JP7244179B2 - Looseness detection system for axial force member and looseness detection method for axial force member - Google Patents

Description

TECHNICAL FIELD The present invention relates to an axial force member loosening detection system and an axial force member loosening detection method for determining an axial force by vibration generated from an axial force member to which an external force is applied.

As disclosed in Japanese Unexamined Patent Application Publication No. 2002-100022, there is known a device for determining looseness of a bolt from a hammering sound generated when the bolt is struck with a hammer. In the bolt looseness determination device of Patent Document 1, a hammering sound is detected with a microphone, separated into time-series signals of each frequency band by an analog bandpass filter, and the time-series signals are rectified and peak-held to obtain each frequency. Extract the instantaneous maximum in the band.

On the other hand, the excitation force when struck by the hammer is detected by the excitation force sensor, and the ratio of the excitation force to the external force reference value is obtained by the ratio calculator. Then, the maximum value extracted from the hammering sound is corrected based on the ratio calculated by the ratio calculator, compared with the preset vibration reference value in the comparator, and the comparison result when it deviates from the predetermined relationship. Output a signal.

Furthermore, when the number of comparison result signals is, for example, three or more, an alarm is issued as an abnormality by an alarm device. In short, this device detects looseness by comparing the magnitude of the input signal with the maximum value of each frequency band, and issues an alarm when looseness is detected.

JP-A-2000-131195

However, since the bolt looseness determination device of Patent Document 1 described above is a device that uses an alarm device to notify that the bolt has loosened, looseness in the initial and middle stages is not detected, and the looseness does not reach a state where an alarm should be issued. If it doesn't progress, the inspector won't notice.

Further, in the method in which the inspector hits the bolt with a hammer to determine the looseness, the inspected object is limited to the bolts that the inspector can approach. Furthermore, if a plurality of sensors are used to detect the excitation force and the hitting sound at the time of hitting, and the input signals thereof are used for determination, the number of measuring devices increases and the signal processing becomes complicated.

Accordingly, it is an object of the present invention to provide an axial force member looseness detection system and an axial force member looseness detection method capable of stepwise determining looseness of an axial force member based on the measurement results of a vibration detection device. purpose.

In order to achieve the above object, a looseness detection system for an axial force member of the present invention is a looseness detection system for an axial force member that determines an axial force by vibration generated from the axial force member to which an external force is applied, comprising: An estimating formula creating unit that creates an axial force estimating formula based on a plurality of natural frequencies obtained by vibration generated from an axial force member whose force is known, and a vibrating device for applying external force to the axial force member to be inspected. a vibration detection device for measuring axial force member vibration of the axial force member to be inspected generated by the external force applied by the vibration excitation device; and frequency analysis of the axial force member vibration detected by the vibration detection device. an axial force estimating unit for estimating the axial force of the axial force member to be inspected by inputting a plurality of values from the peak frequency and amplitude obtained by the above into the axial force estimation formula; and a looseness determination unit that determines looseness of the axial force member to be inspected based on the estimation result of the unit.

Here, a bolt can be applied as the axial force member. Further, it is preferable that two to four peak frequencies occurring in a range of 500 Hz to 6 kHz are used as natural frequencies for creating the axial force estimation formula. Further, the vibrating device may be a keying device for striking the head of the axial force member to be inspected, or a vibrating laser device for irradiating a YAG laser.

Further, the vibration detection device may be a sound collector that measures sound generated from the axial force member to be inspected by applying an external force, or a laser vibrometer that measures vibration.

Further, the invention of a method for detecting looseness of an axial force member is a method for detecting looseness of an axial force member for determining the axial force by vibration generated from the axial force member to which an external force is applied, wherein the axial force is a known axial force member. is used to create an axial force estimation formula based on a plurality of natural frequencies obtained from the generated vibration; applying an external force to the axial force member to be inspected; A step of measuring the axial force member vibration of the axial force member to be inspected; estimating the axial force of the axial force member to be inspected by inputting it into a formula; and determining looseness of the axial force member to be inspected based on the result of estimating the axial force. do. Here, a bolt can be applied as the axial force member. Moreover, it is preferable to use 2 to 4 peak frequencies and amplitudes generated in the range of 500 Hz to 6 kHz to create the axial force estimation formula.

The looseness detection system of the axial force member of the present invention configured as described above is an estimation method that creates an axial force estimation formula based on a plurality of natural frequencies obtained by vibration generated from an axial force member whose axial force is known. It has an expression generator.

Then, the vibration of the axial force member generated by applying an external force to the axial force member to be inspected by the vibrating device is measured by the vibration detection device, and the peak frequency and the peak frequency obtained by analyzing the frequency of the axial force member vibration are measured. By inputting multiple values from the amplitude into the axial force estimation formula, the axial force estimator estimates the axial force of the axial force member to be inspected, and the looseness of the axial force member to be inspected is determined by the looseness determination unit. .

By using the axial force estimation formula created based on a plurality of natural frequencies in this way, it becomes possible to determine the looseness of the axial force member step by step. In addition, since the input data to be detected when inspecting the axial force member to be inspected is only the measurement result of the vibration sensing device, the configuration can be made relatively simple.

Further, in the invention of the looseness detection method of the axial force member, after creating the axial force estimation formula based on a plurality of natural frequencies obtained by the vibration generated from the axial force member whose axial force is known, The axial force member vibration generated from the force member is frequency-analyzed, and the axial force of the axial force member to be inspected is estimated by the axial force estimation formula. Then, looseness of the axial force member to be inspected is determined from the result of estimation of the axial force.

Using the axial force estimation formula created based on a plurality of natural frequencies in this way, the looseness of the axial force member to be inspected is determined in the above steps, so that the looseness of the axial force member is grasped step by step. be able to

1 is a block diagram illustrating the configuration of a bolt looseness detection system according to an embodiment of the present invention; FIG. It is an explanatory view showing the flow of processing of the bolt looseness detection method according to the embodiment of the present invention. It is the figure which showed the frequency-analysis result of bolt vibration, (a) is a frequency-spectrum figure when there is no axial force, (b) is a frequency-spectrum figure when axial force is 20%. FIG. 2 is an explanatory diagram schematically showing the configuration of a tapping sound recording device; FIG. 10 is a perspective view for explaining an inspection situation using a keying sound recording device; FIG. 2 is an explanatory diagram showing an outline of a configuration including a vibration laser device and a laser vibrometer; FIG. 4 is a perspective view for explaining an inspection situation using a vibration laser device and a laser vibrometer; A table comparing looseness determination results by two determination methods, (a) showing the result when using the measurement data by the sound collector, (b) using the measurement data by the laser vibrometer It is the figure which showed the result of a case.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram for explaining the configuration of a bolt looseness detection system that is a looseness detection system for an axial member according to the present embodiment. FIG. It is a figure explaining the flow of processing of a looseness detection method.

The bolt looseness detection system of the present embodiment is a system for determining the axial force (bolt fastening force) by vibration generated from the bolt, which is an axial force member to which an external force is applied. As shown in FIG. 1, this bolt looseness detection system includes a vibrating device 2 for applying an external force to the bolt to be inspected, which is an axial force member to be inspected, and an inspection object generated by the external force applied by the vibrating device 2. It is mainly composed of a vibration detection device 3 for measuring bolt vibration (response vibration), which is the axial force member vibration of the target bolt, a processing calculation unit 4, an input unit 11 and an output unit 12.

The input unit 11 is means for inputting various settings necessary for creating an axial force estimation formula and determining bolt looseness, and data other than the input data described later. The input unit 11 corresponds to a keyboard, a mouse, a storage medium in which input data is recorded, and the like.

On the other hand, the output unit 12 is means for outputting the created axial force estimation formula, the estimated value of the axial force, the looseness determination result, and the like. Examples include displays, printers, and storage media for recording output data. As storage media for the input unit 11 and the output unit 12, an SD memory card, a flash memory such as a USB memory, a solid state drive (SSD), a hard disk, or the like can be used.

The vibrating device 2 may take any form as long as it is a device capable of applying an external force to the bolt. For example, a keying device 2A (see FIG. 4) that directly strikes a bolt with a key 20 such as a hammer or a weight can be used as the vibrating device 2. FIG. The keying device 2A is a device that can move the key 20 at regular time intervals by combining a motor, a cam, a spring, and the like.

A vibrating laser device 2B (see FIG. 6) that irradiates a YAG laser of visible light can also be used as the vibrating device 2. FIG. When a YAG laser (a solid-state laser using Yttrium Aluminum Garnet) is applied to the bolt surface, impact is generated by ablation (ionization) of the surface material, which can cause the bolt to vibrate.

On the other hand, the vibration detection device 3 may be of any type as long as it can measure the vibration (bolt vibration) generated by the impact given to the bolt by the keying device 2A or the vibrating laser device 2B.

For example, a sound collector 3A (see FIG. 4) that detects a hammering sound generated when a bolt is hit can be used as the vibration detector 3. FIG. The sound collector 3A includes a microphone that converts a hammering sound into an electric signal, and an amplifier that amplifies the converted electric signal.

A laser vibrometer 3B (see FIG. 6) using a Doppler laser capable of detecting vibrations from a distance can also be used as the vibration detection device 3. FIG. The laser vibrometer 3B includes a sensor head that emits laser light and a light receiving element that receives the reflected laser light. For example, if the bolt surface vibrates due to impact, the laser light reflected from the vibrating bolt surface becomes Doppler-shifted laser light, and the change in frequency (speed) is converted into voltage and detected as a vibration phenomenon. can do.

The processing calculation unit 4 is configured by a personal computer (PC) such as a notebook computer, a tablet terminal, or the like. The processing calculation unit 4 includes a control unit 40 that performs various controls, a waveform analysis unit 41 that performs frequency analysis, an estimation formula creation unit 42 that creates an axial force estimation formula, and an axial force estimation unit that estimates the axial force of the bolt. It includes an estimation unit 43 and a looseness determination unit 44 that determines looseness of the bolt.

The control unit 40 sends a signal input from the input unit 11 to each calculation unit, sends a hitting operation instruction signal to the vibrating device 2, and sends detection data (measurement data) of the vibration detection device 3 to each calculation unit. control such as sending to

Also, the waveform analysis unit 41 is a calculation unit that divides the signal waveform measured by the vibration detection device 3 and performs frequency analysis, as will be described later. The analysis result by the waveform analysis unit 41 is used by both the estimation formula creation unit 42 and the axial force estimation unit 43 .

Next, the processing of the estimation formula creation unit 42 will be described with reference to FIG. The estimation formula creating unit 42 creates an axial force estimation formula using a bolt B having the same configuration as the bolt B of the joint to be inspected. Here, the bolt B is passed through the hole of the mounting member M and tightened using the nut B2. By hitting the head B1 of the bolt B, the fastening force of the bolt B, that is, the axial force (torque) is estimated.

In creating the axial force estimation formula, the axial strain of the bolt B and the axial vibration acceleration during vibration are measured while changing the axial force of the bolt B. In short, this bolt B can be said to be a bolt with a known axial force since the axial strain can be measured. Axial strain can be measured by attaching a strain gauge to the shaft portion of the bolt B. If the bolt B has a small diameter, the axial force can be known by measuring the tightening torque with a torque wrench.

When creating the axial force estimation formula, for example, a calibration experiment is performed in which the axial force is changed in about four patterns, vibration is applied about five times in each pattern, and the measurement of the axial force and vibration is repeated. Then, the following processing is performed on the obtained measured waveform.

First, from the signal waveform (measurement data) detected and input by the vibration detection device 3, the vibration caused by a plurality of hammering sounds is extracted for each one (one impact). Subsequently, filtering and high-frequency correction (enhancement) processing are performed on each cut-out vibration waveform.

Furthermore, peak frequencies are extracted by the AR method (Yule-Walker method). For example, extract 6 peak frequencies generated in a specific frequency range (500 Hz-6 kHz), set those frequencies to f1-f6, respectively, peak spectrum (peak amplitude) corresponding to those frequencies, the maximum peak Let p1-p6 be values normalized to 1.

Then, looking at the distribution of each frequency, etc., three to four (here, four) frequencies for axial force estimation are selected. The frequency range and the number of frequencies to be extracted as described above vary depending on the form of the bolt B to be inspected. It is known that some of the eight parameters (f1-f4, p1-p4) selected in this way have a correlation with the axial force of the bolt B, and the results of the calibration experiment , find the respective correlations.

FIG. 3 illustrates the results of frequency analysis of bolt vibration. FIG. 3(a) shows a frequency spectrum diagram when there is no axial force (0% of the default value), and FIG. 3(b) shows a frequency spectrum diagram when the axial force is 20% of the default value. As shown in this figure, different peak frequencies and different peak amplitudes are exhibited when the axial force is 0% and when the axial force is 20%. That is, the peak frequency and the peak amplitude can be used as feature quantities when estimating the axial force.

Therefore, using the frequency and amplitude as parameters f _i and p _i , an axial force estimating formula is created using the following formula. Since the measurement data for which the axial strain is known is used here, it can also be said to be an axial strain estimation formula. In order to simplify the explanation, the following equations show a case where the number of peak frequencies (i) used for estimation is two and the axial strain estimation equation is a linear equation.
ε _N =Σ(v _i (a _i f _i +b _i ) + w _i (c _i p _i +d _i )); i=1,2
where _εN is the axial strain, f _i is the i-th peak frequency, a _i f _i +b _i and c _i p _i +d _i are the axial strain estimation formulas based on the ith order vibration parameter, v _i and w _i are Indicates weight.

Note that the number of peak frequencies (i) when actually creating the axial force estimation formula is set to 2 to 4, and the axial strain estimation formula for each parameter is a quadratic formula. In order to determine the axial strain estimation formula (axial force estimation formula), it is necessary to determine the weights (v _i , w _i ) that minimize the error of ε _N in the above formula. is complicated, the weights are determined by machine learning.

In order to create an axial force estimation formula with high estimation accuracy (step S11 in FIG. 2), it is necessary to perform a hammering experiment on the bolt B in advance (step S12). For example, multiple experiments are conducted to measure the vibration generated by impacting the bolt B with the axial force of the bolt B set to 0%, 20%, 50%, and 80% of the default value.

Since the flow of the impact experiment and the inspection of the bolt to be inspected are the same processes, the description will be continued with reference to FIG. 2 . Hitting sound generated by hitting the head B1 of the bolt B with the vibrating device 2 is recorded with the vibration detecting device 3, amplified with the amplifier 31, and recorded.

The recorded (measured) signal waveform is waveform-divided into vibrations caused by each hammering sound (step S1). The divided vibration waveform is subjected to frequency analysis (step S2), and eight variables (f1-f4, p1-p4) indicating frequency and amplitude are taken out (steps S31-S34).

Then, the eight variables are given to the determination program incorporating the once determined axial force estimation formula (step S13) to estimate the axial force (step S4). This estimation work is repeated several times for the bolt B whose axial force is known (including simultaneous measurement with a strain gauge).

In the determination program, machine learning corrects the coefficients (weights (v _i , w _i )) of the axial force estimation formula so that the correct answer can be derived when the axial force determination is incorrect. By repeating this, it is possible to improve the axial force determination accuracy of the determination program. If the variables used for diagnosis are appropriately selected, an accuracy rate of almost 100% can be obtained as described later.

This determination program finally outputs the result of looseness determination (step S5) based on the estimated value of the axial force. The looseness determination can be indicated, for example, by the ratio (R) of the amount of axial strain to the yield strain of the bolt to be inspected.

In short, if the yield strain of the introduced axial force of the bolt to be inspected is r _N0 and the amount of axial strain during inspection is r _N , the ratio (R) that becomes the looseness judgment result is calculated as R = r _N /r _N0 can do.

Next, a specific configuration of the bolt looseness detection system 1 of the present embodiment will be described with reference to the drawings. First, the case of using the keying sound recording device 5 will be described with reference to FIGS.

As schematically shown in FIG. 4, the keystroke sound recording device 5 is integrated with a keystroke device 2A as a vibrating device and a sound collector 3A as a vibration detection device. This keying sound recording device 5 is used for a bolt to be inspected (bolt B) at a short distance of about 2 m.

The tapping sound recording device 5 is attached to the tip of the extension pole 51 and fixed to the attachment member M by a magnet or the like. A motor, a cam, and a spring are incorporated in the keying device 2A, and the peripheral surface of the head portion B1 of the bolt B can be struck by the key 20 that moves at regular time intervals.

A sound generated by striking with the keying device 2A is recorded by a microphone of a sound collecting device 3A provided nearby. With such a configuration, as shown in FIG. can be inspected by the keying sound recording device 5 at the tip of the .

Next, a configuration in which the excitation laser device 2B and the laser vibrometer 3B are combined will be described with reference to FIGS. The configuration schematically shown in FIG. 6 shows the case of using a vibration excitation laser device 2B as a vibration excitation device and a laser vibrometer 3B as a vibration detection device.

This configuration is applied to a bolt B located at a distance of about 3 m to 20 m, which is difficult for an inspector N to approach. A vibrating laser device 2B for vibrating irradiates a YAG laser 21 of visible light. For example, the YAG laser 21 with a wavelength of 532 nm is irradiated with an output of 220 mJ for an irradiation time of 5 nsec.

When the green luminescent spot of the YAG laser 21 is applied to the surface of the head B1 of the bolt B, impact is generated by ablation (ionization) of the surface material. As a result, the bolt B can be vibrated, and the vibration is remotely measured by the laser vibrometer 3B, which is a Doppler laser vibrometer.

With such a configuration, as shown in FIG. 7, the excitation laser device 2B and the laser vibrometer 3B are mounted on a movable carriage N1, and inspection can be performed while moving in the tunnel T. can. That is, the carriage N1 can be moved to a position where the bolt B for fixing the mounting member M of the road sign M1 can be aimed, and the inspection can be performed sequentially. Therefore, it is possible to quickly inspect (looseness determination) the bolt B located at a high place without setting up scaffolding or restricting traffic for a long period of time.

Next, an accuracy confirmation experiment conducted to confirm the effects of the bolt looseness detection system and the bolt looseness detection method of the present embodiment will be described. In the conventional method, it was only possible to determine whether or not the bolt was loose. It will be explained that it is possible to detect that the level is lowered to a certain extent.

In the experiment, a test piece in which the lap joint was tightened with bolt B was used, and the tightening torque (axial force of bolt B) by a torque wrench was changed to 4 types of the default value of 0%, 20%, 50%, and 80%. went. The bolt B was struck by lightly tapping the head B1 with a metal test hammer, and the vibration was recorded from a distance of 3 m with the microphone of the sound collector 3A and from a distance of 20 m with the laser vibrometer 3B. Two patterns of experiments were conducted, one for the case of measurement.

Vibration data measured by the sound collector 3A and the laser vibrometer 3B are input to the waveform analysis unit 41 of the processing operation unit 4 in which the determination program is executed. In the waveform analysis unit 41, the input signal waveform was divided into waveforms, the feature amounts (peak frequency, peak amplitude) of the divided vibration waveforms were determined, and training (machine learning) of the determination program was performed.

Then, using the axial force estimation formula created by the trained estimation formula creation unit 42, an experiment was conducted to determine the slackness from the vibration data, and the accuracy rate of the determination result was calculated. FIG. 8(a) shows the determination result when using the measurement data by the sound collector 3A, and FIG. 8(b) shows the determination result when using the measurement data by the laser vibrometer 3B.

Then, as the discrimination method, two methods, the quadratic discrimination method and the k-nearest neighbor method (kNN), were applied to compare the accuracies. Here, the quadratic discriminant method and the k-nearest neighbor method (kNN) are one of methods for classifying data into a plurality of classes (groups), and the quadratic discriminant method performs discrimination by a quadratic function such as an ellipse. The k-nearest neighbor method (kNN) is a method for performing non-linear discrimination, and both of them are often used in pattern recognition.

In the diagnosis result using the measurement data by the sound collector 3A in FIG. We selected 5 out of 1 and compared the accuracy rate.

As a result, the accuracy rate for all data (axial force 0%, 20%, 50%, 80%) is a maximum of 89% (secondary discrimination method), and even when the axial force is 20% or less, the maximum accuracy rate is 86%. An accuracy rate of (kNN) was obtained.

On the other hand, in the diagnosis result using the measurement data by the laser vibrometer 3B in FIG. %, and when the axial force was 20% or less, a 100% accuracy rate was obtained in both the secondary discrimination method and the kNN discrimination method.

In this way, regardless of whether the configuration of the vibration detection device 3 is the sound collector 3A or the laser vibrometer 3B, four stages of axial force (0%, 20%, 50%, 80% of the default value) can be discriminated with an accuracy rate of 80% or more. In short, it was confirmed that by applying the bolt looseness detection system and the bolt looseness detection method of the present embodiment, it is possible to distinguish whether the bolt looseness is 80% or 50%.

Next, the operation of the bolt looseness detection system and the bolt looseness detection method according to the present embodiment will be described.
The bolt looseness detection system of the present embodiment configured as described above detects the axial force based on a plurality of natural frequencies (peak frequency, peak amplitude) obtained by vibration generated from the bolt B having a known axial force. An estimation formula creating unit 42 for creating a force estimation formula is provided.

Then, the bolt vibration generated by applying an external force to the bolt (B) to be inspected by the vibrating device 2 is measured by the vibration detection device 3, and the bolt vibration is frequency-analyzed by the waveform analysis unit 41. By inputting a plurality of values from the peak frequency and the peak amplitude into the axial force estimation formula, the axial force estimating unit 43 estimates the axial force of the bolt (B) to be inspected and the looseness of the bolt to be inspected. It is determined by the looseness determination unit 44 .

By using the axial force estimation formula created based on a plurality of natural frequencies in this way, it becomes possible to determine the looseness of the bolt B step by step as well as the presence or absence of looseness. In addition, since the input data to be detected when inspecting the bolt to be inspected is only the measurement result of the vibration sensing device 3, the configuration can be made relatively simple.

If it becomes possible to grasp the looseness of bolt B in stages, it will be necessary not only for items with a high risk of falling off, but also for bolt B to be re-tightened or replaced after several months to several years. can be grasped with high accuracy, the frequency of inspection can be reduced. In addition, it becomes easier to formulate plans such as the timing and scale of repairs.

In addition, in the invention of the bolt looseness detection method, an axial force estimation formula is created based on a plurality of natural frequencies (peak frequency, peak amplitude) obtained from vibrations generated from a bolt B with a known axial force (Fig. 2 step S11), frequency analysis is performed on the bolt vibration generated from the inspection target bolt (B) (step S2), and the axial force of the inspection target bolt (B) is estimated by the created axial force estimation formula ( step S4). Then, looseness of the inspected bolt (B) is determined from the estimated axial force (step S5).

By using the axial force estimation formula (step S13) created based on a plurality of natural frequencies in this way, the looseness of the bolt B to be inspected is determined in the above steps, and the looseness of the bolt B is determined. You will be able to understand step by step.

Although the embodiment of the present invention has been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment. Included in the invention.
For example, in the above-described embodiment, the bolt B for attaching the road sign M1 etc. as the axial force member has been described as an example, but it is not limited to this, and high strength bolts, ordinary bolts, anchor bolts, suspension bolts, PC bolts, etc. The present invention can be applied to axial force members such as steel wires.

Further, in the above-described embodiment, the configuration in which the head B1 of the bolt B is hit by the keying device 2A or the vibrating laser device 2B has been described. Any vibrating device may be used as long as it can generate vibration from the bolt B.

Furthermore, the vibration detection device 3 is not limited to the sound collector 3A or the laser vibrometer 3B, and any device capable of measuring the vibration generated from the bolt B may be used. For this reason, the combination of the vibrating device 2 and the vibration detecting device 3 is not limited to the combination described in the above embodiment. A combination of the sound device 3A may be used.

1: Bolt looseness detection system (Axial member looseness detection system)
2: vibration device 2A: keying device (vibration device)
2B: Vibration laser device (vibration device)
3: Vibration detection device 3A: Sound collector (vibration detection device)
3B: Laser vibrometer (vibration detector)
41: Waveform analysis unit 42: Estimation formula creation unit 43: Axial force estimation unit 44: Looseness determination unit B: Bolt (axial force member)

Claims (8)

A looseness detection system for an axial force member that determines an axial force by vibration generated from the axial force member to which an external force is applied,
an estimation formula creating unit for creating an axial force estimation formula based on a plurality of natural frequencies obtained by vibration generated from an axial force member having a known axial force;
a vibrating device for applying an external force to an axial force member to be inspected;
a vibration detection device for measuring axial force member vibration of the axial force member to be inspected generated by the external force applied by the vibration excitation device;
By inputting a plurality of values into the axial force estimation formula from the peak frequency and amplitude obtained by frequency analysis of the axial force member vibration detected by the vibration detection device, the axial force to be inspected an axial force estimation unit that estimates the axial force of the member;
A looseness detection system for an axial force member, comprising: a looseness determination unit that determines looseness of the axial force member to be inspected based on an estimation result of the axial force estimation unit. 2. The looseness detection system for an axial force member according to claim 1, wherein the axial force member is a bolt. 3. The axial force according to claim 1 or 2, wherein two to four peak frequencies generated in a range of 500 Hz to 6 kHz are used as natural frequencies for creating the axial force estimation formula. Material looseness detection system. 4. The vibrating device according to claim 1, wherein the vibrating device is a keying device that hits the head of the axial force member to be inspected, or a vibrating laser device that irradiates a YAG laser. A looseness detection system for an axial force member as described. 5. The vibration detection device is a sound collector for measuring sound generated from the axial force member to be inspected by applying an external force, or a laser vibrometer for measuring vibration. Looseness detection system of the axial force member according to any one of claims 1 to 1. A looseness detection method for an axial force member for determining an axial force by vibration generated from the axial force member to which an external force is applied, comprising:
creating an axial force estimation formula based on a plurality of natural frequencies obtained from the generated vibration using an axial force member with a known axial force;
a step of applying an external force to the axial force member to be inspected;
measuring axial force member vibration of the axial force member to be inspected generated by the external force;
By inputting a plurality of values into the axial force estimation formula from the peak frequency and amplitude obtained by frequency analysis of the measured axial force member vibration, the axial force of the axial force member to be inspected is calculated. an estimating step;
and determining looseness of the axial force member to be inspected based on the axial force estimation result. 7. The method for detecting loosening of an axial force member according to claim 6, wherein the axial force member is a bolt. 8. The looseness of the axial force member according to claim 6 or 7, wherein two to four peak frequencies and amplitudes generated in a range of 500 Hz to 6 kHz are used to create the axial force estimation formula. Detection method.

Images