JP2005337846A - Tension measuring method of belt-like plate - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a tension measuring method capable of detecting accurately the natural frequency of a belt-like plate. <P>SOLUTION: Air pressure fluctuation generated by vibration of a metal plate 1 is measured by two low-frequency noise meters 4, 5 arranged respectively on a first position P1 on the upper surface side of the metal plate 1 to be excited and on a second position P2 on a horizontal plane including the metal plate 1. The difference ΔP between two pressure fluctuation data Pa, Pb at the first position P1 and the second position P2 measured by the two low-frequency noise meters 4, 5 is taken, and the difference data are subjected to frequency analysis, and thereby the natural frequency of the metal plate 1 is extracted, and the tension of the metal plate 1 is determined based on the extracted natural frequency. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a tension measuring method for various strip-shaped plates such as metal, paper, and cloth, and more particularly to a tension measuring method for obtaining the tension based on the natural frequency of the strip-shaped plate.

  In a line that feeds a strip made of metal or the like to the downstream side through a roll or the like, the plate is always subjected to an external load having various frequency components due to fluctuation torques caused by various drive rolls or roll eccentricity. Vibration force is acting, and the plate often vibrates at a natural frequency. In view of this, a method has been conventionally proposed in which the tension is obtained from the natural frequency in the longitudinal direction at both ends in the width direction of the plate between the support rolls using the vibration phenomenon of the plate.

  Here, when measuring the vibration of the plate, a non-contact displacement meter such as a laser displacement meter or an eddy current displacement meter is widely used, but the measurement range of a general laser displacement meter is narrow. Therefore, a vibration detector such as a microphone with a wide measurement range is arranged on one surface side of the plate and at a position separated from the plate, and the vibration detector detects the pressure fluctuation of the air around the plate to detect the fluctuation of the plate. A method for measuring vibration has also been proposed (see, for example, Patent Documents 1 and 2).

  Further, when a clear peak of the natural frequency of the plate cannot be detected, a required mode (generally, a primary mode) is generated by forcibly vibrating the plate by applying an external force to the plate. It is effective to excite the natural vibration of. As a method for generating vibrations on the plate in this way, there is a method of blowing a fluid such as compressed air against the plate in addition to a method of giving a direct impact (see, for example, Patent Document 3).

JP-A-8-120433 JP-A-7-98246 JP-A-6-43051

  However, when measuring the pressure fluctuation caused by the vibration of the plate with a vibration detector located at a position away from the plate, if the surrounding noise level is high, the pressure fluctuation caused by the external noise is also caused by the vibration detector. Therefore, the ratio (SN ratio) between the pressure fluctuation due to the vibration of the plate and the pressure fluctuation due to the external noise is lowered, and it becomes difficult to extract the natural frequency component of the plate. In addition, various external excitation forces are acting on the plate, but when the response of a specific frequency component of such external excitation force is sufficiently large that it cannot be ignored compared to the natural frequency component. Since both the peak of the frequency component due to the external excitation force and the peak of the natural frequency component appear at the same time, it is difficult to extract the natural frequency component of the plate.

  Furthermore, the plate has a number of natural vibration modes. In particular, in a state where the tensions at both ends in the width direction of the plate are different (hereinafter referred to as a biased tension state), more complex natural frequency characteristics are exhibited. It may be difficult to extract a necessary natural frequency component (particularly, a first-order mode) by simply vibrating the plate.

  An object of the present invention is to provide a tension measuring method capable of accurately detecting the natural frequency of a belt-like plate.

Means for Solving the Problems and Effects of the Invention

  The tension measuring method according to the first aspect of the invention comprises two pressure measuring means arranged at a first position on one surface side of a belt-like plate to be vibrated and a second position on the surface including the belt-like plate. A difference between two pressure fluctuation data at the first position and the second position measured by the two pressure measuring means, and a first step of measuring the pressure fluctuation of the air generated by the vibration of the belt-like plate And the second step of extracting the natural frequency of the strip plate by frequency analysis of the difference data, and the third step of obtaining the tension of the strip plate based on the extracted natural frequency. It is characterized by that.

  In this tension measurement method, an external force (forced excitation force) is forcibly applied to the plate and vibrated to excite the natural frequency component of the plate, and pressure fluctuations caused by the vibration of the plate are detected by pressure measuring means. Measure (first step), perform a Fourier transform on the pressure fluctuation data and perform frequency analysis to extract the natural frequency (second step), and then the tension of the strip plate based on the extracted natural frequency Is obtained (third step). By the way, two pressure measuring means are respectively arranged at the first position on one surface side of the belt-like plate and the second position on the surface including the belt-like plate. Then, in the first step, the pressure measuring means arranged at the first position measures the pressure fluctuation of the air generated by both the vibration of the excited strip and the external noise. On the other hand, the pressure measuring means arranged at the second position measures the pressure fluctuation of the air generated by the external noise, but the pressure fluctuation due to the vibration of the plate propagates from both sides of the plate to the second position. Not measured because pressure fluctuations are out of phase with each other and cancel. Therefore, in the second step, the pressure fluctuation caused by external noise can be removed from the pressure fluctuation data by taking the difference between the two pressure fluctuation data respectively measured by these two pressure measuring means. The frequency can be easily extracted.

  A tension measuring method according to a second aspect of the present invention is the first position on one surface side of the belt-like plate to be excited and the other surface side of the belt-like plate, and the first position and the belt-like plate are symmetrical. The first step of measuring the pressure fluctuation of the air generated by the vibration of the strip plate by two pressure measuring means respectively arranged at the position 3, and the first position measured by the two pressure measuring means And the second step of extracting the natural frequency of the strip plate by taking the difference between the two pressure fluctuation data at the third position and analyzing the frequency of the difference data, and the extracted natural frequency. And a third step of determining the tension of the belt-like plate on the basis thereof.

  In the first step, the pressure measuring means disposed at the first position on one surface side of the belt-shaped plate, and the third position disposed on the other surface side and symmetrical with respect to the first position and the belt-shaped plate. In the pressure measuring means, both the pressure fluctuation due to the vibration of the excited strip and the pressure fluctuation due to external noise are measured. However, the pressure fluctuation due to the vibration of the plate is substantially opposite in phase between the first position and the third position, whereas the pressure fluctuation due to external noise is substantially the same at the first position and the third position. It becomes a phase. Therefore, in the second step, the difference between the two pressure fluctuation data respectively measured by the two pressure measuring means arranged at the first position and the third position is taken, so that the vibration of the plate having a substantially opposite phase is obtained. The pressure fluctuation data due to the external noise becomes large (the peak doubles), while the pressure fluctuation data due to the external noise of approximately the same phase becomes small and is almost eliminated, so the SN ratio becomes large and the natural frequency detection accuracy is improved. Get higher.

  In the first or second invention, the tension measuring method of the third invention measures the pressure fluctuation before the vibration and the pressure fluctuation after the vibration by the two pressure measuring means in the first step, respectively. Then, in the second step, the difference between the two pressure fluctuation data measured by the two pressure measuring means is taken, and the difference data is divided into pre-vibration data and post-vibration data. The frequency analysis is performed, and the natural frequency of the strip plate is extracted from the difference or ratio between the data obtained by frequency analysis of the data before and after the vibration. .

  Prior to the excitation of the strip plate, in addition to the external force that is forcibly applied to the plate to excite the natural frequency component (hereinafter referred to as forced excitation force), various external forces that are applied to the plate (hereinafter referred to as external excitation). The pressure fluctuation caused by the vibration due to the vibration force) is measured by two pressure measuring means. On the other hand, after the strip plate is vibrated, both the pressure fluctuation caused by the forced vibration force and the pressure fluctuation caused by the external vibration force are measured by two pressure measuring means. Therefore, the difference data between the two pressure fluctuation data measured by the two pressure measuring means is divided into pre-vibration data and post-vibration data, and frequency analysis is performed for each, and the pre-vibration data and added data are analyzed. By taking the difference or ratio of the data obtained by frequency analysis of the data after vibration, the effect of pressure fluctuation due to external excitation force can be removed from the frequency analysis results, and the natural frequency of the strip plate Can be easily extracted.

  In the tension measurement method according to a fourth aspect of the present invention, in any one of the first to third aspects, the excitation position of the belt-like plate is a central position of a portion supported at two support positions in the longitudinal direction. It is characterized by this. In this way, by vibrating the belt-like plate at the center in the longitudinal direction of the portions supported at two locations, the natural vibration of the primary mode in the longitudinal direction is easily excited. Further, by exciting the belt-like plate at the center in the width direction, the natural vibration of the primary mode is easily excited at both ends in the width direction.

  The tension measurement method according to a fifth aspect of the present invention is the method according to any one of the first to fourth aspects, wherein the belt-like plate is vibrated at two timings at two excitation positions at both ends in the width direction, and the second In the process, by analyzing the frequency of the portions corresponding to the excitation at the two excitation positions of the difference data between the two pressure fluctuation data measured by the two pressure measuring means, the width direction of the strip plate The characteristic frequencies in the longitudinal direction are extracted at both ends, respectively. Thus, the required natural vibration mode in both ends can be excited by exciting the belt-like plate at the two excitation positions at both ends in the width direction at different timings. Therefore, even when the tension at both ends in the width direction of the plate is in a different tension state, the natural frequency component can be greatly excited by exciting one end and the natural vibration at both ends in the width direction. Numbers can be extracted easily.

  In the tension measuring method according to a sixth aspect of the present invention, in any one of the first to fifth aspects of the invention, the first position is the one of the center portions of the portions supported at the two support positions in the longitudinal direction. It is a position on the surface side. For example, in a portion supported by two rolls or the like, the vibration of the belt-like plate is the largest at the central portion thereof, so that the air pressure fluctuation due to the vibration of this plate is also large. Therefore, a large pressure fluctuation can be measured by arranging the pressure measuring means at a position on one surface side of the central portion, so it is easier to extract the natural frequency based on the pressure fluctuation data. become.

  A first embodiment of the present invention will be described. This 1st Embodiment is an example which applied this invention, when measuring the tension | tensile_strength of a metal plate in the line which conveys a strip | belt-shaped metal plate downstream via several rolls.

  FIG. 1 is a schematic configuration diagram of a line according to the first embodiment. The band-shaped metal plate 1 is fed in the left-right direction via a plurality of rolls 2 and is wound up in a coil shape by a take-up reel 3. As shown in FIG. 1 and FIG. 2, the position is spaced from the metal plate 1, and the position P <b> 1 (first position) on the upper surface side of the metal plate 1 and the position P <b> 2 in the horizontal plane including the metal plate 1 (second The two low-frequency sound level meters 4 and 5 are respectively arranged at the position (1). Further, above the metal plate 1, an air nozzle 6 that ejects compressed air that vibrates the metal plate 1 is disposed with its ejection port facing the metal plate 1.

  Air pressure fluctuation data Pa and Pb respectively measured by the two low frequency sound level meters 4 and 5 at the two positions P1 and P2 are amplified by the amplifiers 7 and 8 and then input to the calculator 9. In the calculator 9, difference data ΔP between the two pressure fluctuation data Pa and Pb is calculated and the output thereof is input to the Fourier transform device 10. In the Fourier transform device 10, difference data ΔP between the two pressure fluctuation data Pa and Pb is Fourier transformed to perform frequency analysis, and the frequency analysis result is displayed on the display unit 11.

Next, the tension measuring method for the metal plate 1 in FIG. 1 will be described in more detail with reference to the flowchart in FIG. In FIG. 3, Si (i = 10, 11, 12,...) Indicates each step.
First, the metal plate 1 is vibrated by jetting compressed air from the air nozzle 6 to the metal plate 1 (S10). At this time, it is preferable to vibrate the metal plate 1 by applying a pulsed excitation wave with compressed air. When a pulsed excitation wave is applied, various frequency components are likely to be excited on the metal plate, and therefore excitation of the natural frequency component is expected. Then, the pressure fluctuations Pa and Pb at the positions P1 and P2 are respectively measured by the two low-frequency sound level meters 4 and 5 (S11: first step).

  Here, as shown in FIG. 4, in the low frequency sound level meter 4 arranged at the position P1 above the metal plate 1, both the pressure fluctuation V1 due to vibration of the metal plate 1 and the pressure fluctuation V2 due to external noise are measured. On the other hand, at the position P2 at the same height as the metal plate 1, the pressure fluctuation V1 propagated above the metal plate 1 and the pressure fluctuation V1 propagated downward are opposite to each other and cancel each other. Therefore, in the low frequency sound level meter 5 arranged at the position P2, the pressure fluctuation V2 due to the external noise is measured, but the pressure fluctuation V1 due to the vibration of the metal plate 1 is hardly measured. Further, since the position P1 and the position P2 are arranged close to each other, the pressure fluctuation due to the external noise measured by the low frequency sound level meter 4 at the position P1 and the low frequency level sound level meter 5 at the position P2 is almost the same. It will be a thing.

  The pressure fluctuation data Pa and Pb measured by the two low-frequency sound level meters 4 and 5 are amplified by the amplifiers 7 and 8 and then input to the computing unit 9, and the computing unit 9 makes a difference ΔP (= Pa-Pb) is calculated (S12). Thus, by taking the difference between the pressure fluctuation Pa at the position P1 and the pressure fluctuation Pb at the position P2, the influence of the pressure fluctuation V2 due to the external noise included in both Pa and Pb is removed. Then, the difference ΔP of the pressure fluctuation data is input to the Fourier transform apparatus 10 and the frequency analysis is performed by Fourier transforming ΔP in the Fourier transform apparatus 10 (S13).

  In this frequency analysis result of ΔP, a frequency (hereinafter referred to as a peak frequency) having a larger peak than the other frequency region and having a large peak is extracted (S14). This peak frequency is the natural frequency of the metal plate 1. Here, as shown in FIG. 5 (a), when the tensions at both ends in the width direction (A side, B side) of the metal plate 1 are substantially equal, only one peak frequency appears. On the other hand, as shown in FIGS. 5B and 5C, in the state where the tension on the A side and the tension on the B side are different (bias tension state), the natural vibrations at the A side end and the B side end, respectively. Two peak frequencies appear, and the larger the difference in tension at both ends, the greater the difference between the two peak frequencies. In addition, S12-S14 demonstrated above corresponds to the 2nd process of this invention.

And tension T is calculated | required from the extracted natural frequency (S15-S17: 3rd process). First, when the metal plate 1 is not in a biased tension state and has one peak frequency (S15: Yes), the tension T per unit area is
T = 4 × ρ × L ² × f 1 ²
(S16). Here, f1 is the peak frequency, ρ is the density of the metal plate 1, and L is the length of the metal plate 1.

On the other hand, when the metal plate 1 is in a state of eccentric tension and there are two peak frequencies (S15: No), the tension T per unit area is calculated from the two peak frequencies f1 and f2, respectively. As an average value,
T = 4 × ρ × L ² × (f1 ² + f2 ² ) / 2
(S17).

  The upper position P1 of the metal plate 1 is preferably the upper position of the central portion of the portion supported by the two rolls 2. In such a position, the pressure fluctuation generated by the vibration of the metal plate 1 becomes large, and the peak frequency (that is, the natural frequency) can be easily extracted.

  Moreover, it is preferable that the position to be vibrated by the compressed air from the air nozzle 6 is the position of the central portion of the portion supported by the two rolls 2. The reason is that the natural vibration of the first-order mode in the longitudinal direction is easily excited by exciting the metal plate 1 in the central portion in the longitudinal direction, and further, the metal plate 1 is vibrated in the central portion in the width direction. This is because the natural vibration of the primary mode is likely to be excited at both ends in the width direction (A side and B side).

Next, specific examples of the tension measuring method according to the first embodiment will be described.
As the metal plate 1, a strip-shaped copper plate having a plate width of 615 mm, a length of 2200 mm and a plate thickness of 0.31 mm is used, and a pulse wave of about 1 second is given once to supply compressed air from the air nozzle 6 to the center of the copper plate. The copper plate was vibrated by spraying. FIG. 6 shows the measurement result of the amplitude of the copper plate by a laser displacement meter as a comparative example, FIG. 7 shows the pressure fluctuation data Pa measured by the low frequency sound level meter 4 arranged at the position P1 above the copper plate, and FIG. Pressure fluctuation data Pb measured by the low-frequency sound level meter 5 disposed at the same height P2 as the copper plate, FIG. 9 is the difference ΔP between Pa in FIG. 7 and Pb in FIG. 8, and FIG. 10 is the laser displacement. FIG. 11 shows the frequency analysis results for 10 seconds after excitation of the measurement results (FIG. 9) of the low-frequency sound level meters 4 and 5, respectively. Show.

  The vibration waveform of the copper plate measured by the laser displacement meter in FIG. 6 and the pressure fluctuation waveform at the position P1 above the copper plate measured by one low frequency sound level meter 4 in FIG. It can be seen that the measured value of the frequency sound level meter 4 is greatly affected by pressure fluctuations due to external noise. Moreover, as shown in FIG.7 and FIG.8, in the two low frequency sound level meters 4 and 5, it turns out that the pressure fluctuation by substantially the same external noise is measured.

  6 and 9, it can be seen that by taking the difference ΔP between the measured values Pa and Pb of the two low-frequency sound level meters 4 and 5, pressure fluctuation due to external noise can be removed and pressure fluctuation due to copper plate vibration can be extracted. . Further, as shown in FIG. 10 and FIG. 11, the peak frequency (f1, f2) appears at substantially the same position in the frequency analysis result of the amplitude value by the laser displacement meter and the frequency analysis result of the pressure fluctuation difference ΔP. It can be seen that the natural frequency of the copper plate can be accurately extracted in the tension measurement method of the first embodiment. In FIGS. 10 and 11, two peak frequencies (f1 = 7.4 Hz and f2 = 9.4 Hz) appear because the copper plate is in a biased tension state.

From the results of the above examples, when the tension T per unit area is obtained,
T = 4 × ρ × L ² × (f1 ² + f2 ² ) /2=12.0×10 ⁶ (N / m ² )
It becomes. Here, ρ = 8650 (kg / m ³ ), L = 2.2 (m), f1 = 7.4 (Hz), and f2 = 9.4 (Hz).

  According to the tension measuring method of the first embodiment described above, it is measured by the low frequency sound level meter 4 at the position P1 above the metal plate 1, and includes pressure fluctuation due to vibration of the metal plate 1 and pressure fluctuation due to external noise. The pressure fluctuation is obtained by taking the difference ΔP between the pressure fluctuation data Pa and the pressure fluctuation data Pb measured by the low frequency sound level meter 5 at the position P2 at the same height as the metal plate 1 and containing almost no pressure fluctuation due to the vibration of the metal plate 1. Since the frequency analysis is performed after removing the pressure fluctuation due to the external noise from the data, the peak of the natural frequency component appears clearly, and the natural frequency of the metal plate 1 can be extracted easily and reliably.

Next, a second embodiment of the present invention will be described. However, components having the same configuration as in the first embodiment are denoted by the same reference numerals and description thereof is omitted as appropriate.
According to the tension measurement method of the second embodiment, it is possible to remove the pressure fluctuation due to external noise from the pressure fluctuation data as in the tension measurement method of the first embodiment, and further vibrate the metal plate. Therefore, it is possible to remove the influence of various external forces (external vibration forces) acting on the metal plate other than the external force (forced vibration force) applied for the purpose.

  FIG. 12 is a schematic configuration diagram of a line according to the second embodiment. The configuration of this line is substantially the same as that of the first embodiment described above, and two low-frequency sound level meters are respectively provided at a position P1 on the upper surface side of the metal plate 1 and a position P2 in the horizontal plane including the metal plate 1. 4, 5 are arranged. An air nozzle 6 that ejects compressed air that vibrates the metal plate 1 is disposed above the metal plate 1.

  By the way, in this 2nd Embodiment, the two low frequency sound level meters 4 and 5 measure both the data before vibrating the metal plate 1 and after vibrating. Similarly to the first embodiment, after calculating the difference data ΔP between the two pressure fluctuation data Pa and Pb in the calculator 9, the difference data ΔP in the pressure fluctuation is calculated in the data dividing device 20. The data is divided into data ΔP1 before vibration and data ΔP2 after vibration. The pre-vibration data ΔP1 and post-vibration data ΔP2 are each subjected to Fourier transform in the Fourier transform device 21 and subjected to frequency analysis. Further, the calculator 22 calculates the difference between the data obtained by frequency analysis of the pre-vibration data ΔP1 and post-vibration data ΔP2 of the pressure fluctuation difference, and the result is output to the display unit 11.

Next, with reference to the flowchart of FIG. 13, the tension measuring method of the metal plate 1 in FIG. 12 will be described in more detail.
First, measurement of the air pressure fluctuations Pa and Pb at the positions P1 and P2 is started by the low-frequency sound level meters 4 and 5 (S20). The metal plate 1 is vibrated by spraying compressed air on the surface (S21). And after measuring the pressure fluctuation after vibration for a certain period of time, the measurement is terminated (S22).

  Next, the pressure fluctuation data Pa and Pb measured by the two low-frequency sound level meters 4 and 5 are amplified by the amplifiers 7 and 8 and then input to the calculator 9, and the calculator 9 calculates the difference ΔP (= Pa-Pb) is calculated (S23). Thus, by taking the difference between the pressure fluctuation Pa at the position P1 and the pressure fluctuation Pb at the position P2, pressure fluctuation due to external noise included in both Pa and Pb is removed.

  Further, the data dividing device 20 divides the pressure fluctuation difference data ΔP into data ΔP1 before vibration and data ΔP2 after vibration (S24). Here, the data ΔP1 before vibration is data of the difference in pressure fluctuations generated by various external vibration forces other than the forced vibration force acting on the metal plate 1 by the compressed air that is injected. The post-vibration data ΔP2 is data on the difference in pressure fluctuation generated by the forced excitation force and the external excitation force.

  Then, the divided pre-vibration data ΔP1 and post-vibration data ΔP2 are input to the Fourier transform apparatus 10, and frequency analysis is performed by Fourier transforming ΔP1 and ΔP2 in the Fourier transform apparatus 10 (S25). Further, the computing unit 22 calculates a difference between ΔF1 which is the frequency analysis result of ΔP1 and ΔF2 which is the frequency analysis result of ΔP2 (S26). As described above, the difference between the frequency analysis result ΔF1 of the pre-vibration data ΔP1 and the frequency analysis result ΔF2 of the post-vibration data ΔP2 is included in both the pre-vibration data ΔP1 and the post-vibration data ΔP2. The influence of the external excitation force can be removed from the frequency analysis result.

  Then, in this frequency analysis result ΔF, a peak frequency with a larger amplitude than the other frequency regions is extracted (S27). This extracted peak frequency becomes the natural frequency of the metal plate 1. As in the first embodiment, when the tension at both ends in the width direction of the metal plate 1 is substantially equal, only one peak frequency appears. On the other hand, in the biased tension state, the natural vibrations at both ends of the metal plate 1 appear. Two peak frequencies that are numbers appear.

Finally, the tension T is obtained from the extracted natural frequency (S28 to S30). That is, when the metal plate 1 is not in a biased tension state and the peak frequency is one (S28: Yes), the tension T per unit area is expressed as follows:
T = 4 × ρ × L ² × f 1 ²
(S29). Here, f1 is a peak frequency.

On the other hand, when the metal plate 1 is in an eccentric tension state and there are two peak frequencies (S28: No), the tension T per unit area is calculated from the two peak frequencies f1 and f2. As an average value,
T = 4 × ρ × L ² × (f1 ² + f2 ² ) / 2
(S30).

  Next, specific examples of the tension measuring method according to the second embodiment will be described. In this example, as in the example of the first embodiment, a copper plate was used as the metal plate 1, and compressed air was jetted from the air nozzle 6 to vibrate the central portion of the copper plate. Furthermore, disturbances of 5 Hz and 10 Hz were applied to the copper plate as external excitation forces other than the forced excitation force caused by the compressed air. The results are shown in FIGS. Here, FIG. 14 shows the difference ΔP between the pressure fluctuations Pa and Pb measured by the low frequency sound level meters 4 and 5, and FIG. 15 shows the frequency analysis result ΔF2 of the data for 10 seconds after the excitation of ΔP shown in FIG. FIG. 16 shows the frequency analysis result ΔF1 of the data for 10 seconds before the excitation of ΔP shown in FIG. 14, and FIG. 17 shows the difference ΔF between ΔF2 and ΔF1, respectively.

  As shown in FIG. 15, when frequency analysis is performed on post-excitation data ΔP2 including pressure fluctuation due to an external excitation force, in addition to the natural frequencies (7.4 Hz and 9.4 Hz) at both ends in the width direction, 5 Hz and 10 Hz due to disturbances are obtained. Therefore, the natural frequency cannot be determined only by this result. On the other hand, as shown in FIG. 16, in the frequency analysis result ΔF1 of the pre-vibration data ΔP1, peaks at 5 Hz and 10 Hz due to disturbance appear. Therefore, by taking the difference between the frequency analysis result ΔF1 of the pre-vibration data ΔP1 and the frequency analysis result ΔF2 of the post-vibration data ΔP2, the peaks of 5 Hz and 10 Hz are removed as shown in FIG. Only peak frequencies of several components (7.4 Hz and 9.4 Hz) appear.

  According to the tension measurement method of the second embodiment described above, the low frequency sound level meter 4 at the position P1 above the metal plate 1 and the low frequency sound level meter 5 at the position P2 at the same height as the metal plate 1 The pressure fluctuation Pa before the vibration and the pressure fluctuation Pb after the vibration are respectively measured and the difference ΔP is obtained, and the frequency analysis result ΔF1 of the pre-vibration data ΔP1 of the pressure fluctuation difference ΔP and the vibration after the vibration are obtained. The difference from the frequency analysis result ΔF2 of the data ΔP2 is taken. As a result, as in the first embodiment, the external noise can be removed from the pressure fluctuation data, and in addition, the influence of the external vibration force other than the forced vibration force can be removed. The natural frequency can be easily and reliably extracted.

  In the second embodiment, by taking the difference between the frequency analysis result ΔF1 of the pre-vibration data ΔP1 and the frequency analysis result ΔF2 of the post-vibration data ΔP2, the influence of the external excitation force is determined from the frequency analysis result. Although removed, the influence of the external excitation force may be removed by taking the ratio of the frequency analysis result ΔF1 of the pre-vibration data ΔP1 and the frequency analysis result ΔF2 of the post-vibration data ΔP2.

Next, a third embodiment of the present invention will be described. However, components having the same configuration as those of the first and second embodiments are denoted by the same reference numerals, and description thereof is omitted as appropriate.
According to the tension measurement method of the third embodiment, as in the tension measurement method of the first embodiment, pressure fluctuations due to external noise can be removed from the pressure fluctuation data, and furthermore, the same as the tension measurement method of the second embodiment. In addition, the influence of an external excitation force other than the forced excitation force can be removed. In addition, in this tension measuring method, the both ends of the metal plate in the width direction can be excited at different timings to excite the necessary natural vibration modes at the both ends of the metal plate. As shown in FIG. 18, the configuration of the line in the third embodiment is substantially the same as that of the second embodiment, but the air nozzle 6 has both end portions (A side end portion and B side end portion) of the metal plate 1. Are different from each other in that one is provided above each. Hereinafter, with reference to the flowchart of FIG. 19, the tension | tensile_strength measuring method of the metal plate 1 of 3rd Embodiment is demonstrated in detail.

  First, measurement of the pressure fluctuations Pa and Pb at the positions P1 and P2 is started by the low frequency sound level meters 4 and 5 (S40), and the pressure fluctuation before the vibration is measured for a certain time. And the A side edge part of the metal plate 1 is vibrated by injecting compressed air with respect to the metal plate 1 from the air nozzle 6 above an A side edge part. Next, after a predetermined time has elapsed after the vibration of the A side end, the B side end is similarly vibrated (S41). Here, it is preferable that the B side end portion is vibrated after the pressure vibration at the positions P1 and P2 due to the vibration at the A side end portion is sufficiently converged. In this way, the both end portions A and B of the metal plate 1 are vibrated at different timings to excite the natural vibration of the first-order mode in the longitudinal direction at the both end portions A and B, respectively. In addition, in order to surely excite the natural vibration of the primary mode at both ends of the metal plate 1, it is preferable to vibrate the longitudinal central portions of the A side end and the B side end, respectively. And after measuring the pressure fluctuation after vibrating the B side edge part for a fixed time, a measurement is complete | finished (S42).

  Next, in order to remove pressure fluctuation due to external noise from the pressure fluctuation data Pa and Pb, the pressure fluctuation data Pa and Pb measured by the two low-frequency sound level meters 4 and 5 are amplified by the amplifiers 7 and 8 and then calculated. The difference 9P of pressure fluctuation is calculated by the arithmetic unit 9 (S43). Here, the measurement of the pressure fluctuation data Pa and Pb by the low frequency sound level meters 4 and 5 is continuously performed from before the vibration of the A side end portion to after the vibration of the B side end portion. The difference data ΔP includes data of pressure fluctuation before and after excitation at the A side end and pressure fluctuation before and after excitation at the B side end.

  Then, for each of the vibration at the A-side end and the vibration at the B-side end, the data ΔP of the difference between the pressure fluctuation data Pa and Pb is obtained before the vibration, as in the second embodiment. The frequency analysis is performed on each of the data and the data after vibration, and the difference between the frequency analysis result before vibration and the frequency analysis result after vibration is obtained.

  That is, in the data dividing device 20, the pressure fluctuation difference data ΔP is converted into the pre-vibration data ΔPA1 at the A side end, the pre-vibration data ΔPB1 at the B side end, the post-vibration data ΔPA2 at the A side end, And it divides | segments into after-vibration data (DELTA) PB2 of the B side edge part, respectively (S44). Then, the divided pre-excitation data ΔPA1, ΔPB1 of the A-side end and B-side end, and post-excitation data ΔPA2, ΔPB2 of the A-side end and B-side end are input to the Fourier transform apparatus 10, respectively. Then, frequency analysis is performed by Fourier-transforming ΔPA1, ΔPB1, ΔPA2, and ΔPB2 in the Fourier transform apparatus 10 (S45). Further, the calculator 22 calculates a difference ΔFA between ΔFA1 which is the frequency analysis result of ΔPA1 and ΔFA2 which is the frequency analysis result of ΔPA2, and further, ΔFB1 which is the frequency analysis result of ΔPB1 and the frequency analysis result of ΔPB2. A difference from ΔFB2 is calculated (S46).

  Then, in the frequency analysis result ΔFA when the A side end portion is vibrated and the frequency result ΔFB when the B side end portion is vibrated, the peak frequencies that are larger than the other frequency regions are extracted. (S47). These extracted peak frequencies are the natural frequency f1 of the primary mode at the A-side end of the metal plate 1 and the natural frequency f2 of the primary mode at the B-side end, respectively. In addition, when the tension | tensile_strength in the width direction both ends of the metal plate 1 is substantially equal, f1 and f2 become substantially equal.

Finally, the tension T is obtained from the extracted natural frequencies f1 and f2 (S48 to S50). When the metal plate 1 is not in a biased tension state and the peak frequencies f1 and f2 are equal (S48: Yes), the tension T per unit area is
T = 4 × ρ × L ² × f 1 ²
(S49).

On the other hand, when the metal plate 1 is in a state of eccentric tension and the peak frequencies f1 and f2 are not equal (S48: No), the tension T per unit area is calculated from the two peak frequencies f1 and f2, respectively. As an average value,
T = 4 × ρ × L ² × (f1 ² + f2 ² ) / 2
(S50).

  Next, specific examples of the tension measuring method of the third embodiment will be described. In this embodiment, compressed air is jetted from the air nozzle 6 disposed above the A side end of the copper plate as the metal plate 1, and the central portion between the two rolls 2 at the A side end is vibrated to be constant. After a lapse of time, compressed air was jetted from the air nozzle 6 disposed above the B side end, and the central part between the two rolls 2 at the B side end was vibrated.

  Then, a difference ΔP between the pressure fluctuation data Pa and Pb measured by the two low-frequency sound level meters 4 and 5 is obtained, and further, this difference data ΔP is obtained as data ΔPA1 for 10 seconds before the vibration on the A side and B side. , ΔPB1, and data ΔPA2 and ΔPB2 for 10 seconds after excitation on the A side and B side, and frequency analysis was performed for each. FIG. 20 shows the difference ΔFA between the frequency analysis data ΔFA1 of ΔPA1 at the A side end and the frequency analysis data ΔFA2 of ΔPA2, and FIG. 21 shows the frequency analysis data ΔFB2 of the frequency analysis data ΔFB1 and ΔPB2 of ΔPB1 at the B side end. The difference ΔFB is shown.

  As shown in FIG. 20, when the end on the A side is vibrated, the component of the natural frequency f1 (7.4 Hz) in the longitudinal mode on the A side is excited. On the other hand, as shown in FIG. 21, when the B-side end is vibrated, the component of the natural frequency f2 (9.4 Hz) of the B-side longitudinal mode is excited.

  According to the tension measurement method of the third embodiment described above, the metal plate 1 is vibrated at two timings at two vibration positions at both ends in the width direction, and the difference ΔP between the two pressure fluctuation data Pa and Pb is obtained. The required natural vibration modes at both ends can be excited by frequency analysis of the portions corresponding to the vibration at the two vibration positions. Therefore, even when the tension at both ends in the width direction of the metal plate 1 is in a different tension state and exhibits a complex natural frequency characteristic, the natural frequency component is greatly excited by exciting one end. Thus, the natural frequencies at both ends in the width direction can be easily extracted.

Next, modified embodiments in which various modifications are made to the tension measuring methods of the first to third embodiments will be described.
1] The position P1 of the low frequency sound level meter 4 for measuring pressure fluctuation data Pa including pressure fluctuation caused by vibration of the metal plate 1 and pressure fluctuation caused by external noise may be below the metal plate 1.

  2] As shown in FIGS. 22 and 23, one low frequency sound level meter 4 is disposed at a position P1 above the metal plate 1, and the other low frequency sound level meter 5 is disposed below the metal plate 1, and You may arrange | position in the position P3 (3rd position) symmetrical with respect to the position P1 and the metal plate 1. FIG. In this case, as shown in FIG. 23, the two low-frequency sound level meters 4 and 5 measure both the pressure fluctuation V1 due to vibration of the vibrated metal plate 1 and the pressure fluctuation V2 due to external noise. However, the pressure fluctuation V1 due to the vibration of the metal plate 1 is substantially opposite in phase at the positions P1 and P3 symmetrical with respect to the metal plate 1, whereas the pressure fluctuation V2 due to external noise is the two positions P1 and P1. At P3, they are substantially in phase with each other. Therefore, in the same manner as in the first to third embodiments, if the difference between the two pressure fluctuation data measured by the two low-frequency sound level meters 4 and 5 respectively disposed at symmetrical positions P1 and P3 is taken. While the data of the pressure fluctuation V1 due to the vibration of the metal plate 1 in the opposite phase becomes large (the peak doubles), the pressure fluctuation V2 due to the external noise can be removed, so the SN ratio becomes large and the natural vibration Number detection accuracy increases.

  Further, as in the second embodiment described above, the pressure fluctuation Pa before the vibration and the pressure after the vibration are generated by the low frequency sound level meter 4 disposed at the position P1 and the low frequency sound level meter 5 disposed at the position P3. Each of the fluctuations Pb is measured and the difference ΔP is taken. Among the pressure fluctuation differences ΔP, the difference between the frequency analysis result ΔF1 of the pre-vibration data ΔP1 and the frequency analysis result ΔF2 of the post-vibration data ΔP2 is taken. Thus, the influence of an external excitation force other than the forced excitation force may be removed. Furthermore, in addition to this, as in the third embodiment described above, the metal plate 1 is vibrated at two timings at the two vibration positions at both ends in the width direction, and the two pressure fluctuation data Pa and Pb are It is also possible to excite required natural vibration modes at both ends by analyzing the frequency of the portions corresponding to the vibration at the two vibration positions of the difference ΔP.

  3] As described above, when the metal plate 1 is vibrated by the air nozzle 6, the mode (usually the primary mode) that becomes a belly at the center in the longitudinal direction of the metal plate 1 supported by the two rolls 2 is provided. In order to facilitate excitation, it is preferable to set the excitation position at the center in the longitudinal direction of the plate. However, even in a vibration mode that becomes a node at the center such as the secondary mode, the natural frequency is measured. Therefore, it is not always necessary to set the excitation position at the center position.

  4] As a pressure measuring means for measuring the pressure fluctuation due to the vibration of the metal plate 1, in addition to the above-mentioned low frequency sound level meters 4 and 5, it is possible to detect the atmospheric pressure fluctuation due to the vibration of the metal plate 1 having a low frequency. A simple pressure sensor can also be used.

5] Instead of compressed air, the metal plate may be vibrated by spraying other fluid such as nitrogen gas or water onto the metal plate 1. Further, the metal plate 1 may be vibrated by directly striking the metal plate 1.
6] The belt-like plate to which the present invention can be applied is not limited to the metal plate 1, and the present invention is also applicable to belt-shaped belts of various materials such as paper, cloth, synthetic resin, and rubber. Can be applied to measure the tension.

It is a schematic block diagram of the board conveyance line which concerns on 1st Embodiment of this invention. It is the figure which looked at FIG. 1 from the longitudinal direction of the board. It is a flowchart of the tension | tensile_strength measuring method of 1st Embodiment. It is explanatory drawing regarding propagation of a pressure fluctuation. It is explanatory drawing of the vibration mode of a metal plate, (a) is a state where both ends in the width direction vibrate in the primary mode, (b) is in a biased tension state, and the A side end is in the primary mode. A vibrating state, (c) shows a state of a bias tension state and a state where the B-side end portion vibrates in the primary mode. It is a figure which shows the plate vibration measurement result by the laser displacement meter in the Example of 1st Embodiment. It is a figure which shows the measurement result of the pressure fluctuation Pa. It is a figure which shows the measurement result of the pressure fluctuation Pb. It is a figure which shows difference (DELTA) P of a pressure fluctuation. It is a figure which shows the frequency analysis result of the plate vibration measurement data by a laser displacement meter. It is a figure which shows the frequency analysis result of difference (DELTA) P of pressure fluctuation. It is a schematic block diagram of the board conveyance line which concerns on 2nd Embodiment of this invention. It is a flowchart of the tension | tensile_strength measuring method of 2nd Embodiment. It is a figure which shows the difference (DELTA) P of the pressure fluctuation in the Example of 2nd Embodiment. It is a figure which shows the frequency analysis result (DELTA) F2 of the difference (DELTA) P2 of the pressure fluctuation after vibration. It is a figure which shows the frequency analysis result (DELTA) F1 of the difference (DELTA) P1 of the pressure fluctuation before vibration. It is a figure which shows difference (DELTA) F of a frequency analysis result. It is a schematic block diagram of the board conveyance line which concerns on 3rd Embodiment of this invention. It is a flowchart of the tension | tensile_strength measuring method of 3rd Embodiment. It is a figure which shows difference (DELTA) FA of the frequency-analysis result of the data (DELTA) PA1 before vibration of the difference of the pressure fluctuation at the time of A side edge part vibration, and the data (DELTA) PA2 after vibration. It is a figure which shows the difference (DELTA) FB of the frequency analysis result of the data (DELTA) PB1 before vibration of the difference of the pressure fluctuation at the time of B side edge part vibration, and the data (DELTA) PB2 after vibration. It is a schematic block diagram of the board conveyance line in a change form. It is explanatory drawing regarding propagation of the pressure fluctuation in a change form.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Metal plate 4, 5 Low frequency sound level meter 6 Air nozzle 9 Calculator 10 Fourier transform device 20 Data division device 21 Fourier transform device 22 Calculator

Claims (6)

Air generated by vibration of the belt-like plate by two pressure measuring means arranged at a first position on one surface side of the belt-like plate to be vibrated and a second position on the surface including the belt-like plate. A first step of measuring the pressure fluctuation of
By taking the difference between the two pressure fluctuation data at the first position and the second position measured by the two pressure measuring means and analyzing the frequency of the difference data, the natural frequency of the belt-like plate is obtained. A second step of extracting;
A third step of determining the tension of the strip plate based on the extracted natural frequency;
A method for measuring the tension of a belt-like plate, comprising: Two of the first position on one surface side of the belt-like plate to be vibrated and the other surface side of the belt-like plate and the third position symmetrical with respect to the first position and the belt-like plate, respectively. A first step of measuring a pressure fluctuation of air generated by vibration of the belt-like plate by a pressure measuring means;
By taking the difference between the two pressure fluctuation data at the first position and the third position measured by the two pressure measuring means, and analyzing the frequency of the difference data, the natural frequency of the belt-like plate is obtained. A second step of extracting;
A third step of determining the tension of the strip plate based on the extracted natural frequency;
A method for measuring the tension of a belt-like plate, comprising: In the first step, the pressure fluctuation before vibration and the pressure fluctuation after vibration are measured by the two pressure measuring means, respectively.
In the second step, the difference between the two pressure fluctuation data measured by the two pressure measuring means is taken, and the difference data is divided into pre-vibration data and post-vibration data, and frequency analysis is performed for each. The natural frequency of the strip-shaped plate is extracted from the difference or ratio between the data obtained by frequency analysis of the data before vibration and the data after vibration, respectively. 2. A method for measuring the tension of the belt-shaped plate according to 1.   The method for measuring the tension of a belt-like plate according to any one of claims 1 to 3, wherein the excitation position of the belt-like plate is a central position of a portion supported at two support positions in the longitudinal direction. The belt-like plate is vibrated at different timings at two vibration positions at both ends in the width direction,
In the second step, by analyzing the frequency of the portions corresponding to the excitation at the two excitation positions in the difference data between the two pressure fluctuation data measured by the two pressure measuring means, 5. The method for measuring the tension of a strip-shaped plate according to claim 1, wherein the natural frequency in the longitudinal direction is extracted at both ends in the width direction.   The belt-like shape according to any one of claims 1 to 5, wherein the first position is a position on the one surface side of a central portion of a portion supported at two support positions in the longitudinal direction. A method for measuring the tension of a plate.

Images