corresponding to the strip thickness change point to the roll speed control device 22 and the reduction control device 23. (S204). In step S204, the roll gap of the rolling stand is changed by the roll gap control device 23 by outputting the differential command value of the roll gap change amount. Further, the roll speed of the rolling stand is changed by the roll speed control device 22 by outputting the differential command value of the roll speed.

The processing of steps S200 to S204 is performed for each rolling stand. Note that the tracking process is performed at any time until the joint points pass through at least all the stands. As a result, the positional information of the joints existing between the stands of the tandem rolling mill 2 is generated at any time, and the timing at which the joints reach the next rolling stand can be determined.
Note that the change of the roll gap described above is specifically performed by a running strip thickness change program provided in the rolling control controller 3 . The running strip thickness change program stores the differential command values of the roll gap and roll speed, and commands each rolling stand to change the roll gap and roll speed.

In changing the roll gap by the running strip thickness change program, at the timing when the joining point reaches the stands on the downstream side sequentially from the first stand 2A, the stand information (stand number etc.) through which the joining point passed is changed to the running strip thickness Passed to change program. Then, the running strip thickness change program issues a differential command value corresponding to the joint position information to the roll speed control device 22 and the roll reduction control device 23 according to the received stand information. The information is output to the control information output section provided in the rolling controller 3 . The control information output section issues a command to the roll speed control device 22 and the roll reduction control device 23 as a new control output using the difference command data output from the program for changing the strip thickness between runs. However, the control output from the control information output process is updated at the timing when the joining point passes the rolling stand, and the control output is held until the next control output from the control information output process is updated.

Note that the control output to the roll speed control device 22 and the roll reduction control device 23 temporarily stores the current values of the roll gap and roll speed of each stand before the joining point reaches the first stand 2A of the tandem rolling mill 2 ( lock on) and hold it. The differential command value is added to the lock-on value and used as a control target value for the roll speed control device 22 and the roll reduction control device 23 .

Through the above processing, the running strip thickness is changed. According to the present embodiment, when the preceding material and the succeeding material having different rolling conditions are joined in the tandem rolling mill 2 and the strip thickness between runs is changed, the difference in the rolling operation conditions between the preceding material and the succeeding material is adjusted. Using this, it is possible to predict tension fluctuations, and it is possible to prevent rolling troubles such as strip breakage and narrowing that accompany changes in running strip thickness. And even if tension fluctuations occur due to differences in mother plate thickness, finished plate thickness, deformation resistance, etc. between the preceding and succeeding materials, consider the parameters that are concerned in generating the tension fluctuation prediction model. Therefore, it is possible to accurately predict the tension fluctuation. In addition, since the tension variation prediction model of the present embodiment can consider the rolling order of the preceding material and the succeeding material, it is possible to further improve the prediction accuracy.

In addition, since the fluctuation of the inter-stand tension can be predicted before the joining point enters the tandem rolling mill 2, if it is judged that the fluctuation amount of the inter-stand tension becomes excessive, the rolling speed is reduced or It is possible for the operator to manually correct the thickness of the intermediate stand of the line material. Also, an alarm device may be incorporated so as to issue an alarm when the tension fluctuation is predicted to be excessive. As a result, it is possible to prevent rolling troubles such as strip breakage during change of running strip thickness, so that stable production can be achieved.
Furthermore, by changing the strip thickness between runs according to the above embodiment, it is possible to reduce the variation in strip thickness before and after the joining point, so it is possible to manufacture a steel strip with a small off-gauge and a good yield.

<Modification>
Although the invention has been described with reference to particular embodiments, it is not intended that the invention be limited by these descriptions. Along with the disclosed embodiments, other embodiments of the invention, including various modifications, will be apparent to persons skilled in the relevant art(s) upon reference to the description of the invention. Therefore, the embodiments of the invention set forth in the claims should be construed to cover the embodiments that include these variations described herein singly or in combination.

For example, in the above embodiment, the tandem rolling mill 2 shown in FIG. 3 is a rolling mill having five stands, but the present invention is not limited to such an example. The number of stands of the tandem rolling mill 2 may be two or more, and may be six or more.
Further, in the above embodiment, the process of step S112 is performed when the tension variation amount is equal to or greater than the predetermined value in step S110, but the present invention is not limited to this example. For example, if the tension variation amount is equal to or greater than a predetermined value in step S110, the process of step S100 may be further performed. At this time, in step S100, the plate thickness of the intermediate stand for the succeeding material is corrected, and the pass schedule is reset. In such a case, the processing of steps S100 to S110 is repeated until the tension variation amount becomes less than the predetermined value.
Furthermore, according to the present invention, there is provided a method for manufacturing a steel sheet, including the step of manufacturing a steel sheet using the method for changing the thickness between running strips according to the above embodiment.

Next, examples of the present invention will be described. In this example, the tandem rolling mill 2, which is a cold tandem rolling mill having five stands shown in FIG. 1, was used. The use of the tandem mill 2 is as follows.
Maximum line speed: 2000mpm (m/min)
Work roll diameter: 500-600mmφ
Backup roll diameter: 1300-1400mmφ

The rolling load detector 24, the rolling position detector 25, the tension gauge 26, and the plate thickness gauge 27 were provided at positions as shown in FIG.
Then, according to the tension variation prediction method according to the above-described embodiment, tension variation prediction when passing the welding point was performed under the following conditions.
As actual rolling data, the difference between the maximum and minimum values of tension fluctuation per unit cross-sectional area after the joint point passes through the rolling stand is used as actual output data, and each rolling stand is used as the rolling operation parameter for the preceding material. The thickness of the delivery side, deformation resistance and rolling load were used. Also, the thickness difference and the deformation resistance difference between the preceding material and the succeeding material were selected as the parameters for the amount of change in rolling operation, and these were used as learning data.

Then, 1,000 pieces of operation performance data were prepared as learning data, 900 pieces were used as model creation (learning) data, and the remaining 100 pieces were used to verify the prediction accuracy of tension fluctuations of each rolling stand. As a result, the model prediction accuracy (actual/prediction) was 0.08 in terms of standard deviation.
As a result, fluctuations in tension between rolling stands can be predicted before changing the strip thickness between runs, and by resetting the intermediate stand pass schedule before rolling, operation troubles due to strip breakage and narrowing can be reduced by 10%. %.

1 continuous cold rolling equipment 2 tandem rolling mill 2A to 2E first rolling stand to fifth rolling stand 21 work roll 22 roll speed control device 23 reduction control device 24 rolling load detector 25 reduction position detector 26 tension gauge 27 plate thickness Total 3 Rolling control controller 4 Control computer 51 Change amount calculation unit 52 Tension fluctuation information calculation unit 53 Database 54 Model generation unit

Claims (3)

A method for changing the running thickness of a tandem rolling mill for continuously rolling a material to be rolled in which a preceding material and a succeeding material having different rolling conditions are joined,
When determining the rolling operation parameters of the tandem rolling mill,
a tension variation amount prediction step of estimating a tension variation amount when the junction point of the preceding material and the trailing material passes through the rolling stand before the junction point of the preceding material and the succeeding material reaches the tandem rolling mill;
a tension variation determination step of determining whether the predicted tension variation amount is equal to or greater than a predetermined value;
In the tension fluctuation determination step, if the tension fluctuation amount is equal to or greater than a predetermined value, the plate thickness of the intermediate stand of the succeeding material is readjusted so that the value of the tension fluctuation evaluation function represented by the formula (4) becomes small. a resetting step to set;
with
In the tension variation amount prediction step, a rolling operation parameter of the preceding material and a rolling operation change amount parameter representing a difference value between the rolling operation parameters of the preceding material and the succeeding material are input data, and the preceding material and Using a tension fluctuation prediction model learned by machine learning, the tension fluctuation information when the joint point with the following material passes the rolling stand is used as output data, and the joint point when the joint passes the rolling stand. A running strip thickness change method for predicting the amount of tension fluctuation .

f: tension fluctuation evaluation function, α: weighting coefficient, N: number of stands, i: stand number, Δt: amount of tension fluctuation The rolling operation parameters of the preceding material include at least one of thickness, width, deformation resistance and rolling load of the preceding material,
2. The running strip according to claim 1, wherein said rolling operation change amount parameter includes at least one of a thickness difference, a width difference, a deformation resistance difference and a rolling load difference between said preceding material and said succeeding material. Thickness change method. A method for manufacturing a steel plate, comprising a step of manufacturing a steel plate using the method for changing thickness between running strips according to claim 1 or 2 .

Images