KR100929017B1 - Plate Break Prediction Method and Apparatus Using Plate Break Prediction Parameters in Cold Rolling Mill - Google Patents

Abstract

The present invention relates to a method for predicting plate break in a cold rolling mill, and more particularly, to a method for predicting plate break signs after deriving plate break prediction parameters by measuring vibration signals generated in a steel sheet.

To this end, the present invention, the plate breaking prediction method in the cold rolling mill, the first step of calculating and storing the crest factor, skewness, overall level size and kurtosis for the vibration signal in the normal rolling conditions; A second step of calculating a vibration signal during a rolling time during rolling operation and then calculating crest factor, skewness, total level size and kurtosis using the vibration signal; A third step of comparing the parameters of distortion, overall level size and kurtosis in the second step with the parameters in normal rolling conditions, respectively; And a fourth step of outputting a plate breakage indication when the parameter of any one of the distortion, the overall level size, and the kurtosis in the third step is increased by 50% or more than the corresponding parameter in the normal rolling condition. A plate breaking prevention method is provided.

Description

Plate breaking prediction method and apparatus using plate breaking prediction parameter in cold rolling mill {A foreknowledge method which is a strip party in cold tandem rolling mill}             

1 is a configuration of a cold rolling mill consisting of five typical stand

2 is a block diagram of a plate breaking prediction device according to the present invention

3A to 3D are graphs showing vibration signals according to an embodiment of the present invention.

4A-4D are graphs showing plate breaking predictive parameters for vibration signals in FIGS. 3A-3D.

※ Explanation of code about main part of drawing ※

20a, 20b: vibration measuring sensor

30: signal amplifier

40: data interpreter

50: Fan breaking monitoring system

The present invention relates to a method for predicting plate break in a cold rolling mill, and more particularly, to a method and apparatus for predicting plate break signs by measuring vibration signal generated from a steel sheet to derive plate break prediction parameters. .

Cold rolled sheet is an important industrial steel material used in furniture, automobiles, beverage cans and the like. The process of making the cold rolled plate is a continuous operation such as welding, pickling, cold rolling, winding, etc. The cold rolling process is a key process for plastic working the steel sheet into a product of appropriate thickness.

In the rolling process, cold rolled steel sheets of 1 to 2 mm are cold rolled to 0.5 to 0.9 mm using a cold rolling mill composed of 4 to 5 stands to produce a product. If a problem arises, it can also affect other processes and degrade the product.

On the other hand, in the cold rolling process when the plate breaks, that is, the plate tearing accident occurs, not only can not use the product, but also costs a lot of time and time required to normalize the installation, greatly reducing the productivity of the product.

There are a number of causes of plate breakage in the cold rolling process. There are many causes of plate breakage such as material defects, control signal abnormalities, and mechanical abnormalities.

For example, when the edge of the plate is deformed much by the overload during the rolling and the thickness becomes thin, the plate breakage is likely to occur. In addition, if the roll of the rolling mill is improperly polished and used, the rolling load is changed every time the roll is rotated, thereby causing plate breakage due to vibration of the rolling mill.

Therefore, engineers engaged in cold rolling have proposed various apparatuses and methods for reducing plate breakage, but there are no known techniques. If you notice the signs before they happen and take appropriate action, you can prevent them.

1 is a configuration diagram of a cold rolling mill composed of five general stands.

In Figure 1, the cold rolling mill is composed of five stands (100 to 500), each stand is composed of a reinforcing roll (10a) and the work roll (10b). In addition, a measuring device such as a plate shape measuring instrument 10c, a plate thickness measuring instrument 10d, and the like is formed at the exit side of each stand.

In the continuous rolling process of FIG. 1, when plate break occurs during rolling, it is difficult to separate and remove the steel plate between the rolling mills, and the failure of the rolling mill due to the plate breakage may occur as well as a safety accident. high.

There are two ways to prevent plate breakage in advance.

The first is to eliminate the factors that can cause the plate to break in advance. For example, in the case of a large number of plate breaks due to work roll machining failures, the standard work management in the roll machining operation is continued, and when the work roll is combined into the rolling mill body, great care is taken to prevent eccentricity from occurring. It is.

Secondly, it is necessary to know in advance the signs that plate breakage may occur during rolling and take appropriate measures in advance.

Japanese Patent Laid-Open No. 2002-192208 discloses a method of suppressing the occurrence of vibration in a rolling mill by sensing a change in the height of a steel sheet due to vibration using a sensor and then controlling the hydraulic pressure of the spindle of the rolling roll. In addition, Japanese Patent Laid-Open No. 2002-205107 discloses a method of improving plate thickness precision by measuring the plate tension at the exit side of the stand and then controlling the horizontal component force and the rolling force of the rolling stand. These known techniques are methods for suppressing and removing plate breaking factors in advance.

In fact, there are many causes of plate failure, so it is virtually impossible to eliminate all of the aforementioned causes of plate failure. Therefore, a more practical countermeasure may be a method of catching and taking measures for signs of plate breakage during rolling by an appropriate method.

An object of the present invention for solving the problems of the prior art as described above is to measure the vibration signal in the steel plate using a vibration measuring sensor in order to reduce the plate breaking phenomenon in the cold rolling mill, and predict the plate breaking by using the vibration signal After deriving the parameters, the present invention provides a method and apparatus for predicting a plate breaking indication.

In order to achieve the above object, the present invention provides a method for predicting plate break in a cold rolling mill, comprising: a first step of calculating and storing crest factor, skewness, overall level size, and kurtosis for a vibration signal in a normal rolling condition; A second step of calculating a vibration signal during a rolling time during rolling operation and then calculating crest factor, skewness, total level size and kurtosis using the vibration signal; A third step of comparing the parameters of distortion, overall level size and kurtosis in the second step with the parameters in normal rolling conditions, respectively; And a fourth step of outputting a plate breakage indication when the parameter of any one of the distortion, the overall level size, and the kurtosis in the third step is increased by 50% or more than the corresponding parameter in the normal rolling condition. A plate breaking prevention method is provided.

In addition, in the cold breaking machine, the plate breaking prediction device comprises: an upper portion of the steel sheet adjacent to the rolling rolls on each stand inlet side and a lower portion of the steel sheet adjacent to the rolling rolls on the stand exit side to measure vibration signals generated in the steel sheet. Measuring sensor; A signal amplifier for receiving a vibration signal from the vibration measuring sensor and amplifying the signal; A data analyzer which receives the vibration signal from the signal amplifier and analyzes the vibration signal during the rolling time; And receiving the output data signal from the data analyzer, calculating the skewness, the overall level magnitude and the kurtosis, and comparing the numerical values of the respective parameters with the numerical values of the parameters in the normal rolling conditions to predict and display the plate breaking signs. Provided is a plate breaking prediction device comprising a plate breaking monitoring system.

Hereinafter, a method and apparatus for predicting plate breakage in a cold rolling mill according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram of a plate breaking prediction device according to the present invention.

2 or more vibration measuring sensors 20a and 20b are formed at a position close to the rolling roll in each stand 100 of the housing of the cold rolling mill as shown in FIG. Measure One vibration measuring sensor 20a is formed at the inlet of each stand, and one vibration measuring sensor 20b is formed at the bottom of the stand at the bottom of the steel plate, and the vibration signal is measured more precisely.

The signal amplifier 30, the data analyzer 40, and the plate breaking prediction monitoring system 50 are configured to process the plate breaking prediction parameter by processing the vibration signal measured by the vibration measuring sensor.

The signal amplifier 30 receives and amplifies a vibration signal from the vibration measuring sensors 20a and 20b. In order to perform more accurate data analysis, it is preferable to amplify a fine vibration signal using a signal amplifier.

The data analyzer 40 receives the amplified signal from the signal amplifier and performs time-domain analysis. That is, the amplitude magnitude of the vibration which generate | occur | produced in each time slot during rolling progress is calculated.

The plate breaking prediction monitoring system 50 receives a time domain data analysis result from the data analyzer to derive plate breaking prediction parameters such as crest factor, skewness, overall level size and kurtosis of the vibration signal. In addition, the plate breaking prediction parameter is analyzed to detect signs of plate breaking. If signs of plate breaking are detected, the plate is output through a monitor installed in front of the operator to inform the user of the signs of plate breaking.

In general, plate breaking occurs during high-speed rolling of more than 800mpm (meter per minutes), so if signs of plate breaking are detected, the operator can take measures such as lowering the rolling speed to less than 100mpm and lowering the tension before and after the rolling mill. Therefore, plate breakage can be prevented in advance.

3A to 3D are graphs showing vibration signals according to an embodiment of the present invention. Hereinafter, the process of finding the basis of plate breaking through the vibration signal measured in cold rolling is demonstrated.

3A to 3D are vibration signal data measured between the third stand and the fourth stand, respectively, in the normal state, in the normal state, in which the speed of the steel sheet is increased, in which the roll is replaced, in the state immediately before breaking of the plate. Indicates a vibration signal.

According to each state, the vibration signal generated during the one second time in the third stand represents the rolling phenomenon at each time point. It consists of data measured at intervals of about 30 minutes to 1 hour from the normal rolling data (FIG. 3A), which is a reference, and each state has a sampling frequency of 4000 Hz.

The vibration signal in the normal state (FIGS. 3A and 3B) is not very characteristic, but the vibration signal detected immediately after the roll replacement (FIG. 3C) shows a periodic abnormal signal with a large intensity, and the strength is the breaking of the plate. It can be seen that it continues until just before (FIG. 3D).

Next, a plate breaking predictive parameter is derived using the vibration signal in the steady state measured by the vibration measuring sensor and a vibration signal immediately before the plate breaking, and the plate breaking indication is predicted through the plate breaking predictive parameter. Explain.

4A to 4D are graphs showing plate breaking prediction parameters for the vibration signal in FIGS. 3A to 3D.

First, the plate breaking prediction parameter will be described as follows.

(1) CREST FACTOR

In most cases, the waveform is distorted, and the crest factor is an important point in order to accurately measure the waveform. The crest factor is large when the slope becomes sharp, such as when the vibration signal suddenly increases or decreases.

Here, x _p means the difference between the highest value and the lowest value of the signal, and the larger the range, the more the vibration signal is moving away from the average.

Here, x _rms represents an effective value of a signal and is also called an RMS value. Equation 3 below represents an RMS value.

Where n is the number of signals and x _i is the magnitude of an individual signal.

(2) SKEWNESS

This value represents the degree of asymmetry of the probability distribution. These skew values are dimensionless due to the characteristics divided by the standard deviation.

는 평균치, s는 표준편차로서 다음과 같이 계산된다.here,

Is the mean value, s is the standard deviation, and

If the skewness is a positive value, the peak shape is generally in the positive direction of x. If the skewness is a negative value, the peak shape is in the negative direction of x.

(3) K-FACTOR

The peak value of the signal multiplied by the RMS value. This parameter is an indicator of the overall level of the signal.

(4) KURTOSIS

This value represents the degree of sharpness of the probability distribution.

If the kurtosis is positive, the shape of the peak is sharp compared to the normal distribution, and if it is negative, it is flat. The subtraction of 3 from Equation 7 is for the kurtosis to be 0 in the normal distribution.

4A to 4D, it can be seen that the remaining three parameter values except the crest factor become larger as the time immediately before the plate break occurs. Accordingly, the four parameters are continuously analyzed in real time during rolling, and it is determined that breakage may occur when the characteristic value is excessively increased. In particular, if the value of each parameter is increased by more than 50% from the normal rolling state, it is foreseen that plate breakage is very likely.

While the invention has been described above based on the preferred embodiments thereof, these embodiments are intended to illustrate rather than limit the invention. It will be apparent to those skilled in the art that various changes, modifications, or adjustments to the above embodiments can be made without departing from the spirit of the invention. Therefore, the protection scope of the present invention will be limited only by the appended claims, and should be construed as including all such changes, modifications or adjustments.

According to the present invention as described above, by measuring the vibration signal using a vibration measuring sensor, by deriving the plate failure prediction parameters to predict the signs of plate failure in advance, to prevent the plate failure in advance to improve the product quality, equipment It is effective in preventing breakdown and improving productivity.

Claims (3)

In the plate breaking prevention method in a cold rolling mill, A first step of calculating and storing the crest factor, skewness, overall level size and kurtosis for the vibration signal under the normal rolling condition; A second step of calculating a vibration signal during a rolling time during rolling operation and then calculating crest factor, skewness, total level size and kurtosis using the vibration signal; A third step of comparing the parameters of distortion, overall level size and kurtosis in the second step with the parameters in normal rolling conditions, respectively; And And a fourth step of outputting a plate breaking indication when any one of the distortion, the overall level size, and the kurtosis in the third step is increased by 50% or more than the corresponding parameter in the normal rolling condition. Plate breaking prediction method using the plate breaking prediction parameter in the cold rolling mill. delete In the plate breaking prediction device in a cold rolling mill, A vibration measuring sensor configured at an upper portion of the steel sheet adjacent to the rolling rolls on each stand inlet side and a lower portion of the steel sheet adjacent to the rolling rolls on the stand exit side to measure vibration signals generated in the steel sheet; A signal amplifier for receiving a vibration signal from the vibration measuring sensor and amplifying the signal; A data analyzer which receives the vibration signal from the signal amplifier and analyzes the vibration signal during the rolling time; And The output data signal is inputted from the data analyzer to calculate the skewness, the total level size and the kurtosis, and compare the numerical values of the calculated parameters of the skewness, the total level size and the kurtosis with the parameters under normal rolling conditions. A plate break prediction device using a plate break prediction parameter in a cold rolling mill, characterized by including a plate break prediction monitoring system for predicting and outputting a plate break indication.

Images