JP2934178B2 - Prediction method of thickness accuracy in hot rolling - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for estimating the thickness of a metal plate or a metal strip by hot rolling, for example, when cold rolling the metal sheet or the metal strip. About.

[0002]

2. Description of the Related Art When manufacturing a metal plate or a metal strip, in many cases, first, hot rolling is performed, and then cold rolling is performed to obtain a product having a predetermined thickness. Hot coil obtained by hot rolling (hot rolled metal plate or metal strip wound into a coil)
In some cases, the thickness accuracy of the metal plate or the metal strip may deteriorate.

[0003] Normally, a tolerance is set for the thickness of a rolled product, and if it is not satisfied, the rolled product is defective. Since this tolerance is set for the final thickness of the metal plate or strip after cold rolling, even if the thickness accuracy of the metal plate or strip in the hot coil is slightly lower, it is not necessarily a defective product. . However, when the thickness of a metal plate or a metal strip during hot rolling deteriorates beyond a certain reference value, the final thickness accuracy does not satisfy the tolerance.

[0004] Therefore, the allowable range at the time of the hot coil is set for the metal plate or the metal strip in the hot coil by any method, and is not used as a criterion when the thickness accuracy of the metal plate or the metal strip is low. Must not.

Conventionally, a control standard value for the thickness accuracy and profile in the longitudinal direction of a metal plate or a metal strip is empirically determined in advance, and the actual hot rolling value of the metal plate or the metal strip is determined with respect to this control standard value. The method of comparing is adopted.

[0006]

However, the method for managing the thickness of a metal plate or a metal strip in a hot coil by using such a management reference value empirically determined in advance has the following problems. In other words, the optimal value as the control standard value is not a person who has experience because various factors such as product thickness and its tolerance, material, standard pass schedule of cold rolling, process capability of cold rolling affect, etc. It is difficult to judge.

In particular, since the thickness in the longitudinal direction is greatly affected by the rolling speed, the pass schedule, and the like, it is difficult to manage the thickness in comparison with a profile in which the change during cold rolling is small, and the above-mentioned problem becomes remarkable. .

[0008] Therefore, the accuracy of cold rolling is predicted by a cold rolling thickness accuracy model, and whether the result satisfies the tolerance or not.
If the determination of hot rolling is made based on whether or not it is possible, automation and quantification of management are realized, which is very effective. However, developing this model is usually a difficult task.

As an example, consider the following simple linear model. Regarding the n-th pass of the cold rolling (a single rolling mill is assumed. In the case of a tandem rolling mill, it can be considered as the n-th stand), the rate of change in the longitudinal plate thickness accuracy of the metal plate or the metal strip before and after the rolling is calculated. As the attenuation rate αn, this attenuation rate α
Consider a model that correlates n with each rolling condition.

Here, the attenuation rate: αn will be described.
Defining the attenuation factor αn and ([Delta] H _n over Δh _n) / _{ΔH n.} Where ΔH _n : incoming side (the side where the metal plate or metal strip enters the pass), thickness variation width, Δh _n : out side (the side where the metal plate or metal strip exits the pass)
The thickness variation width is used.

The attenuation rate αn is predicted for each pass of the cold rolling. With the attenuation factor α1 of the first pass, the [Delta] H ₁ when the thickness variation of the hot coil, the first pass out side thickness fluctuation range: Delta] h ₁ = (1 over alpha) · [Delta] H ₁ thickness at delivery side of the second path Variation range: Δh ₂ = (1−α) · Δh …… Nth pass exit side plate thickness variation range: Δh _N = (1−α _N ) · Δh
Assuming that _{N −1} N is the total number of passes, Δh _N is a predicted longitudinal plate thickness of the product plate thickness.

Now, the relevant parameter is set as the total number of paths:
N, reduction amount: r _n , strip width: B _n , rolling speed V _n , value dependent on the material of the material to be rolled, temper: q _n , equipment conditions of the rolling mill (mechanical conditions such as roll bearings, etc. Automatic thickness control capability, etc.): When W is set, the output side thickness variation width:
Delta] h _n is thickness at entrance side fluctuation range can be expressed by the formula from [Delta] H _n below.

Α n = K 1 · r _n + K 2 · B _n + K 3 · V _n + K 4 · q _n + K 5 · W (1) Δh _n = (1−α) · ΔH _n (2) = (1−α) · Δh _{n -1} (2 ′) The final thickness variation width Δh _N is represented by the following equation since the actual accuracy of the hot coil is the first pass entry side thickness variation width: ΔH ₁ in cold rolling.

Δh _N = (1−α ₁ ) · (1−α ₂ ) (1−α _N ) · ΔH ₁ (3) In order to obtain the damping rate, a large number of actual cold rolling data are used.
What is necessary is just to regress the values of each coefficient: K1, K2, K3, K4, K5.

However, it is actually quite difficult to secure sufficient model accuracy by such a method. further,
It is difficult to consider the physical meaning of each coefficient, and not everyone can judge the validity of the model. Therefore, even if a model is created and started to be used for sheet thickness management, it is difficult for anyone other than the person who first determines the coefficient to understand the rationality. This does not improve the situation of relying on the experience of a specific person.

The invention of the present application has been made based on the above-mentioned circumstances, and the thickness accuracy in hot rolling can quickly and easily quantitatively determine the thickness of a metal plate or a metal strip after hot rolling. It is an object to provide a prediction method.

Further, the invention of the present application is to quickly and easily quantitatively determine the thickness of a metal plate or a metal strip after hot rolling, and to determine a metal plate or a metal strip rejected as a result of the determination. To provide a method of predicting thickness accuracy in hot rolling, which can set appropriate and reasonable relief rolling conditions so as not to reduce productivity when performing relief rolling in a subsequent rolling step. Make it an issue.

[0018]

According to a first aspect of the present invention, there is provided a method for predicting thickness accuracy in hot rolling, comprising: when hot rolling a metal plate or a metal strip; Predict the transition of the thickness accuracy of the metal plate or metal strip in rolling in the subsequent process from the sheet thickness results, and predict the pass / fail of the thickness accuracy in the final product of the metal plate or metal strip based on the result of this prediction. Along with the prediction of the longitudinal thickness accuracy of the metal plate or the metal strip, a model for evaluating all factors affecting the longitudinal thickness accuracy as an equivalent mill stiffness coefficient and an equivalent material plasticity coefficient is used. .

From the viewpoint that a model using parameters familiar to the production site is desirable, a plate thickness accuracy model using mill stiffness coefficient and material plasticity coefficient, which are general rolling process data, was considered.

Among the factors affecting the thickness accuracy, a value obtained by modeling equipment-side factors that can be evaluated as mill stiffness, such as natural mill stiffness and automatic thickness control ability, is referred to as an equivalent mill stiffness coefficient. From these, a thickness accuracy model is constructed from the relational expression with the thickness accuracy as follows.

According to a gauge meter formula which is a basic formula of a general rolling theory, h = S + P / K (4) where h: output side plate thickness S: roll gap at no load P: rolling load K: mill stiffness material When the plasticity coefficient M is determined according to the equation (5),

[0022]

(Equation 1)

H: Entry side plate thickness From equations (4) and (5), h = ｛K / (K + M)｝ S + ｛M / (K + M)｝ H (6) Considering only the fluctuation component, Δh = ｛K / (K + M)｝ ΔS + ｛M / (K + M)｝ ΔH (7) where ΔH: entry side thickness variation component ΔS: roll gap variation component Δh: exit side thickness variation component (7) is a widely known relational expression. However, based on this relational expression, a model expression based on an equivalent mill stiffness coefficient and an equivalent material plasticity coefficient expressed as a function of a parameter affecting the thickness accuracy is considered.

It is empirically known that the thickness accuracy changes depending on, for example, the rolling speed. This phenomenon is expressed as the rolling speed dependence of the mill stiffness. That is, it is known that the mill stiffness during actual rolling is affected by the dynamic characteristics of the rolling mill. By expressing this as an equivalent mill stiffness coefficient, the rolling speed and the thickness control ability of the thickness control ability are expressed. The effect on the plate thickness can be evaluated.

The selection of parameters differs depending on the purpose of the model or the conditions of the rolling mill and the process to be performed. For example, when importance is placed on the rolling speed dependency, the following equation (8) is used:
Equation (9) is appropriate.

K ′ = fK (V) · K (8) M ′ = fm (V) · M (9) where fK and fm are functions expressing the effect of rolling speed K ′: Equivalent mill stiffness coefficient M ′ : Equivalent material plasticity coefficient Using K ′ and M ′, the equation (7) is expressed as follows. Δh = ｛K ′ / (K ′ + M ′)｝ ΔS + ｛M ′ / (K ′ + M ′)｝ ΔH (10) Equations (8), (9) and (10) are the mill stiffness coefficient and material plasticity This is a thickness accuracy model that uses coefficients as mediation.

In the same manner as in the equation (1), the attenuation rate α
n = K ′ / (K ′ + M ′) + {M ′ / (K ′ + M ′)} · (ΔS / ΔH _N ) (11) Equation (11) and (3) The thickness accuracy prediction can be realized from the expression (3). In the equations (8) and (10), k is basic data of a rolling mill and is usually known. Therefore, assuming that the value is a value obtained by adding a slight correction to this, the validity of K ′ can be evaluated by comparing with K, and a highly reliable model can be developed relatively easily.

Since M is also basic data for determining the method of the rolling operation, it is sufficient to determine M 'from existing common knowledge. This model can be understood by anyone with common knowledge as long as it is engaged in work related to rolling, so it is not only excellent in developability, but also maintenance and adjustment of models required when process conditions change after development. This is also advantageous.

Further, if the thickness management task is aimed at quality assurance, a high reliability is required for the thickness prediction model. Therefore, what can be understood with common-sense knowledge is desirable also from the viewpoint of early detection of a reduction in design errors during model development and a decrease in model accuracy after development.

As a further application, it is also conceivable to automatically monitor the thickness of a post-process, for example, a cold-rolled plate, similarly by a computer, and to correct the same. The method for predicting thickness accuracy in hot rolling according to the invention of claim 2 is characterized in that when rolling after hot rolling a metal plate or a metal strip, the post-process is performed from the actual thickness of the hot coil at the end of hot rolling. Predict the transition of the thickness accuracy of the metal plate or metal strip in the rolling of, while predicting the pass / fail of the accuracy of the thickness of the metal plate or metal strip in the final product based on the result of this prediction, the metal plate or For the prediction of the longitudinal thickness accuracy of the metal strip, using a model that evaluates all factors affecting the longitudinal thickness accuracy as an equivalent mill stiffness coefficient and an equivalent material plasticity coefficient, and the final product of the metal sheet or metal strip When performing rolling to correct an abnormal portion of the metal plate or the metal strip in a subsequent process, the metal plate or the metal strip determined to be unacceptable in the prediction of the pass / fail of the sheet thickness accuracy. ,
The rolling condition is determined based on the model.

That is, according to the second aspect of the present invention, the metal plate or the metal strip, which is predicted to be unacceptable at the time of predicting the pass / fail of the thickness accuracy of the final product of the metal plate or the metal strip, Reduction of rolling speed and number of passes when performing relief cooling rolling to correct abnormal parts of a metal plate or metal strip based on the optimal quantitative model used to predict the longitudinal thickness accuracy of the metal strip Set conditions such as increase. Then, by cooling and rolling the rejected metal plate or metal strip based on this condition, the abnormal part of the metal plate or metal strip is corrected under reasonable conditions without lowering the productivity, and the metal plate is corrected. Alternatively, the metal strip can be kept within the range of the thickness accuracy of the final product.

That is, according to the second aspect of the present invention, by using the quantitative model used for predicting the accuracy of the thickness of the metal plate or the metal strip in the longitudinal direction, it is possible to appropriately reduce the productivity. It is possible to set a reasonable condition for the relief cooling and rolling.

[0033]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment according to the present invention will be described with reference to FIG. This embodiment is an example in which the method of the present invention is realized using a computer. In the rolling of this embodiment, a metal plate or a metal strip P is rolled in a hot rolling step 1 and then in a cold rolling step 2.

In the hot rolling step 1, a metal plate or a metal strip P
Roll 3 while heating the
, And then wound into a coil. The hot coil HC in which a hot-rolled metal plate or metal strip P is wound into a coil shape is transported to a cold rolling step. In the cold rolling step 2, a hot-rolled metal sheet or metal strip P is unwound from the hot coil HC, passed through a rolling roll 4 and rolled at room temperature, and then wound again as a coil C.

In the hot rolling step 1, before winding the rolled metal plate or metal strip P, a thickness gauge 11 for measuring the thickness of the metal plate or metal strip P in the longitudinal direction, a metal plate or metal strip P, A thickness gauge 12 for measuring the thickness (profile) of P in the width direction is prepared. In the cold rolling step 2, a thickness gauge 13 for measuring the thickness of the rolled metal plate or metal strip P in the longitudinal direction is prepared.

A computer 21 for monitoring the thickness of the hot rolling and a computer 31 for monitoring the thickness of the cold rolling are prepared corresponding to the hot rolling step 1. The hot-rolling thickness monitoring computer 21 includes a longitudinal thickness data collecting / analyzing unit 22, a profile data collecting / analyzing unit 23, and a cold rolling thickness accuracy second.
Model part (longitudinal thickness) 24, cold rolling thickness accuracy first
It has a model section (profile) 25 and a total thickness accuracy calculation / pass / fail judgment section 26. The thickness monitoring computer 31 in the cold rolling step has a longitudinal thickness data collection / analysis unit 32.

Here, the prediction calculation of αn will be described. Based on the above equation (11), the attenuation factor of the i-th path is calculated as follows. αi = Ki / (Ki + Mi) + Ki / (Ki + Mi) · (ΔS / ΔHi) where ΔS (μm): Fixed value set for each rolling mill Mill stiffness coefficient: Ki = CK (Vi) · KoCK (Vi) : Speed correction coefficient. Rolling speed Vi of the i-th pass
(Mpm) Function CK (Vi) = CKv1: 0 <Vi ≦ VokvCK (Vi) = CKv1 · expexp−CKv2 · log (Vi / Vokv)｝: Vokv <Vi ≦ VskvCK (Vi) = CKv1 · exp {-CKv2 • log (Vskv / Vokv)}: Vskv <ViCKv1, CKv2, Vokv, Vskv are semi-fixed values set for each rolling mill. Ko: Reference value (ton / mm). Semi-fixed value set for each rolling mill. Vi is obtained from the cold rolling standard process data.

Next, a plastic elastic coefficient: Mi = Cm (Vi) · Cm2 (i) · MoCm (Vi): a velocity correction coefficient. (Common function for all grades) Cm (Vi) = Cmv1: 0 <Vi ≦ Vomv Cm (Vi) = Cmv1 · exp {Cmv2 · (Vi−Vomv)}: Vomv <Vi ≦ Vsmv Cm (Vi) = Cmv1 · exp {Cmv2 · (Vsmv-Vomv)}: Vsmv <Vi Cmv1, Cmv2, Vomv, Vsmv are positive semi-fixed values. However, common to all rolling mills, however, Cm2 (i): Process correction coefficient (softening etc. when dulling occurs) ｛(0 <Cm2 (i) <1} Mo: Reference value (ton / mm). Mo = Ms · Bi / (Hi-hi) Ms: Reference value for each material (ton / mm ² ) Bi: Strip width (mm) Hi-hi: Reduction amount (mm) is obtained from the cold rolling standard process data. The amount of reduction refers to the difference between the thickness of the entrance side and the thickness of the exit side of the pass.

A method for predicting the thickness accuracy in hot rolling using these factors will be described. In addition. The processing described below is performed every time hot rolling is completed. In the hot rolling step 1, the thickness gauge 11 measures the thickness of the rolled metal plate or metal strip P in the longitudinal direction, and transmits the measured data to the longitudinal plate of the hot rolling thickness monitoring computer 21. The data is sent to the thickness data collection / analysis unit 22. The longitudinal thickness data collection / analysis unit 22 analyzes the measurement data and sends it to the cold rolling thickness accuracy second model unit (longitudinal thickness) 24.

Data of the standard cold rolling process (rolling speed of metal plate or metal strip P, pass schedule, etc.) 41
Enter for. The cold rolling thickness accuracy second model unit 24 creates a product longitudinal thickness prediction value with reference to the data 41 of the cold rolling standard process, and supplies this data to the overall thickness accuracy calculation / pass / fail judgment unit 26. I do.

In the hot rolling step 1, the thickness gauge 12 measures the thickness of the rolled metal plate or metal strip P in the width direction, and the measured data is used by the computer 21 for monitoring the thickness of the hot rolling. The profile data is sent to the profile data collection / analysis unit 23.
The profile data collection / analysis unit 23 analyzes the measurement data and performs a cold rolling thickness accuracy first model unit (profile) 2
Send to 5. The first model section (profile) 25 for the cold-rolled thickness accuracy prepares a predicted thickness value in the product width direction, and supplies this data to the overall thickness accuracy calculation / pass / fail judgment unit 26.

The tolerance specification data 42 of the product thickness is input to the total thickness accuracy calculation / pass / fail judgment unit 26 in advance. The thickness accuracy calculation / pass / fail judgment unit 26 calculates the total product thickness accuracy over the entire length and width of the metal plate or metal strip P based on the input product longitudinal thickness prediction value and product width direction thickness prediction value. Then, the product total thickness accuracy is compared with the tolerance specification to determine whether or not the tolerance is satisfied. If the product total sheet thickness accuracy satisfies the above tolerance specification, the test is passed.

The judgment result is displayed on an operator guidance screen (not shown), and the judgment result is output as a form. In the cold rolling step 2, the thickness gauge 13 measures the thickness of the rolled metal plate or metal strip P in the longitudinal direction, and transmits the measured data to the longitudinal plate of the computer 31 for monitoring the thickness of the cold rolled sheet. The data is sent to the thickness data collection / analysis unit 32. The longitudinal thickness data collection / analysis unit 32 analyzes the measurement data and performs the cold rolling thickness accuracy second model unit 24.
Send to The cold-rolled thickness accuracy second model section 24 uses this data for model correction.

Next, a first embodiment of the present invention will be described with reference to FIGS. As described above, the thickness accuracy calculation / pass / fail judgment unit 26 calculates the product longitudinal thickness estimate (model) created by the cold-rolled thickness accuracy second model unit 24 and the cold-rolled thickness accuracy first value. Based on the predicted product thickness (model) in the product width direction created by the model section (profile) 25, the product total thickness accuracy (model) over the entire length and width of the metal plate or the metal strip P is obtained. The accuracy is compared with the tolerance specification to determine whether the tolerance is satisfied.

According to the present invention, a metal plate or a metal strip predicted to be rejected in the above-mentioned determination is subjected to relief rolling for correcting an abnormal portion of the metal plate or the metal strip in a subsequent step of cold rolling. Rolling conditions that can correct an abnormal portion of a metal plate or a metal strip so as to fall within a tolerance range using a predicted thickness (model) are obtained. That is, based on a quantitative model used for estimating the thickness of the metal plate or the metal strip in the longitudinal direction, it is possible to determine an appropriate relief rolling condition according to an abnormal portion of the metal plate or the metal strip.

As a result, in the relief rolling, the relief rolling is performed without waste under appropriate conditions according to the abnormal portion of the metal plate or the metal strip, that is, without impairing the productivity, and the total thickness accuracy of the product falls within the tolerance range. By obtaining a metal plate or a metal strip, it is possible to prevent a rejected metal plate or metal strip from directly becoming a defective product.

As a method of correcting an abnormal thickness portion of a metal plate or a metal strip, (1) reduction in rolling speed, (2) increase in the number of passes, (3) equipment, and automatic thickness control selection can be considered.
These methods are factors that reduce the productivity in the rolling process. If the rolling speed is reduced more than necessary or the number of passes is increased, the productivity in the rolling process is reduced. As a result, it is impossible to take measures such as lowering the rolling speed more than necessary or increasing the number of passes.

Assuming that the setting of the rolling condition for correcting the abnormal thickness of the metal plate or the metal strip is conventionally performed, the person in charge sets the rolling speed and the number of passes empirically, which is more than necessary. It is conceivable to reduce the rolling speed or increase the number of passes to reduce the productivity in the rolling process.

In order to avoid such unnecessary treatment, it is effective to determine a relief rolling method based on a quantitative model as in the method of the present invention. A specific example of determining a relief rolling method based on the above-described model will be described with reference to flowcharts shown in FIGS. First, when carrying out the method of the present invention, a rescue method setting unit 41 is provided in the computer 21 as shown in FIG. The rescue method is set by obtaining information on the predicted thickness value and performing a calculation. 2, the same parts as those in FIG. 1 are denoted by the same reference numerals.

The embodiment shown in the flowchart shown in FIG. 3 will be described. In this embodiment, a reduction in the rolling speed is first employed as a relief method, and then an increase in the number of passes is employed.

When the hot rolling step 1 is completed, the longitudinal thickness data collection / analysis section 22 of the thickness monitoring computer 21
The thickness measurement data in the longitudinal direction of the metal plate or the metal strip P rolled in the step (1) is analyzed (S1), and the cold rolling thickness accuracy second model unit 24 refers to the data 41 of the standard cold rolling process. A predicted thickness value in the product longitudinal direction is created (S2). Further, the profile data of the rolled metal sheet or metal strip P is analyzed by the profile data collection / analysis section 23, and then the cold rolled sheet thickness accuracy first model section (profile) 2
In step 5, a predicted thickness value in the product width direction is created (S3).

Next, a total thickness accuracy calculation / pass / fail judgment unit 2
6 is based on the product longitudinal direction thickness prediction value and the product width direction thickness prediction value input from the cold rolled thickness accuracy second model part 24 and the cold rolled thickness accuracy first model part (profile) 25. Then, the total thickness accuracy of the product over the entire length and width of the metal plate or the metal strip P is determined, and the overall thickness accuracy of the product is compared with the tolerance specification to determine whether or not the tolerance is satisfied. If the total product thickness accuracy satisfies the tolerance specification, it is accepted, and if not, it is rejected (S4, S5).

Next, an instruction speed is calculated to correct an abnormal portion of the rejected metal plate or metal strip P (S6). The calculation of the speed will be described with reference to the flowchart shown in FIG. In the calculation,
In consideration of the actual work, too small speed designation is not realistic, so calculation is performed in units of 100 mpm (m / min). Also, to minimize the impact on productivity,
The product thickness is recalculated under the condition that the speed is reduced to give priority to the upstream path, and if the tolerance is not satisfied, the speed is further reduced. further,
The path to be decelerated is expanded downstream and recalculated.

In the calculation, the cold rolling pass = the i-th pass, the calculated rolling speed = the standard speed of the pass- (100 mpm ×
j) is a calculation formula. The initial values i0 = 1 and jo = 1 are set, and the cold rolling pass = the i-th pass, the calculated rolling speed = the relevant pass calculation formula is calculated (S11). Next, the attenuation rate of the product of the i-th pass is recalculated (S12), and then a product thickness direction thickness prediction value over the entire length and the entire width of the metal plate or metal strip P is created, and based on the product total thickness prediction value. To obtain the product total thickness accuracy (S13).

Then, the product total thickness accuracy is compared with the tolerance specification to determine whether or not the tolerance is satisfied (S14).
If the tolerance is satisfied, the process ends as the instruction speed determination (S1).
9). If the deviation is out of the tolerance, the target speed is decreased, and the speed is calculated by substituting j = j0 + 1 into the calculated rolling speed = standard speed of the pass- (100 mpm × j) (S15). Next, it is checked whether the calculation speed obtained by this calculation is faster or slower than the minimum allowable speed of the path (S16). If fast, i-th again
The attenuation factor of the product of the pass is calculated, and the speed is calculated by repeating the processes from (S12) to (S16).

If the speed is slow, the path whose speed is to be reduced is moved to the next path, i = i0 + 1. In this case, j sets an initial value j0 = 1 (S1).
7). In this pass, from (S12) to (S
The speed calculation is performed through the processes up to 16).

From the upstream pass to the downstream pass, the speed at which the rejected metal plate or metal strip P in each pass can be rolled with a plate thickness accuracy falling within a predetermined tolerance is determined as the speed. Are sequentially lowered. i is the number of cold rolling passes +
When it becomes 1, it is determined that rescue of the metal plate or the metal strip P is impossible by changing the speed, and the calculation is stopped (S1).
8).

In this way, it is determined whether or not rescue is possible by changing the speed (S7 in FIG. 3). If possible, a speed change instruction is transmitted to the cold rolling process (S8 in FIG. 3). If it is impossible, the number of passes is executed (S9 in FIG. 3).

The remedy method for the rejected metal plate or metal strip P is determined. In this case, as shown in S1 to S3 in the flowchart of FIG. 3, the rejected metal plate or metal strip is based on the quantitative model used for estimating the longitudinal thickness accuracy of the metal plate or metal strip. Calculate the relief rolling conditions for P. For this reason, it is possible to set appropriate and reasonable relief cooling rolling conditions according to the abnormal portion of the metal plate or the metal strip P.

That is, the rolling for correcting the abnormal portion of the metal plate or the metal strip P is performed without difficulty under appropriate conditions without lowering the productivity, and the metal plate or the metal strip is adjusted to the range of the thickness accuracy of the final product. Can fit.

It should be noted that, in addition to the present embodiment, the invention of claim 2 can be applied to the specifications of the rolling mill (sizes of rolls and housings, the presence or absence of a hydraulic pressure reduction device) and the setup of the rolling mill, such as K0 and CK. By back-calculating the state, that is, the selection of the automatic thickness control and the parameters defined by (feed forward, monitor, MMC, etc.), the equipment, the selection of the automatic thickness, and the like can be specifically determined.

Further, the following processing is periodically performed at a time when the actual data of the cold rolling is accumulated. For the same coil, the model deviation value is evaluated from the model prediction of the sheet thickness accuracy calculated in advance at the completion of hot rolling and the actual result of cold rolling. Specifically, CKv1, CKv2,
Vokv, Vskv, Cmv1, Cmv2, Vomv, Vsmv, C
The difference between the current value of each parameter of m2 and the actual value based on the actual data of the cold rolling is regarded as an error, and the amount of correction is determined from this error value through statistical processing to avoid accidental errors.

That is, the error of the model is evaluated using a computer based on the actual results of the thickness of the cold rolling and the actual rolling conditions such as the load and the speed pass schedule of the cold rolling, and the accuracy of the model is improved by correcting the error. It can always be maintained well.

In this embodiment, since the model can be understood by anyone with common sense knowledge as long as the model is engaged in the work related to rolling, the model is not only excellent in developability but also when the process state changes after development. This is also advantageous for maintenance and adjustment of necessary models.

Further, if the thickness management task is aimed at quality assurance, the thickness prediction model requires high reliability. Therefore, what can be understood with common-sense knowledge is desirable also from the viewpoint of early detection of a reduction in design errors during model development and a decrease in model accuracy after development.

Further, in this embodiment, the thickness accuracy after cold rolling is analytically expressed as a model, so that the concept of cold rolling is quantitatively specified, so that rolling is in charge. There is no difference between the people in charge. In addition, by performing the thickness accuracy prediction quantitatively, it is possible to evaluate the validity of the prediction method from the evaluation result of the error. In addition, by comparing and evaluating the predicted value and the measured value of the thickness accuracy of cold rolling, the validity of the thickness accuracy model and the hot rolling judgment using the thickness accuracy model using this model is quantified. Can be grasped.

In the above-described embodiment, it is assumed that the product is shipped via a cold rolling process after the hot rolling process. However, if a plurality of rolling processes are performed, the cold rolling process is performed after the hot rolling process. The present invention can be applied to other cases. As described above, the present invention is not limited to the above-described embodiment, but can be implemented with various modifications.

[0068]

As described above, according to the method for predicting thickness accuracy in hot rolling according to the first aspect of the present invention, when a metal plate or a metal strip is hot-rolled and then rolled, It is possible to obtain an optimal thickness model for performing the thickness accuracy prediction. The obtained model can be understood by anyone with common knowledge as long as it is engaged in work related to rolling, so it is not only excellent in developability but also maintenance of the model required when the process state changes after development. This is also advantageous for adjustment.

Further, if the thickness management task is aimed at quality assurance, the thickness prediction model requires high reliability. Therefore, what can be understood with common-sense knowledge is desirable also from the viewpoint of early detection of a reduction in design errors during model development and a decrease in model accuracy after development.

Further, according to the method for predicting thickness accuracy in hot rolling according to the second aspect of the present invention, the thickness of a metal plate or a metal strip after hot rolling can be quickly and easily quantitatively determined. Metal sheet or metal strip that is predicted to be unacceptable in predicting the pass / fail of the thickness accuracy of the final product of the metal sheet or metal strip is subjected to relief rolling to correct abnormal parts of the metal sheet or metal strip in the subsequent cold rolling. At the time of performing the rolling, a rolling condition capable of correcting an abnormal portion of a metal plate or a metal strip so as to fall within a tolerance range is determined by using a predicted value (model) of a thickness in a product longitudinal direction. That is, based on the quantitative model used for predicting the thickness of the metal plate or the metal strip, it is possible to determine an appropriate relief rolling condition according to the abnormal portion of the metal plate or the metal strip.

Thus, the relief rolling is performed without waste under appropriate conditions according to the abnormal portion of the metal plate or the metal strip, that is, without impairing the productivity, and the metal plate or the metal having a product thickness accuracy within the tolerance range. By obtaining the strip, the rejected metal plate or metal strip can be prevented from becoming a defective product as it is.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining an embodiment of the method according to the present invention;

FIG. 2 is a view for explaining an embodiment of the method of the invention according to claim 2;

FIG. 3 is a flowchart for explaining an embodiment of the method of the present invention according to claim 2;

FIG. 4 is a flowchart for explaining an embodiment of the method of the present invention according to claim 2;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Hot rolling process, 2 ... Cold rolling process, 11, 12, 13 ... Thickness gauge, 21 ... Thickness monitoring computer, 22 ... Longitudinal thickness data collection and analysis part, 23 ... Profile data collection and analysis part, 24: Cold-rolled thickness accuracy second model part (longitudinal thickness), 25: Cold-rolled thickness accuracy first model part (profile), 26: Total thickness accuracy calculation / pass / fail judgment part, 31: Sheet Thickness monitoring computer, 32 ... Longitudinal thickness data collection and analysis unit. P: metal plate or metal strip, HC: hot coil.

Claims (2)

(57) [Claims] 1. When a metal plate or a metal strip is hot-rolled and then rolled, a change in the thickness accuracy of the metal plate or the metal strip in a post-rolling process based on the actual thickness of the hot coil at the end of the hot rolling. And, based on the result of this prediction, predicting the pass / fail of the plate thickness accuracy in the final product of the metal plate or metal strip, and predicting the longitudinal plate thickness accuracy of the metal plate or metal strip, the longitudinal plate A method for predicting thickness accuracy in hot rolling, characterized by using a model that evaluates all factors affecting thickness accuracy as an equivalent mill stiffness coefficient and an equivalent material plasticity coefficient. 2. When a metal plate or a metal strip is hot-rolled and then rolled, a change in the thickness accuracy of the metal plate or the metal strip in the post-rolling process based on the actual thickness of the hot coil at the end of the hot rolling. And, based on the result of this prediction, predicting the pass / fail of the plate thickness accuracy in the final product of the metal plate or metal strip, and predicting the longitudinal plate thickness accuracy of the metal plate or metal strip, the longitudinal plate Using a model that evaluates all factors affecting the thickness accuracy as equivalent mill stiffness coefficient and equivalent material plasticity coefficient, and was judged to be rejected in the prediction of pass / fail of the thickness accuracy in the final product of the metal plate or metal strip When rolling the metal plate or metal strip to correct an abnormal portion of the metal plate or metal strip in a subsequent step, rolling conditions are determined based on the model. Plate thickness accuracy prediction method in hot rolling to.