JP7490164B1 - Repair content determination device, repair device, learning device, inference device, repair content determination system, and repair method - Google Patents

Abstract

The repair device (10), which is a repair content determination device, is characterized by having an inspection unit (14) which is an inspection result acquisition unit that acquires inspection results of a crack (C) that has occurred in an inspected material portion of a structure (S), and a judgment unit (15) that determines the repair content including a repair pattern indicating whether or not to repair the crack (C) with a repair material that is a liquid containing particles, based on the inspection results, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material into the crack (C) before applying the repair material.

Description

The present disclosure relates to a repair content determination device, a repair device, a learning device, an inference device, a repair content determination system, and a repair method for repairing cracks in a structure.

Conventionally, there is a technology that suppresses the growth of cracks in structures by applying a repair material, which is a liquid containing particles, to the cracks. Cracks grow by repeatedly opening and closing, but by applying a repair material to the crack, the particles contained in the repair material act as a wedge, preventing the crack from closing, thereby slowing the rate at which the crack grows. For this reason, in order to increase the effect of suppressing crack growth, it is important to allow the repair material to penetrate into the crack and then have it remain within the crack.

Patent Document 1 describes a repair material for slowing down the rate of fatigue crack growth in metal materials, which has a specified viscosity range that provides excellent penetration into cracks.

Patent No. 4852163

However, the above conventional technology has the problem that, depending on the state of the crack, it may be difficult to make the repair material penetrate into the crack. For example, whether the repair material penetrates into the crack depends on the direction of the crack relative to the direction of gravity. In a crack whose opening faces downward, such as a crack on a ceiling surface, the weight of the repair material does not exert a force to penetrate into the crack, as compared to a crack whose opening faces upward or sideways, so it may be difficult for the repair material to penetrate into the crack.

The present disclosure has been made in consideration of the above, and aims to obtain a repair content determination device that can increase the penetration of repair material into a crack regardless of the condition of the crack.

In order to solve the above-mentioned problems and achieve the objective, the repair content determination device disclosed herein is characterized by comprising an inspection result acquisition unit that acquires inspection results of cracks that have occurred in an inspected material portion of a structure, and a judgment unit that determines repair content including a repair pattern that indicates whether or not to repair the crack with a repair material, which is a liquid containing particles, based on the length of the crack indicated by the inspection results and the angle the crack makes with respect to the direction of gravity, and, if repair with the repair material is to be performed, whether or not to apply a pretreatment material into the crack before applying the repair material.

The present disclosure has the effect of making it possible to obtain a repair content determination device that can increase the penetration of repair material into a crack regardless of the condition of the crack.

FIG. 1 is a diagram showing a functional configuration of a repair device according to a first embodiment; FIG. 2 is a diagram for explaining the function of the repair device shown in FIG. FIG. 2 is a diagram showing an example of a detailed configuration of an inspection unit shown in FIG. 1 . FIG. 2 is a diagram for explaining the function of a determination unit shown in FIG. 1 ; FIG. 1 is an explanatory diagram of the relationship between the repair pattern and the operation of the repair unit; An explanatory diagram of the process in which the determination unit determines the material to be used. FIG. 1 is a diagram illustrating an example of a detailed configuration of a determination unit according to a first embodiment; FIG. 1 is a diagram showing a bridge, which is an example of a structure to be inspected by the repair device. An explanatory diagram of the crack length calculated by the calculation unit An explanatory diagram of the crack angle calculated by the calculation unit FIG. 1 is an explanatory diagram of a first example of a specific method in which a calculation unit calculates a crack length and a crack angle. FIG. 11 is an explanatory diagram of a second example of a specific method in which the calculation unit calculates the crack length and the crack angle. FIG. 11 is an explanatory diagram of a third example of a specific method for the calculation unit to calculate the crack length and the crack angle. A flowchart for explaining the operation of the repair device according to the first embodiment. 1 is a flowchart for explaining details of a determination process according to the first embodiment. FIG. 13 is a diagram showing a functional configuration of a repair device according to a second embodiment. FIG. 13 is a diagram showing a detailed configuration of a determination unit according to a second embodiment; 11 is a flowchart for explaining the operation of a determination unit according to a second embodiment. FIG. 13 is a diagram showing a functional configuration of a learning device according to a third embodiment. A flowchart for explaining the learning process of the learning device. FIG. 13 is a diagram showing a functional configuration of a determination unit of a repair device according to a third embodiment. Configuration diagram of an inference device for repair equipment A flowchart for explaining an inference process of an inference device. FIG. 13 is a diagram showing a functional configuration of a repair device according to a fourth embodiment. FIG. 13 is a diagram showing a configuration of a repair content determination system according to a fifth embodiment. FIG. 13 is a diagram showing a configuration of a repair content determination system according to a sixth embodiment. A flowchart for explaining the operation of the repair content determination system according to the sixth embodiment. FIG. 13 is a diagram showing a configuration of a repair content determination system according to a seventh embodiment. FIG. 13 is a diagram showing a configuration of a repair content determination system according to an eighth embodiment. FIG. 23 is a diagram showing an example of a movement path of a repair device according to an eighth embodiment. FIG. 23 is a diagram showing another example of a route taken by the repair device according to the eighth embodiment to return from the goal point to the start point; FIG. 1 is a diagram showing an example of the configuration of a computer system that realizes a repair content determination device. FIG. 1 is a diagram showing an example of a processing circuit for realizing each function of a repair device or a repair content determination system.

Below, the repair content determination device, repair device, learning device, inference device, repair content determination system, and repair method relating to embodiments of the present disclosure are described in detail with reference to the drawings.

Embodiment 1.
FIG. 1 is a diagram showing a functional configuration of a repair device 10 according to a first embodiment. The repair device 10 has a function of inspecting a crack C generated in a structure S, determining a repair content based on the inspection result, and repairing the crack C according to the determined repair content. The repair device 10 is an example of a repair content determination device having a function of determining the repair content of the crack C. The repair device 10 has a main body 11, a suction means 12 for fixing the main body 11 of the repair device 10 to the structure S, a driving means 13 for moving the repair device 10 along the structure S, an inspection unit 14 for inspecting the crack C to be repaired and outputting the inspection result, a determination unit 15 for determining the repair content of the crack C based on the inspection result, and a repair unit 16 for repairing the target crack C according to the repair content determined by the determination unit 15. The repair unit 16 has a pre-processing unit 17, an intermediate processing unit 18, and a post-processing unit 19.

Here, the material to be inspected is a material in which fatigue cracks occur due to repeated localized plastic deformation on the material surface, and the fatigue cracks grow as the tip of the crack is repeatedly blunted and re-sharpened due to plastic deformation. Such a material is called the "material to be inspected." The structure S is, for example, a bridge, a vehicle, an aircraft, a power generation device, a machine tool, etc. The main examples of the material to be inspected are, for example, steel, stainless steel, and aluminum alloys, but are not limited to metals and also include resins and ceramics. The entire structure S does not need to be made of the material to be inspected, but the crack C to be repaired is a crack C that has occurred in the part of the structure S made of the material to be inspected.

Figure 2 is a diagram for explaining the function of the repair device 10 shown in Figure 1. The repair device 10 can perform repair processing using a pretreatment material 21 and a repair material 22.

The repair material 22 is a liquid containing particles. By applying the repair material 22 to the inside of the crack C and the repair material 22 remaining inside the crack C, the particles contained in the repair material 22 exert a wedge effect, which can suppress the growth of the crack C. The wedge effect is also called a bridging effect. The phenomenon in which the particles contained in the repair material 22 prevent the crack C from closing is called the wedge effect. Since the crack C grows by repeatedly opening and closing, the wedge effect can suppress the growth of the crack C. In order to improve the crack growth suppression effect, it is important to make the repair material 22 penetrate to the tip of the crack C and to make the penetrated repair material 22 remain inside the crack C. The lower the viscosity of the repair material 22, the easier it is to make the repair material 22 penetrate into the crack C, but it is more difficult to make the repair material 22 remain in the crack C. Therefore, by applying the pretreatment material 21 into the crack C before applying the repair material 22, it is possible to increase the permeability of the repair material 22 into the crack C. The pretreatment material 21 is a liquid that has a higher wettability with respect to the structure S than the repair material 22. In other words, the contact angle θ ₁ between the pretreatment material 21 and the surface of the structure S is smaller than the contact angle θ ₂ between the repair material 22 and the surface of the structure S.

The pre-treatment unit 17 of the repair unit 16 has a function of executing a pre-treatment process of applying a pre-treatment material 21 to the inside of the crack C. The intermediate treatment unit 18 has a function of executing an intermediate treatment process of removing the excess of the pre-treatment material 21 applied in the pre-treatment process. The post-treatment unit 19 has a function of executing a post-treatment process of applying a repair material 22 to the inside of the crack C. Note that FIG. 2 shows the state in which the pre-treatment process, intermediate treatment process, and post-treatment process are all performed sequentially, so in the post-treatment process, the crack C is filled with the repair material 22 with a film of the pre-treatment material 21 formed on the inner wall surface of the crack C. However, the repair unit 16 does not need to drive all of the pre-treatment unit 17, intermediate treatment unit 18, and post-treatment unit 19 depending on the state of the crack C. For example, the repair unit 16 may sequentially drive all of the pre-treatment unit 17, intermediate treatment unit 18, and post-treatment unit 19, or may drive only the post-treatment unit 19 without driving the pre-treatment unit 17 and intermediate treatment unit 18, or may not drive any of the pre-treatment unit 17, intermediate treatment unit 18, and post-treatment unit 19 and therefore no repairs may be performed.

3 is a diagram showing an example of a detailed configuration of the inspection unit 14 shown in FIG. 1. The inspection unit 14 has a LiDAR (Light Detection and Ranging) scanner 141 and a camera 142. The LiDAR scanner 141 is a device that irradiates a laser beam on an object and measures the distance, properties, and shape of the object from the time it takes to receive the reflected light. The camera 142 is a device that acquires an image of the crack C. The camera 142 is synchronized with the LiDAR scanner 141. Each of the LiDAR scanner 141 and the camera 142 is equipped with a gyro sensor and can acquire its own attitude. This allows the surface shape of the inspection target material measured by the LiDAR scanner 141 and the camera 142 to be associated with the direction of gravity. The LiDAR scanner 141 can measure the surface of the inspection target material and measure the depth inside the crack C. The camera 142 can output an image of the surface of the inspection target material.

Figure 4 is a diagram for explaining the function of the judgment unit 15 shown in Figure 1. The judgment unit 15 receives the photographed image of the crack C and direction information indicating the direction of gravity in the photographed image as the inspection result from the inspection unit 14, determines the repair content based on the inspection result, and outputs the determined repair content. The direction information is, for example, the output of a gyro sensor. The repair content output by the judgment unit 15 includes at least a repair pattern indicating whether or not to perform repair with the repair material 22, and whether or not to apply the pretreatment material 21 before applying the repair material 22 when performing repair with the repair material 22. In addition, the repair content may further include a material to be used for repair when performing repair. The material to be used for repair includes at least the repair material 22 when performing repair, and further includes the pretreatment material 21 when applying the pretreatment material 21. For example, the determination unit 15 selects one of three repair patterns based on the inspection results: repair pattern (1) in which the pretreatment material 21 is applied and then the repair material 22 is applied, repair pattern (2) in which only the repair material 22 is applied without applying the pretreatment material 21, and repair pattern (3) in which no repair is performed. In addition, when the determination unit 15 selects the repair pattern (1), the determination unit 15 may determine the pretreatment material 21 and the repair material 22 as the materials to be used for repair based on the inspection results, and when the determination unit 15 selects the repair pattern (2), the determination unit 15 may determine the repair material 22 as the material to be used for repair based on the inspection results.

Figure 5 is an explanatory diagram of the relationship between the repair pattern and the operation of the repair unit 16. The judgment unit 15 selects one of repair patterns (1), (2), and (3) based on the inspection results. Repair pattern (1) is a pattern in which the pre-treatment unit 17, the intermediate treatment unit 18, and the post-treatment unit 19 are driven in the order described above, and after applying the pre-treatment material 21 to the crack C, the excess of the pre-treatment material 21 is removed, and then the repair material 22 is applied. Repair pattern (2) is a pattern in which only the post-treatment unit 19 is driven, and the pre-treatment material 21 is not applied, but the repair material 22 is applied to the crack C. Repair pattern (3) is a repair pattern in which the repair unit 16 does not operate, and no repair is performed.

Figure 6 is an explanatory diagram of the process in which the determination unit 15 determines the material to be used. When the determination unit 15 selects repair pattern (1), it determines the combination of the pretreatment material 21 to be used and the repair material 22 to be used. For example, as shown in Figure 6, the determination unit 15 can select the pretreatment material 21 from three candidates, "a", "b", and "c". The determination unit 15 can also select the repair material 22 from three candidates, "A", "B", and "C". In Figure 6, "b" is selected as the pretreatment material 31, and "C" is selected as the repair material 22.

There are various possible methods for determining the repair pattern and the materials to be used for the repair, but as an example, a method for determining the repair content based on the length of crack C and the angle that crack C makes with respect to the direction of gravity will be described. Specifically, the angle that crack C makes with respect to the direction of gravity can be the angle that crack C makes with respect to the direction of gravity in the depth direction. Hereinafter, for simplicity, "the angle that crack C makes with respect to the direction of gravity" will be referred to as "the angle of crack C."

7 is a diagram showing an example of a detailed configuration of the determination unit 15 according to the first embodiment. The determination unit 15 has a calculation unit 151, a repair pattern determination unit 152, and a material to be used determination unit 153.

The calculation unit 151 calculates data used by the judgment unit 15 for judgment from the inspection results output by the inspection unit 14. Specifically, the inspection unit 14 outputs, as the inspection results, a photographed image of the crack C and directional information indicating the direction of gravity in the photographed image, and the calculation unit 151 calculates the length of the crack C and the angle of the crack C from these inspection results. The calculation unit 151 outputs the calculated length of the crack C and the angle of the crack C to the repair pattern determination unit 152 and the material used determination unit 153.

Here, the specific processing of the calculation unit 151 will be explained using a concrete example. Figure 8 is a diagram showing a bridge S1, which is an example of a structure S to be inspected by the repair device 10. In the bridge S1, a pavement 32 is superimposed on a deck plate 31, and vehicles and the like pass over the pavement 32. Beneath the deck plate 31, support members 30 such as U-ribs 33, horizontal ribs 34, and vertical reinforcement members 37 support the deck plate 31.

9 is an explanatory diagram of the length of the crack C calculated by the calculation unit 151. Here, the crack C that occurred in the weld bead 35 between the deck plate 31 and the U-rib 33 of the bridge S1 is shown. Here, the crack C occurs from the starting point P1, and progresses from the side of the weld bead 35 formed between the side of the U-rib 33 and the deck plate 31 to the surface of the weld bead 35 formed between the outer surface of the U-rib 33 and the deck plate 31. Here, the length L1 of the crack C is the length of the crack line on the surface of the weld bead 35. At this time, if the crack line is not a straight line, the length L1 of the crack C may be the actual length of the crack line or the length of an approximate straight line of the crack line. Although not used in the first embodiment, the opening amount of the crack C on the surface where the crack C occurred is the opening amount W1.

10 is an explanatory diagram of the angle of the crack C calculated by the calculation unit 151. The angle of the crack C is an angle _φ1 between the gravity direction D1 and the depth direction D2 of the crack C. Hereinafter, this angle may be referred to as the angle _φ1 of the crack C.

FIG. 11 is an explanatory diagram of a first example of a specific method in which the calculation unit 151 calculates the length L1 of the crack C and the angle φ ₁ of the crack C. The length L1 of the crack C can be calculated by analyzing the image captured by the camera 142 provided in the inspection unit 14. In addition, in order to calculate the angle φ ₁ of the crack C, it is necessary to grasp the gravity direction D1 and the depth direction D2 of the crack C. Here, since each of the LiDAR scanner 141 and the camera 142 of the inspection unit 14 is equipped with a gyro sensor and can acquire its own attitude, it is possible to associate the information acquired by the LiDAR scanner 141 and the camera 142 with the gravity direction D1. In addition, the calculation unit 151 can estimate the depth direction D2 of the crack C from the result of the depth measurement inside the crack C by the LiDAR scanner 141. Alternatively, the calculation unit 151 may calculate the vertical direction D3 at the crack line position on the surface of the weld bead 35, and the vertical direction D3 may be the depth direction D2 of the crack C.

12 is an explanatory diagram of a second example of a specific method in which the calculation unit 151 calculates the lengths L1-1, L1-2 of the cracks C-1, C-2 and the angles φ _1-1 , φ _1-2 of the cracks C-1, C-2. In the first example of FIG. 11, a crack C occurring in a weld bead 35 formed between the deck plate 31 and the U-rib 33 is illustrated. In FIG. 12, a crack C occurring in a butt weld bead 36 formed between the U-rib 33-1 and the U-rib 33-2 is illustrated. Of the cracks C occurring in the butt weld bead 36, the crack C occurring on the side of the U-ribs 33-1, 33-2 is referred to as crack C-1, and the crack C occurring on the underside of the U-ribs 33-1, 33-2 is referred to as crack C-2.

The crack C-1 propagates from the starting point P1-1 near the inner surface of the U-rib 33-1, 33-2 toward the outer side surface of the U-rib 33-1, 33-2, and then propagates in two directions opposite to each other on the surface of the butt weld bead 36. In this case, the calculation unit 151 can set the length of the crack line on the surface of the butt weld bead 36 as the length L1-1 of the crack C-1. In addition, the calculation unit 151 can estimate the depth direction D2-1 of the crack C-1 from the result of the depth measurement inside the crack C-1 by the LiDAR scanner 141. Alternatively, the calculation unit 151 may calculate the vertical direction at the crack line position on the surface of the butt weld bead 36, and set the vertical direction as the depth direction D2-1 of the crack C. The calculation unit 151 can calculate the angle φ _1-1 of the crack C-1 from the gravity direction D1 and the depth direction D2-1.

Like the crack C-1, the crack C-2 progresses from a starting point P1-2 near the inner surface of the U-rib 33-1, 33-2 toward the outer side surface of the U-rib 33-1, 33-2, and then progresses in two directions opposite to each other on the surface of the butt weld bead 36. For the crack C-2, the calculation unit 151 can set the length of the crack line on the surface of the butt weld bead 36 as the length L1-2 of the crack C-2. In addition, the calculation unit 151 can estimate the depth direction D2-2 of the crack C-2 from the result of the depth measurement inside the crack C-2 by the LiDAR scanner 141. Alternatively, the calculation unit 151 may calculate the vertical direction at the crack line position on the surface of the butt weld bead 36, and set the vertical direction as the depth direction D2-2 of the crack C. The calculation unit 151 can calculate the angle _φ1-2 of the crack C-2 from the gravity direction D1 and the depth direction D2-2. Since the crack C-2 occurs on a horizontal plane and the depth direction D2-2 of the crack C-2 is close to the gravity direction, the value of the angle _φ1-2 of the crack C-2 is approximately 0.

FIG. 13 is an explanatory diagram of a third example of a specific method in which the calculation unit 151 calculates the length L1-3 of the crack C-3 and the angle φ _1-3 of the crack C-3. In the third example, a crack C-3 generated from a contact portion between the deck plate 31 and the fillet weld bead 38 formed between the deck plate 31 and the vertical reinforcement member 37 is illustrated. This crack C-3 is an initial crack generated in a direction penetrating the deck plate 31 from the starting point P1-3 of the contact portion between the fillet weld bead 38 and the deck plate 31, and then spreads in the in-plane direction of the deck plate 31. At this time, since force is likely to be applied to the contact portion between the fillet weld bead 38 and the deck plate 31, the crack C-3 often spreads along the end of the fillet weld bead 38. In this case, as shown in FIG. 13, the crack line of the crack C-3 is a curve. In this case, the length L1-3 of the crack C-3 may be the length of the curved crack line, or as shown in the figure, it may be the length of a straight line in the direction in which the crack C-3 spreads within the plane of the deck plate 31. The depth direction D2-3 of the crack C-3 is in the thickness direction of the surface of the deck plate 31, which is a horizontal plane, and is therefore approximately the direction of gravity D1, and the angle φ _1-3 of the crack C-3 is a value close to 0.

Returning to the explanation of Fig. 7, the repair pattern determination unit 152 can determine the repair pattern for the target crack C from among repair pattern (1), repair pattern (2), and repair pattern (3) based on the output of the calculation unit 151. Here, the repair pattern determination unit 152 determines the repair pattern based on the length L1 of the crack C and the angle _φ1 of the crack C calculated by the calculation unit 151.

For cracks C that are short and minute, repair itself may not be necessary yet. Therefore, the judgment unit 15 can judge whether or not to perform repair based on the length L1 of the crack C. Specifically, when the length of the crack C is less than the first threshold value, the judgment unit 15 does not perform repair itself, and when the length of the crack C is equal to or greater than the first threshold value, repair is performed. In addition, the angle φ ₁ of the crack C changes the ease with which the repair material 22 penetrates into the crack C and the ease with which the repair material 22 remains inside the crack C. For example, when cracks C of the same shape are compared with the angle φ ₁ of the crack C being a value between 0 degrees and 180 degrees, the larger the angle φ ₁ of the crack C, such as a crack C with an opening on the upper surface of the structure S facing upward, the easier it is for the repair material 22 to penetrate into the crack C due to gravity, and since the tip of the crack C is in the gravity direction, the repair material 22 is likely to accumulate in the crack C unless the crack C penetrates the structure S. Furthermore, when the angle φ ₁ of the crack C is small, such as a crack C occurring on the underside of the ceiling of the structure S or a crack C occurring on the side of the structure S, which has an opening facing downward or sideways, the repair material 22 is unlikely to penetrate into the crack C and is unlikely to remain inside the crack C. For this reason, the determination unit 15 can determine whether or not to use the pretreatment material 21 based on the angle φ ₁ of the crack C. Specifically, the determination unit 15 uses the pretreatment material 21 when the angle φ ₁ of the crack C is less than the second threshold value, and does not use the pretreatment material 21 when the angle φ ₁ of the crack C is equal to or greater than the second threshold value.

The material to be used determination unit 153 determines the material to be used to repair the target crack C based on the output of the calculation unit 151. The material to be used determination unit 153 determines the material to be used according to the repair pattern determined by the repair pattern determination unit 152. Specifically, in the case of repair pattern (1), the material to be used determination unit 153 determines the pretreatment material 21 and repair material 22 to be used, in the case of repair pattern (2), the repair material 22 to be used, and in the case of repair pattern (3), no repair is performed, so no material to be used is determined.

When determining the material to be used for repair based on the length L1 of the crack C and the angle φ ₁ of the crack C, first, the appropriate shape and size of the particles contained in the repair material 22 are determined based on the length L1 of the crack C. Also, the appropriate viscosity of the repair material 22 is determined based on the angle φ ₁ of the crack C. The viscosity of the repair material 22 can be adjusted by changing the content of the particles contained in the repair material 22. However, if the content of the particles is reduced to lower the viscosity, the amount of particles per unit volume of the repair material 22 is reduced, so that the penetration power into the crack C increases, but the maximum amount of particles that can be accommodated in the space in the crack C decreases. In selecting the repair material 22, the penetration power of the repair material 22 and the amount of particles that penetrate into the crack C are taken into consideration, and a repair material 22 that maximizes the amount of particles in the crack C is selected. The determination unit 15 selects a repair material 22 that contains particles of a shape and size determined according to the length L1 of the crack C and has a viscosity according to the angle φ ₁ of the crack C. Next, the determination unit 15 selects a pretreatment material 21 in accordance with the selected repair material 22 and the material of the structure S.

In the case of repair pattern (1), the material determination unit 153 determines the materials to be used so that the combination of the pretreatment material 21 and the repair material 22 to be used satisfies " _2s1 x cos _θ1 / _ρ1 > _2s2 x cos _θ2 / _ρ2 ." Here, the surface tension _s1 of the pretreatment material 21, the contact angle _θ1 of the pretreatment material 21 with the surface of the structure S, the density _ρ1 of the pretreatment material 21, the surface tension _s2 of the repair material 22, the contact angle _θ2 of the repair material 22 with the surface of the structure S, and the density _ρ2 of the repair material 22 are defined as follows.

Here, the pretreatment material 21 is preferably an oil having a density _ρ1 of 0.83 g/ ^cm3 or more and less than 0.89 g/ ^cm3 , a contact angle _θ1 with the surface of the structure S of 0 degrees or more and less than 10 degrees, and a surface tension _s1 of 20 dyn/cm or more and less than 50 dyn/cm.

Figure 14 is a flowchart for explaining the operation of the repair device 10 according to the first embodiment. The repair device 10 first performs an inspection process by the inspection unit 14 (step S10). Next, the repair device 10 performs a judgment process by the judgment unit 15 based on the inspection results by the inspection unit 14 (step S20). Details of the judgment process will be described later. When the repair content is determined as a result of the judgment process, the repair device 10 determines whether the determined repair content is "repair required" (step S30).

If "repair is required" (step S30: Yes), that is, if the repair pattern included in the repair content is repair pattern (1) or repair pattern (2), the repair device 10 performs repair processing using the repair unit 16 (step S40). Specifically, in the case of repair pattern (1), the repair device 10 applies pretreatment material 21 using the pretreatment unit 17, intermediate treatment unit 18, and post-treatment unit 19 of the repair unit 16, removes excess pretreatment material 21, and then applies repair material 22. At this time, if the repair content includes a material to be used for repair, the repair unit 16 performs repair processing using the pretreatment material 21 and repair material 22 indicated in the repair content. In the case of repair pattern (2), the repair device 10 applies repair material 22 using the post-treatment unit 19 of the repair unit 16. At this time, if the repair content includes a material to be used for repair, the repair unit 16 performs repair processing using the repair material 22 indicated in the repair content.

If "repair required" is not determined (step S30: No), that is, if the repair pattern included in the repair content is repair pattern (3), the processing of step S40 is omitted. Next, the repair device 10 determines whether or not repair is complete (step S50). If repair is complete (step S50: Yes), the repair device 10 ends the processing. If repair is not complete (step S50: No), the repair device 10 moves the repair device 10 using the driving means 13 (step S60) and returns to the processing of step S10.

In addition, in the process of determining whether or not repair is complete in step S50, it can be determined that repair is complete when the user performs an end operation. Alternatively, if the repair device 10 has a function of automatically moving along a specified route, it can be determined that repair is complete when the end point of the specified route is reached.

Figure 15 is a flowchart for explaining the details of the determination process in embodiment 1. The process shown in Figure 15 corresponds to the determination process of step S20 in Figure 14.

The calculation unit 151 of the determination unit 15 first estimates the depth direction D2 of the crack C (step S101). Next, the determination unit 15 calculates the angle between the gravity direction D1 based on the directional information included in the inspection result and the depth direction D2 of the crack C estimated in step S101 (step S102). The calculation unit 151 also calculates the length L1 of the crack C from the captured image of the crack C included in the inspection result (step S103).

The repair pattern determination unit 152 determines whether the length L1 of the crack C calculated in step S103 is equal to or greater than the first threshold value (step S104). If the length L1 of the crack C is equal to or greater than the first threshold value (step S104: Yes), the repair pattern determination unit 152 determines that "repair is performed" (step S105). If the length L1 of the crack C is less than the first threshold value (step S104: No), the repair pattern determination unit 152 determines that "repair is not performed" and selects repair pattern (3) (step S106).

Following step S105, the repair pattern determination unit 152 determines whether the angle φ ₁ of the crack C is less than the second threshold value (step S107). If the angle φ ₁ of the crack C is less than the second threshold value (step S107: Yes), the repair pattern determination unit 152 determines that "pretreatment is performed" and selects the repair pattern (1) (step S108). The material determination unit 153 selects the pretreatment material 21 and the repair material 22 to be used when repairing the target crack C (step S109).

If the angle _φ1 of the crack C is equal to or greater than the second threshold value (step S107: No), the repair pattern determination unit 152 determines that "pre-processing is not performed" and selects the repair pattern (2) (step S110). The material determination unit 153 selects the repair material 22 to be used when repairing the target crack C (step S111).

In the above embodiment 1, the directional information is based on the self-attitude acquired by the gyro sensor, but this embodiment is not limited to such an example. If a gyro sensor is not provided, a vertical indicator such as a pendulum can be used. When a pendulum is used, the direction of gravity can be recorded by capturing an image with the pendulum within the angle of view of the camera 142.

The repair device 10 according to the first embodiment described above is an example of a repair content determination device that determines the repair content of the crack C that has occurred in the inspected material portion of the structure S. The repair device 10 has an inspection unit 14, which is an inspection result acquisition unit that acquires the inspection result of the crack C that has occurred in the inspected material portion of the structure S, and a determination unit 15 that determines the repair content including a repair pattern indicating whether or not to repair the crack C with a repair material 22 that is a liquid containing particles, and whether or not to apply a pretreatment material 21 to the crack C before applying the repair material 22 when repairing with the repair material 22. As a result, it is determined whether or not to repair the crack C and whether or not to apply the pretreatment material 21 before applying the repair material 22 when repairing, according to the inspection result of the crack C. By following this repair content, it is possible to apply the pretreatment material 21 as necessary to increase the permeability of the repair material 22 into the crack C. Therefore, it is possible to increase the permeability of the repair material 22 into the crack C regardless of the state of the crack C. In addition, since the repair device 10 determines the repair content, a higher effect of suppressing the propagation of the crack C can be achieved without requiring the repair worker to be highly skilled.

Here, the judgment unit 15 can determine the repair pattern based on the length L1 of the crack C indicated by the inspection result and the angle _φ1 of the crack C with respect to the gravity direction D1. For a very small crack C with a short length L1, repair may not be required yet. Also, depending on the angle _φ1 of the crack C, the pretreatment material 21 may not be required. It is desirable to determine the repair pattern taking these factors into consideration.

The repair contents further include the material to be used for repairing the crack C, and when performing repair with the repair material 22, the judgment unit 15 determines the repair material 22 to be used for repairing the crack C as the above-mentioned material based on the inspection result, and when applying the pretreatment material 21, the judgment unit 15 can determine the pretreatment material 21 to be used for repairing the crack C based on the inspection result and the repair material 22 to be used. More specifically, it is preferable that the repair material 22 to be used contains particles having a size and shape according to the state of the crack C to be repaired, for example, the length L1 and opening amount W1 of the crack C, and has a viscosity according to the state of the crack C to be repaired, particularly the angle φ ₁ of the crack C. The pretreatment material 21 to be used is determined based on the material of the structure S to be repaired and its compatibility with the selected repair material 22.

More specifically, the determination unit 15 selects the pretreatment material 21 and the repair material 22 so as to satisfy the formula " _2s1 x cos θ1/ρ1 > _2s2 x cos θ2/ρ2," where the surface tension of the pretreatment material 21 is _s1 , the contact angle of the pretreatment material 21 with the surface of the structure S is _θ1 , the density of the pretreatment material ₂₁ is _ρ1 , the surface tension of the repair material 22 is s2, the contact angle of the repair material ₂₂ with the surface of the structure _{S is θ2 ,} and the density of the repair material _{22 is ρ2} . It is also preferable that the density _ρ1 of the pretreatment material 21 is 0.83 g/ ^cm3 or more and less than 0.89 g/ ^cm3 , the contact angle _θ1 with the surface of the structure S is 0 degrees or more and less than 10 degrees, and the surface tension _s1 is 20 dyn/cm or more and less than 50 dyn/cm. This makes it possible to select a pretreatment material 21 that can more reliably increase the permeability of the repair material 22 .

Embodiment 2.
16 is a diagram showing the functional configuration of a repair device 10A according to the embodiment 2. The configuration of the repair device 10A is the same as that of the repair device 10 except that the repair device 10A has a determination unit 15A instead of the determination unit 15 of the repair device 10, and therefore detailed description thereof will be omitted here.

17 is a diagram showing a detailed configuration of the determination unit 15A according to the second embodiment. The determination unit 15A includes a calculation unit 151A, a repair pattern determination unit 152A, and a material used determination unit 153A. The calculation unit 151A has a function of calculating the length L1 of the crack C and the angle _φ1 of the crack C, similar to the calculation unit 151, and also has a function of calculating the opening amount W1 of the crack C based on the inspection results.

In this case, since the inspection results are the captured image and directional information of the structure S, the calculation unit 151A calculates the opening amount W1 by analyzing the captured image. The calculation unit 151A outputs the calculated length L1, angle _φ1 , and opening amount W1 of the crack C to each of the repair pattern determination unit 152A and the material used determination unit 153A.

The repair pattern determination unit 152A determines a repair pattern for the target crack C based on the length L1, angle φ ₁ , and opening amount W1 of the crack C output by the calculation unit 151A. For example, for a crack C in a closed state with a small opening amount W1, even if the repair material 22 is applied, it may not be possible to expect the repair material 22 to penetrate into the crack C. For this reason, the repair pattern determination unit 152A can prevent the crack C with the opening amount W1 less than the third threshold from being repaired. Specifically, the repair pattern determination unit 152A determines that "repair is performed" when the length L1 of the crack C is equal to or greater than the first threshold and the opening amount W1 is equal to or greater than the third threshold. In addition, the repair pattern determination unit 152A determines that "repair is not performed" and selects the repair pattern (3) when the length L1 of the crack C is less than the first threshold or when the length L1 of the crack C is equal to or greater than the first threshold but the opening amount W1 is less than the third threshold.

The material to be used determination unit 153A determines the material to be used to repair the target crack C based on the length L1, angle _φ1 , and opening amount W1 of the crack C output by the calculation unit 151A. Specifically, the material to be used determination unit 153A can select the shape and size of the particles contained in the repair material 22 to be used based on the opening amount W1 in addition to the same criteria as those used by the material to be used determination unit 153.

Figure 18 is a flowchart for explaining the operation of the judgment unit 15A in the second embodiment. Parts similar to the operation of the judgment unit 15 in the first embodiment explained using Figure 15 may be given the same reference numerals and explanations may be omitted. In the omitted parts, "judgment unit 15" shall be read as "judgment unit 15A", "calculation unit 151" shall be read as "calculation unit 151A", and "repair pattern determination unit 152" shall be read as "repair pattern determination unit 152A".

The operations from step S101 to step S103 are the same as those in FIG. 15, and therefore will not be described. After step S103, the calculation unit 151A calculates the opening amount W1 of the crack C (step S201).

After step S201, the repair pattern determination unit 152A determines whether the length L1 of the crack C is equal to or greater than the first threshold value (step S104). If the length L1 of the crack C is less than the first threshold value (step S104: No), the repair pattern determination unit 152A determines that "no repair" is required and selects repair pattern (3) (step S106).

If the length L1 of the crack C is equal to or greater than the first threshold (step S104: Yes), the repair pattern determination unit 152A determines whether the opening amount W1 is equal to or greater than the third threshold (step S202). If the opening amount W1 is less than the third threshold (step S202: No), the process proceeds to step S106.

If the opening amount W1 is equal to or greater than the third threshold value (step S202: Yes), the repair pattern determination unit 152A determines that "repair is required" (step S105). The subsequent steps S107, S108, and S110 are the same as those in FIG. 15, and therefore will not be described here.

When repair pattern (1) is selected in step S108, the material determination unit 153A selects the pretreatment material 21 and repair material 22 to be used based on the length L1, angle _φ1 , and opening amount W1 of the crack C (step S203).

Furthermore, when repair pattern (2) is selected in step S110, the material selection unit 153A selects the repair material 22 to be used based on the length L1, angle _φ1 , and opening amount W1 of the crack C (step S204).

As described above, the repair device 10A according to the second embodiment is an example of a repair content determination device that determines the repair content of a crack C that has occurred in an inspected material portion of a structure S. In addition to the functions of the repair device 10 according to the first embodiment, the repair device 10A can further use the opening amount W1 of the crack C when determining the repair pattern and the material to be used. Therefore, it is possible to prevent repairs from being performed on a crack C whose opening is closed and in which the repair material 22 cannot be expected to penetrate into the crack C. In addition, since the material to be used for repair is determined based on the opening amount W1, it is possible to further improve the effect of suppressing the progression of the crack, regardless of the level of skill of the repair worker.

Embodiment 3.
In embodiment 3, a method for determining appropriate repair content according to the state of crack C using machine learning will be described.

Figure 19 is a diagram showing the functional configuration of a learning device 50 according to embodiment 3. The learning device 50 is a machine learning device related to the repair device 10. The learning device 50 has a learning data acquisition unit 51 and a model generation unit 52.

The learning data acquisition unit 51 acquires the inspection results and the repair details of the repairs performed before acquiring the inspection results as learning data. Here, the inspection results include a photographed image of the structure S and directional information that is information for identifying the direction of gravity in the photographed image.

The model generation unit 52 learns the repair content based on the learning data including the inspection results and the repair content. That is, the model generation unit 52 generates a learned model that infers the repair content from the inspection results of the repair device 10. Here, the repair content includes at least the repair pattern, and may further include the materials to be used when performing the repair.

The learning algorithm used by the model generation unit 52 can be any known algorithm, such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case where reinforcement learning is applied will be described. In reinforcement learning, an agent, which is an agent of action in a certain environment, observes the parameters of the environment that indicate the current state and determines the action to be taken. The environment changes dynamically due to the agent's actions, and the agent is given a reward according to the change in the environment. The agent repeats this process and learns the action policy that will obtain the most reward through a series of actions. Q-learning and TD-learning are known as representative methods of reinforcement learning. For example, in the case of Q-learning, a general update formula for the action value function Q(s, a) is expressed by the following formula (1).

In the above formula (1), s _t represents the state of the environment at time t, and a _t represents the action at time t. The action a _t changes the state to s _t+1 . r _t+1 represents the reward obtained by the change in state, γ represents the discount rate, and α represents the learning coefficient. Note that γ takes values in the range of 0<γ≦1, and α takes values in the range of 0<α≦1. The repair content becomes the action a _t , the inspection result becomes the state s _t , and the best action a _t for the state s _t at time t is learned.

The update formula expressed by formula (1) increases the action value Q if the action value Q of the action a with the highest Q value at time t+1 is greater than the action value Q of the action a executed at time t, and decreases the action value Q in the opposite case. In other words, the action value function Q(s, a) is updated so that the action value Q of action a at time t approaches the best action value at time t+1. As a result, the best action value in a certain environment is propagated sequentially to the action value in the previous environment.

As described above, when generating a learned model by reinforcement learning, the model generation unit 52 includes a reward calculation unit 53 and a function update unit 54.

The reward calculation unit 53 calculates the reward based on the repair content and the inspection results. The reward calculation unit 53 calculates the reward r based on the crack C propagation suppression effect of the repair target. For example, when the crack C propagation suppression effect improves, the reward r is increased by giving a reward of "1", and when the crack C propagation suppression effect decreases, the reward r is decreased by giving a reward of "-1". The method of evaluating the crack C propagation suppression effect is not particularly limited. For example, the crack C propagation suppression effect may be expressed by fracture mechanics parameters or thermoelastic temperature fluctuations. As the fracture mechanics parameters, the effective stress intensity factor range, J value, energy release rate, etc. can be used.

The function update unit 54 updates the function for determining the repair content in accordance with the reward calculated by the reward calculation unit 53, and outputs the function to the learned model storage unit 55. For example, in the case of Q-learning, the action value function Q(s _t , a _t ) expressed by the formula (1) is used as a function for calculating the repair content.

The learning process described above is repeated. The learned model storage unit 55 stores the action value function Q(s _t , a _t ) updated by the function update unit 54, that is, the learned model.

Next, the process of learning by the learning device 50 will be described with reference to Figure 20. Figure 20 is a flowchart for explaining the learning process of the learning device 50.

The learning data acquisition unit 51 acquires the repair details and inspection results as learning data (step S301).

The model generation unit 52 determines whether the effect of suppressing the propagation of crack C is improved based on the repair content and inspection results (step S302).

If the effect of suppressing the propagation of the crack C improves (step S302: Yes), the reward calculation unit 53 increases the reward (step S303). If the effect of suppressing the propagation of the crack C decreases (step S302: No), the reward calculation unit 53 decreases the reward (step S304).

The function update unit 54 updates the action value function Q(s _t , a _t ) represented by equation (1) stored in the trained model storage unit 55, based on the reward calculated by the reward calculation unit 53 (step S305).

The learning device 50 repeatedly executes the above steps S301 to S305, and stores the generated action value function Q(s _t , a _t ) as a learned model.

Although the above-mentioned learning device 50 is configured to store the learned model in a learned model memory unit 55 provided outside the learning device 50, the learned model memory unit 55 may also be provided inside the learning device 50.

Figure 21 is a diagram showing the functional configuration of the judgment unit 15B of the repair device 10B according to the third embodiment. The judgment unit 15B has an inference device 56. Although the configuration of the repair device 10B according to the third embodiment is not shown, it has a judgment unit 15B instead of the judgment unit 15 of the repair device 10. This judgment unit 15B can perform judgment processing using machine learning. The inference device 56 uses the learned model stored in the learned model storage unit 55 to infer repair content from the inspection result obtained by the inspection unit 14, and outputs the repair content that is the inference result.

Figure 22 is a configuration diagram of an inference device 56 for the repair device 10. The inference device 56 has an inference data acquisition unit 57 and an inference unit 58.

The inference data acquisition unit 57 acquires the inspection results as inference data.

The inference unit 58 infers the repair content using the trained model. In other words, by inputting the inspection results acquired by the inference data acquisition unit 57 into this trained model, it is possible to infer repair content appropriate to the inspection results of the structure S.

In this embodiment, the repair device 10 used by the model generation unit 52 when learning is the same as the repair device 10 that is the target of inference by the inference unit 58, but it is also possible to obtain a trained model from another repair device 10 and output repair content based on this trained model.

Next, the process by which the inference device 56 obtains the repair content will be described with reference to Figure 23. Figure 23 is a flowchart for explaining the inference process of the inference device 56.

The inference data acquisition unit 57 acquires the inspection results as inference data (step S401). The inference unit 58 inputs the inspection results acquired by the inference data acquisition unit 57 as inference data to the trained model stored in the trained model storage unit 55 (step S402), and obtains repair content as an output. The inference unit 58 outputs the obtained repair content as an inference result (step S403).

The repair device 10B performs repair using the output repair content (step S404). This allows the repair of the crack C to be performed with the repair content determined according to the inspection result. Since this repair content includes at least a repair pattern, one of the repair pattern (1) in which the pretreatment material 21 and the repair material 22 are applied, the repair pattern (2) in which the repair material 22 is applied, and the repair pattern (3) in which no repair is performed is selected according to the state of the target crack C. In addition, by using the photographed image and direction information of the structure S as the inspection result, it is possible to determine the appropriate repair content simply by inputting the photographed image output by the camera 142 and the output of the gyro sensor provided in the camera 142 into the trained model.

In this embodiment, the case where reinforcement learning is applied to the learning algorithm used by the model generation unit 52 has been described, but the present invention is not limited to this. As for the learning algorithm, it is also possible to apply supervised learning, unsupervised learning, semi-supervised learning, etc. in addition to reinforcement learning.

In addition, the learning algorithm used in the model generation unit 52 can be deep learning, which learns to extract the features themselves, or machine learning can be performed according to other known methods, such as neural networks, genetic programming, functional inference programming, support vector machines, etc.

In this embodiment, the inference device 56 is provided in the judgment unit 15B of the repair device 10B, but this is not limited to the example. The learning device 50 and the inference device 56 may be devices separate from the repair device 10B, for example, connected to the repair device 10B via a network. The learning device 50 and the inference device 56 may also exist on a cloud server.

The model generation unit 52 may learn the repair contents using learning data acquired from multiple repair devices 10. The model generation unit 52 may acquire learning data from multiple repair devices 10 used in the same area, or may learn the repair contents using learning data collected from multiple repair devices 10 operating independently in different areas. It is also possible to add or remove a repair device 10 that collects learning data from the target midway. Furthermore, the learning device 50 that has learned the repair contents for a certain repair device 10 may be applied to another repair device 10, and the repair contents for the other repair device 10 may be re-learned and updated.

As described above, according to the third embodiment, it is possible to determine the repair contents using machine learning. The repair device 10B is an example of a repair content determination device. For example, the judgment unit 15B has an inference data acquisition unit 57 that acquires an inspection result, and an inference unit 58 that outputs the repair contents from the inspection results acquired by the inference data acquisition unit 57 using a trained model for inferring the repair contents of the crack C from the inspection result. The repair contents output by the inference unit 58 include at least a repair pattern, and may further include a material used to repair the crack C. This trained model is generated by a learning device 50 having, for example, a learning data acquisition unit 51 that acquires learning data including the inspection result and the repair contents, and a model generation unit 52 that uses the learning data to generate a trained model for inferring the repair contents from the inspection result. The model generation unit 52 can learn, for example, repair contents that improve the effect of suppressing the progress of the crack C. The input to the trained model is the inspection result, and can include, for example, a photographed image and directional information of the structure S. In the above embodiment, the inspection results are the captured image and directional information of the structure S, but the inspection results input to the trained model may be values calculated from the captured image and directional information. When a value calculated from the captured image and directional information, for example a specific feature value such as "the angle φ ₁ of the crack C with respect to the direction of gravity," is input to the trained model, it becomes possible to reliably reflect the effect of the specific feature value on the repair content.

Embodiment 4.
24 is a diagram showing a functional configuration of a repair device 10C according to the fourth embodiment. The repair device 10C includes a control unit 41 that controls the suction force of the suction unit 12 and the driving force of the driving unit 13. The main body 11 of the repair device 10C moves along the structure S by the driving unit 13 under the control of the control unit 41, and the pre-treatment unit 17, the intermediate treatment unit 18, and the post-treatment unit 19 of the repair unit 16 are arranged in the order of processing, with the moving direction as the head. Specifically, the pre-treatment unit 17, the intermediate treatment unit 18, and the post-treatment unit 19 are arranged in the order of processing, with the moving direction as the head. As a result, the pre-treatment unit 17, the intermediate treatment unit 18, and the post-treatment unit 19 face the crack C in this order as they move, so that the repair work can be performed efficiently.

As described above, the repair device 10C according to the fourth embodiment includes a judgment unit 15 that determines the repair content, including a repair pattern indicating whether or not to repair the crack C with a repair material 22, which is a liquid containing particles, based on the inspection results of the crack C that has occurred in the material portion of the structure S to be inspected, and whether or not to apply a pretreatment material 21 to the crack C before applying the repair material 22 if repair with the repair material 22 is to be performed, and a repair unit 16 that performs the repair of the crack C based on the repair content determined by the judgment unit 15. The repair unit 16 is characterized by having a pretreatment unit 17 that performs a pretreatment process of applying the pretreatment material 21 to the crack C of the structure S, an intermediate treatment unit 18 that performs an intermediate treatment process of removing the excess of the pretreatment material 21 applied in the pretreatment process, and a post-treatment unit 19 that performs a post-treatment process of applying the repair material 22 to the crack C. The repair device 10C further includes suction means 12 that suctions the structure S to fix the main body 11 of the repair device 10C to the structure S, drive means 13 for moving the main body 11 along the structure S, and control means 41 that controls the suction force of the suction means 12 and the drive force of the drive means 13. In addition, the pre-treatment section 17, intermediate treatment section 18, and post-treatment section 19 of the repair section 16 are arranged in order of treatment, with the direction of movement of the repair device 10C by the drive means 13 at the head. This makes it possible to improve the efficiency of the repair work.

The intermediate treatment section 18 can be a suction mechanism for the pretreatment material 21, or industrial oil blotting paper that adsorbs the pretreatment material 21.

Embodiment 5.
25 is a diagram showing the configuration of a repair content determination system 1 according to the fifth embodiment. The repair content determination system 1 includes an inspection device 61, a repair content determination device 64, and a repair device 10D. Hereinafter, parts having the same functions as those in the functional configurations described in the first to third embodiments are denoted by the same reference numerals, and detailed description thereof will be omitted.

The inspection device 61 has an inspection unit 14, a calculation unit 151, a storage unit 62, and a transmission unit 63. The inspection unit 14 outputs the inspection result to the calculation unit 151. The calculation unit 151 calculates the length L1 and angle φ ₁ of the crack C based on the inspection result, and outputs the length L1 and angle φ ₁ of the crack C to the storage unit 62. The storage unit 62 stores the length L1 and angle φ ₁ of the crack C. The transmission unit 63 transmits the length L1 and angle φ ₁ of the crack C stored in the storage unit 62 to the repair content determination device 64.

The repair content determination device 64 has a receiving unit 65, a judgment unit 15D, and an output unit 66. The receiving unit 65 receives the length L1 and angle φ ₁ of the crack C from the inspection device 61 and outputs them to the repair pattern determination unit 152 and the material used determination unit 153 of the judgment unit 15D. The judgment unit 15D is similar to the judgment unit 15 except that it acquires the length L1 and angle φ ₁ of the crack C via the receiving unit 65, and outputs the determined repair content to the output unit 66. The output unit 66 outputs the determined repair content to the repair device 10D. The output unit 66 may output the repair content to the repair device 10D by communication, or may output a display screen including the repair content, for example. In this case, the worker who sees the display screen can operate the repair device 10D to perform repair according to the determined repair content.

The repair device 10D has a suction means 12, a driving means 13, a repair unit 16, and a control means 41. The repair device 10D executes the repair process according to the repair content determined by a repair content determination device 64, which is a device separate from the repair device 10D.

The inspection device 61 and repair device 10D must be operated at the location where the structure S is located, but the repair content determination device 64 can be installed anywhere as long as it has communication capabilities.

As described above, according to the repair content determination system 1 of the fifth embodiment, the repair content including the repair pattern can be determined from the inspection results in the repair content determination device 64, which is separate from the inspection device 61 and the repair device 10D. Therefore, as long as the repair device 10D has a communication function, it is not necessary for the repair device 10D to have a function for performing processing to determine the repair content. According to this repair content determination device 64, similar to the repair device 10, it is possible to easily determine whether or not the application of the pretreatment material 21 is necessary and whether or not repair using the repair material 22 is appropriate, regardless of the worker's level of proficiency. Note that, although the inspection device 61 is a separate device from the repair device 10D here, the function of the inspection device 61 may be provided in the repair device 10D.

Embodiment 6.
26 is a diagram showing the configuration of a repair content determining system 1E according to the embodiment 6. The repair content determining system 1E includes an inspection device 61E, a repair content determining device 64E, and a repair device 10D.

The inspection device 61E has an inspection unit 14, a calculation unit 151A, a storage unit 62E, and a transmission unit 63E. The inspection device 61E according to the sixth embodiment has a calculation unit 151A instead of the calculation unit 151 of the inspection device 61. The calculation unit 151A calculates the opening amount W1 in addition to the length L1 and angle φ ₁ of the crack C. The storage unit 62E stores the length L1, angle φ ₁ , and opening amount W1 of the crack C. The transmission unit 63E transmits the length L1, angle φ ₁ , and opening amount W1 of the crack C to the repair content determination device 64E. In addition, the transmission unit 63E transmits at least the opening amount W1 of the crack C to the repair effect prediction unit 67.

The repair effect prediction unit 67 predicts the repair effect. The repair effect is expressed by the effect of suppressing the progression of the crack C. For example, the effect of suppressing the progression of the crack C may be expressed by a fracture mechanics parameter or by a thermoelastic temperature change. As the fracture mechanics parameter, an effective stress intensity factor range, a J value, an energy release rate, etc. can be used. Note that the amount indicating the repair effect is not particularly limited here. Here, since the opening amount W1 of the crack C can be obtained from the inspection results of the inspection unit 14, the repair effect prediction unit 67 predicts the repair effect based on the opening amount W1 before and after the repair of the crack C.

The repair content determination device 64E has a receiving unit 65, a determination unit 15E, and an output unit 66. The determination unit 15E has a repair pattern determination unit 152A and a material use determination unit 153A.

Figure 27 is a flowchart for explaining the operation of the repair content determination system 1E according to the sixth embodiment. The inspection unit 14 of the repair content determination system 1E performs an inspection process (step S10). Based on the inspection results of the inspection unit 14, the calculation unit 151A and the judgment unit 15E execute a judgment process (step S20). The repair device 10D judges whether or not the result of the judgment process is "repair required" (step S30). If "repair required" (step S30: Yes), the repair unit 16 of the repair device 10D performs a repair process (step S40).

When the repair device 10D performs the repair process, the inspection unit 14 of the inspection device 61E performs a post-repair inspection process (step S41). The post-repair inspection process is an inspection process that is performed after the repair process, and is similar to the inspection process of step S10. In addition, the repair effect prediction unit 67 performs a repair effect prediction process (step S42).

If "no repairs required" (step S30: No) and after performing the repair effect prediction process, the repair device 10D determines whether or not repairs are complete (step S50). If repairs are not complete (step S50: No), the control means 41 of the repair device 10D controls the drive means 13 to move the repair device 10D (step S60) and returns to the process of step S10. If repairs are complete (step S50: Yes), the repair content determination system 1E ends the process.

As described above, according to the repair content determination system 1E of embodiment 6, in addition to the functions of the repair content determination system 1, it is possible to determine the repair content based on the opening amount W1 of the crack C, and it has the function of predicting the effect of the repair treatment in suppressing the progression of the crack C based on the opening amount W1.

Embodiment 7.
28 is a diagram showing the configuration of a repair content determination system 1F according to the seventh embodiment. The repair content determination system 1F includes an inspection device 61F, a learning device 50, a repair content determination device 64F, and a repair device 10D.

The inspection device 61F has an inspection unit 14, a storage unit 62, and a transmission unit 63. In the seventh embodiment, the repair content is inferred from the inspection result using machine learning. Here, since the input to the trained model is the photographed image and directional information, which are the inspection result, the inspection device 61F does not include the calculation units 151 and 151A, and the inspection result is stored as it is in the storage unit 62. Note that, when the input to the trained model is not the inspection result as it is, but values calculated from the inspection result, such as the length L1, angle φ ₁ , and opening amount W1 of the crack C, the inspection device 61F may have a function of calculating these values by the calculation units 151 and 151A. The transmission unit 63 transmits the inspection result to the repair content determination device 64F and the learning device 50.

The repair content determination device 64F has a function of inferring repair content from the inspection result of the inspection device 61F using machine learning. The repair content determination device 64F has a receiving unit 65, a judgment unit 15F, and an output unit 66. The judgment unit 15F has an inference device 56. The receiving unit 65 receives the inspection result from the inspection device 61F and receives the learned model from the learned model storage unit 55. The receiving unit 65 outputs the received inspection result and the learned model to the judgment unit 15F. The inference device 56 of the judgment unit 15F infers the repair content of the repair device 10D using the input inspection result and the learned model. Specifically, the inference device 56 inputs the inspection result to the learned model and obtains the repair content as an output. The inference device 56 outputs the obtained repair content to the output unit 66. The output unit 66 outputs the repair content to the repair device 10D and the learning device 50.

The learning device 50 uses the inspection results before and after repair obtained from the inspection device 61F and the repair content obtained from the repair content determination device 64F to generate a trained model for obtaining the repair content from the inspection results, and stores the generated trained model in the trained model memory unit 55.

As described above, in the repair content determination system 1F according to the seventh embodiment, the repair content can be determined using machine learning. Here, the input to the trained model is the inspection result as is, so the repair content determination system 1F does not need to include the configuration of the calculation units 151 and 151A. Note that, even in the seventh embodiment, the hardware that includes each function of the repair content determination system 1F can be changed as appropriate. For example, the learning device 50 may be included in the inspection device 61F, the repair device 10D, or the repair content determination device 64F. Also, the inspection device 61F and the repair device 10D may be an integrated device.

Embodiment 8.
29 is a diagram showing a configuration of a repair content determination system 1G according to the eighth embodiment. The repair content determination system 1G includes a repair device 10G and a repair content determination device 64. The repair device 10G is an unmanned mobile body having an inspection function and a repair function.

The repair device 10G has an inspection device 61, a receiving unit 68, a repair unit 16, a control means 41G, a driving means 13, a suction means 12, and a route information storage unit 69. The route information storage unit 69 stores route information indicating a movement route in the structure S to be inspected and repaired. The control means 41G controls the driving means 13 and the suction means 12 in accordance with the route information stored in the route information storage unit 69, thereby moving the repair device 10G along the movement route of the structure S.

The repair device 10G transmits the inspection results of the inspection device 61 to the repair content determination device 64, and performs repairs in the repair section 16 according to the repair content received from the repair content determination device 64.

Figure 30 is a diagram showing an example of a movement path of the repair device 10G according to the eighth embodiment. For example, as shown by the solid arrow in Figure 30, it is assumed that a movement path for the structure S is determined in advance. At this time, the repair device 10G inspects the structure S while moving along this movement path, and performs a judgment process based on the inspection results. When a crack C to be repaired is found by the judgment process, the repair device 10G performs the repair process and then resumes movement along the movement path. In this case, the operation of the repair device 10G is as shown in Figure 14 or Figure 27.

In the example of Figure 30, the repair device 10G performs inspection and judgment processes while moving along a predetermined movement path from start point 5-1 to goal point 5-2, and also performs repair processes at points 6-1, 6-2, and 6-3 where a crack C to be repaired is found. When the repair device 10G reaches goal point 5-2, it automatically returns to start point 5-1. For example, as shown in Figure 30, the repair device 10G can return from goal point 5-2 to start point 5-1 via the shortest route.

Alternatively, the repair device 10G may perform the inspection process and the judgment process while moving on a predetermined moving path, and when the crack C to be repaired is found, record the position information of the points 6-1, -2, 6-3 where the crack C occurs. In this case, the repair process may not be performed while moving from the start point 5-1 to the goal point 5-2, and the repair process may be performed at the recorded points 6-1, 6-2, 6-3 while returning from the goal point 5-2 to the start point 5-1. FIG. 31 is a diagram showing another example of the path of the repair device 10G according to the eighth embodiment returning from the goal point 5-2 to the start point 5-1. Note that, here, the repair device 10G is provided with the inspection device 61 and the repair unit 16, but the inspection process and the repair process may be performed separately using the inspection device 61 separate from the repair device 10G. Even in this case, if the inspection device 61 and the repair device 10G are each mounted on an unmanned mobile body, the inspection and repair processes can be performed along a predetermined movement route, making it possible to shorten the work time.

Here, an example of the hardware configuration will be described. The above-mentioned repair content determination devices 64, 64E, 64F are realized, for example, by a computer system. The repair content determination devices 64, 64E, 64F may be realized by one computer system or by multiple computer systems. For example, the repair content determination devices 64, 64E, 64F may be realized by a cloud system. In a cloud system, the hardware of the computer system and devices such as servers for each function can be arbitrarily set. For example, one computer system may have the functions of multiple devices, or multiple computer systems may have the functions of one device.

An example of the configuration of a computer system that realizes the repair content determination devices 64, 64E, and 64F is described below. FIG. 32 is a diagram showing an example of the configuration of a computer system that realizes the repair content determination devices 64, 64E, and 64F. As shown in FIG. 32, this computer system includes a control unit 101, an input unit 102, a memory unit 103, a display unit 104, a communication unit 105, and an output unit 106, which are connected via a system bus 107.

In FIG. 32, the control unit 101 is, for example, a CPU (Central Processing Unit) or the like. The control unit 101 executes a repair content determination program in which each process performed by the repair content determination device 64, 64E, 64F of this embodiment is described. The input unit 102 is composed of, for example, a keyboard, a mouse, etc., and is used by the user of the computer system to input various information. The storage unit 103 includes various memories such as a RAM (Random Access Memory) and a ROM (Read Only Memory) and a storage device such as a hard disk, and stores the program to be executed by the control unit 101, necessary data obtained in the process of processing, etc. The storage unit 103 is also used as a temporary storage area for the program. The display unit 104 is composed of an LCD (Liquid Crystal Display) or the like, and displays various screens to the user of the computer system. The communication unit 105 is a communication circuit or the like that performs communication processing. The communication unit 105 may be composed of a plurality of communication circuits each corresponding to a plurality of communication methods. The output unit 106 is an output interface that outputs data to an external device such as a printer or an external storage device.

Note that FIG. 32 is just an example, and the configuration of the computer system is not limited to the example of FIG. 32. For example, the computer system may not have the output unit 106. In addition, when the repair content determination devices 64, 64E, 64F are realized by multiple computer systems, not all of these computer systems may be the computer systems shown in FIG. 32. For example, some computer systems may not have at least one of the display unit 104, output unit 106, and input unit 102 shown in FIG. 32.

Here, an example of the operation of the computer system until the repair content determination program in which the processing of the repair content determination device 64, 64E, 64F of this embodiment is described becomes executable will be described. In the computer system having the above-mentioned configuration, for example, the repair content determination program is installed in the memory unit 103 from a CD-ROM or DVD-ROM set in a CD (Compact Disc)-ROM drive or DVD (Digital Versatile Disc)-ROM drive (not shown). Then, when the repair content determination program is executed, the repair content determination program read from the memory unit 103 is stored in an area that becomes the main storage device of the memory unit 103. In this state, the control unit 101 executes the processing as the repair content determination device 64, 64E, 64F of this embodiment according to the repair content determination program stored in the memory unit 103.

In the above explanation, a program describing the processing in the repair content determination devices 64, 64E, 64F is provided using a CD-ROM or DVD-ROM as a recording medium, but this is not limited to this, and depending on the configuration of the computer system, the capacity of the program to be provided, etc., it is also possible to use a program provided over a transmission medium such as the Internet via the communication unit 105, for example.

The repair content determination program of this embodiment causes a computer to execute the steps of acquiring inspection results of a crack C that has occurred in a material portion of a structure S to be inspected, determining repair content including a repair pattern indicating whether or not to repair the crack C with a repair material 22, which is a liquid containing particles, based on the inspection results, and, if repair with the repair material 22 is to be performed, whether or not to apply a pretreatment material 21 into the crack C before applying the repair material 22, applying the pretreatment material 21 into the crack C if the repair pattern indicates that repair with the repair material 22 will be performed and that the pretreatment material 21 will be applied, removing excess of the applied pretreatment material 21 if the repair pattern indicates that repair with the repair material 22 will be performed and that the pretreatment material 21 will be applied, and applying the repair material 22 into the crack C if the repair pattern indicates that repair with the repair material 22 will be performed.

The receiving unit 65 shown in Figures 25, 26, 28 and 29 is realized by the communication unit 105 shown in Figure 32, the judgment units 15D, 15E and 15F shown in Figures 25, 26, 28 and 29 are realized by the control unit 101 shown in Figure 32, and the output unit 66 shown in Figures 25, 26, 28 and 29 is realized by at least one of the communication unit 105 and the output unit 106 shown in Figure 32.

The learning device 50 and the inference device 56 are realized by one or more computer systems, similar to the repair content determination devices 64, 64E, and 64F. The program for the learning device 50 and the inference device 56 to perform the operations described in this embodiment is provided by a storage medium, a transmission medium, etc., and installed in a computer system, similar to the above-mentioned repair content determination program. This realizes the above-mentioned operations in the learning device 50 and the inference device 56. In addition, the repair content determination devices 64, 64E, and 64F and the learning device 50 and the inference device 56 may be realized by one computer system. In this case, the receiving units 65, 68 and the transmitting units 63, 63E may not be provided, and these functions are performed by data exchange within the computer system.

Note that the division of functions in each device is just one example, and as long as the repair content determination systems 1, 1E, 1F, and 1G can perform the operations described above, the division of functions in each device is not limited to the example shown in the figure.

Figure 33 is a diagram showing an example of a processing circuit 110 for realizing each function of the repair device 10 or the repair content determination system 1. The judgment units 15, 15A, 15B of the repair devices 10, 10A, 10B, 10C, the calculation units 151, 151A of the inspection devices 61, 61E, and the control means 41, 41G of the repair devices 10C, 10D, 10G are each realized by the processing circuit 110. These processing circuits 110 may be dedicated hardware or may be control circuits using a CPU.

When the above processing circuit 110 is realized by dedicated hardware, the processing circuit 110 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination of these.

When the processing circuit 110 is realized by a control circuit using a CPU, the control circuit includes a processor and a memory. The processor is a CPU, and is also called a processing device, an arithmetic device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), etc. The memory is, for example, a non-volatile or volatile semiconductor memory such as a RAM, a ROM, a flash memory, an EPROM (Erasable Programmable ROM), an EEPROM (registered trademark) (Electrically EPROM), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, etc.

When the processing circuit 110 is realized by a control circuit, the processing circuit 110 is realized by the processor reading and executing a program corresponding to the processing of each component stored in the memory. The memory is also used as a temporary memory for each processing executed by the processor.

The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or the embodiments may be combined with each other. Also, parts of the configurations may be omitted or modified without departing from the spirit of the invention.

For example, in the above embodiment, the inspection result is the captured image and directional information, but is not limited to such an example. For example, the inspection result may be any result that indicates the state of the structure S, in particular, the state of the crack C that has occurred in the inspected material portion of the structure S.

Furthermore, the allocation of functions shown in the above embodiment is just one example. As long as the above functions can be realized, each process may be executed in any device. Inspection processes and repair processes must be executed at the location where the structure S exists, but calculation processes such as judgment processes may be executed in any device, and there are no particular restrictions on where the devices are installed.

In addition, in the above embodiment, the repair processing is performed by the repair device 10, etc., but the repair worker may perform the repair manually in accordance with the repair content determined by the repair content determination device 64, etc.

1, 1E, 1F, 1G Repair content determination system, 5-1 Start point, 5-2 Goal point, 6-1, 6-2, 6-3 Points, 10, 10A, 10B, 10C, 10D, 10G Repair device, 11 Main body, 12 Suction means, 13 Driving means, 14 Inspection unit, 15, 15A, 15B, 15D, 15E, 15F Judgment unit, 16 Repair unit, 17 Pre-treatment unit, 18 Intermediate treatment unit, 19 Post-treatment unit, 21 Pre-treatment material, 22 Repair material, 30 Support member, 31 Deck plate, 32 Pavement, 33, 33-1, 33-2 U-rib, 34 Horizontal rib, 35 Weld bead, 36 Butt weld bead, 37 Vertical reinforcement material, 38 Fillet weld bead, 41, 41G Control means, 50 Learning device, 51 Learning data acquisition unit, 52 Model generation unit, 53 Reward calculation unit, 54 Function update unit, 55 Learned model storage unit, 56 Inference device, 57 Inference data acquisition unit, 58 Inference unit, 61, 61E, 61F Inspection device, 62, 62E Storage unit, 63, 63E Transmission unit, 64, 64E, 64F Repair content determination device, 65, 68 Reception unit, 66, 106 Output unit, 67 Repair effect prediction unit, 69 Path information storage unit, 101 Control unit, 102 Input unit, 103 Storage unit, 104 Display unit, 105 Communication unit, 107 System bus, 110 Processing circuit, 141 LiDAR scanner, 142 Camera, 151, 151A Calculation unit, 152, 152A Repair pattern determination unit, 153, 153A Material usage determination section, C, C-1, C-2, C-3 crack, D1 gravity direction, D2, D2-1, D2-2, D2-3 depth direction, D3 vertical direction, L1, L1-1, L1-2, L1-3 length, P1-1, P1-2, P1-3 starting point, S structure, S1 bridge, W1 opening amount, φ ₁ , φ _1-1 , φ _1-2 , φ _1-3 angle.

Claims (24)

an inspection result acquisition unit that acquires inspection results of cracks occurring in an inspection target material portion of a structure;
a determination unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and whether or not to apply a pretreatment material to the crack before applying the repair material, if repair with the repair material is to be performed, based on the length of the crack indicated by the inspection result and the angle of the crack with respect to the direction of gravity;
A repair content determination device comprising: an inspection result acquisition unit that acquires inspection results of cracks occurring in an inspection target material portion of a structure;
a determination unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material to the inside of the crack before applying the repair material; and
Equipped with
The repair content determination device, wherein the inspection result includes a photographed image of the structure and directional information indicating a direction of gravity in the photographed image. an inspection result acquisition unit that acquires inspection results of cracks occurring in an inspection target material portion of a structure;
a determination unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material to the inside of the crack before applying the repair material; and
Equipped with
The repair content further includes a material to be used to repair the crack;
A repair content determination device characterized in that, when repairing using the repair material, the judgment unit determines the repair material to be used to repair the crack based on the length of the crack indicated by the inspection results and the angle of the crack with respect to the direction of gravity . The repair content determination device according to claim 3, characterized in that, when applying the pretreatment material, the determination unit determines the pretreatment material to be used to repair the crack based on the inspection results and the repair material to be used. 4. The repair content determination device according to claim 3 , wherein the determination unit selects the repair material that includes particles having a shape corresponding to the length of the crack and has a viscosity corresponding to the angle. 2. The repair content determining device according to claim 1 , wherein the determining unit determines the repair pattern based on an opening amount of the crack in addition to the length and angle of the crack. 4. The repair content determining device according to claim 3 , wherein the determining unit determines the material based on an opening amount of the crack in addition to the length and angle of the crack. The repair content determination device according to claim ₁ , characterized in that, when the surface tension of the pretreatment material is _s1 , the contact angle of the pretreatment material with the surface of the structure is _θ1 , the density of the pretreatment material is _ρ1 , the surface tension of the repair material is _{s2, the contact angle of the repair material with the surface of the structure is θ2} , and the density of the repair material is _ρ2 , the formula _2s1 × cos _θ1 / _ρ1 > _2s2 × cos _θ2 / _ρ2 is satisfied. The repair content determination device according to claim 1, characterized in that the pretreatment material has a density of 0.83 g/cm3 or ^more and ^less than 0.89 g/cm3, a contact angle with the surface of the structure of 0 degrees or more and less than 10 degrees, and a surface tension of 20 dyn/cm or more and less than 50 dyn/cm. an inspection result acquisition unit that acquires a photographed image of the structure as an inspection result of a crack occurring in an inspection target material portion of the structure and directional information indicating the direction of gravity in the photographed image;
a determination unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material to the inside of the crack before applying the repair material; and
Equipped with
The determination unit is
an inference unit that outputs the repair content from the inspection result by using a trained model for inferring the repair content of the crack from the inspection result;
A repair content determination device comprising: The repair content further includes a material to be used to repair the crack;
The repair content determination device described in claim 10, characterized in that when repair is performed using the repair material, the inference unit outputs the repair material to be used to repair the crack as the material, and when the pretreatment material is applied, the inference unit outputs the pretreatment material to be used to repair the crack as the material. an inspection result acquisition unit that acquires a photographed image of the structure as an inspection result of a crack occurring in an inspection target material portion of the structure and directional information indicating the direction of gravity in the photographed image;
a determination unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material to the inside of the crack before applying the repair material; and
a learning data acquisition unit that acquires learning data including the inspection result and the repair content;
A model generation unit that generates a trained model for inferring the repair content from the inspection result by using the learning data;
A repair content determination device comprising : The repair content determination device according to claim 12, characterized in that the model generation unit learns the repair content that improves the crack growth suppression effect. 10. The repair content determining device according to claim 1, wherein the inspection result includes a photographed image of the structure and direction information indicating a direction of gravity in the photographed image. a judgment unit that determines repair details including a repair pattern that indicates whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material into the crack before applying the repair material; and
A repair unit that executes repair of the crack based on the repair content determined by the judgment unit;
Equipped with
The repair portion includes:
a pretreatment unit that performs a pretreatment step of applying the pretreatment material to the crack of the structure;
an intermediate treatment section that performs an intermediate treatment process to remove an excess of the pretreatment material applied in the pretreatment process;
a post-treatment section that performs a post-treatment step of applying the repair material to the crack;
A repair device comprising: a suction means for suctioning the main body of the repair device to the structure and fixing the main body of the repair device to the structure;
a drive means for moving the body along the structure;
a control means for controlling the suction force of the suction means and the drive amount of the drive means;
Further equipped with
The repair device according to claim 15, wherein the pre-treatment section, the intermediate treatment section, and the post-treatment section are arranged in a processing order with a direction in which the repair device is moved by the driving means as a leading edge. The repair device according to claim 15, characterized in that the intermediate treatment section is a suction mechanism that sucks in the pretreatment material, or industrial oil blotting paper that absorbs the pretreatment material. a learning data acquisition unit that acquires learning data including a length of a crack that has occurred in a material portion of a structure to be inspected, the length of the crack indicated by an inspection result, an angle of the crack with respect to the direction of gravity , and repair details including a repair pattern indicating whether or not the crack should be repaired with a repair material that is a liquid containing particles, and whether or not a pretreatment material should be applied to the crack before applying the repair material when repairing with the repair material;
A model generation unit that generates a trained model for inferring the repair content from the inspection result by using the learning data;
A learning device comprising: The learning device according to claim 18, characterized in that the inspection result includes a photographed image of the structure and directional information indicating the direction of gravity in the photographed image. The learning device according to claim 18, characterized in that the model generation unit learns the repair content that improves the crack growth suppression effect. an inference data acquisition unit that acquires a photographed image of a structure as an inspection result of a crack occurring in an inspection target material portion of the structure and directional information indicating a gravity direction in the photographed image ;
an inference unit that outputs the repair content from the inspection results acquired by the inference data acquisition unit, using a trained model for inferring the repair content of a repair device that repairs the crack from the inspection results;
An inference device comprising: an inference data acquisition unit that acquires inspection results of cracks occurring in an inspection target material portion of a structure;
an inference unit that uses a trained model for inferring repair details of a repair device that repairs the crack from the length of the crack indicated by the inspection result and the angle of the crack with respect to the direction of gravity, and outputs the repair details from the inspection result acquired by the inference data acquisition unit;
An inference device comprising: a repair content determination device that determines repair content including a repair pattern that indicates, based on an inspection result of a crack that has occurred in an inspected material portion of a structure, whether or not to repair the crack with a repair material that is a liquid containing particles, and, if repair is to be performed with the repair material, whether or not to apply a pretreatment material into the crack before applying the repair material;
a repair device that repairs the crack based on the repair content determined by the repair content determination device;
Equipped with
The repair device comprises:
a pretreatment unit that performs a pretreatment step of applying the pretreatment material to the crack of the structure;
an intermediate treatment section that performs an intermediate treatment process to remove an excess of the pretreatment material applied in the pretreatment process;
a post-treatment section that performs a post-treatment step of applying the repair material to the crack;
A repair content determination system comprising: Obtaining an inspection result of a crack occurring in an inspected material portion of a structure;
determining a repair content including a repair pattern indicating whether or not to repair the crack with a repair material that is a liquid containing particles, and whether or not to apply a pretreatment material to the crack before applying the repair material if repair with the repair material is to be performed based on the inspection results;
applying the pretreatment material to the crack when the repair pattern indicates that repair is to be performed using the repair material and that the pretreatment material is to be applied;
removing an excess of the applied pretreatment material when the repair pattern indicates that repair is to be performed using the repair material and that the pretreatment material is to be applied;
applying the repair material into the crack when the repair pattern indicates that repair is to be performed using the repair material;
A repair method comprising the steps of:

Images