JP6432308B2 - Measuring device and measuring range switching method - Google Patents

Description

  The present invention relates to a measuring apparatus and a measuring range switching method that can automatically switch a measuring range.

  In order to diagnose the state of a structure, the free damped vibration of the structure is measured and analyzed. This is because free-damping vibration does not depend on the excitation force (cause of vibration generation) but greatly depends on mechanical properties such as mass and rigidity of the structure.

  It is desirable that a measurement apparatus that measures such free-damping vibration has a wide measurement range so that measurement data can be used in many applications. In addition, it is desirable to have high resolution so that accurate analysis can be performed even for minute signals.

  The resolution is determined mainly by analog / digital conversion (A / D conversion) when measuring data, which is an analog signal, is converted into a digital signal. Therefore, Patent Document 1 proposes an input / output characteristic automatic adjustment method for arbitrarily setting an input range in an A / D converter when converting an analog signal into a digital signal.

  That is, in this input / output characteristic automatic adjustment method, output data sequentially output from the A / D converter is sampled, and the amount of change per unit time of the analog input signal is calculated based on the sampled data. Then, from the calculated change amount of the analog input signal, it is predicted whether or not all bits of the output data within a certain period are valid, and the measurement range of the A / D converter is adjusted.

JP-A-1-174123

  However, in the input / output characteristic automatic adjustment method according to Japanese Patent Laid-Open No. 1-174123, the range is accurately switched for an input signal including noise such as high-frequency electromagnetic noise caused by the surrounding environment or the device itself. There was a problem that could not be done.

  Therefore, a main object of the present invention is to provide a measuring device and a measuring range switching method capable of appropriately switching a measuring range even when noise is included in an input signal.

  In order to solve the above problems, an invention relating to a measurement device capable of switching a measurement range is based on a vibration detection unit that detects vibration of a structure and outputs the detection result as measurement data of an analog signal, and the measurement data. A range switching processing unit that determines and switches the measurement range, and the range switching processing unit converts analog signal measurement data into a digital signal within a set measurement range, and attenuation of the measurement data Attenuation characteristic calculating means for calculating characteristics, a measurement range determining means for determining the measurement range based on the attenuation characteristics, and a measurement range switching means for switching the measurement range according to the determination result of the measurement range switching means It is characterized by.

  In addition, the invention relating to the measurement range switching method when automatically setting the measurement range detects the vibration of the structure and outputs it as measurement data of an analog signal, calculates the attenuation characteristics of the measurement data, and measures based on the attenuation characteristics A range is determined, and analog signal measurement data is converted into a digital signal within the determined measurement range.

  According to the present invention, the measurement range can be appropriately switched even if noise is included in the input signal.

It is a block diagram of the measuring device concerning the present invention. It is a flowchart which shows a measurement range switching procedure. It is a figure which shows the vibration curve which showed the measurement data of the analog signal with respect to time. It is a figure which shows the vibration curve and damping characteristic curve applied to description of Example 1. FIG. It is a figure which shows the comparison result of the conventional measuring method and the measuring method concerning this embodiment.

  An embodiment of the present invention will be described. FIG. 1 is a block diagram of a measurement apparatus 2 capable of automatically setting a measurement range according to the present embodiment.

  The measuring device 2 includes a vibration detection unit 10, a range switching processing unit 20, and a power supply 30.

  The range switching processing unit 20 includes an analog filter 21, a microcomputer (MPU) 22, an output unit 23, and a plurality of regulators 24 (24a to 24c).

The vibration detection unit 10 is fixed to a structure that is a vibration measurement target by a fixing member such as an adhesive, a permanent magnet, or a mechanical joining method. And the vibration of the said structure is measured and the measurement data of an analog signal are output.
In addition, as the vibration detection unit 10, it is preferable to use a piezoelectric element that can measure minute vibration with high sensitivity.

  The analog filter 21 removes noise components included in the measurement data from the vibration detection unit 10. Examples of the noise component include high-frequency components such as environmental vibrations that are likely to occur in a low frequency range.

  The analog filter 21 can be provided as necessary. Even when the analog filter 21 is provided, when raw measurement data from which noise has not been removed is required, path switching is performed so that the measurement data is input to the MPU 22 by bypassing the analog filter 21. It is also possible.

The MPU 22 includes analog / digital conversion means (A / D conversion means) 22a, attenuation characteristic calculation means 22b, measurement range determination means 22c, and measurement range switching means 22d.
The A / D conversion means 22a converts analog measurement data into a digital signal.
The attenuation characteristic calculation means 22b and the measurement data are used to calculate the attenuation characteristic of the measurement data.
The measurement range determination unit 22c determines the measurement range based on the attenuation characteristic.
The measurement range switching unit 22d switches the measurement range according to the determination result of the measurement range switching unit.

  Such an MPU 22 may be configured by a general-purpose product having a small built-in memory capacity, but may be configured by a digital signal processor capable of high-speed arithmetic processing.

  The output unit 23 outputs measurement data converted into a digital signal by the MPU 22 to an external device (server or the like). At this time, it does not matter whether the output method is wireless or wired, and also in the case of wireless, whether it is electromagnetic waves or light for infrared rays. FIG. 1 illustrates the case of wireless output.

  The regulator 24 supplies power to the MPU 22. At this time, the supply voltage of each regulator 24 is different. The measurement range is determined by this supply voltage. Therefore, in this embodiment, the measurement range is switched by selecting the regulator 24.

  For example, the regulator 24a supplies the first voltage 5 [V], and the resolution at that time is 0.020 [V]. The regulator 24b supplies the second voltage 3.3 [V], and the resolution at that time is 0.013 [V]. The regulator 24c supplies the third voltage 1.8 [V], and the resolution at that time is 0.007 [V].

  In the present embodiment, a configuration in which the voltage supplied to the MPU 22 by the plurality of regulators 24 is changed to change the measurement range is described as an example, but the present invention is not limited to such a configuration.

  When the measuring device 2 is used in an environment where a commercial power source such as a bridge cannot be used, a storage battery can be used as the power source 30.

  Next, an automatic determination range switching method for the measurement range performed by the measurement apparatus 2 having such a configuration will be described. For convenience of explanation, it is assumed that the structure to be subjected to vibration measurement is a road bridge, and the vibration to be measured is vibration generated on the road bridge due to vehicle traffic. The vibration generated in the road bridge as the vehicle passes is a damped vibration that attenuates over time.

  FIG. 2 is a flowchart showing an automatic determination switching method for the measurement range.

  Step S1: The measurement range switching means 22d of the MPU 22 sets the measurement range to the maximum range and starts vibration measurement. This corresponds to selecting the regulator 24 that supplies the maximum supply voltage when the measurement range switching process corresponds to the process of selecting the regulator 24 (the process of selecting the supply voltage). The reason why the maximum range is set is that the maximum amplitude value needs to be detected even though the maximum amplitude value of the vibration to be measured is unknown.

  Then, the measurement data from the vibration detection unit 10 passes through the analog filter 21, so that noise components due to environmental vibration and the like included in the measurement data are removed and input to the A / D conversion means 22 a of the MPU 22.

  The A / D converter 22a converts the measurement data of the analog signal into a digital signal in the measurement range corresponding to the currently set regulator 24. That is, the analog signal is converted into a digital signal with a resolution corresponding to the measurement range, and is output from the output unit 23 to an external device.

  Step S2: The attenuation characteristic calculation means 22b of the MPU 22 detects the maximum amplitude value Amax from the vibration curve, and sets the time at that time as the reference time. FIG. 3 is a vibration curve K1 showing measurement data of an analog signal with respect to time. In the same figure, an attenuation characteristic curve (envelope) K2 described later is indicated by a bold line.

  The vibration measurement is continued for a predetermined time (characteristic derivation time) so that at least the maximum amplitude value Amax can be reliably measured and the attenuation characteristic curve K2 can be derived. This characteristic derivation time is, for example, several tens of milliseconds, and is stored in advance in the MPU 22 as an experience value.

  Steps S2 and S3: Next, the damping characteristic calculation unit 22b performs time differentiation of the vibration curve K1, and extracts a point where the sign of the differentiation value changes. The point where the sign of the differential value changes corresponds to the inflection point of the vibration curve K1.

Step S4: Then, the attenuation characteristic calculation means 22b
A (t) = Amax × EXP (−σ × t) (1)
According to Equation 1, the attenuation rate σ is calculated with the value of the inflection point as A (t). Note that t is time. Hereinafter, this envelope is referred to as an attenuation characteristic curve K2, and the value thereof is referred to as an amplitude value A (t).

  A line connecting the amplitude values A (t) forms an attenuation characteristic curve K2 of the vibration curve K1. The probability that the inflection point is caused by noise such as high-frequency electromagnetic noise caused by the surrounding environment or the device itself in the measurement data is not zero. However, since the frequency of these high frequency noises is greater than the low frequency of the damped vibration, the probability due to noise is very small. Therefore, the influence of noise on the shape of the attenuation characteristic curve K2 is small. Therefore, as will be described later, in this embodiment, the measurement range is switched based on the attenuation characteristic curve K2. However, since the noise is not affected by the attenuation characteristic curve K2, the measurement range can be switched appropriately. .

  Although the probability that the inflection point is caused by noise is small, in order to make it smaller, it is possible to derive the attenuation rate σ a plurality of times and use the average value thereof.

  Furthermore, if an approximate value of the period of vibration to be measured is calculated in advance and a measurement time range of about three times the period is used, for example, an accurate attenuation rate with little influence of noise can be derived. Here, the period of vibration to be measured is the reciprocal of the resonance frequency of the vibration measurement object or the reciprocal of the resonance frequency of the bending vibration in the axial direction of the road bridge.

  Step S5: Next, the measurement range determination means 22c predicts an amplitude value A (t) when a predetermined time has elapsed from the reference time using the attenuation characteristic curve K2.

  That is, the measurement range determination unit 22c predicts and determines whether or not the amplitude value A (t) is smaller than a preset range switching condition Vr. This range switching condition Vr is preset in the MPU 22 as an experience value, for example, 30% of the maximum amplitude value Amax. As a result, a plurality of measurement ranges can be sequentially switched according to the attenuation state. Needless to say, when there are a plurality of measurement ranges, a plurality of range switching conditions Vr are also provided.

  In the above description, the range switching condition is a reduction rate with respect to a preset maximum amplitude value, but may be a reduction range. That is, in the above description, the measurement range is switched at the timing when it becomes 30% or less of the maximum amplitude value Amax. However, the present embodiment is not limited to such a configuration, and may be a range switching condition Vr with a decrease range (numerical value range).

  For example, the range switching condition Vr is set to 100% to 81%, 80% to 30%, 29% to 5%. When the predicted amplitude value A (t) is included in the range switching condition Vr of 100% to 81% with respect to the maximum amplitude value Amax, the current measurement range is maintained. If the predicted amplitude value A (t) is 80% to 30% with respect to the maximum amplitude value Amax, the measurement range is switched to one lower (narrow area) or one higher (wide area). Furthermore, if the predicted amplitude value A (t) is 29% to 5% with respect to the maximum amplitude value Amax, the measurement range is switched to the next lower (or upper) measurement range. Thereby, a more appropriate measurement range can be switched.

  Step S6: When the amplitude value A (t) is smaller than the range switching condition Vr (A (t) ≦ Vr), the measurement range is set to the measurement range one level lower (the narrow range) than the current measurement range. The measurement range is switched when the measurement range switching means 22d of the MPU 22 selects the regulator 24.

  For example, when the regulator 24 currently connected to the MPU 2 is the regulator 24b that supplies the second voltage 3.3 [V], the regulator 24c that supplies the third voltage 1.8 [V] is selected. Become.

  Step S7: On the other hand, when the amplitude value A (t) is not smaller than the range switching condition Vr (A (t)> Vr), the current measurement range is maintained.

  Step S8: The range switching process is completed by such a procedure, and it is then determined whether or not to continue the measurement (measurement continuation determination process). In the measurement continuation determination, when the amplitude value A (t) becomes smaller than the measurement continuation determination threshold Ct (A (t) ≦ Ct), it is determined that the measurement is finished. The measurement continuation determination threshold Ct is an experience value set in advance in the MPU, and can be exemplified by 5% of the maximum amplitude value Amax, for example.

  In the above description, the attenuation characteristic curve K2 is derived based on the measurement data being measured. However, it is also possible to use data (hereinafter referred to as average measurement data) obtained by adding and averaging the measurement data obtained by performing the measurement a plurality of times. Since the high-frequency vibration component and the electromagnetic noise component are removed (suppressed) by this averaging process, a more accurate attenuation characteristic curve K2 can be derived.

  The damping rate σ was calculated using the inflection point of the vibration curve, but it was obtained by performing a simulation that considered the mechanical properties of the structure (mass, stiffness, viscosity, etc. determined from the structure and material). Or may be obtained by theoretical calculation.

In addition, for example, an AR model parameter may be calculated using the amplitude value A of measurement data using an autoregressive model (hereinafter referred to as an AR model), and thereby the attenuation rate σ may be set.
Calculation of the AR model parameters can be performed sequentially, and there is an advantage that the parameters can be updated each time the amplitude value A is input.

<Example 1>
Next, Example 1 of the present invention will be described. In the first embodiment, the vibration of the road bridge is measured by the vibration detection unit 10 using a concrete road bridge as a structure to be measured.

At this time, the vibration detection unit 10 is composed of a piezoelectric element incorporating an amplifier circuit, and has a sensitivity of 20 [mV / (m / s ² )]. The MPU 22 includes an analog-to-digital converter having a processing bit number of 8 bits, and a commercially available product having a maximum clock number of 25 MHz. The regulator 24a supplies the first voltage 5 [V], the resolution at that time is 0.020 [V], the regulator 24b supplies the second voltage 3.3 [V], and the resolution at that time is 0. 0.013 [V], the regulator 24c supplied the third voltage 1.8 [V], and the resolution at that time was 0.007 [V].

  And the vibration detection part 10 was fixed to the back surface of the concrete floor board of width 3000 [mm], length 20000 [mm], and thickness 4000 [mm], and the measurement data when a vehicle passed the upper surface of the floor board was acquired. .

Under such conditions, the measurement range was set to a maximum of 5.0 [V] (regulator 24a was selected), and measurement was performed at a sampling frequency of 1 kHz. This measurement range can be measured up to 250 [m / s ² ] in terms of acceleration.

In the analysis, we focused on the primary bending vibration of the concrete floor slab in the axial direction. In the calculation using a simple model by the finite element method, the period of the primary bending vibration in the axial direction is about 30 [ms].
Therefore, the characteristic derivation time for deriving the attenuation characteristic curve K2 is set to 100 [ms].

  FIG. 4 is a diagram showing measurement data measured under such setting conditions. The horizontal axis represents measurement time, and the vertical axis represents measurement data.

  The measurement data showed a maximum amplitude value Amax at a time of 350 [ms]. That is, the reference time was 350 [ms]. The measurement data was attenuated after this reference time.

  Therefore, the attenuation characteristic curve K2 was calculated using measurement data from the reference time to the characteristic derivation time (100 [ms]). This attenuation characteristic curve K2 is referred to as a first attenuation characteristic curve K2_1.

From the first attenuation characteristic curve K2_1, the amplitude value A (t) at time 540 [ms] could be predicted to be 1.1 [V].
Currently, a regulator 24 a having a first voltage of 5 [V] is connected to the MPU 22. Therefore, the predicted voltage value (1.1 [V]) is 22% (= 1.1 / 5 * 100) with respect to the power supply voltage 5 [V]. Since the power supply voltage is proportional to the measurement range, the predicted measurement data is 30% or less of the maximum amplitude value (Amax).

  Thus, when the measurement time reaches time 540 [ms], the regulator 24 having a voltage lower than the current power supply voltage is selected, and the measurement range is switched. That is, the regulator 24a having the first voltage 5 [V] is switched to the regulator 24b having the second voltage 3.3 [V].

  Then, the maximum amplitude value detected after switching of the measurement range was detected, and the attenuation characteristic curve K2_2 was calculated using the measurement data from the reference time at this time to the characteristic derivation time (100 [ms]).

  Using this attenuation characteristic curve K2_2, the amplitude value A (t) at time 750 [ms] from the start of measurement was expected to be 0.2 [V].

  This predicted amplitude corresponds to 6% (= 0.2 / 3.3 * 100) of the second voltage 3.3 [V]. Therefore, the MPU 22 switches the measurement range by switching to the regulator 24c having a voltage lower than the current power supply voltage. That is, the regulator 24b having the second voltage 3.3 [V] is switched to the regulator 24c having the third voltage 1.5 [V].

As a result, after time 750 [ms] from the start of measurement, the resolution is 0.007 [V], and measurement data with an amplitude value of 0.01 [V] can be measured. The amplitude value of 0.01 [V] corresponds to an acceleration of 0.5 [m / s ² ], which enables highly accurate analysis.

  For comparison, data analysis was also attempted for the conventional method under the measurement conditions described above. And it was investigated whether the minute vibration after the time 750 [ms] was acquired correctly. The result is shown in FIG.

From FIG. 5, it can be seen that minute vibrations (acceleration 0.5 [m / s ² ]) could be detected by the measurement method according to the present embodiment, but could not be detected by the conventional measurement method. This indicates the effectiveness of the measurement method according to the present embodiment.

<Example 2>
Next, a second embodiment of the present invention will be described. In Example 2, the attenuation characteristic curve K2 was derived prior to the actual measurement, and the measurement range was switched based on the attenuation characteristic curve K2.

  That is, prior to the actual measurement, a plurality of measurements were performed, and the average value of the measurement data obtained at that time was used as the average measurement data. The attenuation characteristic curve K2 was derived using the average measurement data. Since the high frequency noise component is suppressed by averaging or processing, the attenuation characteristic curve K2 is also a curve close to the true value. Therefore, accurate measurement range switching can be performed.

  The regulator 24 is selected so that the measurement range becomes the maximum range. Then, measurement was performed 5 times in this measurement range, and averaged data was obtained by averaging. The attenuation characteristic curve K2 was calculated from the average measurement data according to Equation 1.

  In this attenuation characteristic curve K2, the amplitude value A is attenuated to 1.0 [V] after 200 [ms] from the reference time of the maximum amplitude value Amax, and further attenuated to 0.2 [V] after 400 [ms]. .

  Therefore, the regulator 24a is switched to the regulator 24b after 200 [ms] so that the second voltage 3.3 [V] is supplied. In addition, the third voltage 1.8 [V] is supplied by switching to the regulator 24b after 400 [ms].

As a result, after time 800 [ms], the resolution becomes 0.007 [V], and measurement data with an amplitude value of 0.01 [V] can be measured. Since the amplitude value of 0.01 [V] corresponds to the acceleration of 0.5 [m / s ² ], it was possible to show that high-precision analysis is possible.

<Example 3>
Next, a third embodiment in which the attenuation characteristic curve K2 is derived by giving the attenuation rate of the structure to be measured as a known constant will be described.

When the concrete floor board is a structure, the attenuation factor σ in the primary bending vibration in the axial direction is generally between 0.2 and 0.4.
Therefore, an attenuation characteristic curve K2 was derived with an attenuation rate of 0.3. Using this attenuation characteristic curve K2, the amplitude value A (t) of 200 [ms] from the reference time is 1.0 [V], and the amplitude value A (t) of 400 [ms] is 0.2 [V]. I was able to predict.

  Therefore, based on this prediction, the regulator 24a is switched to the regulator 24b at 200 [ms] so that the second voltage 3.3 [V] is supplied. Further, the third voltage 1.8 [V] is supplied by switching to the regulator 24c at 400 [ms].

As a result, after 800 [ms], the resolution is 0.007 [V], and measurement data with an amplitude value of 0.01 [V] can be measured. Since the amplitude value of 0.01 [V] corresponds to the acceleration of 0.5 [m / s ² ], it was possible to show that high-precision analysis is possible.

<Example 4>
Next, a case will be described in which FastFourierVrform (FFT) processing is performed on previously acquired measurement data to derive the natural frequency of the structure. The measurement data is passed through a band-pass filter having the natural frequency as the center frequency, thereby removing unnecessary frequency components such as high-frequency noise. Then, the attenuation characteristic curve K2 was derived using the measurement data from which unnecessary components were removed, and the measurement range was switched using the attenuation characteristic curve K2.

  As a result of performing FFT processing on the measurement data, the natural frequency of the structure (concrete floorboard) was estimated to be 30 Hz. In addition, using the finite element method, the natural frequency of the primary bending vibration in the axial direction of the concrete floor board was 32 Hz. Therefore, unnecessary components were removed from the measurement data using a bandpass filter having a passband of 20 Hz to 40 Hz.

  Using the measurement data obtained in this way, the attenuation characteristic curve K2 was derived with the characteristic deriving time of 100 [ms]. From this attenuation characteristic curve K2, the amplitude value of 200 [ms] was predicted to be 0.9 [V], and the amplitude value of 400 [ms] was predicted to be 0.1 [V].

  Therefore, the regulator 24a is switched to the regulator 24b after 200 [ms] so that the second voltage 3.3 [V] is supplied. Further, after switching to 400 [ms], switching to the regulator 24c is made so that the third voltage 1.8 [V] is supplied.

As a result, after 800 [ms], the resolution is 0.007 [V], and measurement data with an amplitude value of 0.01 [V] can be measured. Since the amplitude value of 0.01 [V] corresponds to the acceleration of 0.5 [m / s ² ], it was possible to show that high-precision analysis is possible.

  While the present invention has been described with reference to the embodiments (and examples), the present invention is not limited to the above embodiments (and examples). Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

2 Measuring Device 10 Vibration Detection Unit 20 Range Switching Processing Unit 21 Analog Filter 22 Microcomputer (MPU)
22a Analog / digital conversion means 22b Attenuation characteristic calculation means 22c Measurement range judgment means 22d Measurement range switching means 23 Output unit 24 (24a to 24c) Regulator 30 Power supply

Claims (10)

A measuring device capable of switching the measurement range,
A vibration detector that detects the vibration of the structure and outputs the detection result as measurement data of an analog signal;
A range switching processing unit that determines and switches the measurement range based on the measurement data, and
The range switching processing unit
Analog / digital conversion means for converting the measurement data of the analog signal into a digital signal in a set measurement range;
An attenuation characteristic calculating means for calculating an attenuation characteristic of the measurement data;
Measurement range determining means for determining the measurement range based on the attenuation characteristics;
And a measurement range switching means for switching the measurement range in accordance with a determination result of the measurement range switching means. The measuring device according to claim 1,
The attenuation characteristic calculating means obtains an envelope of the measurement data and uses it as an attenuation characteristic curve. The measuring device according to claim 1,
The attenuation characteristic calculation means calculates an envelope of average measurement data obtained by performing an averaging process on the measurement data obtained by a plurality of measurements performed in advance prior to the actual measurement as an attenuation characteristic curve. A measuring device. The measuring device according to claim 1,
The measurement apparatus characterized in that the attenuation characteristic calculation means calculates an envelope of the measurement data using an attenuation rate of the structure acquired and set in advance, and calculates this as an attenuation characteristic curve. The measuring device according to claim 1,
The measurement apparatus characterized in that the attenuation characteristic calculation means calculates an attenuation rate using an autoregressive model and derives the attenuation characteristic curve using the attenuation rate. It is a measuring device given in any 1 paragraph of Claims 1 thru / or 5,
The measurement range determination means determines a time at which an amplitude value after a predetermined time is predicted to satisfy a preset range switching condition based on the attenuation rate characteristic, and performs range switching at a timing corresponding to the time. A characteristic measuring device. It is a measuring device of Claim 6, Comprising:
The range switching condition is a reduction rate or a reduction range with respect to a preset maximum amplitude value. The measurement device according to any one of claims 1 to 7,
The measurement apparatus according to claim 1, wherein the measurement range switching means performs range switching by changing a voltage supplied to the analog / digital conversion means. A measurement range switching method for automatically setting the measurement range,
Detects vibration of the structure and outputs it as analog signal measurement data.
The measurement data of the analog signal is converted into a digital signal with the maximum measurement range at the start of measurement,
Calculate the attenuation characteristics using the measurement data converted into a digital signal in the maximum measurement range ,
Determining the measurement range based on the attenuation characteristics;
A measurement range switching method, wherein the measurement data of an analog signal is converted into a digital signal in the determined measurement range. The measurement range switching method according to claim 9, wherein
A measurement range switching method, wherein an envelope of the measurement data is obtained and used as an attenuation characteristic curve.

Images