JP6347078B2 - Deflection measuring method and deflection measuring apparatus - Google Patents

Description

  The present invention relates to a deflection measuring method and a deflection measuring apparatus that dynamically measure the deflection of a paved road surface.

  In recent years, road pavement has shifted from the new construction to the era of maintenance and repair, and a method for quickly and accurately evaluating the soundness of the paved road surface is required. The paved road surface needs to have less damage such as “wad digging” and “cracking”, but in addition to this, the bearing capacity of the paved road surface needs to satisfy the design value. In other words, the paved road surface may be cavities or inequalities in the roadbed, roadbed, or road body due to accidental effects such as damage to the roadbed, roadbed, earthquake motion, or groundwater due to repeated loads associated with vehicle travel. Deformations such as settlement may occur. For this reason, it is necessary to grasp the deformation and maintain the soundness of the paved road surface by measuring the bearing capacity of the paved road surface. For this reason, the soundness evaluation method of the paved road surface which measures the bearing capacity of a paved road surface nondestructively is required.

  An FWD (Falling Weight Deflector) which is an in-vehicle deflection measuring machine is widely used as a non-destructive measuring machine used for evaluating the soundness of a paved road surface. This FWD is a device that drops a weight onto a paved road surface in a stationary state and measures the shape of the road surface that is deformed by the impact load. A load meter that measures the impact load and a displacement meter that measures the deflection amount of the road surface. Based on the measurement results of these measuring instruments, the soundness of the paved road surface is evaluated.

  However, since FWD measures the deflection of the paved road surface for each measurement point, it cannot continuously evaluate the deflection of the paved road surface. Therefore, when evaluating the soundness of the entire paved road surface over a wide area, it is necessary to repeat the measurement for each point, and it is necessary to move, re-install, and measure the deflection of the FWD each time. It takes time and money. In addition, since the measurement by FWD is performed in a stationary state, traffic regulation is necessary, and the burden of road surface management is large. Furthermore, in order to perform a leak-free evaluation, it is necessary to measure the deflection with the measurement points set at narrow intervals, but there is a limit to the measurement, and there is a possibility that local deformation cannot be confirmed. is there.

  Under such circumstances, for example, Patent Document 1 discloses a mobile deflection measuring device as a measuring device that continuously measures the deflection of a paved road surface while moving. This mobile deflection measuring device uses a displacement sensor or Doppler sensor mounted on the vehicle body of a traveling vehicle to measure the distance between the sensor and the road surface, so that the amount of deflection of the paved road surface continuously while moving, etc. Is measured.

  Further, for example, in Non-Patent Documents 1 and 2, when a sensor is attached to a vehicle body and the deflection of a pavement road surface is measured, a device such as attaching a vibration control device to a mount to which the sensor is attached is devised.

Japanese National Patent Publication No. 11-503520

Samer W. Katicha et al., Estimation of Pavement TSD Slope Measurements Repeatability from a Single Measurement Series, TRB 2012 Annual Meeting Soren, Rasmussen et al., A comparison of two years of network level measurements with the Traffic Speed Deflectometer, TRA Europe 2008, Ljubljana

  However, when the vehicle travels on an uneven paved road surface, it is affected by shocking vibrations generated in the vehicle body due to the unevenness. This shocking vibration is converted into vibration having a relatively long period by the seismic isolation effect due to the suspension and weight of the vehicle body and becomes vibration noise. This vibration noise appears differently for each vehicle position depending on the vehicle body and the road environment. Therefore, as disclosed in Patent Document 1, in a method of measuring deflection by installing sensors at a plurality of positions. The influence of vibration at each vehicle position is irregularly included in the measurement result, and it is difficult to remove vibration noise. For this reason, in the mobile deflection measuring device described in Patent Document 1, in evaluating the soundness of a realistic pavement road surface with unevenness, accurate measurement cannot be performed between the sensor and the road surface, which hinders the evaluation results. There is a risk of coming.

  In addition, as disclosed in Non-Patent Documents 1 and 2, in order to mount the vibration damping device on the vehicle, it is necessary to use a special large vehicle, and the construction of the measurement system becomes large. Also, in Japan, there are many complicated roads that such large vehicles cannot enter, so it is not easy to evaluate the soundness of paved road surfaces.

  Furthermore, a hard vibration countermeasure method using a vibration isolating material has been proposed for the sensor mount. However, depending on the characteristics of the vibration isolating material, for example, a soft vibration isolating material may be used. Irregular vibration of frequency will occur. When such irregular vibration occurs, the beam irradiation direction of the Doppler vibrometer provided on the sensor mount becomes unstable, and the accurate deflection speed cannot be measured.

  The present invention has been made in view of such circumstances, and a deflection measurement method and a deflection that can improve the accuracy of deflection measurement by removing the vehicle vibration speed of the loading vehicle from the measured deflection speed. It aims at providing a measuring device.

  (1) In order to achieve the above object, the present invention takes the following measures. In other words, the deflection measurement method of the present invention is a deflection measurement method for dynamically measuring the deflection of a pavement road surface, wherein a sensor frame is rigidly connected to the vehicle body, and a laser Doppler vibrometer and a vibration accelerometer are installed on the sensor frame. A step of running the loaded vehicle, and a deflection speed of each measurement point is obtained by the laser Doppler vibrometer while the loading vehicle is running, and a vibration acceleration of each measurement point is obtained by the vibration accelerometer. A step of performing discrete wavelet transform on a vehicle vibration waveform indicating vibration acceleration at each measurement point to determine a filter that substantially flattens the vehicle vibration waveform; and Removing the vehicle vibration speed from the deflection speed of the point.

  As described above, since the sensor mount is rigidly connected to the vehicle body of the loading vehicle, the influence of vibration generated from the vibration-proof material is eliminated when the vibration-proof structure is adopted, and the vehicle vibration speed can be directly acquired. The vehicle vibration waveform indicating the vibration acceleration at each measurement point is subjected to discrete wavelet transform to determine a filter that substantially flattens the vehicle vibration waveform. Using this filter, the vehicle is determined from the deflection speed at each measurement point. Since the vibration speed is removed, it is possible to remove the influence of the vehicle running vibration without adopting the vibration isolation structure, and it is possible to improve the accuracy of deflection measurement.

(2) In addition, the deflection measurement method of the present invention uses the wavelet filter and the scaling filter to calculate the wavelet coefficient w _{i, t} expressed by Equation (1) and the scaling coefficient v _{i, t} expressed by Equation (2). The discrete wavelet transform is performed by calculating, and the wavelet inverse transform is performed by calculating the scaling coefficient v _{i, t represented} by Equation (3), and the vehicle vibration speed is removed from the deflection speed of each measurement point. It is characterized by.

  In this way, by performing discrete wavelet transform and inverse wavelet transform, the vehicle vibration speed is removed from the deflection speed at each measurement point, so that it is possible to remove the influence of vehicle running vibration without adopting a vibration isolation structure. It is possible to improve the accuracy of deflection measurement.

  (3) Further, the deflection measuring method of the present invention defines a local coordinate system having the origin directly below the wheel of the cargo vehicle as described above, the traveling direction and loading direction of the cargo vehicle as described above as positive directions, and the deflection amount. w, the travel speed of the cargo vehicle described above is V, the distance is x, the time is t, the deflection angle is dw / dx, the maximum deflection is a, the coefficient for the deflection is b, and the maximum deflection occurs The position is set as d, the deflection angle represented by Formula (5) is calculated from Formula (4), Formula (7) is calculated from this deflection angle and Formula (6), c is set to 2abV, and g (x) is set to dw / Dx is taken as the logarithm of both sides of Equation (7) to obtain Equation (8), and the evaluation function J (c, b, d) is set as shown in Equation (9) using the least square method. D when the evaluation function J (c, b, d) is minimized, a and the previous By determining the b, and calculates the deflection shape and maximum deflection amount.

  Thus, by calculating d, a, and b when the evaluation function J (c, b, d) is minimized, the deflection shape and the maximum deflection amount are calculated. The influence of vehicle running vibration can be removed, and the accuracy of deflection measurement can be improved.

  (4) Further, in the deflection measuring method of the present invention, a laser displacement meter gantry is provided in the cargo vehicle described above, and the laser displacement meter gantry includes a plurality of parallelly arranged in the traveling direction of the cargo vehicle described above. The laser displacement meter measures the distance between each laser displacement meter and the paved road surface while the cargo vehicle is traveling.

  As described above, a plurality of laser displacement meters arranged in parallel in the traveling direction of the loading vehicle of the laser displacement meter gantry, the distance between each laser displacement meter and the pavement surface during traveling of the loading vehicle can be determined. Since the measurement is performed, it is possible to eliminate the influence of the vehicle running vibration without adopting the vibration isolation structure, and it is possible to improve the accuracy of the deflection measurement.

  (5) Further, the deflection measuring device of the present invention is a deflection measuring device that dynamically measures the deflection of the pavement surface, and is installed on the sensor platform rigidly connected to the vehicle body of the loading vehicle. A laser Doppler vibrometer and a vibration accelerometer, wherein the laser Doppler vibrometer acquires a deflection speed at each measurement point while the cargo vehicle is traveling, and the vibration accelerometer vibrates at each measurement point. The vehicle vibration waveform indicating the vibration acceleration at each measurement point is subjected to discrete wavelet transform to determine a filter that substantially flattens the vehicle vibration waveform, and each measurement point is measured using the filter. The vehicle vibration speed is removed from the deflection speed of the vehicle.

  As described above, since the sensor mount is rigidly connected to the vehicle body of the loading vehicle, the influence of vibration generated from the vibration-proof material is eliminated when the vibration-proof structure is adopted, and the vehicle vibration speed can be directly acquired. The vehicle vibration waveform indicating the vibration acceleration at each measurement point is subjected to discrete wavelet transform to determine a filter that substantially flattens the vehicle vibration waveform. Using this filter, the vehicle is determined from the deflection speed at each measurement point. Since the vibration speed is removed, it is possible to remove the influence of the vehicle running vibration without adopting the vibration isolation structure, and it is possible to improve the accuracy of deflection measurement.

  (6) Moreover, the deflection measuring apparatus of the present invention further includes a laser surface velocimeter for measuring the traveling speed of the cargo vehicle described above.

  As described above, since the laser surface speed meter for measuring the traveling speed of the loading vehicle is provided, it is possible to measure the vehicle traveling speed of the loading vehicle necessary for measuring the deflection speed.

  (7) Moreover, the deflection measuring apparatus of the present invention includes a plurality of laser displacement meter mounts provided in the load vehicle, and a plurality of the displacement measurement devices arranged in parallel in a traveling direction of the load vehicle described above. The laser displacement meter is further provided, and the distance between the laser displacement meter and the paved road surface is measured by the laser displacement meter while the cargo vehicle is traveling.

  As described above, a plurality of laser displacement meters arranged in parallel in the traveling direction of the loading vehicle of the laser displacement meter gantry, the distance between each laser displacement meter and the pavement surface during traveling of the loading vehicle can be determined. Since the measurement is performed, it is possible to eliminate the influence of the vehicle running vibration without adopting the vibration isolation structure, and it is possible to improve the accuracy of the deflection measurement.

  According to the present invention, since the sensor mount is rigidly connected to the vehicle body of the loading vehicle, the vibration effect generated from the vibration-proof material is eliminated when the vibration-proof structure is adopted, and the vehicle vibration speed can be directly acquired. . In addition, it is possible to remove the influence of the vehicle running vibration without adopting the vibration isolation structure, and to improve the accuracy of the deflection measurement.

It is a figure which shows the concept of wavelet analysis (Kohei Arai: Basic theory of wavelet analysis, extracted from Morikita Publishing Co., Ltd., 2005). It is a figure which shows the concept of Fourier analysis (Kohei Arai: Basic theory of wavelet analysis, Morikita Publishing Co., Ltd., extracted from 2005). It is a diagram showing true data and data including noise (Excerpt from Shoichi Inada, Koichiro Kamada: Economic Analysis by Wavelet, Financial Research, pp.1-62, 2004). It is a figure which shows the data after the noise removal by Fourier analysis, and the data after the noise removal by wavelet analysis (Shoichi Inada, Koichiro Kamada: Economic analysis by wavelet, financial research, pp.1-62, 2004 excerpt). It is a figure which shows a scaling filter (left) and a wavelet filter (right) of Haar (Excerpt from Shoichi Inada, Koichiro Kamada: Economic analysis by wavelet, financial research, pp.1-62,2004). It is a figure which shows the scaling filter (left) and wavelet filter (right) of D (4) (Seiichi Inada, Koichiro Kamada: Excerpt from economic analysis by wavelet, financial research, pp.1-62,2004). It is a figure which shows the scaling filter (left) and wavelet filter (right) of D (12) (Seiichi Inada, Koichiro Kamata: Economic analysis by wavelet, financial research, pp.1-62, 2004 excerpt). It is a figure which shows the concrete numerical value of D (4) and D (12) (Excerpt from Shoichi Inada, Koichiro Kamada: Economic analysis by wavelet, financial research, pp.1-62,2004). It is a figure which shows the pyramid algorithm of a discrete wavelet transform of Haar. It is a figure which shows the observed vibration waveform. It is a figure which shows distribution of the wavelet coefficient in J = 1. It is a figure which shows distribution of the wavelet coefficient in J = 2. It is a figure which shows distribution of the wavelet coefficient in J = 3. It is a figure which shows distribution of the wavelet coefficient in J = 4. It is a figure which shows distribution of the wavelet coefficient in J = 5. It is a figure which shows distribution of the wavelet coefficient in J = 6. It is a figure which shows distribution of the wavelet coefficient in J = 7. It is a figure which shows distribution of the wavelet coefficient in J = 8. It is a figure which shows distribution of the wavelet coefficient in J = 9. It is a figure which shows distribution of the wavelet coefficient in J = 10. It is a figure which shows the wavelet inverse transformation result which made the wavelet coefficient of J = 1-5 0. It is a figure which shows the concept of the deflection speed measuring apparatus by a laser Doppler vibrometer. It is a figure which shows vibration acceleration measurement data and the data after wavelet analysis. It is a figure which shows the data after a deflection velocity measurement data and wavelet analysis. It is a figure which shows the mode of a laser irradiation position and the measurement of a laser irradiation angle. It is a figure which shows the laser irradiation position and angle of a laser Doppler vibrometer. It is a figure which shows the influence of the speed by the inclination of a laser Doppler vibrometer. It is a figure which shows the distance from a sensor mount (horizontal shaft) to a road surface in case a road surface state differs. It is a figure which defines a local coordinate system and shows the running direction and loading direction of a vehicle. It is a figure which shows the measurement conditions of a driving | running | working experiment. It is a figure which shows the vibration acceleration of the sensor mount with respect to the travel distance when a travel speed is 10 km / h. It is a figure which shows the bending speed of the paved road surface with respect to the travel distance when a travel speed is 10 km / h. It is a figure which shows the bending speed of the pavement surface with respect to the travel distance after a wavelet analysis in case a travel speed is 10 km / h. It is a figure which shows the vibration acceleration of the sensor mount with respect to the travel distance in case a travel speed is 30 km / h. It is a figure which shows the bending speed of the paved road surface with respect to the travel distance when a travel speed is 30 km / h. It is a figure which shows the bending speed of the pavement surface with respect to the travel distance after a wavelet analysis in case a travel speed is 30 km / h. It is a figure which shows the vibration acceleration of the sensor mount with respect to the travel distance when a travel speed is 50 km / h. It is a figure which shows the bending speed of the paved road surface with respect to the travel distance when a travel speed is 50 km / h. It is a figure which shows the bending speed of the pavement surface with respect to the travel distance after a wavelet analysis in case a travel speed is 50 km / h. It is a figure which shows the comparison result of MWD deflection and FWD deflection in case traveling speed is 10 km / h. It is a figure which shows the comparison result of MWD deflection and FWD deflection in case traveling speed is 30 km / h. It is a figure which shows the comparison result of MWD deflection and FWD deflection in case traveling speed is 50 km / h. It is a figure which shows schematic structure of the modification 1 of this invention. It is explanatory drawing which shows the outline | summary of the measurement principle of the deflection amount which concerns on the modification 2. FIG. 10 is a schematic diagram showing an outline of a deflection measuring device according to Modification 2. FIG. It is a schematic diagram which shows the arrangement | positioning condition of each laser displacement meter. It is a schematic diagram which shows the state in which the sensor mount was inclined. It is a figure which shows the example of a bending condition and its measurement result.

  The present inventors have attached a sensor mount having a high-precision laser Doppler vibrometer to a loading vehicle, and can measure the deflection speed of a paved road surface non-destructively while running the loading vehicle. (MWD: Moving Wheel Deflectometer) “Research for hardware vibrations for sensor mounts, focusing on the fact that anti-vibration materials cause irregular vibrations in the sensor mounts. It is possible to improve the accuracy of deflection measurement by directly acquiring the vehicle vibration velocity by rigidly connecting to the vehicle and removing the vehicle vibration velocity from the deflection velocity measured by the laser Doppler vibrometer using discrete wavelet analysis. And led to the present invention.

  In other words, the deflection measurement method of the present invention is a deflection measurement method for dynamically measuring the deflection of a pavement road surface, wherein a sensor frame is rigidly connected to the vehicle body, and a laser Doppler vibrometer and a vibration accelerometer are installed on the sensor frame. A step of running the loaded vehicle, and a deflection speed of each measurement point is obtained by the laser Doppler vibrometer while the loading vehicle is running, and a vibration acceleration of each measurement point is obtained by the vibration accelerometer. A step of performing discrete wavelet transform on a vehicle vibration waveform indicating vibration acceleration at each measurement point to determine a filter that substantially flattens the vehicle vibration waveform; and Removing the vehicle vibration speed from the deflection speed of the point.

  As a result, the present inventors can eliminate the influence of vehicle running vibration without adopting a vibration isolation structure, and as a result, can improve the accuracy of deflection measurement. Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings.

<Soft vibration countermeasures by time series frequency analysis>
(A) Reasons for Applying Discrete Wavelet Analysis Time series frequency analysis includes Fourier analysis that combines Fourier transform and inverse transform, and wavelet analysis that combines wavelet transform and inverse transform. In this specification, of the wavelet analysis, a soft vibration countermeasure is performed using discrete wavelet analysis that handles discrete data.

  FIG. 1A is a diagram showing the concept of wavelet analysis, and FIG. 1B is a diagram showing the concept of Fourier analysis (Keihei Arai: Basic theory of wavelet analysis, extracted from Morikita Publishing Co., Ltd., 2005). In the wavelet analysis, as shown in FIG. 1A, arbitrary time-series data is expressed as a sum of wavelets as “small waves expressed by time and frequency”. On the other hand, in the Fourier analysis, as shown in FIG. 1B, arbitrary time-series data is expressed as a sum of infinite permanent waves having different frequency components.

  Both of them are not always superior. For example, for steady data whose time-series characteristics are invariant over time and repeats regular fluctuations, ignore the temporal information and details It is thought that the Fourier analysis which performs a simple frequency analysis is efficient.

  On the other hand, since the wavelet transform analyzes data as a collection of short waves, it is effective for data showing irregular fluctuations. Conversely, it is not necessary to apply wavelet transform to regular data. Wavelet analysis includes continuous wavelet analysis and discrete wavelet analysis. In continuous wavelet analysis, data is analyzed as continuous data, whereas in discrete wavelet analysis, data is discretized and analyzed, so that time-frequency analysis can be performed more efficiently than continuous wavelet analysis. In discrete wavelet analysis, the original signal is decomposed into a high-frequency component and a low-frequency component, and the decomposed low-frequency component is further decomposed into a high-frequency component and a low-frequency component. The original signal is reproduced by performing reverse conversion (inverse conversion). Therefore, if the frequency component of an arbitrary region is set to 0 in the inverse transformation, a signal from which the frequency component of that region has been removed (filtered) is generated.

  Such filtering processing is also possible in Fourier analysis, and conventionally Fourier analysis has been widely used. However, the vibration data targeted by the present invention is irregular vibration data of the vehicle caused by unevenness of the paved road surface.

  FIG. 2A is a diagram showing an example of observation data including true data and noise (Excerpt from Shoichi Inada, Koichiro Kamada: Economic Analysis by Wavelet, Financial Research, pp. 1-62, 2004). As shown on the left side of FIG. 2A, it is assumed that there is a step signal that takes a peak value of 2.0 at a point where the coordinate of the horizontal axis is 0.5, and this is the true data. On the other hand, as shown on the right side of FIG. 2A, it is assumed that there is data mixed with noise, and this is actually observed data. It tries to extract true data by removing noise from this observation data. At this time, when Fourier analysis is performed on the observation data shown on the right side of Fig. 2A, the left side of Fig. 2B (Excerpt from Shoichi Inada, Koichiro Kamada: Economic Analysis by Wavelet, Financial Research, pp.1-62, 2004) As shown in the figure, a gentle mountain-shaped signal is extracted. On the other hand, when wavelet analysis is performed on the observation data shown on the right side of FIG. 2A, a signal close to true data can be extracted fairly accurately as shown on the right side of FIG. 2B. This is because in wavelet analysis, filters corresponding to various waveforms are prepared. From this, it can be seen that wavelet analysis is suitable for performing filtering processing on data including irregular noise.

  The vibration data targeted by the present invention is irregular vibration data of the vehicle caused by unevenness of the paved road surface, etc., and knows time information of vibration characteristics, that is, how much vehicle vibration has occurred at which position. Rather, it is more important how quickly noise caused by vehicle vibration can be removed. That is, it is considered that discrete wavelet analysis capable of performing time-frequency analysis more efficiently than continuous wavelet analysis is suitable for removing noise in deflection measurement.

(B) Discrete wavelet analysis (B-1) Wavelet filter and scaling filter In the discrete wavelet analysis used in this specification, discrete wavelet transform is performed on measurement data, and unnecessary vibration frequency components (noise) are set to zero. Later, a signal from which noise has been removed is generated by inverse transformation. That is, the wavelet transformation is a kind of linear filtering, and the wavelet used for the transformation is called a “wavelet filter”. The wavelet filter is generally represented by h and must satisfy the following three conditions.

  In the above formulas (10) to (12), L is a filter length (referred to as a support length). In other words, Equation (10) means that when all the filter elements are added, it becomes zero, and means that “nothing is added or nothing” is added to the original sequence by filtering. Further, in the formula (11), the filter energy (sum of squares) is standardized to 1. Finally, equation (12) means that the even-shifted filter and the original filter are orthogonal. Formula (11) indicates that the inner product when the shift is not performed is 1, and together with Formula (12), it is referred to as “regularity condition”.

  In order to execute the wavelet transform, a scaling filter g that is paired with the wavelet filter is required. Both are connected by the following relational expression called orthogonal mirror image relation.

From Equation (13), it can be confirmed that the scaling filter satisfies the following relational expression.

  Further, the following relationship is established between the wavelet filter and the scaling filter. That is, two pairs of filters are orthogonal to each other with respect to an even multiple shift.

(B-2) Various Wavelets As widely used wavelet filters, there are Haar and Daubechies wavelets. The Haar wavelet is a filter having a support length of L = 2, and a wavelet filter and a scaling filter can be obtained by the above formulas (10) to (14), and is represented by the formula (18).

  Fig. 3 shows Haar's scaling filter and wavelet filter (excerpted from Shoichi Inada and Koichiro Kamada: Economic Analysis with Wavelets, Financial Research, pp. 1-62, 2004). The left side of FIG. 3 is a Haar scaling filter, and the right side of FIG. 3 is a Haar wavelet filter.

  The Daubechies wavelet is determined so as to satisfy the following equation (19) in addition to the equations (10) to (12). In general, a Daubechies wavelet with a support length of L is denoted as D (L), and a Daubechies wavelet with L = 4 and 12 is used.

  Fig. 4A shows the scaling filter and wavelet filter of D (4) (Excerpted from Shoichi Inada and Koichiro Kamada: Economic Analysis by Wavelet, Financial Research, pp. 1-62, 2004). The left side of FIG. 4A is a scaling filter of D (4), and the right side of FIG. 3 is a wavelet filter of D (4). FIG. 4B shows the scaling filter and wavelet filter of D (12) (Excerpted from Shoichi Inada and Koichiro Kamada: Economic Analysis by Wavelet, Financial Research, pp. 1-62, 2004). The left side of FIG. 4B is a D (12) scaling filter, and the right side of FIG. 4B is a D (12) wavelet filter. FIG. 4C is a diagram showing specific numerical values of the scaling filter and wavelet filter of D (4) and D (12) (Inada Shoichi, Kamada Koichiro: Economic Analysis by Wavelet, Financial Research, pp.1- (Extracted from 62,2004).

(B-3) Wavelet Transform / Inverse Transform In discrete wavelet transform, “wavelet coefficient” and “scaling coefficient” are obtained by passing observation data through a wavelet filter and a scaling filter. As shown in FIGS. 3, 4A, and 4B, the scaling filter has a convex shape, whereas the wavelet filter has a vibrational uneven shape. That is, the scaling filter represents a moving average component (data trend), whereas the wavelet filter represents a difference component (deviation from the trend), in other words, a vibration component. And the information of the vibration component and the trend component included in the observation data is stored by the coefficients obtained through these filters.

FIG. 5 is a diagram showing a pyramid algorithm. In the discrete wavelet transform, calculation is efficiently performed by the pyramid algorithm shown in FIG. FIG. 5 shows an example of calculation using Haar wavelets. In this calculation method, when the number of data is N, filtering is performed on two discrete data that do not overlap each other in the first calculation to obtain N / 2 wavelet coefficients and scaling coefficients. Next, (N / 2) / 2 = (N / 2) ² wavelet coefficients and scaling coefficients are obtained using the N / 2 scaling coefficients obtained in the first calculation. Finally, this operation is repeated until a set of wavelet coefficients and scaling coefficients is obtained.

This is referred to as downsampling. At this time, the number of data required is N = ²ⁿ , and n repetitive calculations are performed. For example, if the data of 2 ¹⁸ (262144), by 18 times about repetitive operation can be efficiently obtained wavelet coefficients and scaling coefficients. If the observed discrete data is X = (x ₁ , x ₂ ,... X _N−1 , x _N ), the wavelet coefficient and scaling coefficient of the first calculation are obtained by the following equations.

Here, “A mod N” is an operator that subtracts (or adds) a multiple of N from A to create a number from 0 to N−1. Further, as described above, the second and subsequent calculations perform wavelet transform using the scaling coefficient obtained in the previous calculation. That is, in the i-th calculation, the (i−1) -th scaling coefficient is used. Therefore, when the equations (20) and (21) are generalized and displayed, they are as follows.

In Equations (22) and (23), “v _{0, N} = x _N ”.

In the wavelet inverse transform, the mathematical formulas (20) to (23) are calculated in the reverse direction. Now, the level of (i + 1) wavelet coefficients w _{i + 1} and the scaling factor v _{i + 1} of, if reconstruct the scaling factor v _i level i, in order to supplement the number of data reduced by downsampling, the "0" w _{i + 1} And v _{i + 1} data. That is, the data shown by the mathematical formulas (24) and (25) is created. This is called upsampling.

When the data up-sampled in this way is passed through a wavelet filter and a scaling filter, a scaling factor v _{i of} level _i is obtained as shown in Equation (26).

  As described above, since the wavelet filter is oscillatory, it can be used as a path filter by setting an arbitrary wavelet coefficient to 0 during inverse transformation.

(C) Example of discrete wavelet analysis FIG. 6A is a diagram illustrating an observed vibration waveform. As an example, the wavelet coefficient of the vibration waveform shown in FIG. 6A is obtained. Note that the number of data is 2 ¹⁴ (16384), and Daubechies wavelet D (12) is used.

As described above, in the discrete wavelet transform, downsampling is performed according to a pyramid algorithm. Number of data shown in FIG. ^6A, since it is ^{N = 2} J = 2 ^14, repetition number of calculations becomes 14 times. The wavelet coefficients obtained by the discrete wavelet transform of J = 1 to 10 are shown in FIGS. 6B to 6K. From these results, it can be seen that when downsampling is repeated, the vibration characteristic of the wavelet coefficient becomes gentle and becomes a long frequency component.

  Further, according to the results shown in FIGS. 6B to 6K, a large irregular vibration component appearing in the vicinity of the horizontal axis 400 m or 1100 m in FIG. 6A is inversely transformed with the wavelet coefficients of J = 1 to 5 being 0 (filtering). ), The result shown in FIG. 6L is obtained. When FIG. 6L is compared with FIG. 6A, it can be seen that large irregular vibrations are excluded and comparatively smooth signals can be extracted.

  From the above results, it can be seen that noise can be easily removed by discrete wavelet analysis. In this specification, such a procedure is adopted, and the irregular vibration component included in the measurement result of the laser Doppler vibrometer is removed.

(D) Software Vibration Countermeasure by Discrete Wavelet Analysis FIG. 7 is a diagram showing an outline of a deflection velocity measuring apparatus using a laser Doppler vibrometer. As shown in FIG. 7, the laser Doppler vibrometer according to the present embodiment is attached to a sensor mount installed on a vehicle body near the rear wheel of the loading vehicle. Since the laser Doppler vibrometer measures the relative speed between the paved road surface and the vehicle body, the measurement data at the time of the running experiment includes the vehicle vibration speed V _v associated with road surface unevenness and the deflection speed V _def associated with loading. Both will be included. Therefore, in this embodiment, a vibration accelerometer is installed on the sensor gantry, a discrete wavelet analysis is performed on the vibration waveform of the accelerometer, and a filter that makes the vehicle vibration waveform substantially flat as shown in FIG. 8A. (The upper limit value of J that makes the wavelet coefficient 0) was determined by trial and error. FIG. 8A is a diagram showing vibration acceleration measurement data and data after wavelet analysis. In FIG. 8A, black lines are measurement data, and white lines near the center are data after filtering.

The deflection speed obtained by the wavelet inverse transform is equal to the deflection speed V _def obtained by loading the rear wheel, using the same level filter as the vibration acceleration for the deflection speed measurement data. The maximum deflection was sought.

  In the present embodiment, analysis was performed using a Daubechies wavelet. The reason why the Daubechies wavelet was selected is that, as shown in FIG. 4B, although it is not a symmetric filter, almost the same result as that obtained when using a Coifflet wavelet close to the symmetric shape devised by Daubechies is obtained. This is because its usefulness is widely recognized in analysis.

  FIG. 8B is a diagram illustrating an example of measurement data and filtered data. In FIG. 8B, black lines are measurement data, and white lines are data after filtering. As can be seen from FIG. 8B, even when the filtering processing is performed at a level such that the measurement data obtained by the vibration accelerometer becomes zero, the trend of the deflection speed is not lost, and a large vibration component is removed. From the above, it can be seen that the discrete wavelet analysis is effective as a countermeasure against soft vibration of the sensor mount.

  In the present embodiment, the sensor mount is rigidly connected to the vehicle body of the loading vehicle. `` Rigid connection '' means that the sensor mount and the vehicle body are joined together so as to be a rigid body, for example, various methods such as fastening with bolts and nuts, fitting, welding, adhesion, etc. can be taken, The method is not limited. In carrying out the running experiment using the loading vehicle, the laser irradiation position and the irradiation angle of the laser Doppler vibrometer were determined as follows. The measurement was performed on the steel plate floor in the large universal testing machine hangar in the civil engineering research institute. Since the steel plate is extremely rigid, it was assumed that no deflection occurs on the floor.

  FIG. 9 is a diagram showing how the laser irradiation position and angle are measured. As shown in FIG. 9, with the red laser emitted from the laser Doppler vibrometer, the distance L (laser irradiation position) from the axle center to the laser irradiation position, the height H from the left end of the laser Doppler vibrometer, on the floor The distance B from the left end of the laser Doppler vibrometer to the laser irradiation position was measured and determined. Note that the axle center position and the left end position of the laser Doppler vibrometer on the floor are positioned using a survey pendulum, and the laser irradiation angle θ uses “b” and “d” shown in FIG. And calculated by the equation (27).

  As shown in FIG. 9, when three laser Doppler vibrometers were provided, measurement results as shown in FIG. 10 were obtained. When the laser Doppler vibrometer is installed to be inclined with respect to the vertical direction, the deflection speed dwm / dt to be measured is affected by the vehicle running speed V as shown in FIG. Bigger than. Therefore, correction is performed using the traveling speed V measured by the vehicle speedometer (laser surface speedometer) according to the equation (28).

Next, the maximum deflection estimation method in this embodiment will be described. FIG. 12 is a diagram illustrating an example of the distance from the sensor mount (horizontal shaft) to the road surface when the road surface state is different. In the present embodiment, the deflection speed of the paved road surface is measured using three laser Doppler vibrometers. Here, in the case shown in FIG. 12A, since the road surface is parallel to the shaft, the distance from the flat shaft to the road surface is equal, and d ₁ = d ₂ = d ₃ . On the other hand, in the case shown in FIG. 12B, since the road surface is uneven, the distance from the horizontal shaft to the road surface varies depending on the location. That is, d ₄ ≠ d ₅ ≠ d ₆ . However, when the horizontal shaft moves up and down in the same manner as in FIGS. 12A and 12B and the speeds are equal, Expression (29) is satisfied regardless of the unevenness of the road surface.

Here, dd _i / dt indicates the relative speed between the horizontal shaft and the road surface. This relative speed is equal to the deflection speed measured by the laser Doppler vibrometer when there is no vehicle vibration. However, since the vehicle vibrates during traveling, assuming that the vertical speed of the horizontal shaft is dv / dt, the deflection speed dw / dt of the pavement surface measured by the laser Doppler vibrometer at this time is given by Equation (30). It becomes as follows.

  Therefore, by subtracting the vehicle vibration speed dv / dt from the deflection speed measured by the laser Doppler vibrometer, the deflection speed associated with loading can be obtained.

  Next, the deflection shape due to loading is approximated by an exponential function shown in Expression (12). It is known that when the vehicle speed increases, the wheel position does not coincide with the maximum deflection occurrence position. Therefore, in Equation (31), calculation is performed in consideration of this fact by introducing the parameter d. In the present embodiment, as shown in FIG. 13, a local coordinate system with the origin directly below the wheel is defined. In this local coordinate system, the traveling direction and loading direction of the loading vehicle are set as positive directions.

Let w be the amount of deflection.
Let V be the traveling speed of the loading vehicle,
Let x be the distance
Let time be t,
The deflection angle is dw / dx,
The maximum deflection is a,
The coefficient relating to the deflection shape is b,
Assuming that the maximum deflection occurrence position is d, the deflection angle, the deflection speed, and the vehicle traveling speed are as follows.

When the equation (31) is differentiated by the distance x to obtain the deflection angle, the equation (33) is obtained. Substituting equation (33) into equation (32) yields equation (34), where
c is 2abV,
Let g (x) be dw / dx,
Taking the logarithm of both sides of Equation (34), Equation (35) is obtained.

  Using the least square method, the evaluation function J (c, b, d) is set as shown in Equation (36), and d, a, and b when the evaluation function J (c, b, d) is minimized Using the equations (37) and (38), the deflection shape and the maximum deflection amount are calculated.

However, in Formula (37), X and Y are as follows.
However, in Formula (38), P and Q are as follows.

  The present inventors conducted a running experiment of a “Dynamic Deflection Measuring Device (MWD)”. In conducting the running experiment, measurement conditions were set as shown in FIG. Data measurement uses a multi-recorder (Tokyo Sokki Kenkyujo Co., Ltd., TMR-200) to synchronize the laser Doppler vibrometer, laser surface velocimeter, and vibration accelerometer, and send measurement data to the computer in real time via a USB cable. I took it in. In this running experiment, the loading vehicle was run at a speed of 10 km / h, 30 km / h, and 50 km / h, and the deflection speed of the paved road surface was measured. Discrete wavelet transform / inverse transform was performed on the measured data as described above to remove noise caused by vehicle vibration. 15A to 15I are diagrams showing the results. FIG. 15A shows the vibration acceleration measured by the vibration accelerometer when the traveling speed is 10 km. FIG. 15B shows a laser Doppler vibrometer No. when the traveling speed is 10 km. 1-No. 3 shows the deflection rate measured. Laser Doppler Vibrometer No. 1-No. 3, the irradiation position L is 25.6 cm, 41.2 cm, and 66.7 cm, respectively. In FIG. 15B, the measured deflection speed is shown in a state where it cannot be distinguished by each laser Doppler vibrometer. FIG. 15C shows the deflection speed after wavelet analysis when the traveling speed is 10 km. At the positions of 41.2 cm and 66.7 cm, almost the same waveform is obtained. In both cases, the vehicle vibration acceleration, which is noise, is removed.

  FIG. 15D shows the vibration acceleration measured by the vibration accelerometer when the traveling speed is 30 km. FIG. 15E shows a laser Doppler vibrometer No. when the traveling speed is 30 km. 1-No. 3 shows the deflection rate measured. In FIG. 15E, the measured deflection speed is shown in a state where it cannot be distinguished by each laser Doppler vibrometer. FIG. 15F shows the deflection speed after wavelet analysis when the traveling speed is 30 km. At the positions of 41.2 cm and 66.7 cm, almost the same waveform is obtained. In both cases, the vehicle vibration acceleration, which is noise, is removed.

  FIG. 15G shows the vibration acceleration measured by the vibration accelerometer when the traveling speed is 50 km. FIG. 15H shows a laser Doppler vibrometer No. when the traveling speed is 50 km. 1-No. 3 shows the deflection rate measured. In FIG. 15H, the measured deflection speed is shown in a state where it cannot be distinguished by each laser Doppler vibrometer. FIG. 15I shows the deflection speed after wavelet analysis when the traveling speed is 50 km. At the positions of 41.2 cm and 66.7 cm, almost the same waveform is obtained. In both cases, the vehicle vibration acceleration, which is noise, is removed.

  In this embodiment, the wavelet coefficient (filter level) at the time of wavelet inverse transformation is set so that the vibration acceleration of the sensor mount can be almost removed, and the deflection speed noise is removed by the filter of the same level. In addition, the deflection speed at each traveling speed is the one before the speed correction is performed by Expression (28), and one time out of the traveling experiments at each traveling speed is shown.

<Maximum deflection analysis result>
As shown in FIGS. 15A to 15I, noise was removed by discrete wavelet analysis, the maximum deflection was calculated for each traveling speed, and compared with the FWD test results. The results are shown in FIGS. 16A to 16C. In addition, the legend 2012 of each figure shows the MWD deflection in the same place measured in 2012, 2013 (1) and 2013 (2) show the MWD deflection measured in 2013.

  Comparing these results, by removing noise by discrete wavelet analysis, the estimation system for maximum deflection in 2013 is much improved over that in 2012, and the FWD deflection and MWD deflection are roughly consistent. It can be seen that the reproducibility of the driving experiment is also high. In particular, the places where the supporting force is reduced over a relatively wide range around 500 m and around 1350 m, that is, the places where the FWD deflection is large have the same value as the FWD regardless of the traveling speed.

  As a result, it was found that the MWD according to this example can obtain measurement accuracy comparable to that of FWD.

<Modification 1>
Next, a modified example in the case where a laser Doppler vibrometer and a laser displacement meter are used in combination will be described. FIG. 17 is a diagram illustrating a schematic configuration of the first modification. In Modification 1, a laser displacement meter base 121 is rigidly connected to a vehicle body of a loading vehicle. A plurality of laser displacement meters 120 are arranged in parallel in the vehicle traveling direction of the laser displacement meter mount 121. In FIG. 17, the mounting position of the laser Doppler vibrometer 3 is the rear end of the vehicle body, but this does not affect the essence of the present invention.

  The loading vehicle is made to travel in such a configuration, and each laser displacement meter 120 measures the distance from the paved road surface. Since the laser displacement meter base 121 is rigidly connected to the vehicle body, the measurement data includes vehicle running vibration. As described above, the vibration removal is performed by the time-series frequency analysis in the present invention. . More specifically, as described above, at a low speed, the deflection speed is obtained by the laser Doppler vibrometer 3, while at a high speed, the displacement is obtained by the laser displacement meter 120. Then, from the obtained deflection speed and displacement, the vehicle running vibration during vehicle running is softly removed by wavelet analysis to obtain the deflection amount of the paved road surface. As a result, it is possible to eliminate the influence of the vehicle running vibration without adopting the vibration isolation structure, and to improve the accuracy of deflection measurement.

<Modification 2>
Next, a second modification of the present embodiment will be described. In the second modification, vibration is prevented by attaching a laser displacement meter to the axle. FIG. 18A is a diagram illustrating a measurement principle of the amount of deflection of a paved road surface according to a modification. In FIG. 18A, it is assumed that the vehicle travels at a speed v in the x-axis direction. R represents the radius of the wheel. Considering an ideal situation where there is no unevenness or inclination on the paved road surface, and there is no vibration or inclination of the car body itself, the deflection just below the axle is given by the following equation.

w ₀ = Δ ₀ −Δ _d
In the above equation, w ₀ indicates the deflection of the paved road surface below the axle (x = 0) when the vehicle body is loaded with weight, and Δ ₀ is the y-axis direction orthogonal to the x axis when the vehicle body is loaded with weight. It indicates the distance between the laser displacement meter and pavement, the delta _d, between the y-axis direction of the laser displacement meter and pavement at a position apart in the x-axis direction from the axle position (x = d) Indicates the distance. Δ _d indicates a reference distance in the deflection measurement, and the deflection amount w ₀ can be measured by taking a difference from the distance Δ ₀ including the deflection amount of the paved road surface caused by the weight load. .

  18B and 18C are diagrams showing a second modification of the present embodiment. As shown in FIG. 18B, a deflection measuring device 10 in which a plurality of laser displacement meters are connected to a sensor base is attached to a rear wheel 110 of the vehicle 100. As shown in FIG. 18C, laser deflection meters D <b> 1 to D <b> 7 are supported on the deflection measuring machine 10 on the sensor mount so as to be arranged in parallel in the traveling direction of the vehicle 100. In FIG. 18B, the laser Doppler vibrometer is omitted.

  The laser displacement meter D1 is provided at the axle position of the rear wheel 110 in the traveling direction of the vehicle 100, and measures the distance (first distance) from the loading direction orthogonal to the traveling direction. Further, the laser displacement meter D2 is provided at a position away from the laser displacement meter D1 by a predetermined distance in the traveling direction with respect to the radius R of the wheel attached to the axle. This laser displacement meter D2 plays the role of measuring the reference distance, and the lower limit of the distance away from the laser displacement meter D1 is preferably 0.4R or more, more preferably 2R or more, where R is the radius of the wheel. .

  If the lower limit is less than 0.4R, the difference between the first distance and the reference distance measured by the laser displacement meter D2 is small, and meaningful deflection measurement may not be performed. If the lower limit is 2R or more, the distance between the laser displacement meter and the paved road surface at a sufficiently distant position where there is no deflection due to the load applied to the wheel or the deflection is negligibly small is the reference distance. It can be. Then, by taking the difference between the first distance and the reference distance, the absolute amount of pavement surface deflection can be measured.

  Further, the upper limit of the distance between the laser displacement meter D2 and the laser displacement meter D is preferably 5R or less, and more preferably 4R or less. When the upper limit exceeds 5R, the length of the sensor mount on which the laser displacement meter D2 is installed becomes long, and the reference distance may not be correctly measured due to deflection due to the weight of the sensor mount. The upper limit of the distance between the laser displacement meter D2 and the laser displacement meter D is the influence of deflection generated by a load received from a wheel (for example, the front wheel) different from the wheel (for example, the rear wheel) to which the deflection measuring device 10 is attached. In order to eliminate the above, it is further preferable that the distance is 1/2 or less of the distance between the front wheel and the rear wheel.

  In the present embodiment, the reference distance is measured by the laser displacement meter D2. By measuring the reference distance with the laser displacement meter arranged in the sensor mount in this way, it is possible to accurately measure the amount of displacement according to the state of the paved road surface and to easily perform the deflection measurement. it can. However, the reference distance does not necessarily need to be measured by the laser displacement meter D2, and may be any distance that can evaluate the deflection amount as the difference between the laser displacement meter D1 and the pavement surface. For example, the distance between the laser displacement meter and the steel plate, which is measured in a state where the vehicle is stopped on the steel plate or the like and there is no deflection by the laser displacement meter D1, may be used as the reference distance. In addition, another laser displacement meter is attached to a position sufficiently away from the laser displacement meter D1 in the traveling direction (for example, the center of the vehicle body between the front wheel and the rear wheel), and the laser type is measured while the vehicle is stationary. The distance between the displacement meter and the paved road surface may be measured, and this may be used as the reference distance.

  By the way, when the vehicle travels, the sensor gantry vibrates in the direction perpendicular to the traveling direction (pitch) according to the state of the paved road surface and the rotational movement of the wheels, like the axle, and vibrates in the left-right direction of the traveling direction. (Rolling). Furthermore, the position with respect to the paved road surface changes in an inclined manner from the horizontal state due to the load applied to the paved road surface due to deceleration and acceleration of the vehicle and torsion of the axle due to the bending motion of the vehicle. For example, if the traveling vehicle is decelerated, the center of gravity shifts to the front of the vehicle and the sensor mount tilts downward in the traveling direction, and if the vehicle is accelerated, the center of gravity shifts to the rear of the vehicle. The sensor mount tilts to the right with respect to the running direction.

  Accordingly, the reference distance based on the measurement by the laser displacement meter D2 always changes during traveling. For this reason, it is necessary to set the reference distance at a certain time each time. Further, it is necessary to calibrate the reference distance so as to eliminate the influence of the tilt variation of the sensor mount from the reference distance measured from the laser displacement meter D2.

  Therefore, in the present embodiment, the laser displacement meter D3 is provided at a symmetrical position with respect to the laser displacement meter D2 around the laser displacement meter D1 (deflection measurement displacement sensor) in the traveling direction. As shown in FIG. 18D, the laser displacement meters D2 and D3 serve as a pair of pair displacement sensors, and the pair displacement sensor and the pavement measured by one pair of displacement sensors (laser displacement meter D2) of the pair of pair displacement sensors. When the distance between the road surface is d2 and the distance between the set displacement sensor measured by another set displacement sensor (laser displacement meter D3) and the pavement road surface is d3, the following formula, d2- (d2 Since the length represented by -d3) / 2 is the influence of the tilt of the sensor mount, the reference distance is calibrated with this length.

Accordingly, the deflection w _{0 in} this case is w ₀ = d ₁ − {d ₂ , where d ₁ is the distance (first distance) between the laser displacement meter measured by the laser displacement meter D1 and the pavement surface. - is calculated by _{(d 2 -d 3) / 2} }.

  Further, when the vehicle is stopped, the paved road surface deflection is axisymmetric about the vertical axis passing through the axle, and the maximum deflection occurs directly under the axle. On the other hand, when the vehicle is running, the maximum deflection position differs depending on the vehicle running speed, the level of the paved road surface, and the type of pavement. Arise. Further, the occurrence of deflection differs between the front side and the rear side of the vehicle with respect to the vertical axis and is asymmetric.

  Therefore, in order to estimate the maximum deflection position, in the present embodiment, laser displacement meters D4 to D7 are provided as deflection state measurement displacement sensors, and the deflection amount at the maximum deflection position is measured.

  That is, as shown in FIG. 18E, an approximate straight line (or approximate curve) based on the measurement results of the laser displacement meters D2, D4, and D5 and an approximation based on the measurement results of the laser displacement meters D3, D6, and D7 on the coordinate axes. A straight line (or approximate curve) is calculated, the intersection of these approximate straight lines (or approximate curves) is estimated as the maximum deflection position, and the laser displacement meter D1 and the pavement are distanced from the laser displacement meter D1 to the maximum deflection position. The distance between the road surface (first distance) is corrected and the amount of deflection at the maximum deflection position is measured.

  This approximate straight line can be obtained by providing at least two deflection state displacement measurement sensors at the front and rear positions in the traveling direction with the deflection measurement displacement sensor (laser displacement meter D1) as the center. Further, the deflection state measurement displacement sensor is preferably provided at a position away from the deflection measurement displacement sensor by a distance of less than 1R with respect to the radius R of the wheel in the traveling direction. If it is provided at a position of 1R or more, measurement is performed at a position where there is little deflection, and it may be difficult to estimate the maximum deflection position using an approximate line (or approximate curve).

  If the distance is less than 1R, the combined displacement sensor (laser displacement meters D2 and D3) can also function as a deflection state measurement displacement sensor.

  When estimating the maximum deflection position using the approximate curve as described above, by arranging the deflection state measurement displacement sensors densely, for example, a downwardly convex high-order function corresponding to the deflection shape of the paved road surface on the coordinate axis Alternatively, an approximate curve using a special function may be calculated, and a more accurate maximum deflection position may be estimated from position information at the bottom. Further, by disposing the deflection state measurement displacement sensor in a symmetrical position around the deflection measurement displacement sensor (laser displacement meter D1) in the traveling direction, a more accurate maximum deflection position can be estimated.

  If there is a significant deflection at a position of 1R or more depending on the magnitude of the load applied from the wheel, a deflection state measurement sensor at a position of less than 1R is obtained from the viewpoint of obtaining a highly accurate approximate curve. In addition, a deflection state measuring sensor can be provided at a position of 1R or more. However, also in this case, from the viewpoint of correctly estimating the maximum deflection position, it is preferable to arrange at least two deflection state measurement sensors at positions less than 1R in the front-rear position in the traveling direction.

  Further, in the deflection measurement according to the second modification, by providing a rangefinder such as a rotary encoder in the vehicle 100, the deflection measurement and the positional information where the deflection has occurred are synchronized, and the deflection information on the paved road surface is obtained. Can be obtained. By simultaneously measuring the deflection amount and position information of the paved road surface at regular time intervals, deflection information corresponding to the measurement position can be obtained. In this way, by using the laser Doppler vibrometer and the laser displacement meter in combination, it is possible to improve the deflection measurement accuracy.

  As described above, according to the dynamic deflection measuring apparatus (MWD) according to the present embodiment, the sensor mount is rigidly connected to the vehicle body of the loading vehicle. The influence of the generated vibration is eliminated, and the vehicle vibration speed can be directly acquired. In addition, it is possible to remove the influence of the vehicle running vibration without adopting the vibration isolation structure, and to improve the accuracy of the deflection measurement.

DESCRIPTION OF SYMBOLS 1 Loading vehicle 3 Laser Doppler vibrometer 5 Vibration accelerometer 10 Deflection measuring device 100 Vehicle 110 Rear wheel 120 Laser displacement meter 121 Laser displacement meter mount

Claims (7)

A deflection measurement method for dynamically measuring the deflection of a paved road surface,
Running a loading vehicle in which a sensor mount is rigidly connected to the vehicle body, and a laser Doppler vibrometer and a vibration accelerometer are installed on the sensor mount;
While the cargo vehicle described above is running, the laser Doppler vibrometer on the sensor mount rigidly connected to the vehicle body acquires the deflection speed of each measurement point, and the vibration on the sensor mount rigidly connected to the vehicle body Directly acquiring the vehicle 's vibration acceleration at each measurement point with an accelerometer;
Performing a discrete wavelet transform on the vehicle vibration waveform indicating the vibration acceleration at each measurement point to determine a filter that substantially flattens the vehicle vibration waveform;
Using the filter, removing a vehicle vibration speed from a deflection speed of each measurement point;
Calculating a maximum deflection from a deflection speed at each of the measurement points from which the vehicle vibration speed has been removed . Using the wavelet filter and the scaling filter, the discrete wavelet transform is performed by calculating the wavelet coefficients wi, t represented by the formula (1) and the scaling coefficients vi, t represented by the formula (2), and the formula (3) The deflection measurement method according to claim 1, wherein wavelet inverse transformation is performed by calculating a scaling coefficient vi, t indicated by: and a vehicle vibration velocity is removed from a deflection velocity at each measurement point.
Define a local coordinate system with the origin directly below the wheel of the cargo vehicle as described above, the traveling direction and loading direction of the cargo vehicle as described above as positive directions, the deflection amount as w, and the traveling speed of the cargo vehicle as described above as V. , The distance is x, the time is t, the deflection angle is dw / dx, the maximum deflection is a, the coefficient relating to the deflection shape is b, the position of occurrence of the maximum deflection is d, and the equations (4) to (5) ), The equation (7) is obtained from this deflection angle and the equation (6), c is 2abV, g (x) is dw / dx, and the logarithm of both sides of the equation (7) is obtained. Equation (8) is obtained, and the evaluation function J (c, b, d) is set as shown in Equation (9) using the least square method, and the evaluation function J (c, b, d) is minimum. By calculating d, a, and b when Claim 1 or claim 2 deflection measuring method wherein calculating the fine maximum deflection amount.
  The load vehicle is provided with a laser displacement meter gantry, and the laser displacement meter gantry includes a plurality of laser displacement meters arranged in parallel in the traveling direction of the load vehicle. The deflection measuring method according to any one of claims 1 to 3, wherein a distance between each laser displacement meter and a paved road surface is measured. A deflection measuring device that dynamically measures the deflection of a paved road surface,
A sensor base rigidly coupled to the body of the loading vehicle;
A laser Doppler vibrometer and a vibration accelerometer installed on the sensor mount;
While the cargo vehicle described above is running, the laser Doppler vibrometer on the sensor mount rigidly connected to the vehicle body acquires the deflection speed of each measurement point, and the vibration on the sensor mount rigidly connected to the vehicle body Acquire the vibration acceleration of the vehicle directly at each measurement point with the accelerometer,
The vehicle vibration waveform indicating the vibration acceleration at each measurement point is subjected to discrete wavelet transform to determine a filter that substantially flattens the vehicle vibration waveform,
Using the filter, the vehicle vibration speed is removed from the deflection speed of each measurement point ,
A deflection measuring apparatus , wherein a maximum deflection is calculated from a deflection speed at each of the measurement points from which the vehicle vibration speed has been removed .   6. The deflection measuring apparatus according to claim 5, further comprising a laser surface velocimeter for measuring a traveling speed of the cargo vehicle described above. A laser displacement gauge mount provided on the cargo vehicle described above;
A plurality of laser displacement meters arranged in parallel in the traveling direction of the cargo vehicle described above of the laser displacement meter mount;
The deflection measuring apparatus according to claim 5 or 6, wherein the distance between each laser displacement meter and a paved road surface is measured by the laser displacement meter while the cargo vehicle is traveling.