JPH02129508A - Shape detecting device - Google Patents

Abstract

PURPOSE:To obtain an output from noises due to the irregular vibration of a body to be measured are removed irrelevantly to the period of a driving signal by operating the difference in the mean processing value of the quantity of displacement in each period wherein the body to be measured is applied with an external force and each period wherein the external force is not applied. CONSTITUTION:A gap detection signal obtained from a gap converter 6 is sampled and held by a sample and hold circuit 20 and A/D-converted 21; and the signal in the period wherein the body 1 to be measured is applied with the external force is stored in a memory 22 and the signal in the period wherein the external force is not applied is stored in a memory 23. The signal read out of the memory 22 is processed by a signal smoothing circuit 24 and the signal read out of the memory 23 as a noise component is processed by a noise smoothing circuit 25 for movement averaging processing respectively. Movement mean signals obtained from the circuits 24 and 25 are sent to a difference arithmetic circuit 26 to operate the difference. A displacement mean value computing element 31 inputs M displacement quantities found by the circuit 26 and operates their mean value and an elongation percentage computing element 32 inputs the mean value from the computing element 31 and operates the elongation percentage from a preset arithmetic expression.

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention is applicable to detecting the shape (flatness) of a strip-shaped object such as a thin steel plate by knowing the tension distribution in the width direction of the strip. The present invention relates to a shape detection device used.

[Conventional technology]

In general, when cold rolling a strip-shaped object such as a thin steel plate, the shape (also referred to as flatness) is important as well as the accuracy of the plate thickness. However, in cold rolling, high tension is applied during rolling, so the elastic elongation of the rolled material, that is, the strip, may cause flatness defects such as mid-elongation or ear waves in the strip. , the displacement (unevenness) usually decreases or disappears and cannot be detected. Therefore, when high tension is applied as described above, it is not possible to directly detect the flatness defects in the strip, but by knowing the tension distribution in the width direction of the strip, the shape can be detected indirectly. is K as shown in Special Publication No. 53-17071.
well known. In other words, the tension is weak in areas with poor flatness, so the shape can be determined from the tension distribution.

FIG. 7 is a block diagram showing a conventional shape detection device described in, for example, Japanese Patent Application No. 119029/1982. A support roll, such as a deflector roll, is shown for the application. Reference numeral 3 indicates a square wave oscillator as a drive signal generator, and 4 indicates an amplifier to which the drive signal is applied.

Reference numeral 5 denotes a detection head, which is provided along the width direction of the strip 10 and separated from the surface of the strip 10 by appropriate locking means at appropriate intervals, and is composed of an external force applying device 5A and a gap detector 5B. be done. The external force applying device 5A is an electromagnet provided with a magnetic pole AK excitation coil B having a U-shaped cross section along the width direction of the strip 10, and the gap detector 5B is attached to the base portion C along the width direction of the strip 10. A plurality of gap detection electrodes are provided facing the surface of the strip 10, and are provided integrally with the external force applying device 5A. Note that the gap detector 5
B may be provided separately from the external force applying device 5A, and the external force applying device 5A is not limited to the electromagnet described above, but may be any device capable of applying force without contact, such as compressed air.

6 is a gap converter such as a capacitance-to-voltage converter connected to the gap detector 5B, T is a signal processing circuit, and bandpass filter 8. Polarity switch9. Integrating circuit 10
．． Sample hold circuit 11 and the polarity switch 9~
It consists of a timing generation circuit 12 connected to a sample hold circuit 11. The timing generation circuit 12 is also connected to the square wave oscillator 3. Reference numeral 13 indicates a display device control circuit connected to the signal processing circuit 1, and 14 indicates a display device, such as a CRT monitor, connected to the display device control circuit 13. Note that the signal processing circuit 7 is connected to the roll crown adjustment device 1 via the roll crown control circuit 15.
It can also be connected to 6.

Further, the double-framed portions 6.8 to 11 in the figure are provided for M gap detection electrodes, respectively.

Next, the operation will be explained.

First, a rectangular wave with a period Tc is generated as a drive signal by the rectangular wave oscillator 3, this drive signal is amplified by the amplifier 4, and this amplified signal is converted into an external force by the external force applying device 5A to apply it to the surface of the strip 10. to be applied.

This causes a displacement P(xt) to be generated in the strip 1. The displacement P(xllt) generated on the surface of the strip 10 is detected as capacitance by a gap detection electrode provided in the gap detector 5B, and converted into a voltage signal by the gap converter 6.

A plurality of (M) gap detection electrodes provided along the width direction of the strip 10 detect the gap at each part with the corresponding strip 1, that is, the displacement of the strip 10 in the width direction. , these gap detection values are input into the signal processing circuit 7 after being converted into voltage signals. Here, the gap detection signal passes through a bandpass filter 8 and is converted into a displacement signal, and then the polarity is switched by a polarity switch 9, and then the integration circuit 1
It is input to 0 and integrated every rectangular wave period. Through this operation, noise components other than signals related to tension are removed, and only the portions related to the tension of each part of the strip 10 are calculated, and the integral value thereof is input to the sample and hold circuit 11 and sampled and held.

Furthermore, the timing generation circuit 12 controls various timings such as the polarity switching timing of the polarity switch 9, the reset timing of the integrating circuit 10, and the sample hold timing of the sample hold circuit 110, based on the reference signal from the square wave oscillator 3. ing. Sample hold circuit 110
The output is transmitted via the display device control circuit 13 to, for example, CR.
It is displayed on a display device 14 such as a T-monitor, and is used for monitoring by an operator and as data for shape adjustment.

In addition, when performing automatic shape control of the strip 10, the output of the sample hold circuit 110 is input to the roll crown adjustment device 16 through an automatic control system, that is, a roll crown control circuit 15, separately from the display system. can be reached.

In the above description, a case has been described in which a rectangular wave signal is used as the drive signal, but the drive signal is not limited to a rectangular wave signal, and may be an M-sequence signal, for example.

A signal waveform such as a random signal or a sine wave signal can also be used, and if these signal waveforms are employed, a multiplication circuit may be provided in place of the polarity switch 9 in the signal processing circuit 7.

The above series of operations will be further explained using mathematical formulas. 7th
In the diagram, Tension L: Roll span KI: Coefficient in force balance. Considering the force balance for the band width (hereinafter simply referred to as plate width) ΔX, the displacement P(x-
t) is expressed as follows.

The displacement detection signal from (1) 2 is integrated for each period of the rectangular wave, and the postmortem sample and hold circuit 110 output C (x-
nTc) is expressed as follows. However, Tc: period of the rectangular wave n: number of cycles of the rectangular wave, or the driving external force f(t) per unit width is a rectangular wave with an amplitude a.

Therefore, assuming that a is an electrical quantity, the output value C (xllnTc) from the sample and hold circuit 11 is the above-mentioned (
It consists of equation 1), equation (2), and equation (3). Here, the second term in Equation (4) is the effect of disturbance, but this is considered to be sufficiently smaller than the stationarity of the irregular vibration d(t) of the strip body 10 as the rectangular wave period Tc becomes larger. Therefore, the above equation (4) can be expressed as follows. From this equation (5), the tension T(X) per unit width at the width direction coordinate X can be expressed as follows, and the tension is measured.

Next, the signal waveforms of each signal processing section in the circuit of FIG. 7 which realizes the content of the above equation (4) will be explained with reference to FIG.

Figure 8) shows a rectangular wave excitation signal generated from the rectangular wave oscillator 3, where n is a positive integer, and so on.

m-2,...9m+1 are the six values that n takes.

Fig. 8) shows the period Tc turned ON by the external force applying device 5A,
It shows a displacement signal when the strip 1 is displaced by an applied external force that is OFF.

FIG. 8(e) shows the disturbance signal component at the zero point due to the irregular vibration of the strip body 10, and the noise signal due to this irregular vibration is superimposed on the displacement signal of FIG. 8(b) described above. The gap detection signal shown in FIG. 8(d) is input to the signal processing circuit 7. This signal with the waveform of FIG. 8(d) is converted into the alternating current waveform of the displacement signal of FIG. 8(e) by passing through the bandpass filter 8, and when the polarity is switched during the period from then to (n+1)Tc, Through this polarity switch 9, a waveform signal (zero point signal) as shown in FIG. 8(f) is obtained. When this is integrated by the integrating circuit 10 over the period from nTc to (n+1)Tct, the waveform signal (0 point signal) shown in FIG. 8(g) is obtained as the integrated output. This waveform signal (
n+1) When the sample and hold circuit 11 samples and holds the signal at Tc0, the signal shown in FIG. 8 (can) (0 point signal) is obtained.

As is clear from the above equation (4) and the signal processing waveform in FIG.
Obtained every c.

[Problem to be solved by the invention]

Since the conventional shape detection device is configured as described above, the output response is determined by the rectangular wave period Tc for periodically applying an external force to the strip, and the rectangular wave period Tc can be shortened. The response must be made faster by
Considering the effect of removing noise caused by irregular vibration of the strip body 10 by increasing the rectangular wave period Tc and the displacement response of the strip body 10, the lower limit of the rectangular wave period Tc is limited, so that the response is improved. You will not be able to speed up the process.

Furthermore, as expressed by equation (5), the output signal is inversely proportional to the tension distribution T of the strip 10) and proportional to the vibration @a due to the applied external force, so even if the flatness of the strip 10 is the same, However, the output varies depending on the applied tension and external force, which is extremely inconvenient when the purpose is to control the flatness of the strip 10.

This invention was made in order to solve the above-mentioned problems, and it is possible to obtain an output that eliminates noise caused by irregular vibrations of the object to be measured, regardless of the period of the drive signal, and is also suitable for shape detection. Converts the output to the elongation rate of the object to be measured, and even if the external force applied during tension changes, if the flatness of the object is the same, the same elongation rate can be obtained, and shape detection with excellent responsiveness. The purpose is to obtain equipment.

[Means to solve the problem]

The shape detection device according to the present invention obtains the amount of displacement by sampling the gap detection signal obtained during the tenth period in which an external force is applied to the object to be measured and the second period in which no external force is applied.
After calculating the average processed value of the displacement amount for each period, calculate the difference between the average processed values for the rainy period, or after calculating the difference in the displacement amount for the rainy period, calculate the average processed value of this difference value. This is how it was done.

[Effect]

In the shape detection device according to the present invention, by performing the above-mentioned difference calculation, the noise component due to the irregular vibration of the object to be measured, which is present evenly in the first and second periods, is canceled out, and the drive signal of the object to be measured is improved. Only the displacement caused by is detected.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, parts indicated by 1 to 6 correspond to parts with the same reference numerals in FIG. 7, so explanations thereof will be omitted. A signal processing circuit 17 processes the gap detection signal obtained from the gap converter 6. In this signal processing circuit 1T, 20 is a sampling≠hold circuit that samples and holds the gap detection signal, 21 is an A/D converter that digitizes the sampled and held signal, and 22 is an A/D converter that digitizes the sampled and held signal. A tenth memory stores data obtained from the transducer 21 during the tenth period during which the external force is applied;
a second memory that stores data obtained from the D converter 21; 24 is a signal smoothing circuit as a tenth smoothing circuit that performs moving average processing on the signal read out from the tenth memory 22;
25 is a noise smoothing circuit as a second smoothing circuit that performs moving average processing on the signal as a noise component read from the second memory 23; 26 is a moving average signal obtained from the signal smoothing circuit 24 and the noise smoothing circuit 25; 27 is a reference clock generation circuit; 28 is a timing generation circuit that generates a clock to be supplied to the sample hold circuit 20 and the A/D converter 21 based on the reference clock; 29 is a timing generation circuit based on the reference clock; First and second memories 22, 23, and a signal smoothing circuit 24. This is a timing generation circuit that generates a clock to be supplied to the noise smoothing circuit 25 and the difference calculation circuit 26. It is assumed that the signal processing circuit 17 and the gap converter 6 are provided for M gap detection electrodes, respectively.

30 indicates an extension wheel signal processing circuit, which is connected to the above-mentioned difference calculation circuit 2.
A displacement average value calculator 31 that inputs the M displacement amounts obtained in step 6 and calculates the average value, and calculates the elongation rate using the average value from this displacement average value calculator 31 according to a preset calculation formula. and an elongation rate calculator 32.

The elongation wheel signal obtained by the elongation rate calculator 32 is applied to the display device control circuit 13 and, if necessary, the roll crown control circuit 15.

FIG. 2 shows the signal waveform of the first cycle of the gap detection signal shown in FIG. 8(d), where the horizontal axis is time t and the vertical axis is voltage V.

FIG. 3 is a diagram showing a moving average processing method in the signal smoothing circuit 24 and the noise smoothing circuit 25, and an arithmetic processing method in the difference calculation circuit 26.

In Figure 4, the width of the strip 10 is plotted on the horizontal axis, and the width of the strip 10 is plotted on the vertical axis.
FIG. 3 is a diagram showing the relationship between the tension T (b) in unit ab and the elongation rate βω.

Next, the operation will be explained.

The gap detection signal at point ■ in FIG. 1 has a signal waveform shown in FIG. 8(d), and only the first period of this signal waveform is shown in FIG. In Figure 2, (m-2)Tc” (
m-2) The first half period of Tc+Tc/2 is the 10th period in which an external force 1 (1) is applied to the strip 1 according to the half period of the rectangular wave drive signal shown in FIG. 8(a). .

The amount of displacement of the strip body 10 in this i@10 period includes the 8th
As mentioned above, the disturbance signal shown in FIG. 3(c) is included. This amount of displacement is sampled by the sample and hold circuit 20. That is, as shown in FIG.
The gap detection signal is sampled by a sampling clock having a cycle that is sufficiently short than the 0 period, and P1s
Subdivided displacement amounts of P1 to Pn are obtained. These displacement amounts are digitized by the A/D converter 21 to become displacement data] and stored in the tenth memory 22.

Next (m-2) Tc + Tc/2 ~ (m-1) Tc
The second half of the period is the second period in which no external force f (t) is applied.

The displacement amount in this second period also includes the above-mentioned disturbance signal. The gap detection signal of this second period is also sampled by the sample and hold circuit 20, thereby obtaining the subdivided displacement amount N1. N, ~Nn are obtained. These displacement amounts are digitized by the A/D converter 21 and stored in the second memory 23 as displacement amount data.

Next, the first and second memories 22.23 are read out, and the read displacement amounts P, ~Pn and N, ~Nn are respectively read out from the signal smoothing circuit 24.23. Moving average processing is performed in the noise smoothing circuit 25 as shown in FIG. FIG. 3 shows a case where moving average processing is performed on three sampled displacement data pieces at a time. First, the three data P1゜p, , p, read out from the tenth memory 22 are averaged, and the averaged value is called P.

shall be. The three pieces of data N, , N, , N read out from the second memory 23 are averaged, and the averaged value is taken as the rice yield. Next, the average processed value P of P*, P, a, P4 is determined, and the average processed value *N of N, , N, , N is determined. In this way, the average processed value is sequentially calculated for each three pieces of data and stored.

Next, the above average processing values rice P1 to rice Pn and rice N1 to rice Nn
-*p by the difference calculation circuit 26.

*N5-HP! By performing calculations to obtain the differences between −US N, ~USPn, and USNn, the difference data S, , S, ~S
n is obtained.

As described above, the displacement amounts of the gap detection signals sampled in the first and second periods are sequentially subjected to moving average processing, and
When calculating the difference between the average processed values for the second period and the second period, the irregular noise components that are evenly present in the first and second periods are canceled out, and the external force f(t) on the strip body 10 is It is possible to extract only the amount of displacement from K (No. 81). Further, the output of the difference calculation circuit 26 can be obtained at the shortest period of the sampling clock, regardless of the drive period Tc.

Note that for the 10th period in the second period following the first period, the difference from the second period of the first period is calculated. Also,
If a certain temporal stationarity exists in the gap detection signal of the second period, an average processed value of all data of this second period may be used. Furthermore, there is no need to set the drive cycle Tc to a deity every Tc/2, and the 10th period is made sufficiently long.
The second period may be made as short as possible within a range that allows moving average processing.

Next, to express the degree of defective shape (flatness) of the strip 10,
Consider the elongation rate β (b). This elongation rate β(g) is defined as the amount of elongation of the defective portion expressed as a ratio of the average portion of the defective shape to the length of the strip 10. There is a difference between the tension distribution T←) of the defective shape part and the average tension T.
A force equation relationship holds true, and FIG. 4 illustrates it.

When E is the elastic modulus of the strip 10, Tω=T±β(g)・E
......(Force) (If βω is determined from the force equation, the above equation (8) is obtained.

Similarly, the average tension T is obtained by the above equation (9).

Corresponds to the amount of displacement. This C is obtained from M displacement signals detected by a plurality of (M) ejection electrodes arranged at appropriate intervals in the width direction of the strip 10.
The average value of the outputs of the difference calculation circuits 26 can be calculated by the displacement average value calculation unit 31.

By substituting T (g) and T into the above equation (8), the following fourth equation is obtained.

However, C(X) is an abbreviation of C(x@t), and indicates the amount of displacement of the strip 10 outputted every time.

As is clear from the above equation, the elongation rate β(t) of the strip 10 can be determined from the above equation 01.

Therefore, the elongated book adder 32 into which the displacement average value C calculated by the displacement average value calculator 31 is inputted,
By executing the calculation of Equation 11, it is possible to obtain an output of elongation rate β).

Note that L-a*E is a value set in advance in the stretch book adder 32, and is detected by the electrode located at the X position in the gap detection electrode plate, and is detected by the electrode at the position of It becomes possible to calculate the above equation (10) using the difference signal C (g) and the average difference signal C given as they are through the elongation calculator 31. Then, the elongation rate calculated by the elongation adder 32 βω is displayed on the display device 14 via the display device control circuit 13, is provided for monitoring by the operator, and is used as data for shape adjustment.

In addition, when performing automatic control of the strip 10, elongation rate β (b)
is input to the roll crown adjusting device 16 via the roll crown control circuit 15 to achieve the purpose.

Note that in the above embodiment, the first and second memories 22.2
The signal smoothing circuit 24 and the noise smoothing circuit 25 perform moving average processing on the displacement data read from 3, and then the difference calculation circuit 26 calculates the difference between the average processed values, as shown in FIG. So 1 first and second memo 1722.
The difference calculation circuit 26 may first calculate the difference between the displacement data read from the displacement data 23, and then the smoothing circuit 33 may perform a moving average process on the difference data.
The same effect as in the above embodiment is lost.

FIG. 3 shows a method of difference calculation processing and moving average processing in the case of the embodiment shown in FIG. First, the tenth memory 22
The displacement amounts P, ~Pn read from the second memory 23
Displacement amount N read from.

Each difference s with ~Nn, =:p, -N@, S! =P!
-N! ~S n =Pn-Nn is calculated. Next, moving average processing is sequentially performed on three pieces of the difference data S, -Sn. That is, first, S1, S8. By calculating the average processed value of Sa and then calculating the average processed value of St, 5s-8+, the difference calculation processed value S1. Rice S2・
...*-8n is obtained.

The difference calculation processing values 81 to 8Sn are substantially equal to the difference data S and 81 to Sn shown in FIG.

〔Effect of the invention〕

As described above, according to the present invention, the amount of displacement is calculated by sampling the gap detection signals obtained for the tenth period in which an external force is applied to the object to be measured, such as a strip, and the second period in which no external force is applied. After performing a moving average process on these displacement data, either calculate the difference between the average processed values for the first and second periods, or calculate the difference between the displacement values for the first and second periods. After that, the moving average processing is performed on each difference data, so irregular vibration components of the object to be measured can be removed and only the amount of displacement of the object to be measured can be detected.
In addition, since the amount of displacement of the object to be measured is obtained at least every sampling period, it is possible to perform shape detection with quick response. By determining the elongation rate β (g) of the object to be measured, the shape of the object to be measured can be clearly understood as the degree of physical defective shape, and therefore the tension and external force applied to the object to be measured can be clearly understood. This prevents the shape detection output from changing, which is advantageous when managing the flatness of the object to be measured.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a shape detection device according to an embodiment of the present invention, FIG. 2 is a waveform diagram for explaining sampling processing in the device, and FIG. 3 is a waveform diagram for explaining signal processing in the device. Fig. 4 is a characteristic diagram showing the relationship between tension distribution and elongation rate, Fig. 5 is a block diagram showing a shape detection device according to another embodiment of the present invention, and Fig. 6 explains the sampling process in the same device. FIG. 7 is a block diagram showing a conventional shape detection device, and FIG. 8 is an output waveform diagram of each part of the device. 1 is a band-shaped body, 3 is a square wave oscillator, 5A is an external force application device,
5B is a gap detector, 20 is a sample hold circuit, 22
23 is a 10th memory, 23 is a second memory, 24 is a signal smoothing circuit, 25 is a noise smoothing circuit, and 26 is a difference calculation circuit. In addition, in the figures, the same reference numerals indicate the same or equivalent parts. 1st WA

Claims (2)

[Claims] (1) An external force applying device that is driven by a drive signal in which one period is composed of a first and second period and applies an external force to a surface of the object to be measured along a predetermined direction during the tenth period, and the external force is applied. A shape detecting device having a gap detector for detecting a gap between a plurality of points on a surface, and a sample hold circuit to which a gap detection signal obtained from the gap detector is supplied during the first and second periods; a first memory that stores a signal output from the sample and hold circuit during the first period; a second memory that stores a signal that is output from the sample and hold circuit during the second period; a first smoothing circuit that performs moving average processing on the signal read from the first memory; a second smoothing circuit that performs moving average processing on the signal read from the second memory; and a second smoothing circuit that performs moving average processing on the signal read from the second memory; A shape detection device comprising: a difference calculation circuit that calculates a difference between a signal output from the smoothing circuit and a signal output from the second smoothing circuit. (2) an external force applying device that is driven by a drive signal in which one period consists of a first and a second period and applies an external force to a surface along a predetermined direction of the object to be measured during the first period; and a gap detector for detecting a gap between a plurality of points on a surface to be detected, the sample and hold circuit being supplied with a gap detection signal obtained from the gap detector during the first and second periods. , a first memory that stores a signal output from the sample and hold circuit during the first period; a second memory that stores a signal that is output from the sample and hold circuit during the second period; a difference calculation circuit that calculates the difference between the signal read from the first memory and the signal read from the second memory; and a smoothing circuit that performs moving average processing of the signal output from the difference calculation circuit. A shape detection device characterized by being provided with.