JP2020115153A - Crack diagnosis method and device - Google Patents

Abstract

To improve the accuracy in diagnosis of the presence or absence of the generation of cracks.SOLUTION: A crack diagnosis device 1 comprises: a distortion sensor 2 and distortion sensors 3a, 3b that are installed in two areas 11 and 12 defined on a crack diagnosis object 100; and an arithmetic unit 4. With stress analysis using a crack absent analysis model and a crack present analysis model for the crack diagnosis object 100, two areas are determined where the relative rate of change in the amount of change in distortion associated with the generation of cracks becomes a preset value or more. As an arrangement corresponding to the two areas, the area 11 and area 12 are set on the crack diagnosis object 100. The arithmetic unit 4 compares, through statistical processing, a diagnosis data group consisting of pieces of data input from the distortion sensors 2, 3a, 3b in a crack diagnosis period, relatively with a normal data group obtained by accumulating data in a normal state where a crack 104 is not generated in the crack diagnosis object 100, and determines the presence or absence of the generation of the crack 104 in the crack diagnosis object 100 based on a result of the comparison.SELECTED DRAWING: Figure 1

Description

The present invention relates to a crack diagnosing method and apparatus for diagnosing a crack to be crack-diagnosed.

Structures such as bridges, steel towers, buildings, hulls, car bodies, airframes of aircraft, and frames of cranes and other transportation devices are mainly formed of metal structural members.

This type of structural member is often formed by joining a plurality of members, and most of the joining of these members is performed by welding.

In structural members, fatigue cracks may occur at joints and the like. When a crack generated in a structural member propagates, the performance of the structural member may be deteriorated. Further, the deterioration of the performance of the structural member may affect the retention of the structure and the maintenance of the function of the structure.

Therefore, it is desired to be able to diagnose the presence or absence of cracks in a structure.

Conventionally, a structure monitoring system has been proposed. This is an equivalence test between the response surface of the measurement data by a plurality of sensors attached to the structure and the response surface of the measurement data when the structure is healthy in order to detect damage to the structure. When the equivalence of the response surface is statistically rejected, it is determined that an abnormality has occurred in the structure.

Further, for this monitoring system, it is proposed to use a strain gauge as a sensor, create a response surface between two strain gauges, and judge the occurrence of cracks by a significant difference test of the equivalence of the response surfaces. (For example, see Patent Document 1).

Japanese Patent No. 3777453

However, in Patent Document 1, although the idea of determining the occurrence of cracks using a plurality of strain gauges is shown, the arrangement of the plurality of strain gauges is not specifically shown. .. Further, Patent Document 1 does not show the relationship between the occurrence of cracks and the measurement result of the strain gauge.

Therefore, it is difficult to increase the accuracy of the determination of crack initiation with the technique disclosed in Patent Document 1.

Therefore, the present invention is intended to provide a crack diagnosis method and apparatus capable of diagnosing the presence or absence of a crack in a crack diagnosis target and improving the accuracy of the diagnosis. is there.

In order to solve the above problems, the present invention provides a first strain sensor installed in a first region of a crack diagnosis target and a second strain sensor installed in a second region of the crack diagnosis target. A sensor and an arithmetic unit to which data is input from the first and second strain sensors are provided, and the first region and the second region have a structure corresponding to the crack diagnosis target. The result of distribution of strain change amount by the stress analysis using the first analysis model in the state without cracks, and the second analysis model in the state corresponding to the crack diagnosis target and in the state with cracks were used. Comparing with the result of the distribution of strain change amount by the stress analysis, two regions determined to satisfy the condition that the relative change ratio between the strain change amounts in each of the analysis models is equal to or more than a set value. In the arrangement corresponding to the above, there are two regions set as the crack diagnosis target, and the arithmetic unit is configured to operate in the first and the second areas when the crack diagnosis target is in a normal state in which no crack has occurred. The function of accumulating the data input from each of the strain sensors No. 2 as a normal data group and the data input from each of the first and second strain sensors at the time of the crack diagnosis of the crack diagnosis target are diagnosed. With a function to set in the data group, for the diagnostic data group, relative comparison with the normal data group by statistical processing, based on the comparison result, the function of determining the presence or absence of cracks in the crack diagnosis target and The crack diagnostic device is equipped with.

In the two regions defined so as to satisfy the condition that the relative change ratio between the strain change amounts in each of the analysis models is equal to or greater than the set value, the strain change amount in either one of the regions is strain. And the amount of strain change in the other region is a decrease in strain.

The arithmetic unit is a filter unit having a function of passing the frequency components corresponding to the set load mode for the data input from the first and second strain sensors while cutting the other frequency components. It is configured to have.

Distribution of strain change amounts by stress analysis using a first analysis model in a crack-diagnostic target with a first region and a second region having a structure corresponding to the crack-diagnostic target and having no crack And the result of the distribution of strain change amount by the stress analysis using the second analytical model having the structure corresponding to the crack diagnosis target and having the crack, the strain in each analytical model is compared. The first area set in the first area is set by the arrangement corresponding to the two areas that are set so as to satisfy the condition that the relative change rate between the change amounts becomes equal to or more than the set value. A strain sensor and a second strain sensor installed in the second area are prepared, and when the crack diagnosis target is in a normal state in which no crack has occurred, each of the first and second strain sensors is prepared. A process of accumulating the data input from the strain sensor as a normal data group, and the data input from the first and second strain sensors at the crack diagnosis timing of the crack diagnosis target into the diagnosis data group. With respect to the processing to be set and the diagnosis data group, relative comparison is performed with the normal data group by statistical processing, and based on the comparison result, processing for determining the presence or absence of a crack in the crack diagnosis target is performed. Use the crack diagnosis method.

According to the crack diagnosing method and apparatus of the present invention, it is possible to diagnose whether or not a crack has occurred in a crack diagnosing target and to improve the accuracy of the diagnosis.

It is a schematic diagram showing a 1st embodiment of a crack diagnostic device. It is a figure for demonstrating arrangement|positioning of the strain sensor in 1st Embodiment. It is a schematic diagram which shows the modification of 1st Embodiment. It is a schematic diagram showing a 2nd embodiment of a crack diagnostic device. It is a figure for demonstrating arrangement|positioning of the strain sensor in 2nd Embodiment. It is a schematic diagram showing a 3rd embodiment of a crack diagnostic device. It is a figure for demonstrating arrangement|positioning of the strain sensor in 3rd Embodiment. It is a schematic diagram showing a 4th embodiment of a crack diagnostic device.

Embodiments for carrying out the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 is a schematic diagram showing a first embodiment of a crack diagnostic device. FIG. 2 is a diagram for explaining the arrangement of strain sensors (strain gauges) in the crack diagnostic device of FIG. 1, and FIGS. 2(a) and 2(b) are schematic diagrams showing an analysis model for stress analysis, respectively. 2(c) is a schematic diagram showing a region set based on the strain change rate.

The crack diagnosing device of the present embodiment is shown by reference numeral 1 in FIG. 1, and is shown as an example in which it is applied to a crack diagnosing target 100.

Here, first, the crack diagnosis target 100 shown in FIG. 1 will be described.

The crack diagnosis target 100 according to the present embodiment is, for example, a structure, in which the end of another flat plate-shaped structural member 102 is joined to the surface of the flat plate-shaped structural member 101 by welding, and a welded joint 103 is formed. It is as a formed part. Furthermore, a uniaxial load of tension and compression is repeatedly applied to the structural member 101 in the direction indicated by the arrow X.

Due to the influence of this load, in the crack diagnosis target 100, a crack (fatigue crack) 104 due to fatigue as shown by a chain double-dashed line may occur at a position near the welded joint 103 in the structural member 101. There is. The locations where the cracks 104 occur in the crack diagnosis target 100 due to fatigue are shown in FIG. 1 empirically or by analysis based on the structure of the crack diagnosis target 100, the joint type, the direction of the applied load, and the like. It is clear that the position will be as shown.

Therefore, the crack diagnosing device 1 of the present embodiment is applied to the crack diagnosing target 100 as described above in order to diagnose the presence or absence of the crack 104.

The crack diagnostic device 1 of the present embodiment includes a strain sensor (strain gauge) 2 and strain sensors (strain gauges) 3a and 3b installed in the crack diagnostic target 100 in an arrangement described below, and further, It is configured to include an arithmetic unit 4 that processes data input from each strain sensor 2, 3a, 3b.

The arrangement of the strain sensor 2 and the strain sensors 3a and 3b installed on the crack diagnosis target 100 is determined by the following method.

That is, first, as shown in FIG. 2A, as an analysis model for stress analysis (stress/strain analysis), an analysis model 5A having a structure corresponding to the structure of the crack diagnosis target 100 (see FIG. 1) is provided. To prepare.

At this time, it is preferable to use, as the analysis model 5A, an analysis model 5A having a structure in which the structure of the crack diagnosis target 100 is standardized, rather than a model of the structure of the crack diagnosis target 100.

In view of this point, the analysis model 5A has a structure shown in FIG. 2A as a structure for standardizing the structure including the structural member 101, the structural member 102, and the welded joint 103 of the crack diagnosis target 100 shown in FIG. As shown in FIG. 3, the plate member 6 is provided with a so-called out-of-plane gusset welded joint structure in which the end of the out-of-plane gusset plate 7 is joined by fillet welding or full penetration welding.

This out-of-plane gusset welded joint is widely referred to as a fatigue design standard for structures, "Guideline and commentary on structural fatigue design by the Japan Steel Structural Association," and by the Japan Road Association. It is known as one of the welded joints shown in the "Fatigue design guidelines for steel road bridges".

Further, as shown in FIG. 2B, it has a structure similar to that of the analysis model 5A shown in FIG. An analysis model 5B having a structure having a crack 8 at a location is prepared.

At this time, the length L of the crack 8 included in the analysis model 5B is determined by the length of the crack 104 that actually occurs in the crack diagnosis target 100 (see FIG. 1). The size may be set according to whether or not detection is desired.

The length L of the crack 8 is preferably set to 5 mm or more. This is because when the length L of the crack 8 is set to less than 5 mm, the result of stress analysis using the analysis model 5B and the analysis model 5A of FIG. This is because the difference from the result of stress analysis becomes small, and thus it becomes difficult to obtain information on the amount of strain change described later.

The length L of the crack 8 is preferably set to 50 mm or less. According to, for example, "steel bridge pier corner portion damage on Hanshin Expressway, 2002 survey research report", this is 100 mm in length for fatigue crack damage determination of steel pier corner portions. The cracks of less than 1 are considered to be repaired in units of several years instead of immediate repair. Therefore, if the crack 104 of the crack diagnosis target 100 shown in FIG. 1 can be detected at a stage of a length of 5 mm to 50 mm using the crack diagnosis apparatus 1 of the present embodiment, on-site confirmation and repair are possible. It will be possible to secure sufficient planning time.

Furthermore, the length L of the crack 8 is more preferably set to less than 30 mm, and further preferably set to 10 mm to 20 mm. This is because the length of cracks that can be found by visual inspection in an actual structure is generally 30 mm or more. Therefore, according to this configuration, the crack diagnosing device 1 of the present embodiment can detect the crack 104 generated in the crack diagnosing target 100 even at a stage where it cannot be detected by visual inspection. ..

Although it is preferable to use a finite element method analysis (FEM analysis) as a method of stress analysis performed on the analysis model 5A and the analysis model 5B, the stress analysis is performed to analyze the strain distribution in the analysis model 5A and the analysis model 5B. If the information can be obtained, any stress analysis method other than the finite element method analysis may be adopted. Therefore, the analysis model 5A and the analysis model 5B may be created as models according to the stress analysis method used.

After the analysis model 5A and the analysis model 5B are created as described above, the stress when the stress acts on each of the analysis model 5A and the analysis model 5B in the direction indicated by the arrow x in FIGS. 2A and 2B. Analysis is performed to obtain the distribution of strains generated in the analysis model 5A and the analysis model 5B.

Next, the analysis result of the strain distribution obtained by the analysis model 5B with the crack 8 (see FIG. 2B) is obtained with the analysis model 5A without the crack (see FIG. 2A). Compare with the analysis result of the strain distribution obtained. By this comparison, how the strain at each position of the analysis model 5B with the crack 8 changes with reference to the strain at the same position of the analysis model 5A without the crack is obtained as a strain change amount. Therefore, this strain change amount is such that the strain at each position of the analysis model 5A in the crack-free state changes to a state in which the crack 8 having the predetermined length L exists, that is, with the occurrence of the crack 8, It is an indicator of how it changes. In the following description, the value of the strain change amount is indicated by r.

Then, based on the information of the strain change amount r, as shown in FIG. 2C, two regions 9 and 10 satisfying the following conditions are determined for the analysis model 5B. Although not shown, it goes without saying that the areas 9 and 10 may be set in the analysis model 5A.

In the present embodiment, the conditions for determining the two regions 9 and 10 are that the strain change amount r ₁ in the first region 9 is an increase in strain and the strain change amount r ₂ in the second region 10 is a strain. And the relative change rate of the strain change amount r ₁ and the strain change amount r ₂ (hereinafter referred to as strain change rate) is such that a significant difference can be detected by various statistical methods. The condition is that there is a difference in the rate of change. In this case, for example, the condition that the equivalence is rejected at the significance level of 5% to 1% may be adopted, but it goes without saying that the condition is not necessarily limited to this condition. As described above, the difference between the strain change rates is set to be equal to or larger than a predetermined value so that the difference between the strain change amount r ₁ in the region 9 and the strain change amount r ₂ in the region 10 becomes clear. This is to improve the detection accuracy of the crack 104 (see FIG. 1) described later.

After the areas 9 and 10 in the analysis model 5B are determined, as shown in FIG. 1, the actual crack diagnosis target 100 is arranged corresponding to the arrangements of the areas 9 and 10 in the analysis model 5B. Two areas 11 and 12 are defined. Note that, in FIG. 1, for convenience of illustration, the two regions 11 and 12 are indicated by a chain double-dashed line, but in the actual crack diagnosis target 100, the regions 11 and 12 may be arranged in a known manner. , There is no need to display it.

After that, one of the two areas 11 and 12 set as described above is set as the crack detection data acquisition area, and the other is set as the comparison data acquisition area.

FIG. 1 shows a case where the area 11 is set as the crack detection data acquisition area and the area 12 is set as the comparison data acquisition area.

The strain sensor 2 for acquiring the crack detection data a is installed in the area 11 which is the crack detection data acquisition area. Further, in the area 12 which is the comparison data acquisition area, two strain sensors 3a and 3b for acquiring the comparison data b are arranged and installed at two locations in the area.

The strain sensor 2 and the strain sensors 3a and 3b installed on the crack diagnosis target 100 are not shown in order to suppress and protect the influence of light, temperature, moisture, oil, oxidizing gas and the like received from the environment. You may make it cover with resin or a cover.

The crack detection data a is input from the strain sensor 2 to the arithmetic unit 4, and the comparison data b is input from the strain sensors 3a and 3b.

Next, the function of the arithmetic unit 4 will be described.

The arithmetic unit 4 uses the crack detection data a input from the strain sensor 2 and the comparison data input from the strain sensors 3a and 3b when the crack diagnosis target 100 is in a normal state in which no crack is generated. It has a function of accumulating a data group including the data b as a normal data group.

Further, the arithmetic unit 4 inputs from the strain sensor 2 at a crack diagnosis time when it is necessary to diagnose whether or not the crack 104 has occurred in the crack diagnosis target 100 after accumulating the normal data group. It has a function of setting a data group including the crack detection data a and the comparison data b input from the strain sensors 3a and 3b as a diagnostic data group.

Further, the arithmetic unit 4 has a function of comparing the diagnostic data group with the normal data group by statistical processing, and determining whether or not the crack 104 has occurred in the crack diagnosis target 100 based on the comparison result. ing.

As a method of relative comparison of the diagnostic data group with the normal data group by the arithmetic unit 4, for example, the following method may be used.

As shown in FIG. 1, in the present embodiment, the strain sensor 2 and the strain sensors 3a and 3b are arranged relatively close to the crack diagnosis target 100. Therefore, the strain sensor 2 and the strain sensors 3a and 3b are substantially similarly affected by a disturbance such as a temperature change or a load change in the crack diagnosis target 100.

Therefore, the relative relationship (correlation) between the crack detection data a by the strain sensor 2 and the comparison data b by the strain sensors 3a and 3b, for example, a proportional relationship, is a disturbance such as the temperature change or the load change. It is hardly affected by and is kept almost constant.

On the other hand, the area 11 provided with the strain sensor 2 and the area 12 provided with the strain sensors 3a and 3b are subjected to stress analysis using analysis models 5A and 5B (see FIGS. 2(a)(b)(c)). Thus, the strain change rate associated with the occurrence of the crack 8 is set in correspondence with the two regions 9 and 10 having a difference in strain change rate such that a significant difference can be detected by various statistical methods. Therefore, when the crack 104 is generated in the crack diagnosis target 100, the proportional relationship, which is the relative relationship between the crack detection data a by the strain sensor 2 and the comparison data b by the strain sensors 3a and 3b, is: It is estimated that changes occur at a rate related to the rate of strain change.

Therefore, in the arithmetic unit 4, the proportional relationship between the crack detection data a and the comparison data b in the diagnostic data group is equal to the proportional relationship between the crack detection data a and the comparison data b in the normal data group. The sex is processed by the F-test, which is one of the statistical tests (also referred to as the statistical temporary test). In this verification process, an equivalence test may be performed between the response surface for the proportional relationship in the diagnostic data group and the response surface for the proportional relationship in the normal data group.

As a result, when the equivalence is rejected, the arithmetic unit 4 determines that the crack 104 has occurred in the crack diagnosis target 100.

On the other hand, when the equivalence is not rejected, the arithmetic unit 4 determines that the crack 104 has not occurred in the crack diagnosis target 100.

The determination result of the presence or absence of the crack 104 in the crack diagnosis target 100 by the arithmetic unit 4 is output to a recording device (not shown) for recording, or output to a display device (not shown) or an audio output device (not shown) as necessary. It may be possible to notify the observer by using the above method.

The data transmission from the strain sensor 2 and the strain sensors 3a and 3b to the arithmetic unit 4 may be performed via a communication line such as a wire, a wireless or a telephone, or a network such as the Internet. Similarly, the output from the arithmetic device 4 to the recording device, the display device, and the audio output device may be performed via a communication line such as a wire, wireless, or telephone, or a network such as the Internet.

Therefore, according to the crack diagnostic device 1 of the present embodiment, it is possible to diagnose the presence or absence of the crack 104 in the crack diagnostic target 100.

Further, at arbitrary plural points in the region 11 of the crack diagnosis target 100, even if the distance from the crack 104 is different, when the crack 104 occurs, the strain shows a similar increasing tendency. Similarly, even if the distances from the cracks 104 are different at arbitrary plural points in the region 12 of the crack diagnosis target 100, when the cracks 104 occur, the strain shows a similar decreasing tendency.

Therefore, even if the crack diagnosis target 100 is provided with a plurality of strain sensors, the installation positions of all of them are included in the area 11, the area 12, or the crack 104 is generated. However, when the strain is arranged in a region where the change in strain is small, it is difficult to accurately detect the occurrence of the crack 104 based on the detection data of each strain sensor.

On the other hand, in the crack diagnostic device 1 of the present embodiment, based on the analysis results using the analysis models 5A and 5B, the region in which the strain change rate associated with the occurrence of the crack 104 is estimated to be the predetermined value The strain sensor 2 and the strain sensors 3a and 3b are separately installed in 11 and the region 12. Therefore, according to the crack diagnostic apparatus 1 of the present embodiment, it is possible to improve the accuracy of the diagnosis of the presence or absence of the crack 104 in the crack diagnostic target 100. When defining the regions 11 and 12, the strain magnitude and distribution may be preliminarily measured for strain so as to accurately grasp the strain change rate, and may be considered together with the analysis result. By doing so, it is possible to determine the installation areas of the strain sensor 2 and the strain sensors 3a and 3b that realize more accurate diagnosis.

Further, in the present embodiment, the area 12 is provided with the two strain sensors 3a and 3b, and the comparison data b can be acquired respectively.

For example, when there is only one comparison data b, when the comparison data b and the crack detection data a change in the same direction of increase or decrease, the relative difference in strain change rate is statistically superior. In order to have such a large difference, it is necessary to have a large difference, so that the detection sensitivity becomes low.

On the other hand, in the present embodiment, since two pieces of comparison data b are acquired, even if both pieces of comparison data b change in the same direction as the crack detection data a, the comparison The relative difference in the strain change rate between the data for data b and the data for crack detection a can be detected at a smaller stage. Further, if one of the two comparison data b changes in the opposite direction to the crack detection data a, the detection accuracy can be further improved.

In the present embodiment, as the strain sensor for acquiring the comparison data b, the configuration including the two strain sensors 3a and 3b is illustrated, but three or more strain sensors for acquiring the comparison data b are provided. Of course, the configuration may be adopted.

[Modification of First Embodiment]
FIG. 3 shows a modification of the first embodiment. Note that, in FIG. 3, the same components as those shown in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

In the first embodiment, as shown in FIG. 1, the area 11 is set as the crack detection data acquisition area and the area 12 is set as the comparison data acquisition area, but as shown in FIG. The area 11 may be set as the comparison data acquisition area and the area 12 may be set as the crack detection data acquisition area.

In this case, if the strain sensor 2 for obtaining the crack detection data a is installed in the region 12 and the strain sensors 3a, 3b for obtaining the comparison data b are installed in the region 11 Good.

Also in this modified example having such a configuration, it can be used in the same manner as in the first embodiment, and the same effect as in the first embodiment can be obtained.

[Second Embodiment]
In the first embodiment, as shown in FIGS. 2A, 2</b>B, and 2</b>C, strain accompanying the occurrence of the crack 8 is based on the analysis result using the analysis model 5</b>A and the analysis model 5</b>B. The area 9 in which the strain increases and the area 10 in which the strain decreases are obtained under the condition that the strain change rates of the both have a difference in strain change rate such that a significant difference can be detected by various statistical methods. .. In the crack diagnosis target 100 shown in FIG. 1, the sensor installation areas 11 and 12 are defined in the arrangements corresponding to the areas 9 and 10 obtained for the analysis model 5B.

However, in the actual crack diagnosis target 100, there are cases where a surface or space for sensor installation cannot be secured at a location corresponding to the region 10 in the analysis model 5B due to structural restrictions or the like.

In this case, the crack diagnostic device may have the configuration shown in FIG.

FIG. 4 is a schematic diagram showing a second embodiment of the crack diagnostic device. FIG. 5 is a diagram for explaining the arrangement of strain sensors in this embodiment. In FIGS. 4 and 5, the same components as those shown in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.

The crack diagnosing device of the present embodiment is shown by reference numeral 1A in FIG. 4, and similarly to the first embodiment, the strain sensor 2 installed on the crack diagnosing target 100, and the strain sensors 3a and 3b, Although it has a configuration including the arithmetic unit 4, the arrangement of the strain sensor 2 and the strain sensors 3a and 3b is determined by a method different from that of the first embodiment.

That is, in the present embodiment, similarly to the first embodiment, the analysis model 5A in the state without cracks shown in FIG. 2A and the analysis model 5B in the state with cracks 8 shown in FIG. 2B are provided. Based on the analysis result used, how the strain at each position of the analysis model 5B changes with the strain at the same position of the analysis model 5A as a reference is obtained as a strain change amount r.

Next, based on the information of the strain change amount r, as shown in FIG. 5, two regions 13 and 14 that satisfy the following conditions are determined for the analytical model 5B. Although not shown, it goes without saying that the regions 13 and 14 may be set in the analysis model 5A.

The condition for defining these two regions 13 and 14 is that the strain change amount r ₃ in the first region 13 is an increase in strain and the strain change amount r ₄ in the second region 14 is greater than the set value. Is a condition that the change (increase/decrease) in strain is small. In addition, regarding the strain change rate between the strain change amount r ₃ and the strain change amount r ₄ , there is a condition that there is a difference in strain change rate such that a significant difference can be detected by various statistical methods. The reason why the difference between the strain change rates is set to be equal to or larger than a predetermined value is the same as in the first embodiment.

After the areas 13 and 14 in the analysis model 5B are defined, as shown in FIG. 4, in the actual crack diagnosis target 100, the area 13 (see FIG. 5) in the analysis model 5B is arranged at a location corresponding to the arrangement. An area 15 for obtaining the crack detection data a is defined. Further, in the crack diagnosis target 100, a region 16 for acquiring the comparison data b is set at a position corresponding to the arrangement of the region 14 (see FIG. 5) of the analysis model 5B.

Note that, in FIG. 4, for convenience of illustration, the two regions 15 and 16 are shown by a chain double-dashed line, but in the actual crack diagnosis target 100, the regions 15 and 16 only have to be arranged. , There is no need to display it.

The strain sensor 2 similar to that of the first embodiment is installed in the region 15, and the two strain sensors 3a and 3b are installed in the region 16 at two positions in the region. Note that the present embodiment may employ a configuration including three or more strain sensors for acquisition of the comparison data b, similar to the first embodiment.

The crack diagnostic device 1A of the present embodiment having the above configuration can also be used in the same manner as in the first embodiment, and the same effects as in the first embodiment can be obtained.

Further, the installation area of the strain sensor 2 and the strain sensors 3a and 3b can be set while avoiding the structural limitation of the crack diagnosis target 100.

[Third Embodiment]
In the first embodiment, as shown in FIG. 2C, for the sensor installation in the crack diagnosis target 100 as shown in FIG. 1, with an arrangement corresponding to the regions 9 and 10 obtained for the analysis model 5B. The areas 11 and 12 of the above are defined.

However, in the actual crack diagnosis target 100, there are cases where a surface or space for sensor installation cannot be secured at a location corresponding to the region 9 in the analysis model 5B due to structural restrictions or the like.

In this case, the crack diagnostic device may have the configuration shown in FIG.

FIG. 6 is a schematic diagram showing a third embodiment of the crack diagnostic device. FIG. 7 is a diagram for explaining the arrangement of strain sensors in this embodiment. In FIGS. 6 and 7, the same parts as those shown in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.

The crack diagnosing device of the present embodiment is shown by reference numeral 1B in FIG. 6, and similarly to the first embodiment, the strain sensor 2 installed in the crack diagnosing target 100, and the strain sensors 3a and 3b, Although it has a configuration including the arithmetic unit 4, the arrangement of the strain sensor 2 and the strain sensors 3a and 3b is determined by a method different from that of the first embodiment.

That is, in the present embodiment, as in the first embodiment, the analysis model 5A in the state without cracks shown in FIG. 2A and the analysis model 5B in the state with cracks 8 shown in FIG. 2B are provided. Based on the analysis result used, how the strain at each position of the analysis model 5B changes with the strain at the same position of the analysis model 5A as a reference is obtained as a strain change amount r.

Next, based on the information of the strain change amount r, as shown in FIG. 7, two regions 17 and 18 satisfying the following conditions are determined for the analytical model 5B. Although not shown, it goes without saying that the areas 17 and 18 may be set in the analysis model 5A.

The conditions for defining these two regions 17 and 18 are that the strain change amount r ₅ in the first region 17 is a decrease in strain, and the strain change amount r ₆ in the second region 18 is greater than the set value. Is a condition that the change (increase/decrease) in strain is small. In addition, regarding the strain change rate between the strain change amount r ₅ and the strain change amount r ₆ , there is a condition that the strain change rate has a difference such that a significant difference can be detected by various statistical methods. The reason why the difference between the strain change rates is set to be equal to or larger than a predetermined value is the same as in the first embodiment.

After the area 17 and the area 18 in the analysis model 5B are determined, as shown in FIG. 6, in the actual crack diagnosis target 100, a position corresponding to the arrangement of the area 17 (see FIG. 7) of the analysis model 5B is obtained. An area 19 for obtaining the crack detection data a is defined. Further, in the crack diagnosis target 100, an area 20 for acquiring the comparison data b is set at a position corresponding to the arrangement of the area 18 (see FIG. 7) of the analysis model 5B.

Note that, in FIG. 6, for convenience of illustration, the two regions 19 and 20 are shown by a chain double-dashed line, but in the actual crack diagnosis target 100, the regions 19 and 20 only have to be arranged. , There is no need to display it.

The strain sensor 2 similar to that of the first embodiment is installed in the region 19, and the two strain sensors 3a and 3b are installed in the region 20 at two positions in the region. Note that the present embodiment may employ a configuration including three or more strain sensors for acquisition of the comparison data b, similar to the first embodiment.

The crack diagnostic device 1B of the present embodiment having the above configuration can also be used in the same manner as in the first embodiment, and the same effects as in the first embodiment can be obtained.

Further, the installation area of the strain sensor 2 and the strain sensors 3a and 3b can be set while avoiding the structural limitation of the crack diagnosis target 100.

[Fourth Embodiment]
FIG. 8A is a schematic diagram showing a fourth embodiment of the crack diagnostic device. 8B, 8C, and 8D are schematic diagrams showing the relationship between the load acting on the structural member to be crack-diagnosed and the stress distribution.

8(a), (b), (c), and (d), the same components as those shown in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

In an actual structure, in addition to the case where only the uniaxial load as shown by the arrow X in FIG. 1 works, a plurality of load modes such as the uniaxial load and bending load may work in combination.

As shown in FIG. 8B, for example, in the flat plate-shaped structural member 101, when the uniaxial load mode in the direction indicated by the arrow acts, a uniform stress distribution occurs in the plate width direction.

On the other hand, as shown in FIG. 8C, when the bending load mode as indicated by the arrow acts on the structural member 101, an uneven stress distribution occurs in the plate width direction.

As described above, since the stress distributions are different in the different load modes, as shown in FIG. 8D, when the uniaxial load mode and the bending load mode are superposed on the structural member 101, the stress distribution becomes uneven. turn into. Further, generally, the frequencies are different between the uniaxial load mode and the bending load mode, so that even if each load mode is a simple vibration, the stress acting at a certain position is not a simple vibration.

Therefore, the crack detection data a acquired by the strain sensor 2 installed in the area 11 of FIG. 8A and the comparison data b acquired by the strain sensors 3 a and 3 b installed in the area 12 The relative relationship includes a proportional relationship regarding the uniaxial load mode and a proportional relationship regarding the bending load mode.

In this case, the proportional relationship between the individual load modes is not easily affected by disturbances such as temperature change and load change, and even if the proportional relationship is maintained at a substantially constant value, in the state where the two proportional relationships exist, cracks are generated. It is difficult to detect a change in the proportional relationship that occurs when the crack 104 occurs in the diagnosis target 100.

Therefore, as shown in FIG. 8A, the crack diagnostic device 1C of the present embodiment has a configuration similar to that of the first embodiment, in which the arithmetic unit 4 is provided with the filter portion 21.

The filter unit 21 allows the frequency component corresponding to the set load mode to pass between the crack detection data a input from the strain sensor 2 and the comparison data b input from the strain sensors 3a and 3b. It has the function of cutting other frequency components.

The filter unit 21 may appropriately select and use a bandpass filter, a lowpass filter, or a highpass filter according to the wavelength of the frequency component to be passed.

According to the crack diagnostic apparatus 1C of the present embodiment having the above-described configuration, even when a plurality of load modes work on the crack diagnostic target 100 in superposition, the arithmetic unit 4 passes through the filter unit 21. It is possible to diagnose the presence or absence of the crack 104 in the crack diagnosis target 100 based on the data of the frequency component corresponding to the predetermined load mode.

Therefore, also in the crack diagnostic device 1C of the present embodiment, the same effect as that of the first embodiment can be obtained.

[Arrangement of strain sensor]
In the second embodiment, in the crack diagnosis target 100, the region 16 in which the strain sensors 3a and 3b for obtaining the comparison data b are installed corresponds to the arrangement of the region 14 in the analysis model 5B shown in FIG. Specified. As one of the conditions for obtaining the region 14, the condition that the strain change amount r ₄ in the region 14 is a strain change (increase/decrease) smaller than the set value is provided.

However, in the analysis model 5B, if equal to or greater than the value distortion change ratio is set between the strain variation r ₄ of strain variation r ₃ and the region 14 of the region 13, the strain variation r ₄ regions 14 Does not necessarily require the condition that the strain changes (increases or decreases) smaller than the set value. Even in this case, when the crack 104 is generated in the crack diagnosis target 100, the proportional relationship between the crack detection data a by the strain sensor 2 and the comparison data b by the strain sensors 3a and 3b is: The same changes as in the case of the first embodiment occur.

In the third embodiment, in the crack diagnosis target 100, the region 20 in which the strain sensors 3a and 3b for obtaining the comparison data b are installed corresponds to the arrangement of the region 18 in the analysis model 5B shown in FIG. Specified. As one of the conditions for obtaining the region 18, a condition that the strain change amount r ₆ in the region 18 is a strain change (increase/decrease) smaller than a set value is provided.

However, in the analysis model 5B, if equal to or greater than the value distortion change ratio is set between the strain variation r ₆ of strain variation r ₅ and region 18 of the region 17, the strain variation r ₆ regions 18 Does not necessarily require the condition that the strain changes (increases or decreases) smaller than the set value. Even in this case, when the crack 104 is generated in the crack diagnosis target 100, the proportional relationship between the crack detection data a by the strain sensor 2 and the comparison data b by the strain sensors 3a and 3b is: The same changes as in the case of the first embodiment occur.

Here, the points of the arrangement of the strain sensor in the present invention will be summarized from the above points and the above-described respective embodiments and modified examples. The arrangement of the area in which the strain sensor 2 for acquiring the crack detection data a is installed and the area for installing the strain sensors 3a, 3b in which the comparison data b is acquired are made to correspond to the arrangement of the two areas in the analysis model 5B. If these settings are made, the basic conditions for defining these two areas are as follows.

This is because the strain change rate between the strain change amount r _x in the first region and the strain change amount r _y in the second region is such that a significant difference can be detected by various statistical methods. The condition is that the difference is equal to or greater than the difference.

If this condition is satisfied, with regard to the strain change amount r _x and the strain change amount r _y , when one is an increase in strain and the other is a decrease in strain, both are an increase in strain, and both are It may be any case where the strain is reduced.

[Analysis model]
In each of the above-described embodiments and modified examples, the analysis models 5A and 5B are not a model of the structure of the crack diagnosis target 100 as it is, but have a structure in which the structure of the crack diagnosis target 100 is standardized. In addition, this standardized structure is shown in "Fatigue design guidelines and explanations for structures" of the Japan Steel Structure Association and "Fatigue design guidelines of steel road bridges" of the Japan Road Association. One of the existing welded joints is an out-of-plane gusset welded joint structure.

Therefore, for various structures, when the portion created as the structure of the out-of-plane gusset welded joint is referred to the fatigue design guideline as a reference of the fatigue design standard as the crack diagnosis target 100, the analysis model 5A, A common one can be used as 5B. Further, regarding the analysis model 5B, the regions 9 and 10 shown in FIG. 2C, the regions 13 and 14 shown in FIG. 5, and the regions 17 and 18 shown in FIG. 7 can be obtained in advance. ..

Of course, the analysis model may be changed according to the structure of the crack diagnosis target. In that case, as an analytical model of a structure in which the structure of the crack diagnosis target is standardized, for example, a load-transfer type cross weld joint, an in-plane gusset weld joint, a joint in which a cover plate is attached by fillet welding, etc. Of course, the structure of the welded joint shown in the design guide may be provided.

Further, an analysis model of a crack-free state provided with the structure of the welded joint shown in each of the fatigue design guidelines, and an analysis model of a cracked state are prepared, and the strain change rate is equal to or more than a set value. By previously obtaining information about the plurality of regions, the crack diagnostic devices 1, 1A, 1B, 1C can be quickly applied to various crack diagnostic targets.

Incidentally, as the analysis model of the state without cracks and the analysis model of the state with cracks, it is preferable to use one having a standardized structure as described above, but a structure simulating the structure of the crack diagnosis target as it is. Of course, it is also possible to use the one. Even in this case, the crack diagnosing devices 1, 1A, 1B, and 1C can obtain the same effects as those of the above-described embodiments and modifications.

[Application example]
In each of the above-described embodiments and modified examples, the crack diagnostic devices 1, 1A, 1B, and 1C have been described as diagnosing the presence or absence of the crack 104 in the crack diagnostic target 100. It may be applied to detect the progress of the crack 104 that has occurred.

Here, considering the phenomenon of crack 104 growth, it is understood that the growth of crack 104 is a phenomenon in which crack 104 occurs at a position on the extension of the tip side of crack 104 that has already occurred. be able to.

Therefore, if the presence or absence of the crack 104 is diagnosed at the location of the existing crack 104 by focusing only on the extension of the crack 104 on the tip side, the existing crack diagnosis target 100 as a whole can be detected. Whether or not the crack 104 has propagated can be detected.

In view of this point, in the same configuration as the crack diagnostic devices 1, 1A, 1B, 1C of the above-described respective embodiments and modified examples, a portion to be diagnosed for the presence or absence of the crack 104 is set as an existing crack. It is set at a position on the extension of the tip side of 104. That is, in the same configuration as the crack diagnostic apparatus 1, 1A, 1B, 1C of each of the embodiments and the modified examples, areas 11, 12, areas 15, 16 and areas where the strain sensor 2 and the strain sensors 3a, 3b are installed. Nos. 19 and 20 have a configuration in which a portion to be diagnosed for the presence or absence of the crack 104 is set to be shifted to a portion on the extension of the tip side of the existing crack.

The data output from the strain sensor 2 and the strain sensors 3a and 3b may be processed by an arithmetic device having the same function as the arithmetic device 4 of each of the embodiments and the modified examples.

According to this configuration, the arithmetic device can detect whether or not the existing crack 104 has propagated in the crack diagnosis target 100.

It should be noted that the present invention is not limited to the above-described respective embodiments and modified examples, and the arithmetic unit 4 compares the diagnostic data group with the normal data group with the crack detection data a when performing a relative comparison. A relative relationship other than the proportional relationship with the comparison data b may be used.

As a method of relative comparison of the diagnostic data group with the normal data group performed by the arithmetic unit 4, the crack detection data a and the comparison data b in the diagnostic data group are proportional to each other. The equivalence of the proportional relationship between the data a and the comparison data b may be tested by any statistical test method other than the F test.

Further, as a method of relative comparison of the diagnostic data group with the normal data group performed by the arithmetic unit 4, a method of verifying anomaly of the diagnostic data group with respect to the normal data group by the MT method (Mahalanobis Taguchi method) is adopted. Good. In this case, the arithmetic unit 4 may be provided with a function of determining that the crack 104 has occurred in the crack diagnosis target 100 when it is determined that the abnormality is present.

You may make it use the arithmetic unit 4 provided with the filter part 21 shown in 4th Embodiment in the modification of 1st Embodiment, 2nd Embodiment, and 3rd Embodiment.

1 to 8, the shapes of the regions 9 and 10, the regions 11 and 12, the regions 13 and 14, the regions 15 and 16, the regions 17 and 18, and the regions 19 and 20 are shown by ellipses and polygons, respectively. Of course, it may have any shape. The sizes of the areas 9 and 10, the areas 11 and 12, the areas 13 and 14, the areas 15 and 16, the areas 17 and 18, and the areas 19 and 20 may be other than the illustrated sizes. ..

The crack diagnosis target 100 may be any portion other than the joint portion between the structural members by the weld joint as long as the position where the crack is generated is known in the structure.

Needless to say, various changes can be made without departing from the scope of the present invention.

2 strain sensor, 3a, 3b strain sensor, 4 arithmetic unit, 5A, 5B analysis model, 8 crack, 9 area, 10 area, 11 area, 12 area, 13 area, 14 area, 15 area, 16 area, 17 area , 18 area, 19 area, 20 area, 21 filter part, 100 crack diagnosis target, 104 crack, a crack detection data (data), b comparison data (data)

Claims (11)

A first strain sensor installed in a first region of the crack diagnosis target;
A second strain sensor installed in a second region of the crack diagnosis target;
A crack detection data acquired by the first strain sensor and an arithmetic unit to which the comparison data acquired by the second strain sensor is input;
The arithmetic unit is
A function of accumulating, as a normal data group, the crack detection data and the comparison data that are input when a normal state in which no crack has occurred in the crack diagnosis target,
A function for setting the crack detection data and the comparison data input at the crack diagnosis time of the crack diagnosis target to a diagnosis data group,
For the proportional relationship between the crack detection data and the comparison data in the diagnostic data group, the equivalence of the proportional relationship between the crack detection data and the comparison data in the normal data group, statistically A crack diagnostic device, comprising: a function of determining whether or not a crack has occurred in the crack diagnostic target by performing a verification by a temporary verification. The crack according to claim 1, wherein the test of the equivalence is a test of equivalence between a response surface regarding the proportional relationship in the diagnostic data group and a response surface regarding the proportional relationship in the normal data group. Diagnostic device. The arithmetic unit is a filter unit having a function of passing the frequency components corresponding to the set load mode with respect to the data input from the first and second strain sensors, while cutting the other frequency components. The crack diagnosis device according to claim 1 or 2. In each of the first region and the second region, a stress acts in a predetermined direction in the first analysis model having a structure corresponding to the crack diagnosis target and having no crack in each region. According to the result of the strain distribution by the stress analysis and the stress analysis when the stress acts in the predetermined one direction in the second analysis model having the structure corresponding to the crack diagnosis target and the state with the crack. Depending on the amount of strain change that is the result of comparison with the result of strain distribution, the relative change rate of the amount of strain change in the first region and the amount of strain change in the second region is greater than or equal to the set value. In the arrangement corresponding to the two areas determined to satisfy the condition, the two areas are set as the crack diagnosis target.
The crack diagnostic device according to claim 1. The two regions defined to satisfy the condition that the relative change rate of the strain change amount of the first region and the strain change amount of the second region is equal to or greater than a set value, The crack diagnostic device according to claim 4, wherein the strain change amount in one of the regions is an increase in the strain, and the strain change amount in the other region is a decrease in the strain. The two regions defined to satisfy the condition that the relative change rate of the strain change amount of the first region and the strain change amount of the second region is equal to or greater than a set value, The crack diagnostic device according to claim 4, wherein the strain change amount in one of the regions is a strain increase, and the strain change amount in the other region is a strain change smaller than a predetermined set value. The two regions defined to satisfy the condition that the relative change rate of the strain change amount of the first region and the strain change amount of the second region is equal to or greater than a set value, The crack diagnostic device according to claim 4, wherein the strain change amount in one of the regions is a decrease in the strain, and the strain change amount in the other region is a change in the strain smaller than a predetermined set value. The data for crack detection is acquired by the first strain sensor installed in the first area of the crack diagnosis target, and compared with the second strain sensor installed in the second area of the crack diagnosis target. Process for acquiring data for
A process of accumulating the crack detection data and the comparison data in a normal state in which no crack has occurred in the crack diagnosis target, as a normal data group,
A process of setting the crack detection data and the comparison data of the crack diagnosis target of the crack diagnosis target to a diagnosis data group,
Regarding the proportional relationship between the crack detection data and the comparison data in the diagnostic data group, the equivalence of the proportional relationship between the crack detection data and the comparison data in the normal data group is statistically A method for diagnosing a crack, comprising: performing a process for determining whether or not a crack has occurred in the crack diagnosis target by performing a verification by a temporary verification. The crack according to claim 8, wherein the test of the equivalence is a test of equivalence between a response surface regarding the proportional relationship in the diagnostic data group and a response surface regarding the proportional relationship in the normal data group. Diagnostic method. The data input from each of the first and second strain sensors is further processed to pass a frequency component corresponding to a set load mode while cutting other frequency components. Crack diagnosis method. In each of the first region and the second region, a stress acts in a predetermined direction in the first analysis model having a structure corresponding to the crack diagnosis target and having no crack in each region. According to the result of the strain distribution by the stress analysis and the stress analysis when the stress acts in the predetermined one direction in the second analysis model having the structure corresponding to the crack diagnosis target and the state with the crack. Depending on the amount of strain change that is the result of comparison with the result of strain distribution, the relative change rate of the amount of strain change in the first region and the amount of strain change in the second region is greater than or equal to the set value. In the arrangement corresponding to the two areas determined to satisfy the condition, the two areas are set as the crack diagnosis target.
The crack diagnosis method according to any one of claims 8 to 10.

Images