JP7097738B2 - Void determination method and void determination system - Google Patents

Description

The techniques disclosed herein relate to void determination methods and void determination systems.

Conventionally, a method for determining a void formed at an interface between two members has been known.

For example, Patent Document 1 discloses a method of detecting a void formed at an interface between a first member and a second member by using an ultrasonic probe. In this method, an ultrasonic probe is installed on the surface of the first member opposite to the interface, ultrasonic waves are transmitted from the opposite surface in the thickness direction of the first member, and the reflected wave from the interface is transmitted. I'm receiving. Then, the presence or absence of voids at the interface is determined based on the intensity of the received reflected wave.

Japanese Unexamined Patent Publication No. 2013-88304

However, in the above-mentioned method, it is necessary to properly receive the reflected wave from the interface, and for that purpose, an ultrasonic probe is installed on the surface of the first member opposite to the interface, and the first member is provided with an ultrasonic probe. Ultrasound must be transmitted in the thickness direction. When another member is present on the opposite surface, it becomes difficult to carry out the above-mentioned method, or pretreatment such as exposing the opposite surface is required. That is, there are many restrictions on implementing the above method.

The technique disclosed herein has been made in view of this point, and an object thereof is to determine an interface void by a method with less restrictions.

The void determination method disclosed herein is a void determination method for determining a void formed at the interface in a structure having an inner member and an outer member forming an interface together with the inner member, and the inner member. An input step for inputting elastic waves, a receiving step for receiving elastic waves propagating in the inner member, a calculation step for calculating evaluation parameters related to the strength of elastic waves received by the receiving step, and the above calculation. Including a determination step of determining at least one of the position and size of the void based on the evaluation parameter calculated by the step, in the calculation step, the evaluation parameter of the elastic wave vibrating in a predetermined first vibration direction is used. A certain first evaluation parameter is calculated, and in the determination step, the first evaluation parameter for reference, which is the first evaluation parameter obtained in advance from a plurality of samples having voids having different positions and sizes, is referred to. At least one of the position and size of the void is determined from the first evaluation parameter calculated in the calculation step.

Another void determination method disclosed herein is a void determination method for determining a void formed at the interface in a structure having an inner member and an outer member forming an interface together with the inner member. Calculation to calculate the specific mode of vibration of the structure based on the input step of inputting the elastic wave to the body, the receiving step of receiving the elastic wave propagating in the structure, and the elastic wave received by the receiving step. The determination step includes a determination step of determining at least one of the position and size of the void based on the specific mode calculated by the calculation step, and the determination step includes a plurality of voids having different positions and sizes. By referring to the reference eigenmode which is the eigenmode obtained in advance from the sample, at least one of the position and the size of the void is determined from the eigenmode calculated in the calculation step.

Further, the void determination system disclosed herein is a void determination system for determining a void formed at the interface in a structure having an inner member and an outer member forming an interface together with the inner member, and is the inner side. An input unit that inputs an elastic wave to the member, a receiving unit that receives the elastic wave propagating through the inner member, a calculation unit that calculates evaluation parameters related to the strength of the elastic wave received by the receiving unit, and a calculation unit. The calculation unit includes a determination unit for determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation unit, and the calculation unit evaluates the elastic wave vibrating in a predetermined first vibration direction. The first evaluation parameter, which is a parameter, is calculated, and the determination unit refers to the reference first evaluation parameter, which is the first evaluation parameter obtained in advance from a plurality of samples having voids having different positions and sizes. At least one of the position and the size of the void is determined from the first evaluation parameter calculated by the calculation unit.

Another void determination system disclosed herein is a void determination system for determining voids formed at the interface in a structure having an inner member and an outer member forming an interface with the inner member, and the structure. Calculation to calculate the specific mode of vibration of the structure based on the input unit that inputs the elastic wave to the body, the receiving unit that receives the elastic wave propagating in the structure, and the elastic wave received by the receiving unit. The determination unit includes a unit and a determination unit that determines at least one of the position and size of the void based on the intrinsic mode calculated by the calculation unit, and the determination unit has a plurality of voids having different positions and sizes. By referring to the reference eigenmode which is the eigenmode obtained in advance from the sample, at least one of the position and the size of the void is determined from the eigenmode calculated by the calculation unit.

According to the void determination method, the void at the interface can be determined with few restrictions.

According to the void determination system, the voids at the interface can be determined with few restrictions.

FIG. 1 is a perspective view of a superstructure of a bridge. FIG. 2 is an enlarged cross-sectional view centered on a hollow rib. FIG. 3 is a block diagram of the void determination system according to the first embodiment. FIG. 4 is a functional block diagram of the arithmetic unit. FIG. 5 is a cross-sectional view of the first sample. FIG. 6 is a cross-sectional view of the second sample. FIG. 7 is a cross-sectional view of the third sample. FIG. 8 is a cross-sectional view of the fourth sample. FIG. 9 is a cross-sectional view of the fifth sample. FIG. 10A is a received waveform of the first vibration sensor in the first sample. FIG. 10B is a received waveform of the second vibration sensor in the first sample. FIG. 11A is a received waveform of the first vibration sensor in the second sample. FIG. 11B is a received waveform of the second vibration sensor in the second sample. FIG. 12A is a received waveform of the first vibration sensor in the third sample. FIG. 12B is a received waveform of the second vibration sensor in the third sample. FIG. 13A is a received waveform of the first vibration sensor in the fourth sample. FIG. 13B is a received waveform of the second vibration sensor in the fourth sample. FIG. 14A is a received waveform of the first vibration sensor in the fifth sample. FIG. 14B is a received waveform of the second vibration sensor in the fifth sample. FIG. 15 is a graph in which the evaluation parameters I calculated from each received waveform are plotted. FIG. 16 is a determination map. FIG. 17 is a flowchart of a void determination method using evaluation parameters. FIG. 18 is a block diagram of the void determination system. FIG. 19 is an enlarged cross-sectional view centered on a hollow rib in a state where a vibration sensor is installed. FIG. 20 is a functional block diagram of the arithmetic unit. FIG. 21A is a received waveform of the third vibration sensor in the first sample. FIG. 21B is a received waveform of the fourth vibration sensor in the first sample. FIG. 22A is a received waveform of the third vibration sensor in the second sample. FIG. 22B is a received waveform of the fourth vibration sensor in the second sample. FIG. 23A is a received waveform of the third vibration sensor in the third sample. FIG. 23B is a received waveform of the fourth vibration sensor in the third sample. FIG. 24A is a received waveform of the third vibration sensor in the fourth sample. FIG. 24B is a received waveform of the fourth vibration sensor in the fourth sample. FIG. 25A is a received waveform of the third vibration sensor in the fifth sample. FIG. 25B is a received waveform of the fourth vibration sensor in the fifth sample. 26 shows the eigenmode of the first sample, (A) is the primary mode, (B) is the secondary mode, (C) is the tertiary mode, and (D). The figure shows the 4th order mode, the figure (E) shows the 5th order mode, and the figure (F) shows the 6th order mode. 27 shows the eigenmode of the second sample, FIG. 27A is a primary mode, FIG. 27B is a secondary mode, FIG. 27C is a tertiary mode, and FIG. 27D. The figure shows the 4th order mode, the figure (E) shows the 5th order mode, and the figure (F) shows the 6th order mode. 28 shows the eigenmode of the third sample, FIG. 28A is a primary mode, FIG. 28B is a secondary mode, FIG. 28C is a tertiary mode, and FIG. 28D. The figure shows the 4th order mode, the figure (E) shows the 5th order mode, and the figure (F) shows the 6th order mode. 29 shows the eigenmode of the fourth sample, (A) is the primary mode, (B) is the secondary mode, (C) is the tertiary mode, and (D). The figure shows the 4th order mode, the figure (E) shows the 5th order mode, and the figure (F) shows the 6th order mode. 30 shows the eigenmode of the fifth sample, FIG. 30A is a primary mode, FIG. 30B is a secondary mode, FIG. 30C is a tertiary mode, and FIG. 30D is. The figure shows the 4th order mode, the figure (E) shows the 5th order mode, and the figure (F) shows the 6th order mode. FIG. 31 is a model of a neural network. FIG. 32 is a flowchart of a void determination method using a unique mode.

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

<< Embodiment 1 >>
The void determination system 100 according to the first embodiment is used, for example, to determine the filling state of mortar or concrete. Here, the void determination system 100 and the void determination method thereof will be described by taking as an example the case of determining the filling state of the mortar filled in the deck of the bridge.

FIG. 1 is a perspective view of the superstructure 10 of the bridge. The superstructure 10 includes a plurality of decks 11 extending in the bridge axis direction, a plurality of main girders 12 extending in the bridge axis direction, and a plurality of laterals extending in a direction orthogonal to the bridge axis direction, that is, in the width direction of the deck 11. It has a digit 13. In the figure, only one floor slab 11, main girder 12, and cross girder 13 are shown. The plurality of decks 11 are arranged in the direction of the bridge axis and are joined to each other. The plurality of main girders 12 are provided on the lower surface of the deck 11 and are arranged at intervals in the width direction of the deck 11. The plurality of cross girders 13 are provided on the lower surface of the floor slab 11 and are arranged at intervals in the bridge axis direction between each of the two adjacent main girders 12. The main girder 12 and the cross girder 13 are joined to the floor slab 11 and support the floor slab 11. The superstructure 10 configured in this way is supported by a pier (not shown).

The deck 11 is filled in the flat deck plate 2, the plurality of hollow ribs 3 extending in the bridge axis direction, the reinforcing plate 4 for reinforcing the joint strength between the deck plate 2 and the hollow ribs 3, and the hollow ribs 3. It has a mortar M and the like. The deck 11 is an example of a structure. A pavement 14 is laid on the upper surface of the deck plate 2. Vehicles pass on the pavement 14. The plurality of hollow ribs 3 are provided on the lower surface of the deck plate 2 and are arranged at intervals in the width direction of the deck plate 11. The arrangement spacing of the hollow ribs 3 is smaller than the arrangement spacing of the main girder 1. That is, a plurality of hollow ribs 3 are arranged between the two adjacent main girders 12. The deck 11 may further have a plurality of lateral ribs extending in the width direction of the deck 11 across the plurality of hollow ribs 3. The plurality of horizontal ribs may be provided on the lower surface of the deck plate 2 and may be arranged at intervals in the bridge axis direction between each of the two adjacent cross girders 13.

The main girder 12, the cross girder 13, and the hollow rib 3 are made of steel and are joined to the deck plate 2 by welding.

The hollow rib 3 forms a closed cross section together with the deck plate 2. FIG. 2 is an enlarged cross-sectional view centered on the hollow rib 3. Specifically, the hollow rib 3 has a pair of first vertical wall 31A and second vertical wall 31B facing each other, and a bottom wall 33. The bottom wall 33 extends substantially parallel to the deck plate 2. The first vertical wall 31A and the second vertical wall 31B extend upward from both end edges of the bottom wall 33. The first vertical wall 31A and the second vertical wall 31B are inclined with respect to the vertical direction so that the distance between them increases upward. The upper ends of the first vertical wall 31A and the second vertical wall 31B are welded to the deck plate 2. In this way, a closed space extending in the direction of the bridge axis is formed between the hollow rib 3 and the deck plate 2. The space inside the hollow rib 3 is divided into a plurality of spaces divided in the bridge axis direction by a partition plate (not shown). When the first vertical wall 31A and the second vertical wall 31B are not distinguished, they are simply referred to as "vertical wall 31". The deck plate 2 and the hollow rib 3 are examples of outer members. That is, the outer member has a divided structure of the deck plate 2 and the hollow rib 3. The deck plate 2 is an example of the first member, and the hollow rib 3 is an example of the second member.

The space inside the hollow rib 3 is filled with mortar M. The bottom wall 33 is formed with an injection port and a discharge port (not shown) for filling the space inside the hollow rib 3 with the mortar M. One inlet and one outlet are provided in each space in the hollow rib 3 separated by the partition plate. After the hollow rib 3 is joined to the deck plate 2, the mortar M is injected into the hollow rib 3 from the injection port, and the excess mortar M is discharged from the discharge port. In this way, the space inside the hollow rib 3 is filled with the mortar M. Finally, the inlet and outlet are closed. The mortar M is an example of an inner member.

The reinforcing plate 4 is formed of a steel plate extending in the direction of the bridge axis. Two reinforcing plates 4 are provided for one hollow rib 3. The reinforcing plate 4 is bent along the vertical wall 31 of the deck plate 2 and the hollow rib 3. The reinforcing plate 4 is bolted to the deck plate 2 and the vertical wall 31. That is, the hollow rib 3 is connected to the deck plate 2 by welding and the reinforcing plate 4. The reinforcing plate 4 is an example of a connecting member.

Specifically, on the deck plate 2, a plurality of threaded studs 41 are arranged at equal intervals in the bridge axis direction. As shown in FIG. 1, the reinforcing plate 4 is attached to the deck plate 2 by screwed studs 41 and nuts 42 at a plurality of positions in the bridge axis direction.

On the other hand, the reinforcing plate 4 is attached to the vertical wall 31 from the outside of the hollow rib 3 with a one-side bolt 43. The one-side bolt 43 penetrates the hollow rib 3, one end of which is exposed to the outside of the hollow rib 3, and the other end of which is pierced into the mortar M. The other end of the one-side bolt 43 (hereinafter referred to as "tip portion") protrudes into the hollow rib 3. The reinforcing plate 4 is attached before filling the mortar M. That is, since the mortar M is filled with the tip of the one-side bolt 43 protruding into the hollow rib 3, the tip of the one-side bolt 43 is in a state of being pierced by the mortar M. As shown in FIG. 1, the reinforcing plate 4 is attached to the vertical wall 31 with one-side bolts 43 at a plurality of positions at equal intervals in the bridge axis direction.

Hereinafter, for convenience of explanation, the one-side bolt 43 attached to the first vertical wall 31A is referred to as a "first one-side bolt 43A", and the one-side bolt 43 attached to the second vertical wall 31B is referred to as a "second one-side bolt". It is called "bolt 43B". The first one-side bolt 43A is an example of a first penetrating member, and the second one-side bolt 43B is an example of a second penetrating member.

The second one-side bolt 43B is also located on the cross section orthogonal to the bridge axis direction where the first one-side bolt 43A is located. Further, on this cross section, the threaded stud 41 and the nut 42 of one reinforcing plate 4 and the threaded stud 41 and the nut 42 of the other reinforcing plate 4 are also located.

In the superstructure 10 configured in this way, the floor slab 11 may bend due to the passage of a vehicle or the like. In the superstructure 10, the deck plate 2 is provided with the hollow ribs 3, and the hollow ribs 3 are filled with the mortar M to reduce the bending of the floor slab 11. If only the hollow rib 3 is provided without filling the mortar M, the bending of the deck 11 can be reduced to some extent, but the stress is concentrated on the joint portion of the vertical wall 31 of the deck plate 2, and the stress is concentrated. Cracks can occur. On the other hand, by filling the hollow rib 3 with the mortar M, it is possible to reduce the bending of the portion of the deck plate 2 between the first vertical wall 31A and the second vertical wall 31B. As a result, stress concentration on the joint portion of the vertical wall 31 of the deck plate 2 can be relaxed.

The floor slab 11 is not always manufactured with the above-mentioned configuration from the beginning. For example, a hollow rib 3, a reinforcing plate 4, and a mortar M may be later installed on an existing deck having a deck plate 2 for reinforcement. Alternatively, the reinforcing plate 4 and the mortar M may be later installed on the existing deck having the deck plate 2 and the hollow rib 3 for reinforcement.

However, if the void G is present at the interface between the deck plate 2 and the mortar M, the deflection of the deck plate 2 becomes large at the portion of the void G, and the effect of relaxing the stress concentration is reduced. Air may be mixed when the hollow rib 3 is filled with the mortar M, and in that case, the gap G is mainly formed at the interface between the deck plate 2 and the mortar M due to the difference in the specific gravity between the air and the mortar M. Can occur.

Therefore, the void determination system 100 determines the position and size of the void G. FIG. 3 is a block diagram of the void determination system 100. The void determination system 100 determines the void G formed at the interface between the deck plate 2 and the mortar M.

The void determination system 100 inputs an elastic wave to the mortar M, receives an elastic wave propagating in the mortar M, and determines the position and size of the void G based on the received elastic wave. The void determination system 100 includes an impulse hammer 51, a vibration sensor 53, an amplification unit 54, a data logger 61, and an arithmetic unit 62.

The impulse hammer 51 is a shaker that gives a blow to the hollow rib 3, strictly speaking, the mortar M. The impulse hammer 51 has a built-in sensor for detecting the exciting force, and detects and outputs the exciting force at the time of hitting.

The vibration sensor 53 is an acceleration pickup and detects the acceleration of an object. As shown in FIG. 2, two vibration sensors 53 are attached to the outer end of the second one-side bolt 43B, which is exposed to the outside of the hollow rib 3. When distinguishing between the two vibration sensors 53, one is referred to as a first vibration sensor 53A and the other is referred to as a second vibration sensor 53B. The first vibration sensor 53A detects the acceleration in the axial direction (hereinafter, also referred to as “first vibration direction”) of the second one-side bolt 43B. The second vibration sensor 53B detects acceleration in a direction orthogonal to both the axial direction and the bridge axial direction of the second one-side bolt 43B (hereinafter, also referred to as “second vibration direction”). That is, the vibration sensor 53 receives the elastic wave propagating in the mortar M via the second one-side bolt 43B.

The output of the vibration sensor 53 is input to the amplification unit 54. The amplification unit 54 amplifies the detection signal of the vibration sensor 53 and outputs it to the data logger 61.

The output of the impulse hammer 51 and the output of the vibration sensor 53 are input to the data logger 61. For example, the data logger 61 sets a trigger based on the exciting force of the impulse hammer 51, and records the output from the vibration sensor 53 at a predetermined sampling frequency for a predetermined sampling period. At this time, the data logger 61 may set a pre-trigger.

The arithmetic unit 62 has an arithmetic unit 63, a memory 64, and a storage device 65.

The arithmetic unit 63 has a processor and performs various processes such as executing a program. For example, the calculation unit 63 analyzes the data of the data logger 61 to determine the position and size of the gap G. The memory 64 is used to temporarily store data when the arithmetic unit 63 executes a program. The storage device 65 stores various programs and data.

Subsequently, the functional block of the arithmetic unit 62 will be described. FIG. 4 is a functional block diagram of the arithmetic unit 62. The arithmetic unit 62 has a calculation unit 66 for calculating evaluation parameters, and a determination unit 67 for determining the position and size of the gap G based on the calculated evaluation parameters.

The calculation unit 66 calculates the evaluation parameter I, which is a parameter related to the strength of the elastic wave received by the vibration sensor 53. The evaluation parameter I is represented by the equation (1).

Here, a _i is each amplitude [mm / s ² ] in the acceleration waveform, and n is the number of amplitudes.

More specifically, the calculation unit 66 applies a filter process to remove components other than the predetermined frequency band to the acceleration signal recorded by the data logger 61 and received by the first vibration sensor 53A, and adds an impulse hammer 51. Standardize the acceleration signal based on the vibration force.

Since the physical property values of the mortar M are known, the approximate propagation velocity of the elastic wave propagating in the mortar M can be obtained. Further, since the dimensions of the floor slab 11, more specifically, the hollow rib 3 and the like are known, the frequency band of the elastic wave multiplely reflected in the hollow rib 3 can be roughly obtained. In the filtering process, the frequency band thus determined is used. That is, the filter processing is a processing for extracting components of elastic waves related to multiple reflections in the mortar M. Further, since the impact by the impulse hammer 51 is performed manually, the exciting force of the impulse hammer 51 may vary. By performing the above-mentioned standardization, it is possible to eliminate the variation in the amplitude of the acceleration signal caused by the variation in the exciting force of the impulse hammer 51.

The calculation unit 66 calculates the evaluation parameter I of the processed signal using the equation (1). The calculation unit 66 performs the same calculation on the acceleration signal received by the second vibration sensor 53B recorded in the data logger 61 to obtain the evaluation parameter I. When distinguishing between the two evaluation parameters I, the evaluation parameter I obtained from the acceleration signal in the first vibration direction, that is, the acceleration signal of the first vibration sensor 53A is referred to as the first evaluation parameter Ia, and is in the second vibration direction. The evaluation parameter I obtained from the acceleration signal, that is, the acceleration signal of the second vibration sensor 53B is referred to as a second evaluation parameter Ib.

The determination unit 67 determines the position and size of the gap G based on the evaluation parameter I. The determination unit 67 refers to a reference evaluation parameter, which is an evaluation parameter I previously acquired from a plurality of samples having voids G having different positions and sizes, from the evaluation parameter I calculated by the calculation unit 66. Determine the position and size of G.

Specifically, the evaluation parameter I is obtained in advance from the sample S having the void G whose position and size are known. The sample S has the same configuration as the actual deck plate 2, hollow rib 3, reinforcing plate 4, one-side bolt 43, mortar M, etc., and a gap G is formed at the interface between the mortar M and the deck plate 2. There is. A plurality of samples S having different positions and sizes of the voids G have been created.

5 to 9 are cross-sectional views of the first to fifth samples S1 to S5. The positions and sizes of the voids G are different between the first to fifth samples S1 to S5. Here, the positions of the voids G in the width direction of the hollow ribs 3 are different. In the following, unless otherwise specified, the "end" means the end in the width direction of the hollow rib 3, and the "center" means the center in the width direction of the hollow rib 3. Further, the "size of the gap G" is a concept including the case where the size of the gap G is zero, that is, there is no gap G.

Specifically, as shown in FIG. 5, the void G is not formed in the first sample S1. That is, in the first sample S1, the hollow rib 3 is completely filled with the mortar M, and the deck plate 2 and the mortar M are in close contact with each other over the entire interface between the deck plate 2 and the mortar M. As shown in FIG. 6, the second sample S2 has a gap G formed at the end of the interface between the deck plate 2 and the mortar M near the first vertical wall 31A. As shown in FIG. 7, the third sample S3 has a gap G formed at the end of the interface between the deck plate 2 and the mortar M near the second vertical wall 31B. As shown in FIG. 8, the fourth sample S4 has a gap G formed in the center of the interface between the deck plate 2 and the mortar M. As shown in FIG. 9, the fifth sample S5 has a relatively large void G formed in the center of the interface between the deck plate 2 and the mortar M.

The dimensions (width, height, length; the same applies hereinafter) of the voids G of the second sample S2, the third sample S3, and the fourth sample S4 are the same. The width of the void G of the fifth sample S5 is larger than the width of the void G of the second sample S2, the third sample S3, and the fourth sample S4. The height and length of the void G of the fifth sample S5 are the same as the height and length of the void G of the second sample S2, the third sample S3, and the fourth sample S4.

10 to 14 are received waveforms when impact elastic waves are input to the first to fifth samples S1 to S5. In each figure, FIG. (A) is a waveform received by the first vibration sensor 53A, and FIG. (B) is a waveform received by the second vibration sensor 53B. The received waveform is a waveform that has been standardized as described above. The horizontal axis of the graph is time, and the vertical axis is acceleration.

As can be seen from each figure, the elastic wave input to the mortar M is multiplely reflected in the hollow rib 3. These received waveforms change depending on the position and size of the gap G. Specifically, the reflectance of elastic waves at the interface between two members depends on the difference in acoustic impedance between the two members. The interface of the portion where the void G is not formed is formed by the mortar M and the deck plate 2. The reflectance at this interface depends on the difference between the acoustic impedance of the mortar M and the acoustic impedance of the deck plate 2. On the other hand, the interface of the portion where the void G is formed is formed by the mortar M and air. The reflectance at this interface depends on the difference between the acoustic impedance of the mortar M and the acoustic impedance of the air. Since the difference between the acoustic impedance of the mortar M and the acoustic impedance of the air is larger than the difference between the acoustic impedance of the mortar M and the acoustic impedance of the deck plate 2, the reflectance at the interface between the mortar M and the air is larger. Become. In this way, the reflection state of the elastic wave propagating in the mortar M changes depending on the presence or absence of the void G.

Further, the portion of the mortar M in which the void G is formed is less constrained than the portion without the void G, and the boundary conditions are different. This can also be a factor that changes the condition of the elastic wave propagating in the mortar M depending on the presence or absence of the void G.

As a result of these, it is considered that the received waveform changes according to the position and size of the gap G.

However, it is difficult to discriminate the difference between the received waveforms, and it is difficult to discriminate the position and size of the gap G based on the received waveform itself.

On the other hand, FIG. 15 is a graph in which the evaluation parameters I calculated from each received waveform are plotted. As described above, the first evaluation parameter Ia is calculated from the received waveform of the first vibration sensor 53A, and the second evaluation parameter Ib is calculated from the received waveform of the second vibration sensor 53B. The first evaluation parameter Ia and the second evaluation parameter Ib of each sample are plotted on a graph with the first evaluation parameter Ia on the horizontal axis and the second evaluation parameter Ib on the vertical axis. The first evaluation parameter Ia obtained from the first sample S1 to the fifth sample S5 is an example of the reference first evaluation parameter, and the second evaluation parameter Ib obtained from the first sample S1 to the fifth sample S5 is. This is an example of the second evaluation parameter for reference.

As can be seen from the figure, the position (that is, the coordinates) on the graph of each sample differs depending on the position and size of the gap G. The difference is remarkable as compared with the difference in the received waveform depending on the position and size of the gap G. As described above, information regarding the position and size of the void G appears in the evaluation parameter I. That is, the information regarding the position and size of the gap G, which did not clearly appear in the received waveform, can be made apparent by the evaluation parameter I.

A determination map is created in advance based on the first evaluation parameter Ia and the second evaluation parameter Ib obtained from the sample S in this way. The determination map is a map for determining the position and size of the void G based on the first evaluation parameter Ia and the second evaluation parameter Ib. FIG. 16 shows a determination map. The horizontal axis is the first evaluation parameter Ia, and the vertical axis is the second evaluation parameter Ib. This judgment map represents the evaluation parameters for reference in the form of a map. The determination map is stored in the storage device 65.

The region A is a region including the evaluation parameters of the first sample S1 and the fourth sample S4, and is a region having a void G having no void G or having a void G of less importance. If the void G is present at the interface between the deck plate 2 and the mortar M, the deflection of the deck plate 2 becomes large at the portion of the void G. When the gap G is located in the center, the adverse effect on the joint portion of the vertical wall 31 of the deck plate 2 is small. On the other hand, when the gap G is located at the end portion, the portion of the deck plate 2 in the vicinity of the joint portion of the vertical wall 31 tends to bend. Therefore, the effect of relaxing the stress concentration by the mortar M is reduced. That is, the void G existing in the center is less important than the void G existing at the end. The void G of the fourth sample S is classified as a less important void because it is located in the center and its size is not so large.

On the other hand, the regions B and C are regions having a void G having a relatively high importance. The region B is a region including the evaluation parameters of the second sample S2 and the third sample S3, and is a region in which the void G exists at the end. The region C is a region including the evaluation parameter of the fifth sample S5, and is a region in which a large void G exists in the center. The size of the void G in the second sample S2 and the third sample S3 is substantially the same as the size of the void G in the fourth sample S4, but the void G existing at the end is more important than the void G in the center. Therefore, the voids G of the second sample S2 and the third sample S3 are classified into voids having high importance. The void G of the fifth sample S5 is located in the center like the void G of the fourth sample S4, but is larger than the void G of the fourth sample S4, so that the importance is increased accordingly. Therefore, the void G of the fifth sample S5 is classified as a void having high importance.

In short, the region A is a region having no void G or the void G is not of high importance, the region B is a region in which the void G of high importance is present at the end, and the region C is the center. This is a region in which a void G having a high importance is present.

Even if the evaluation parameters change depending on the position and size of the void G, the ranges of the first evaluation parameter Ia and the second evaluation parameter Ib are limited to some extent as long as the evaluation parameters are appropriately acquired. For example, if both the first evaluation parameter Ia and the second evaluation parameter Ib are very small, it is highly possible that the input and / or reception of elastic waves was not performed properly. Therefore, the region closer to the origin 0 than the alternate long and short dash line in the figure may be set as a region where the evaluation parameters could not be appropriately acquired.

The determination unit 67 determines the position and size of the void G from the evaluation parameter I calculated by the calculation unit 66 by referring to the determination map. That is, the determination unit 67 compares the first evaluation parameter Ia and the second evaluation parameter Ib calculated by the calculation unit 66 with the determination map. The determination unit 67 determines the position and size of the void G depending on which region of the determination map the first evaluation parameter Ia and the second evaluation parameter Ib are included in. Since this determination map is created based on the first evaluation parameter Ia and the second evaluation parameter Ib of a plurality of samples, it is possible to use the determination map from a plurality of samples having voids G having different positions and sizes. It is equivalent to indirectly referencing the evaluation parameter for reference acquired in advance.

Next, a void determination method using the void determination system 100 will be described. FIG. 17 is a flowchart of a void determination method using evaluation parameters.

First, in step Sa1, the operator inputs an impact elastic wave into the mortar M using the impulse hammer 51. Specifically, the operator installs the first vibration sensor 53A and the second vibration sensor 53B on the second one-side bolt 43B, and the position in the bridge axis direction is the same as the second one-side bolt 43B. The end edge of the bolt 43A is hit with the impulse hammer 51. As shown by the arrow in FIG. 2, the operator applies a blow mainly in the axial direction of the first one-side bolt 43A. The tip of the first one-side bolt 43A is in a state of being pierced into the mortar M, and the impact of the impulse hammer 51 applied to the first one-side bolt 43A is transmitted to the mortar M.

In this way, an impulse-like elastic wave (that is, an impact elastic wave) is input to the mortar M by the impact of the impulse hammer 51. Step Sa1 corresponds to an input step of inputting an elastic wave to the inner member.

At this time, the exciting force applied from the impulse hammer 51 to the first one-side bolt 43A is detected by the sensor built in the impulse hammer 51, and the detected exciting force signal is input to the data logger 61.

Next, in step Sa2, the first vibration sensor 53A and the second vibration sensor 53B receive elastic waves propagating through the mortar M. The first vibration sensor 53A and the second vibration sensor 53B are attached to the second one-side bolt 43B as described above. The tip of the second one-side bolt 43B is in a state of being pierced into the mortar M. Therefore, the elastic wave propagating in the mortar M is transmitted to the first vibration sensor 53A and the second vibration sensor 53B via the second one-side bolt 43B. The output signals of the first vibration sensor 53A and the second vibration sensor 53B are input to the data logger 61. The data logger 61 records the output signals of the first vibration sensor 53A and the second vibration sensor 53B as a response corresponding to the elastic wave input by the impulse hammer 51. Step Sa2 corresponds to a receiving step of receiving elastic waves propagating in the inner member.

In step Sa3, the calculation unit 66 calculates the first evaluation parameter Ia and the second evaluation parameter Ib from the acceleration signal received by the first vibration sensor 53A recorded in the data logger 61. Step Sa3 corresponds to a calculation step of calculating evaluation parameters related to the strength of elastic waves received by the receiving step.

After that, in step Sa4, the determination unit 67 determines the position and size of the gap G by comparing the first evaluation parameter Ia and the second evaluation parameter Ib with the determination map. Step Sa4 corresponds to a determination step of determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation step.

In this way, the determination unit 67 determines the position and size of the void G from the evaluation parameter I. However, this void determination is made for a cross section including a set of the first one-side bolt 43A and the second one-side bolt 43B. Since the elastic wave in the mortar M is also diffused in the bridge axis direction, a region including a set of the first one-side bolt 43A and the second one-side bolt 43B and having a certain length in the bridge axis direction. The position and size of the void G in the inside are determined.

The operator changes the set of the first one-side bolt 43A and the second one-side bolt 43B for inputting and receiving elastic waves, and repeats the above-mentioned steps Sa1 to Sa4. In this way, the position and size of the gap G are determined over the entire area in the bridge axis direction.

According to this void determination method, unlike the method of Patent Document 1, it is not necessary to input elastic waves from the upper surface of the deck plate 2 (the surface on which the pavement 14 is laid), so that the upper surface of the deck plate 2 is exposed. Pretreatment such as is not required. That is, the voids at the interface can be determined by a method with less restrictions. In particular, in the superstructure 10 of the bridge, since the pavement 14 is laid on the upper surface of the deck plate 2, it is difficult to input elastic waves from the upper surface of the deck plate 2 toward the interface between the deck plate 2 and the mortar M. However, even if it can be done, the preprocessing becomes very complicated. However, according to this void determination method, the void can be determined by inputting an elastic wave from a portion other than the upper surface of the deck plate 2 and receiving the elastic wave at a portion other than the upper surface of the deck plate 2. can. For example, the void can be determined from below the floor slab 11.

Further, since the void can be determined from below the floor slab 11, it is possible to maintain the state in which the vehicle passes on the pavement 14 when determining the void. That is, since the pavement 14 functions as a road, vehicles normally pass over the pavement 14. When an elastic wave is input from the upper surface of the deck plate 2 as in Patent Document 1, it is necessary to restrict traffic. On the other hand, according to this void determination method, the void can be determined from below the floor slab 11 while the vehicle is passing on the pavement 14. Also in this respect, this void determination method has few restrictions.

However, according to this void determination method, it is difficult to receive only a simple reflected wave from the interface as in Patent Document 1. Therefore, it is not possible to determine the presence or absence of voids based on the reception intensity of the reflected wave from the interface. However, by calculating the evaluation parameter I of the received signal of the elastic wave propagating in the mortar M, the information regarding the position and size of the gap G included in the received signal can be revealed. Then, by referring to the reference evaluation parameters obtained in advance from the samples S1 to S5 in which the position and size of the gap G are known, the position and size of the gap G are determined from the evaluation parameter I obtained from the received signal. can do.

As described above, the void determination system 100 is formed at the interface of the floor slab 11 (structure) having the mortar M (inner member), the deck plate 2 forming the interface together with the mortar M, and the hollow rib 3 (outer member). An impulse hammer 51 (input unit) that inputs an elastic wave to the mortar M, a vibration sensor 53 (reception unit) that receives the elastic wave propagating through the mortar M, and a void determination system that determines the void G. At least one of the position and size of the gap G is determined based on the calculation unit 66 that calculates the evaluation parameter I related to the strength of the elastic wave received by the vibration sensor 53 and the evaluation parameter I calculated by the calculation unit 66. The determination unit 67 includes a determination unit 67, the calculation unit 66 calculates the first evaluation parameter Ia, which is the evaluation parameter I of the elastic wave vibrating in the predetermined first vibration direction, and the determination unit 67 has a position and a size. Calculated by the calculation unit 66 by referring to the reference first evaluation parameter, which is the first evaluation parameter Ia obtained in advance from the first to fifth samples S1 to S5 (plural samples) having different voids G. At least one of the position and the size of the gap G is determined from the first evaluation parameter Ia.

In other words, the void determination method using the void determination system 100 is a floor slab 11 (structure) having a mortar M (inner member), a deck plate 2 forming an interface with the mortar M, and a hollow rib 3 (outer member). It is a void determination method for determining a void G formed at an interface in the above, and is received by an input step of inputting an elastic wave to the mortar M, a receiving step of receiving an elastic wave propagating in the mortar M, and a receiving step. Calculation including a calculation step of calculating the evaluation parameter I related to the strength of the elastic wave and a determination step of determining at least one of the position and the size of the void G based on the evaluation parameter I calculated by the calculation step. In the step, the first evaluation parameter Ia, which is the evaluation parameter I of the elastic wave vibrating in the predetermined first vibration direction, is calculated, and in the determination step, from a plurality of samples S1 to S5 having voids G having different positions and sizes. By referring to the reference first evaluation parameter which is the first evaluation parameter Ia acquired in advance, at least one of the position and the size of the void is determined from the first evaluation parameter Ia calculated in the calculation step.

According to these configurations, if the elastic wave is input to the mortar M and the elastic wave propagating in the mortar M can be received, the position and size of the gap G can be determined. That is, unlike the method described in Patent Document 1, there is no strict restriction that ultrasonic waves must be transmitted from the surface of the first member toward the interface, so that the position and size of the gap G can be determined. It can be easily realized. For specific void determination, information on the position and size of the void G contained in the elastic wave is revealed by calculating the first evaluation parameter Ia related to the strength of the elastic wave propagating in the mortar M. Can be made to. Further, the first evaluation parameters for reference are acquired in advance from a plurality of samples S1 to S5 having voids G having different positions and sizes, and by referring to these, the voids G are obtained from the above-mentioned first evaluation parameters Ia. The position and size of can be determined.

Further, in the calculation step, the second evaluation parameter Ib, which is the evaluation parameter I of the elastic wave vibrating in the second vibration direction different from the first vibration direction, is further calculated, and in the determination step, the first evaluation parameter Ia and the first evaluation parameter Ia are further calculated. The first evaluation parameter calculated in the calculation step by referring to the reference second evaluation parameter, which is the second evaluation parameter Ib obtained in advance from a plurality of samples S1 to S5 having voids G having different positions and sizes. At least one of the position and the size of the void G is determined from Ia and the second evaluation parameter Ib.

According to this configuration, the position and size of the void G are determined using the evaluation parameter I obtained from a plurality of elastic waves having different vibration directions, specifically, the first evaluation parameter Ia and the second evaluation parameter Ib. Therefore, the accuracy of determining the void can be improved.

Further, the first vibration direction corresponds to the vibration direction of the elastic wave (hereinafter, also referred to as “input elastic wave”) input in the input step.

According to this configuration, the position and size of the gap G can be determined using the evaluation parameter I of the elastic wave in the main propagation direction among the elastic waves propagating in the deck 11. That is, since the elastic waves are multiplely reflected in the deck 11, the elastic waves propagate in various directions. Even under such circumstances, there are many elastic waves propagating in the vibration direction of the input elastic waves. Since the first vibration direction corresponds to the vibration direction of the input elastic wave, that is, it roughly coincides with the vibration direction of the input elastic wave, the first evaluation parameter Ia propagates more in the floor slab 11. It reflects the information of elastic waves. By using such a first evaluation parameter Ia, the accuracy of determining the void G can be improved.

Further, the floor slab 11 penetrates the hollow rib 3, one end of which is exposed to the outside of the hollow rib 3, and the other end of the first one-side bolt 43A (first penetration member) and the first one pierced into the mortar M. It further has 2 one-side bolts 43B (second penetration member), and in the input step, elastic waves are input to the mortar M from the outside of the hollow rib 3 via the first one-side bolt 43A, and in the receiving step, the elastic waves are input to the mortar M. The elastic wave propagating is received from the outside of the hollow rib 3 via the second one-side bolt 43B.

According to this configuration, the first one-side bolt 43A and the second one-side bolt 43B pierced into the mortar M penetrate the hollow rib 3 and are exposed to the outside of the hollow rib 3. Therefore, by inputting an elastic wave to the mortar M via the first one-side bolt 43A, the elastic wave can be appropriately input to the mortar M from the outside of the hollow rib 3. Further, by receiving the elastic wave propagating in the mortar M via the second one-side bolt 43B, the elastic wave propagating in the mortar M can be appropriately received from the outside of the hollow rib 3.

Further, the outer member has a divided structure including at least a deck plate 2 (first member) and a hollow rib 3 (second member), and the deck 11 connects the deck plate 2 and the hollow rib 3. Further having a reinforcing plate 4 (connecting member), the first one-side bolt 43A and the second one-side bolt 43B are members for fastening the reinforcing plate 4 to the hollow rib 3.

According to this configuration, elastic wave input and reception can be performed by using the first one-side bolt 43A and the second one-side bolt 43B for fastening the reinforcing plate 4 to the hollow rib 3.

Further, the floor slab 11 is divided into a deck plate 2, a hollow hollow rib 3 (rib) joined to the lower surface of the deck plate 2 and forming a closed cross section together with the deck plate 2, and a deck plate 2 and a hollow rib 3. It is a deck of a bridge having a mortar M filled in the space to be formed, the outer member is a deck plate 2 and a hollow rib 3, and the inner member is a mortar M.

According to this configuration, the deck 11 of the bridge is filled with the mortar M in the space partitioned by the deck plate 2 and the hollow rib 3. Then, the position and size of the gap G formed at the interface between the deck plate 2 and the hollow rib 3 and the mortar M are determined.

<< Embodiment 2 >>
Subsequently, the void determination system 200 according to the second embodiment will be described. FIG. 18 is a block diagram of the void determination system 200.

The void determination system 200 inputs elastic waves to the deck 11 (particularly, a portion including the mortar M and the hollow rib 3), while receiving elastic waves propagating through the deck 11 and based on the received elastic waves. The position and size of the void in the plate 11 are determined. When an elastic wave is input to the mortar M and the hollow rib 3 or the like covering the mortar M, the elastic wave propagates in the mortar M to generate multiple reflections. The void determination system 200 captures elastic waves that are multiple reflected inside the mortar M and the member that covers it as vibrations of the mortar M and the member that covers it, and determines the position and size of the voids based on the intrinsic mode of the vibration. judge. The void determination system 200 differs from the void determination system 100 in that the position and size of the void are determined based on the intrinsic mode of vibration of the deck 11 instead of the evaluation parameter I. Therefore, of the void determination system 200, the same configuration as the void determination system 100 will be described by adding the same reference numerals and omitting the description, focusing on different parts.

The void determination system 200 includes an impulse hammer 51, six vibration sensors 53, an amplification unit 54, a data logger 61, and an arithmetic unit 262.

FIG. 19 is an enlarged cross-sectional view centered on the hollow rib 3 in a state where the vibration sensor 53 is installed. The configuration of each vibration sensor 53 is the same as that of the vibration sensor 53 of the void determination system 100. The six vibration sensors 53 are attached to the outer surface of the hollow rib 3 and the reinforcing plate 4. The six vibration sensors 53 are arranged in the same cross section orthogonal to the bridge axis direction. When distinguishing each vibration sensor 53, they are referred to as a first vibration sensor 53A, a second vibration sensor 53B, ..., And a sixth vibration sensor 53F.

Specifically, the first vibration sensor 53A is an outer surface of a portion of the reinforcing plate 4 that overlaps with the first vertical wall 31A, and is arranged in the vicinity of the first one-side bolt 43A. The second vibration sensor 53B is arranged on the outer surface of the first vertical wall 31A in the vicinity of the bottom wall 33. The third vibration sensor 53C is arranged on the outer surface of the bottom wall 33 in the vicinity of the first vertical wall 31A. The fourth vibration sensor 53D is arranged on the outer surface of the bottom wall 33 in the vicinity of the second vertical wall 31B. The fifth vibration sensor 53E is arranged on the outer surface of the second vertical wall 31B in the vicinity of the bottom wall 33. The sixth vibration sensor 53F is an outer surface of a portion of the reinforcing plate 4 that overlaps with the second vertical wall 31B, and is arranged in the vicinity of the second one-side bolt 43B. Each vibration sensor 53 detects the acceleration in the normal direction of the attached outer surface.

The arithmetic unit 262 includes an arithmetic unit 263, a memory 264, and a storage device 265.

The arithmetic unit 263 has a processor and performs various processes such as executing a program. For example, the calculation unit 263 analyzes the data of the data logger 61 to determine the position and size of the gap G. The memory 264 is used to temporarily store data when the arithmetic unit 263 executes the program. The storage device 265 stores various programs and data.

Subsequently, the functional block of the arithmetic unit 262 will be described. FIG. 20 is a functional block diagram of the arithmetic unit 262. The arithmetic unit 262 has a calculation unit 266 that calculates the specific mode of vibration of the floor slab 11, and a determination unit 267 that determines the position and size of the gap G based on the calculated specific mode.

The calculation unit 266 calculates the specific mode of vibration of the deck 11 in the cross section including the six vibration sensors 53.

More specifically, the calculation unit 266 applies a filter process to remove components other than the predetermined frequency band to each of the acceleration signals recorded by the data logger 61 and received by the six vibration sensors 53, and also performs an impulse hammer. The acceleration signal is standardized based on the exciting force of 51. The filtering process and standardization are performed for the same purpose as the calculation unit 66 of the void determination system 100.

The calculation unit 266 obtains the eigenmode by using the eigen-orthogonal function expansion (Proper Orthogonal Decomposition, hereinafter referred to as “POD”) for the six signals after processing. POD is a known analysis method as described in "Recommendation of Application of Intrinsic Orthogonal Function Expansion to Random Fluctuation Field" in October 1995, Journal of the Japanese Society of Wind Engineering. The POD searches for a reference coordinate system that is a coordinate system suitable for expressing a random fluctuation field, and transforms the data of the fluctuation field into a reference coordinate system. As for the reference axis, the axis that maximizes the variance of the data is the primary reference axis, the axis that maximizes the variance among the axes orthogonal to this is the secondary reference axis, and the axis orthogonal to these is the axis. The axis that maximizes the variance is set as the third-order reference axis, and is determined by the procedure. If there are n measurement points, n reference axes are set. The vector representing the direction of each reference axis is the eigenvector. Each eigenvector represents each eigenmode.

In this example, since acceleration (acceleration time series data) is measured at 6 measurement points, 6 (1st to 6th) new reference coordinate axes are set, and 6 (1st to 6th) are set. The eigenvector (eigenmode) of (next) is obtained.

The determination unit 267 determines the position and size of the void G based on the eigenvector, that is, the eigenmode. The determination unit 267 determines the position and size of the void G from the intrinsic mode calculated by the calculation unit 266 by referring to the intrinsic mode previously acquired from a plurality of samples having voids G having different positions and sizes. do.

Specifically, the eigenmode is acquired in advance from the sample S having the void G whose position and size are known. The sample S is the same as the sample of the first embodiment.

21 to 25 are received waveforms when impact elastic waves are input to the first to fifth samples S1 to S5. In each figure, for reference, the received waveforms of the third vibration sensor 53C and the fourth vibration sensor 53D are shown. (A) is the received waveform by the third vibration sensor 53C, and (B) is the received waveform by the fourth vibration sensor 53D. The received waveform is a waveform that has been standardized as described above. The horizontal axis of the graph is time, and the vertical axis is acceleration.

The elastic wave input to the mortar M is multiplely reflected in the hollow rib 3. As can be seen from each figure, these received waveforms change depending on the position and size of the gap G. Although not shown, the waveforms received by the vibration sensors 53 other than the third vibration sensor 53C and the fourth vibration sensor 53D also change depending on the position and size of the gap G. However, since the received waveform itself is complicated, it is difficult to understand the difference between the received waveforms, and it is difficult to determine the position and size of the gap G based on the received waveform itself.

On the other hand, FIGS. 26 to 30 are unique modes calculated from the respective received waveforms. In each figure, (A) is a primary mode, (B) is a secondary mode, (C) is a tertiary mode, (D) is a quaternary mode, and (E). ) The figure is the 5th order mode, and the figure (F) is the 6th order mode. The shape of each eigenmode is represented by a spline curve based on the eigenvector. In each figure, voids are also shown for reference. Table 1 is the eigenvector of the first sample S1, Table 2 is the eigenvector of the second sample S2, Table 3 is the eigenvector of the third sample S3, and Table 4 is the eigenvector of the fourth sample S4. It is an eigenvector, and Table 5 is an eigenvector of the fifth sample. The eigenmodes or eigenvectors obtained from these first sample S1 to the fifth sample S5 are examples of reference eigenmodes.

As can be seen from the figure, the specific mode of each sample differs depending on the position and size of the void G. Since the eigenmode has a simpler shape than the received waveform, the difference is easier to discriminate than the difference in the received waveform according to the position and size of the gap G.

As described above, the information regarding the position and size of the void G appears in the intrinsic mode. That is, information on the position and size of the gap G, which did not clearly appear in the received waveform, can be made apparent by obtaining the unique mode.

The storage device 265 is a neural network constructed based on a unique mode previously acquired from a plurality of samples (for example, the above-mentioned first to fifth samples S1 to S5) having different positions and sizes of the gap G. The model (or classifier) is stored. FIG. 31 shows a model of a neural network. The model of FIG. 31 has an input layer, an intermediate layer, and an output layer.

The input layer contains 36 units. In this example, since the eigenvectors of the first to sixth orders are obtained, there are 36 input data for one sample. 36 input data are input to each unit of the input layer. The number of units in the intermediate layer is set to an arbitrary number. There is one unit in the output layer. The output layer outputs whether or not there is a highly important void G. The importance of the void G depends on the position and size of the void G as described above. Even if the void G exists as in the fourth sample S4, if the position is in the center and the size is not so large, it is determined that the void G is not of high importance. On the other hand, as in the case of the second sample S2 and the third sample S3, even if the void G has substantially the same size as the fourth sample S4, if the position is at the end, the void is considered to be of high importance. It is judged. That is, this model determines the position and size of the void G.

The eigenvectors obtained from each sample are input as teacher data, and the outputs of the first and fourth samples S1 and S4 have no voids of high importance, while the second, third and fifth samples S2 and S3 The coupling load (weight) between the units is adjusted so that the output of S5 has a gap of high importance. In this way, the model is pre-built. It should be noted that a plurality of striking tests may be performed for each sample to acquire a plurality of eigenvectors, and a model may be constructed using a plurality of teacher data for each sample. This neural network model represents the reference specific mode in the form of a model.

Then, the determination unit 267 inputs the eigenvectors of the first to sixth orders calculated by the calculation unit 266 into the model of the neural network stored in the storage device 265, and determines the presence or absence of the void G having a high importance. Since this model is built on the eigenmodes of multiple samples, using the model indirectly refers to eigenmodes previously obtained from multiple samples with voids G of different positions and sizes. Is equivalent to doing.

It should be noted that the reference of the eigenmode acquired in advance is not limited to using the model of the neural network. Any process can be adopted as long as it is a process that directly or indirectly refers to the eigenmode obtained in advance from a plurality of samples having voids G having different positions and sizes. For example, a deep learning model constructed by a reference specific mode previously obtained from a plurality of samples having voids G having different positions and sizes may be used.

Next, a void determination method using the void determination system 200 will be described. FIG. 32 is a flowchart of a void determination method using a unique mode.

First, in step Sb1, the operator inputs an impact elastic wave into the mortar M using the impulse hammer 51. Specifically, the operator installs six vibration sensors 53 at the above-mentioned positions in a cross section orthogonal to the bridge axis direction, including the first one-side bolt 43A that strikes with the impulse hammer 51. After that, the operator applies an impulse hammer 51 to the end edge of the first one-side bolt 43A in the direction indicated by the arrow in FIG. 19 (that is, the axial direction of the first one-side bolt 43A). The tip of the first one-side bolt 43A is in a state of being pierced into the floor slab 11, and the impact of the impulse hammer 51 applied to the first one-side bolt 43A is the floor slab 11, specifically, the mortar. It is transmitted to M.

In this way, an impulse-like elastic wave (that is, an impact elastic wave) is input to the mortar M by the impact of the impulse hammer 51. Step Sb1 corresponds to an input step of inputting an elastic wave to the structure.

At this time, the exciting force applied from the impulse hammer 51 to the first one-side bolt 43A is detected by the sensor built in the impulse hammer 51, and the detected exciting force signal is input to the data logger 61.

Next, in step Sb2, the first vibration sensor 53A to the sixth vibration sensor 53F receive the elastic wave propagating in the deck 11. The output signals of the first vibration sensor 53A to the sixth vibration sensor 53F are input to the data logger 61. The data logger 61 records the output signals of the first vibration sensor 53A to the sixth vibration sensor 53F as a response corresponding to the elastic wave input by the impulse hammer 51. Step Sb2 corresponds to a receiving step of receiving elastic waves propagating through the structure.

In step Sb3, the calculation unit 266 calculates the eigenmode, that is, the eigenvector from the acceleration signals received by the first vibration sensor 53A to the sixth vibration sensor 53F recorded in the data logger 61. Step Sb3 corresponds to a calculation step of calculating the specific mode of vibration of the structure based on the elastic wave received by the receiving step.

After that, in step Sb4, the determination unit 267 determines the presence or absence of the highly important void G by inputting the eigenvector into the model of the neural network. Step Sb4 corresponds to a determination step of determining at least one of the position and size of the void based on the intrinsic mode calculated by the calculation step.

In this way, the determination unit 267 determines the void G from the eigenmode. Since the elastic waves in the deck 11 are diffused not only in the cross section where the six vibration sensors 53 are arranged but also in the bridge axis direction, the voids in the region having a certain length in the bridge axis direction. G is determined.

The operator changes the cross section of the deck 11 that inputs and receives elastic waves, and repeats the above-mentioned steps Sb1 to Sb4. In this way, the void G is determined over the entire area in the bridge axis direction.

According to this void determination method, the voids at the interface can be determined by a method with less restrictions, similar to the method of the void determination system 100. In particular, according to this void determination method, an elastic wave can be determined by inputting an elastic wave from a portion other than the upper surface of the deck plate 2 and receiving an elastic wave at a portion other than the upper surface of the deck plate 2. can. For example, the void can be determined from below the floor slab 11.

Then, according to this void determination method, by obtaining the unique mode of the floor slab 11, it is possible to reveal the information regarding the position and size of the void G included in the received signal. Then, by referring to the reference specific mode previously acquired from the samples S1 to S5 in which the position and size of the gap G are known, the position and size of the gap G are determined from the unique mode obtained from the received signal. be able to.

As described above, the void determination system 200 is formed at the interface of the deck plate 11 (structure) having the deck plate 2 and the hollow rib 3 (outer member) forming an interface with the mortar M (inner member) and the mortar M. An impulse hammer 51 (input unit) that inputs elastic waves to the deck 11 and a vibration sensor 53 (reception unit) that receives elastic waves propagating through the deck 11 in a void determination system that determines the void G. And the calculation unit 266 that calculates the specific mode of vibration of the deck 11 based on the elastic wave received by the vibration sensor 53, and the position and size of the gap G based on the specific mode calculated by the calculation unit 266. The determination unit 267 includes a determination unit 267 for determining at least one of the determination units 267, and the determination unit 267 provides a reference specific mode which is a unique mode previously acquired from the first to fifth samples S1 to S5 having gaps G having different positions and sizes. By reference, at least one of the position and the size of the gap G is determined from the eigenmode calculated by the calculation unit 266.

In other words, the void determination method using the void determination system 200 is a floor slab 11 (structure) having a mortar M (inner member), a deck plate 2 forming an interface with the mortar M, and a hollow rib 3 (outer member). It is a void determination method for determining the void G formed at the interface in the above, and is received by an input step of inputting an elastic wave to the deck 11, a receiving step of receiving the elastic wave propagating in the deck 11, and a receiving step. A calculation step of calculating the natural mode of vibration of the deck 11 based on the generated elastic wave, and a determination step of determining at least one of the position and the size of the gap G based on the natural mode calculated by the calculation step. In the determination step, the calculation step is performed by referring to the reference unique mode, which is a unique mode previously acquired from the first to fifth samples S1 to S5 having voids G having different positions and sizes. At least one of the position and the size of the void G is determined from the intrinsic mode.

According to these configurations, if the elastic wave is input to the deck 11 and the elastic wave propagating in the deck 11 can be received, the position and size of the gap G can be determined. That is, unlike the method described in Patent Document 1, there is no strict restriction that ultrasonic waves must be transmitted from the surface of the first member toward the interface, so that the position and size of the gap G can be determined. It can be easily realized. As for the specific void determination, by calculating the specific mode of the elastic wave propagating in the deck 11, the information regarding the position and size of the void G included in the elastic wave can be clarified. Further, the reference specific mode is acquired in advance from a plurality of samples S1 to S5 having voids G having different positions and sizes, and by referring to these, the position and size of the void G are obtained from the calculated intrinsic mode described above. Can be determined.

Further, in the receiving step, elastic waves are received at a plurality of locations of the deck 11, and in the calculation step, the unique mode is calculated based on the elastic waves received at the plurality of locations by the receiving process.

According to this configuration, since elastic waves are received at a plurality of locations of the floor slab 11, the specific mode can be acquired in more detail. As a result, the determination accuracy of the void G can be improved.

<< Other Embodiments >>
As described above, the above-described embodiment has been described as an example of the technique disclosed in the present application. However, the technique in the present disclosure is not limited to this, and can be applied to embodiments in which changes, replacements, additions, omissions, etc. are made as appropriate. It is also possible to combine the components described in the above embodiment to form a new embodiment. Further, among the components described in the attached drawings and the detailed description, not only the components essential for solving the problem but also the components not essential for solving the problem in order to illustrate the technique. Can also be included. Therefore, the fact that those non-essential components are described in the accompanying drawings or detailed description should not immediately determine that those non-essential components are essential.

For example, the target for performing the void determination is not limited to the deck 11 of the bridge. Any structure having an inner member and an outer member forming an interface together with the inner member can determine the voids at the interface in any structure.

Further, the configuration of the floor slab 11 is only an example, and may be a configuration other than the above-mentioned configuration. For example, the reinforcing plate 4 may be omitted. Further, the shape of the hollow rib 3 is not limited to the above-mentioned shape. The hollow rib 3 can be formed in any shape forming a closed cross section, such as a U-shaped cross section or a V-shaped cross section. When the hollow rib 3 has such a shape, the reference first and / or second evaluation parameters or the reference specific mode are obtained in advance from the sample having the same shape. Furthermore, the number of inlets and outlets of the hollow rib 3 is not limited to the above-mentioned number. Further, concrete may be filled instead of the mortar M.

Further, the input of elastic waves to the mortar M or the floor slab 11 does not have to use the first one-side bolt 43A. A penetrating member (for example, a bolt dedicated to determining a gap) is provided only for inputting elastic waves, with one end exposed to the outside of the hollow rib 3 and the other end piercing the mortar M through the hollow rib 3. You may. Alternatively, elastic waves may be input to the mortar M by directly hitting the hollow rib 3 or the reinforcing plate 4.

The vibration sensor as a receiving unit is not limited to the acceleration pickup. For example, an AE sensor, a receiving sensor for ultrasonic flaw detection, or the like may be adopted as the vibration sensor.

Further, although the received signal of the vibration sensor 53 is an acceleration signal, it may be converted into a speed signal. That is, any physical quantity can be adopted as long as it is a physical quantity representing an elastic wave propagating in the inner member (mortar M) or the structure (deck 11).

In the void determination systems 100 and 200 and the void determination method thereof, both the position and the size of the void G are determined, but the position or the size may be determined.

Further, the number of samples S is not limited to 5, and may be any number. As a matter of course, as the number of samples S increases, the accuracy of determining the voids improves.

Further, the void determination method of the void determination system 100 and the void determination method of the void determination system 200 may be combined. The difference in the hardware configuration between the void determination system 100 and the void determination system 200 is basically only the installation position and the number of vibration sensors 53. Therefore, if a number of vibration sensors 53 that can cover both the void determination system 100 and the void determination system 200 are arranged at predetermined positions, one system can be used to determine the voids of both the void determination system 100 and the void determination system 200. Can be carried out. Since each method has a position and size of a gap G that can be easily determined, the accuracy of determining the position and size of the gap G can be improved by combining both methods. For example, the position and size of the void G may be roughly determined by the void determination method of the void determination system 100, and the position and size of the void G may be determined in detail by the void determination method of the void determination system 200.

In the void determination system 100, elastic waves propagating in the mortar M may be received at a place other than the second one-side bolt 43B. That is, since the mortar M is in contact with the deck plate 2 and the hollow rib 3, the elastic wave propagating in the mortar M is transmitted to the deck plate 2 and the hollow rib 3. Therefore, elastic waves propagating in the mortar M can also be received by detecting the vibration of the deck plate 2 or the hollow rib 3. For example, the vibration sensor 53 may be attached to the screwed stud 41 or the nut 42. Even the evaluation parameters obtained from the received signal detected at this position represent the information regarding the position and size of the gap G. Alternatively, a penetrating member (for example, a bolt dedicated to void determination) that penetrates the hollow rib 3 and has one end exposed to the outside of the hollow rib 3 and the other end pierced into the mortar M is used only for receiving elastic waves. It may be provided in.

Further, the method of determining the position and size of the void G based on the evaluation parameter I is not limited to the above-mentioned method. For example, the first vibration direction may be a direction other than the axial direction of the second one-side bolt 43B, and the second vibration direction is orthogonal to both the axial direction and the bridge axial direction of the second one-side bolt 43B. It may be in a direction other than the direction. For example, two vibration sensors 53 are installed on the bottom wall 33 of the hollow rib 3, and one vibration sensor 53 receives an elastic wave vibrating in the normal direction of the bottom wall 33 as an elastic wave vibrating in the first vibration direction. The other vibration sensor 53 may receive an elastic wave that vibrates in a direction orthogonal to the bridge axis direction and parallel to the bottom wall 33 as an elastic wave vibrating in the second vibration direction. Further, the vibration sensor 53 may be one sensor capable of detecting accelerations of two orthogonal axes instead of two.

Furthermore, instead of determining the void G based on both the first evaluation parameter Ia and the second evaluation parameter Ib, the void G may be determined based on either evaluation parameter. As can be seen from FIG. 15, each of the first evaluation parameter Ia and the second evaluation parameter Ib changes according to the position and size of the void G. That is, the position and size of the void G can be determined by using only one of the evaluation parameters.

Further, the determination map is not limited to the map of FIG. The determination map may be further divided into areas or simpler areas.

The parameters related to the strength of elastic waves are not limited to the evaluation parameter I represented by the equation (1). Any parameter can be adopted as long as it is a parameter representing the strength of the elastic wave and, by extension, the energy.

In the void determination system 200, the method for obtaining the eigenmode is not limited to the eigen-orthogonal function expansion, and any method can be adopted. Further, the number of vibration sensors 53 of the void determination system 200 is not limited to six. Any number of vibration sensors 53 of two or more can be adopted. Further, the position where the vibration sensor 53 is installed is not limited to the position shown in FIG. For example, when the number of vibration sensors 53 is four, the primary to fourth specific modes are required. In that case, the pre-built neural network model is also built based on the four eigenmodes for each sample. When actually determining the void G, the void G is determined by inputting the four unique modes obtained by the impact into the model.

Further, the method of determining the void G from the intrinsic mode is not limited to the above-mentioned method. For example, it is not necessary to use all the obtained unique modes. For example, when the contribution rate of the eigenmode with a high order is low, a model of the neural network is constructed with the eigenmode of another order, and the eigenmode of the other order is input at the time of actual judgment to enter the void G. May be determined. In addition, according to the eigen-orthogonal function expansion, the contribution rate of each eigenmode can also be obtained. Therefore, in addition to the eigenmodes, the contribution rate of each eigenmode may be used as an input to construct a model and determine the void G. Further, the output is not limited to the presence or absence of the void G having a high importance. The model may be constructed so as to output the result of determining the position and size of the gap G in more detail.

Further, the void G may be determined using the shape of the eigenmode instead of the numerical value of the eigenvector. For example, the degree of symmetry / asymmetry of the eigenmode may be quantified to determine the void G.

Further, the determination of the void G is not limited to the method using the neural network. For example, the gap G may be determined by image recognition processing from the shape of the unique mode.

As described above, the techniques disclosed herein are useful for void determination methods and void determination systems.

100,200 Void judgment system 11 Floor slab (structure)
2 Deck plate (outer member, first member)
3 Hollow rib (outer member, second member)
4 Reinforcing plate (connecting member)
43A 1st one side bolt (1st penetrating member, penetrating member)
43B 2nd one side bolt (2nd penetrating member, penetrating member)
51 Impulse hammer (input section)
53 Vibration sensor (receiver)
66,266 Calculation unit 67,267 Judgment unit M Mortar (inner member)
S sample

Claims (11)

A void determination method for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input process for inputting elastic waves to the inner member,
A receiving step of receiving elastic waves propagating through the inner member,
A calculation step of calculating evaluation parameters related to the strength of elastic waves received by the reception step, and a calculation step.
Including a determination step of determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation step.
In the calculation step, the first evaluation parameter, which is the evaluation parameter of the elastic wave vibrating in the predetermined first vibration direction, and the evaluation parameter of the elastic wave vibrating in the second vibration direction different from the first vibration direction are used. Calculate with a certain second evaluation parameter ,
In the determination step, the first evaluation parameter for reference and the second evaluation parameter for reference, which are the first evaluation parameter and the second evaluation parameter , which are obtained in advance from a plurality of samples having voids having different positions or sizes, are referred to. A void determination method, characterized in that at least one of the position and size of the void is determined from the first evaluation parameter and the second evaluation parameter calculated in the calculation step. In the void determination method according to claim 1 ,
The void determination method, wherein the first vibration direction corresponds to the vibration direction of the elastic wave input in the input step. A void determination method for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input process for inputting elastic waves to the inner member,
A receiving step of receiving elastic waves propagating through the inner member,
A calculation step of calculating evaluation parameters related to the strength of elastic waves received by the reception step, and a calculation step.
Including a determination step of determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation step.
The structure further includes a first penetrating member and a second penetrating member that penetrate the outer member, one end of which is exposed to the outside of the outer member, and the other end of which is pierced into the inner member.
In the input step, elastic waves are input from the outside of the outer member to the inner member via the first penetrating member.
In the receiving step, elastic waves propagating through the inner member are received from the outside of the outer member via the second penetrating member.
In the calculation step, the first evaluation parameter, which is the evaluation parameter of the elastic wave vibrating in the predetermined first vibration direction, is calculated.
In the determination step, the first evaluation parameter calculated in the calculation step is performed by referring to the reference first evaluation parameter, which is the first evaluation parameter obtained in advance from a plurality of samples having voids having different positions or sizes. (1) A void determination method comprising determining at least one of the position and size of a void from an evaluation parameter . In the void determination method according to claim 3 ,
The outer member has a divided structure including at least a first member and a second member.
The structure further includes a connecting member that connects the first member and the second member.
A void determination method, wherein the first penetrating member and the second penetrating member are members for fastening the connecting member to the second member. A void determination method for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input step of inputting elastic waves to the structure and
A receiving step of receiving elastic waves propagating in the structure,
A calculation step of calculating the natural mode of vibration of the structure based on the elastic wave received by the reception step, and a calculation step.
Including a determination step of determining at least one of the position and size of the void based on the specific mode calculated by the calculation step.
The structure further has a penetrating member that penetrates the outer member, one end of which is exposed to the outside of the outer member, and the other end of which is pierced into the inner member.
In the input step, elastic waves are input from the outside of the outer member to the inner member via the penetrating member.
In the determination step, by referring to the reference eigenmode which is the eigenmode obtained in advance from a plurality of samples having voids having different positions or sizes, the voids from the eigenmode calculated in the calculation step are used. A void determination method comprising determining at least one of a position and a size. In the void determination method according to claim 5 ,
In the receiving step, elastic waves are received at a plurality of points of the structure, and the elastic waves are received.
The calculation step is a void determination method, characterized in that the unique mode is calculated based on elastic waves received at a plurality of locations by the reception step. In the void determination method according to claim 5 ,
The outer member has a divided structure including at least a first member and a second member.
The structure further includes a connecting member that connects the first member and the second member.
A void determination method, wherein the penetrating member is a member for fastening the connecting member to the second member. In the void determination method according to any one of claims 1 to 7 .
The structure is joined to a deck plate, a hollow rib joined to the lower surface of the deck plate to form a closed cross section together with the deck plate, and a mortar or mortar filled in the space partitioned by the deck plate and the ribs. A deck of a bridge with concrete
The outer members are the deck plate and the ribs.
The void determination method, wherein the inner member is the mortar or the concrete. A void determination system for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input unit that inputs elastic waves to the inner member,
A receiving unit that receives elastic waves propagating through the inner member,
A calculation unit that calculates evaluation parameters related to the strength of elastic waves received by the reception unit, and a calculation unit.
A determination unit for determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation unit is provided.
The calculation unit uses the first evaluation parameter, which is the evaluation parameter of the elastic wave vibrating in the predetermined first vibration direction, and the evaluation parameter of the elastic wave vibrating in the second vibration direction different from the first vibration direction. Calculate with a certain second evaluation parameter ,
The determination unit refers to the first evaluation parameter for reference and the second evaluation parameter for reference, which are the first evaluation parameter and the second evaluation parameter , which are obtained in advance from a plurality of samples having voids having different positions or sizes. The void determination system is characterized in that at least one of the position and the size of the void is determined from the first evaluation parameter and the second evaluation parameter calculated by the calculation unit. A void determination system for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input unit that inputs elastic waves to the inner member,
A receiving unit that receives elastic waves propagating through the inner member,
A calculation unit that calculates evaluation parameters related to the strength of elastic waves received by the reception unit, and a calculation unit.
A determination unit for determining at least one of the position and size of the void based on the evaluation parameter calculated by the calculation unit is provided.
The structure further includes a first penetrating member and a second penetrating member that penetrate the outer member, one end of which is exposed to the outside of the outer member, and the other end of which is pierced into the inner member.
The input unit inputs an elastic wave from the outside of the outer member to the inner member via the first penetrating member.
The receiving unit receives elastic waves propagating through the inner member from the outside of the outer member via the second penetrating member.
The calculation unit calculates the first evaluation parameter, which is the evaluation parameter of the elastic wave vibrating in the predetermined first vibration direction.
The determination unit is calculated by the calculation unit by referring to a reference first evaluation parameter, which is the first evaluation parameter obtained in advance from a plurality of samples having voids having different positions or sizes. 1 A void determination system characterized in that at least one of the position and size of a void is determined from an evaluation parameter. A void determination system for determining a void formed at an interface in a structure having an inner member and an outer member forming an interface together with the inner member.
An input unit that inputs elastic waves to the structure,
A receiver that receives elastic waves propagating through the structure,
A calculation unit that calculates the natural mode of vibration of the structure based on the elastic wave received by the reception unit, and a calculation unit.
A determination unit for determining at least one of the position and size of the void based on the specific mode calculated by the calculation unit is provided.
The structure further has a penetrating member that penetrates the outer member, one end of which is exposed to the outside of the outer member, and the other end of which is pierced into the inner member.
The input unit inputs an elastic wave from the outside of the outer member to the inner member via the penetrating member.
The determination unit refers to the reference eigenmode, which is the eigenmode previously acquired from a plurality of samples having voids having different positions or sizes, from the eigenmode calculated by the calculation unit. A void determination system comprising determining at least one of position and size.

Images