JP2017037052A - Vibration detection device, vibration detection method and vibration detection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vibration detection device, a vibration detection method and a vibration detection program capable of improving the accuracy in vibration detection using an acceleration sensor.SOLUTION: The vibration detection device comprises: a contactor which is in contact with a vibration object, and is preset with a contact resonance frequency band; an acceleration sensor provided to the contactor; and a detection section that detects a signal by reducing the strength of the signal which is amplified in the contact resonance frequency band in the signals output from the acceleration sensor.SELECTED DRAWING: Figure 3

Description

  The present case relates to a vibration detection device, a vibration detection method, and a vibration detection program.

  In the abnormality diagnosis of equipment equipped with a mechanical mechanism such as a rotating machine, vibration diagnosis is used to determine whether abnormality is possible or not by a signal obtained by detecting vibration in the vicinity of a point where the frequency of failure such as a bearing is high.

  For example, as the vibration detection device, an acceleration sensor may be installed at a portion that contacts the vibrating object, and a sensor jig may be directly screwed into the vibrating object and firmly fixed. By increasing the contact resonance frequency, it is possible to have flat frequency characteristics in a wide range of frequency bands to be detected (for example, from 10 Hz to several tens of kHz).

  Other fixing methods include using a magnet or an adhesive to make contact between vibrating objects, and forming a moving part in contact with the vibrating object with a shape having high contact resonance frequency characteristics (see Patent Documents 1 and 2). Various things are used according to the environment and circumstances.

JP 2004-125676 A Japanese Patent Laid-Open No. 11-64093

  However, frequency characteristics fluctuate in the high frequency band, and signal strength is emphasized in some frequency bands, so that erroneous detection occurs.

  In one aspect, an object of the present invention is to provide a vibration detection device, a vibration detection method, and a vibration detection program that can improve the accuracy of vibration detection using an acceleration sensor.

  In one aspect, the vibration detecting device is in contact with a vibrating object, a contact body in which a contact resonance frequency band is set, an acceleration sensor provided in the contact body, and the contact resonance among output signals of the acceleration sensor. A detection unit that detects the signal by reducing the intensity of the signal amplified in the frequency band.

  The accuracy of vibration detection using the acceleration sensor can be improved.

(A)-(c) illustrates the output signal of an acceleration sensor. (A)-(c) illustrates the output signal of an acceleration sensor. 1 is a block diagram illustrating an overall configuration of a vibration detection device according to a first embodiment. (A) And (b) is a figure which illustrates a jig | tool. It is a figure which illustrates the relationship between the curvature radius of a front-end | tip part, and a contact resonance frequency. It is a flowchart showing an example of operation | movement of a vibration detection apparatus. It is a figure which illustrates the output signal of a MEMS type acceleration sensor, and the output signal of a piezoelectric type acceleration sensor. It is a block diagram for demonstrating the hardware constitutions of a calculating part and a determination part.

  Hereinafter, embodiments will be described with reference to the drawings.

  First, the outline | summary of Example 1 is demonstrated. FIG. 1A illustrates the output of an acceleration sensor attached to a vibrating object. As illustrated in FIG. 1A, if noise is not superimposed on the output, an electrical signal having a predetermined intensity is obtained for each frequency according to the vibration of the vibrating object. By using the relationship between these frequencies and signal intensity, it is possible to determine the failure of the vibrating object.

  Noise is superimposed on the output of the acceleration sensor. FIG. 1B illustrates an output on which noise is superimposed. As illustrated in FIG. 1B, it is difficult to distinguish between a signal having a strength smaller than the noise strength and the noise, and thus it is difficult to separate the signal and the noise. Since the output intensity of the acceleration sensor tends to decrease at a high frequency, it is particularly difficult to separate a signal and noise at a high frequency.

  Therefore, the contact resonance frequency band is adjusted by adjusting the contact mode between the acceleration sensor and the vibrating object. By adjusting the contact resonance frequency band so as to include the frequency of the signal having a strength smaller than the noise strength, the strength of the signal can be amplified. In the example of FIG. 1C, the intensity of the signal buried in the noise is amplified larger than the noise intensity. Thereafter, the signal buried in the noise can be restored by removing the noise and reducing the intensity of the amplified signal. For example, by multiplying the intensity of the signal amplified in the contact resonance frequency band by the inverse function of the transfer function (frequency response function) of the contact resonance frequency band, the signal can be restored more accurately. By restoring the signal buried in the noise, the accuracy of vibration detection using the acceleration sensor can be improved.

  If a plurality of acceleration sensors having different contact resonance frequency bands are used, the frequency band that can be amplified can be widened. Note that the difference in the contact resonance frequency band indicates that the respective peaks are different, and some of the frequency bands may overlap each other. FIG. 2A illustrates an output signal before amplification of the acceleration sensor. FIG. 2B illustrates a case where two acceleration sensors having different resonance frequency bands are used. If a signal that is difficult to separate from noise is included in these resonance frequency bands, the signal can be easily detected. That is, the accuracy of vibration detection using the acceleration sensor can be further improved. FIG. 2C illustrates a signal restored by multiplying the inverse function of the transfer function of each contact resonance frequency band after amplification.

  Details of the first embodiment will be described below. FIG. 3 is a block diagram illustrating the overall configuration of the vibration detection apparatus 100 according to the first embodiment. As illustrated in FIG. 3, the vibration detection apparatus 100 includes four acceleration sensors 10 a to 10 d, a bandpass filter 20, an input reception unit 30, a calculation unit 40, a database 50, a determination unit 60, and an output unit 70. The database 50 includes a sensor sensitivity database 51, a contact resonance database 52, a theoretical abnormal frequency band database 53, and a past measurement database 54.

  The acceleration sensor 10 a and the acceleration sensor 10 b are connected to the input receiving unit 30 via the band pass filter 20. The acceleration sensor 10c and the acceleration sensor 10d are connected to the input receiving unit 30 without a band pass filter. The acceleration sensors 10a and 10b are provided on the jig 80a. The acceleration sensors 10c and 10d are provided on the jig 80b. The jigs 80a and 80b are signal amplification jigs for the acceleration sensors 10a to 10d.

  FIG. 4A is a diagram illustrating the structure of the jig 80a. As illustrated in FIG. 4A, the jig 80a includes a fixed portion 81 fixed to the adherend and a movable portion 82 movable with respect to the adherend. The fixed part 81 is disposed so as to cover the movable part 82. In addition, the fixing portion 81 uses, for example, aluminum as a base member as a material having a small specific gravity and excellent workability in order to increase the contact resonance frequency and avoid fluctuations.

  The movable part 82 is an example of a contact body that contacts the adherend. The movable portion 82 is connected to the fixed portion 81 via a vibration insulating spring 83. The movable part 82 includes a flat plate part 84. The spring 83 is connected near the center of the upper surface of the flat plate portion 84. In the vicinity of the center of the lower surface of the flat plate portion 84, a protruding portion 85 that protrudes from the adherend is provided. The protruding portion 85 is in contact with the adherend via the tip portion 86. Thus, the movable part 82 has a T-shaped cross section. The acceleration sensors 10a and 10b are disposed symmetrically on the flat plate portion 84 with the spring 83 interposed therebetween. Thereby, the inclination of the flat plate portion 84 is suppressed. As a result, the measurement accuracy of the acceleration sensors 10a and 10b is improved. The acceleration sensors 10a and 10b convert the vibration of the adherend into an electric signal. The tip end portion 86 has a radius of curvature at the contact point with the adherend. The distal end portion 86 is, for example, stainless steel that can be easily processed. The radius of curvature of the tip 86 can be changed by processing.

  The jig 80b has the same structure as the jig 80a. In the jig 80b, acceleration sensors 10c and 10d are fixed to the upper surface of the flat plate portion 84 of the movable portion 82. As illustrated in FIG. 4B, the jig 80 a and the jig 80 b are fixed to the rotating body 90. The rotating body 90 is a vibrating object that includes therein a rotor such as a motor or a bearing. The jigs 80a and 80b are not rotated because they are fixed to the casing of the rotating body 90, but receive vibration from the rotating body 90. That is, the jigs 80a and 80b receive the same vibration (synchronized vibration). For example, the jig 80a and the jig 80b are arranged at two locations so as to be symmetric at the same angle (for example, 10 °) with respect to the vertical upward direction from the central axis of the rotating body 90. Thereby, it arrange | positions so that the same vibration condition may be obtained from the rotary body 90 with respect to the perpendicular direction. Further, the pressing pressure of the movable portion 82 is adjusted in the jigs 80a and 80b.

  FIG. 5 is a diagram illustrating the relationship between the radius of curvature of the tip 86 and the contact resonance frequency band. The example of FIG. 5 shows the response (dB) to the frequency when the radius of curvature is 50 mm, 300 m, and ∞ (flat). As illustrated in FIG. 5, the contact resonance frequency band of the movable portion 82 varies by changing the radius of curvature of the distal end portion 86. With flat pressing, the resonance point is 20 kHz, and flat characteristics are obtained up to 10 kHz, and it can be seen that amplification is not possible. When the curvature radius is as small as 50 mm, the resonance point is lowered and an amplification of 10 dB is possible at 7 kHz. With a curvature radius of 300 mm, amplification of 13 dB at 15 kHz is possible. In this way, the contact resonance frequency band of the movable portion 82 can be adjusted by adjusting the radius of curvature of the distal end portion 86. Further, signal amplification of about 10 dB is possible.

  The radius of curvature of the tip portion 86 of the jig 80a is different from the radius of curvature of the tip portion 86 of the jig 80b. Thereby, the jig 80a and the jig 80b have different contact resonance frequency bands. For example, the contact resonance frequency band of the jig 80a and the contact resonance frequency band of the jig 80b are preferably set to frequency bands where the output signals of the acceleration sensors 10a to 10d are assumed to be smaller than noise. .

  Referring to FIG. 3 again, the sensor sensitivity database 51 stores sensor sensitivities (output magnification, amplification factor, etc.) of the acceleration sensors 10a to 10d as frequency response characteristics. The sensitivity of the sensor is stored as a transfer function (frequency response function). For example, you may memorize | store as a map showing the relationship between a frequency and a sensitivity. The contact resonance database 52 stores frequency response characteristics after setting the contact resonance frequency band, and stores frequency response characteristics reflecting the contact resonance frequency bands in the jigs 80a and 80b. The frequency response characteristic is stored as a transfer function (frequency response function). For example, you may memorize | store as a map showing the relationship between a frequency and an amplification factor.

  The theoretical abnormal frequency band database 53 stores the cause of failure of the rotating body 90 and the frequency band and signal intensity of signal abnormality that appears in the acceleration sensors 10a to 10d when the failure occurs. The past measurement database 54 stores measurement results related to past normal operations of the rotating body 90. It is possible to determine whether or not an abnormality has occurred by comparing the measurement result related to normal operation with the current measurement result.

  Next, details of the operation of the vibration detection apparatus 100 will be described. FIG. 6 is a flowchart illustrating an example of the operation of the vibration detection apparatus 100. As illustrated in FIG. 6, the output signal of one of the acceleration sensors 10a and 10b of the jig 80a is input to the input receiving unit 30 through the bandpass filter 20 (step S1). In addition, when using both output signals of the acceleration sensors 10a and 10b, an average of both output signals may be used. In this embodiment, as an example, the output signal of the acceleration sensor 10a is used.

  Next, the band pass filter 20 extracts the contact resonance frequency band of the jig 80a from the output signal of the acceleration sensor 10a. Next, the calculating part 40 removes noise from the output signal of the acceleration sensor 10a (step S2). Next, the computing unit 40 checks the presence or absence of a signal in the output signal after noise removal (step S3). For example, the presence or absence of a signal can be confirmed by confirming whether or not an intensity of a predetermined value or more has been detected.

  The output signal of one of the acceleration sensors 10c and 10d of the jig 80b is input to the input receiving unit 30. Note that when both output signals of the acceleration sensors 10c and 10d are used, an average of both output signals may be used. In the present embodiment, as an example, the output signal of the acceleration sensor 10c is used. The calculation unit 40 removes noise from the output signal of the acceleration sensor 10c (step S4).

  Next, the calculation unit 40 superimposes the output signal of the acceleration sensor 10a after noise removal in step S3 and the output signal of the acceleration sensor 10c after noise removal in step S4 (step S5). At this time, the calculation unit 40 corrects the output signal of the acceleration sensor 10a and the output signal of the acceleration sensor 10b using the transfer function of the sensor sensitivity database 51.

  Next, the calculation unit 40 extracts an amplification signal (fluctuation part) in the contact resonance frequency band of the jig 80a and the jig 80b (step S6). Next, the calculation unit 40 refers to the contact resonance database 52 and multiplies the intensity of the amplified signal in the contact resonance frequency band of the jig 80a by the inverse function of the transfer function in the contact resonance frequency band. In addition, the calculation unit 40 multiplies the intensity of the amplified signal in the contact resonance frequency band of the jig 80b by the inverse function of the transfer function in the contact resonance frequency band. As a result, the amplified signals in both contact resonance frequency bands can be restored. The calculation unit 40 reflects the restored signal on the output signal obtained in step S5 (step S7).

  Next, the determination unit 60 refers to the past measurement database 54 and determines whether or not the output signal obtained in step S7 matches the past measurement results (signal intensity and frequency) related to normal operation ( Step S8). It is not necessary to be completely coincident, and a predetermined allowable range may be set. When it determines with "Yes" at step S8, the determination part 60 makes the output part 70 output the signal which concerns normally (step S9). When it determines with "No" at step S8, the determination part 60 makes the output part 70 output the signal which concerns on a failure (step S10). At that time, the determination unit 60 may cause the output unit 70 to output a corresponding failure factor with reference to the theoretical abnormal frequency band database 53.

  In this embodiment, the signal is detected by reducing the intensity of the amplified signal in the contact resonance frequency band. According to this configuration, even if noise is superimposed on the output signal of the acceleration sensor, the signal buried in the noise can be restored. Thereby, the accuracy of vibration detection using the acceleration sensor can be improved. By multiplying the intensity of the signal amplified in the contact resonance frequency band by the inverse function of the transfer function in the contact resonance frequency band, the signal can be restored with higher accuracy. Moreover, the frequency band which can be amplified can be widened by using the some acceleration sensor from which a contact resonance frequency band differs. Thereby, the accuracy of measurement using the acceleration sensor can be further improved.

  Here, a small and lightweight acceleration sensor that can be manufactured by a mass production method called a MEMS type is widespread, and is promising from the viewpoint of cost reduction. However, since noise is larger than that of a piezoelectric element and it is difficult to separate a signal and noise, it is difficult to detect low-vibration acceleration which is important for predictive detection. FIG. 7 is a diagram illustrating an output signal of the MEMS acceleration sensor and an output signal of the piezoelectric acceleration sensor. As illustrated in FIG. 7, noise is large in the output signal of the MEMS type acceleration sensor. By using such a noisy acceleration sensor in this embodiment, it is possible to perform highly accurate measurement while suppressing costs.

  In this embodiment, a plurality of jigs 80a and 80b are used, but one jig may be used. In this case, the signal buried in the noise can be restored by amplifying the signal in the contact resonance frequency band set in the jig and reducing the intensity of the amplified signal.

  When a plurality of jigs are used, the acceleration sensors provided in each jig are the same type of sensors. This is because output signals can be superimposed by using the same type of sensor. The same type of sensor is a sensor having the same measurement function, for example, a sensor having the same model number. In the above example, the contact resonance frequency band is adjusted by adjusting the radius of curvature of the distal end portion 86, but is not limited thereto. For example, the contact resonance frequency band may be adjusted by changing the material of the spring 83 and the structure of the jigs 80a and 80b.

  In the above example, the movable portion 82 functions as an example of a contact body that is in contact with the vibrating object and in which the contact resonance frequency band is set. The calculation unit 40 functions as an example of a detection unit that detects the signal by reducing the intensity of the signal amplified in the contact resonance frequency band in the output signal of the acceleration sensor.

(Other examples)
FIG. 8 is a block diagram for explaining another hardware configuration of the calculation unit 40, the database 50, and the determination unit 60. As illustrated in FIG. 8, the calculation unit 40, the database 50, and the determination unit 60 include a CPU 101, a RAM 102, a storage device 103, an interface 104, and the like. Each of these devices is connected by a bus or the like. A CPU (Central Processing Unit) 101 is a central processing unit. The CPU 101 includes one or more cores. A RAM (Random Access Memory) 102 is a volatile memory that temporarily stores programs executed by the CPU 101, data processed by the CPU 101, and the like. The storage device 103 is a nonvolatile storage device. As the storage device 103, for example, a ROM (Read Only Memory), a solid state drive (SSD) such as a flash memory, a hard disk driven by a hard disk drive, or the like can be used. When the CPU 101 executes the vibration detection program stored in the storage device 103, the calculation unit 40, the database 50, and the determination unit 60 are realized in the vibration detection device 100.

  Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. It can be changed.

10a to 10d Acceleration sensor 20 Bandpass filter 30 Input reception unit 40 Calculation unit 50 Database 51 Sensor sensitivity database 52 Contact resonance database 53 Theoretical abnormal frequency band database 54 Past measurement database 60 Determination unit 70 Output unit 80a, 80b Jig 81 Fixing unit 82 Movable part 83 Spring 84 Flat plate part 85 Projection part 86 Tip part 90 Rotating body 100 Vibration detection device

Claims (9)

A contact body that contacts a vibrating object and has a contact resonance frequency band set;
An acceleration sensor provided on the contact body;
A vibration detection apparatus comprising: a detection unit that detects the signal by reducing the intensity of the signal amplified in the contact resonance frequency band of the output signal of the acceleration sensor.   The detection unit reduces the intensity of the signal by multiplying the intensity of the signal amplified in the contact resonance frequency band by an inverse function of a transfer function of the contact resonance frequency band. 1. The vibration detection apparatus according to 1.   The vibration detecting apparatus according to claim 1, wherein the contact body is connected in a vibration-insulated state to a fixed portion fixed to the vibrating object while being in contact with the vibrating object. A plurality of the contact bodies are provided,
An acceleration sensor is provided for each of the plurality of contact bodies,
The vibration detection apparatus according to claim 1, wherein contact resonance frequency bands of the plurality of contact bodies are different from each other.   The vibration detection apparatus according to claim 4, wherein the plurality of contact bodies have different contact resonance frequency bands due to different curvature radii of tips at which the plurality of contact bodies contact the vibrating object. Acquire the output signal of the acceleration sensor provided on the contact body that is in contact with the vibrating object and the contact resonance frequency band is set,
The vibration detection method characterized by detecting the said signal by reducing the intensity | strength of the signal amplified in the said contact resonance frequency band among the said output signals.   7. The vibration detection according to claim 6, wherein the intensity of the signal is reduced by multiplying the intensity of the signal amplified in the contact resonance frequency band by an inverse function of a transfer function of the contact resonance frequency band. Method. On the computer,
Processing for obtaining an output signal of an acceleration sensor provided on a contact body that is in contact with a vibrating object and has a contact resonance frequency band set;
And a process of detecting the signal by reducing the intensity of the signal amplified in the contact resonance frequency band in the output signal.   In the process of detecting a signal, the intensity of the signal is reduced by multiplying the intensity of the signal amplified in the contact resonance frequency band by the inverse function of the transfer function of the contact resonance frequency band. The vibration detection program according to claim 8.

Images