JP3785577B1 - Crack detection system and crack detection method - Google Patents

Abstract

An object of the present invention is to easily detect the occurrence of a crack on the surface of an object to be detected with high sensitivity.
[Solution]
The element 3 (virtual region) is formed by connecting at least three points of the plurality of reference points 2 provided in the measurement range 10 on the surface of the detection object, and the measurement range 10 is discretized into a group of the plurality of elements 3. , Calculating the maximum principal strain of each element 3, calculating the difference strain of each element 3 by differentially processing each maximum principal strain, and detecting an element having a difference strain larger than an arbitrary threshold value And a crack detection means for detecting the occurrence of a crack in the body.
[Selection] Figure 2

Description

  The present invention relates to a novel crack detection system and crack detection method for detecting a damage state due to a crack in a concrete structure or the like.

By grasping the cracking situation of concrete structures such as tunnels and buildings, the degree of deterioration of the structures can be evaluated. By this evaluation, it is possible to prevent a concrete structure from being peeled off, and to determine a coping method such as repair, reinforcement, and demolition of the structure. For this reason, various devices and methods have been proposed for detecting cracks in structures.
As a device for detecting cracks, an electrical resistor is embedded in concrete and the measured value of this resistor is analyzed (see Patent Document 1), or elastic vibration is applied to the structure and data from its propagation vibration is used. Have been developed (see Patent Document 2).
JP 2003-107030 A JP-A-10-142200

However, since an apparatus for embedding an electric resistor in concrete is troublesome to prepare and troublesome, and there is a limit to the dimensional configuration of an object to be measured, it is very difficult to measure a micro object such as a micro device. And the method of applying a vibration to a structure has the problem that the propagation distance of a vibration is short and a measurement range is narrow. In addition, the measured data has a problem in the analysis accuracy because the surface of the structure and the internal reflected wave are mixed.
Therefore, the problem to be solved by the present invention is that a crack detection system can easily detect a crack occurrence location with high sensitivity by discretizing a concrete structure or the like into a plurality of elements and differentially processing the maximum principal strain of each element. And providing a method.

  In order to solve the above-described problem, a crack detection system according to the present invention measures an image pickup device that picks up a plurality of marks provided in a measurement range on the surface of a detected object, and the marks picked up by the image pickup device. An image processing and computing device for determining the coordinates of each target point, and connecting at least three points of each target point to form one element (virtual region) to discretize the measurement range into a group of a plurality of elements, The maximum principal strain of each element is calculated from the distortion of the surface of the detection object caused by the arbitrary load applied to the detection body and the coordinates of each gauge point before and after the load application obtained by the image processing and calculation device, and A calculating means for calculating the differential strain of each element by differential processing the maximum principal strain, and a crack detecting means for detecting an element having a differential strain larger than an arbitrary threshold as a crack occurrence location of the detected object It is equipped with a.

  Further, the crack detection method according to the present invention includes a step of imaging a plurality of target points existing in the measurement range of the surface of the continuous object to be detected, and a step of obtaining the coordinates of each target point by measuring the acquired target points. , A step of connecting at least three points of each mark to form one element (virtual area) and discretizing the measurement range into a collection of a plurality of elements (virtual areas), and an arbitrary load applied to the detected object Calculating the maximum principal strain of each element from the distortion of the surface of the object to be detected generated by the coordinates of the target points before and after the load is applied to the object to be detected; And a step of comparing the differential strain with a preset threshold value and detecting an element having a differential strain larger than the threshold value as a crack occurrence location of the detected object.

  Preferably, each element is discretized in a triangular shape, and the differential processing is performed by comparing the maximum principal strain of an arbitrary central element with the maximum principal strains of three peripheral elements sharing one vertex of the central element.

  More preferably, each element is formed into a right isosceles triangle having substantially the same shape, each element is made into a plurality of different arrays, and the direction of cracking is detected from the differential strain based on each array.

  Moreover, you may comprise so that each element may become a regular triangle of substantially the same shape.

  The crack detection system and method according to the present invention discretizes the surface of an object to be detected such as a concrete structure into a plurality of elements, performs differential processing on the maximum principal strain of each element, and sets this differential strain and a preset threshold value. Therefore, it is not necessary to embed a sensor or apply vibration to the concrete, so that cracks can be detected easily. Moreover, since the crack width etc. according to the kind etc. of a to-be-detected body etc. can be detected by setting a threshold value arbitrarily, the deterioration degree of a structure, etc. can be grasped | ascertained accurately.

  In addition, each element is discretized into a triangular shape, and the maximum principal strain of the peripheral element with respect to the central element is calculated by performing a difference process by comparing an arbitrary central element with three peripheral elements sharing one vertex of the central element. Since the influence can be reduced, the sensitivity of crack detection is significantly improved.

  Then, by configuring each element into a right isosceles triangle having substantially the same shape and performing differential processing from each element having a different arrangement, the directionality of the crack can be grasped. That is, when each element is a right-angled isosceles triangle, the maximum principal strain when the crack passes through two sides sandwiching a right angle is remarkably increased. You can understand and understand the direction of cracks.

  Furthermore, by making each element an equilateral triangle having substantially the same shape, the value of the differential strain falls within a substantially constant range regardless of the directionality of the crack, so that the crack detection sensitivity can be stabilized.

Hereinafter, a crack detection system and method according to the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 is a diagram for explaining a crack detection system. FIG. 1 (a) is a side view showing a state in which a load is applied to a detection target, and FIG. 1 (b) is an overall view of the crack detection system. FIG. As shown in FIG. 1, the detected object (1) of the present embodiment is a concrete beam in which a CFRP (carbon fiber reinforced plastic) sheet is bonded to the lower surface and bent and reinforced, and is in a simple beam state. The dimension of the detection object (1) is a square section having a width of 100 mm and a height of 100 mm, and a length of 400 mm. Then, a plurality of marks (2, 2,...) At intervals of 10 mm were attached to a measurement range (10) of 100 mm × 100 mm at the center of the surface of the detection object (1). In this example, a black circle mark (2, 2,...) Is attached to the surface of the detection object (1) by hand, but the mark (2, 2,...) Is the center. As long as the position is clear, an existing one provided in advance may be used. For example, tile joints provided on the surface of a building, a plurality of rivets provided on an airplane or the like, or a metal tissue pattern visualized with a microscope can be used.

  In this state, the measurement range (10) of the detection target (1) is imaged by an imaging device (20) such as a CCD camera. Then, using the image processing device (21), the center of each target point (2, 2,...) Of the captured image is set as a node (2 ′, 2 ′,...), And the plurality of nodes. A virtual region formed by connecting three points (2 ′, 2 ′, 2 ′) of (2 ′, 2 ′,...) Is defined as one element (3). Then, the measurement range (10) is divided into a group of a plurality of elements (3, 3,...) That do not interfere with each other and discretized. That is, the measurement range (10) is divided from each node (2 ', 2, ...) into a plurality of triangular virtual areas (3, 3, ...). FIG. 2 is a diagram showing a state in which the measurement range (10) is discretized into a plurality of elements (3, 3,...).

In this state, the coordinates of each node (2 ′, 2 ′,...) Are measured. Further, as shown in FIG. 1, an arbitrary load (P) is applied from the center upper surface of the detected object (1), and each node (2) in the measurement range (10) of the detected object (1) after being deformed thereby. Also measure the coordinates of ', 2' ...).
Then, using the arithmetic device (22) connected to the image processing device (21), each virtual region of the measurement range (10) is calculated from the coordinates of the nodes (2 ′, 2 ′,...) Before and after the deformation. That is, the maximum principal strain of each element (3, 3,...) Is calculated. Thereafter, the maximum principal strain is subjected to differential processing, and the differential strain of each element (3, 3,...) Is calculated. Then, using a crack detection device (23) such as a computer, the differential distortion is compared with a preset threshold value to detect the cracked element (3). Hereinafter, this calculation method will be described in detail.

FIG. 3 is an enlarged view showing one arbitrary element for explaining the calculation method. In order to explain the calculation method of the maximum principal strain for each element, the arbitrary element (3) is assigned the element (3) number (e) and the node (2 ′) number (i, j, k). ). Then, in the element number (e), the coordinates of each node (i, j, k) before deformation are {X ₀ ^e }, and the coordinates of each node (i, j, k) after deformation are {X ₁ ^e }. As shown below (Formula 1). These coordinates {X ₀ ^e } and {X ₁ ^e } are generalized, but in actuality, values of coordinates obtained by measuring each node (i, j, k) with an arbitrary point as the origin (0, 0). It is.

Using the coordinates {X ₀ ^e } and {X ₁ ^e } of the nodes before and after the deformation, a node displacement vector {U ₀ } of the node coordinates before and after the deformation is calculated from the following expression (Formula 2).

Then, in order to remove the rotational component from the nodal displacement vector {U ₀ }, the coordinate transformation matrix [T] obtained from the following expressions (Expression 3) and (Expression 4), and the {U ₀ } are expressed as follows (Expression 5). To calculate the true nodal displacement vector {U _e }.

Further, the strain displacement matrix [B ^e ] obtained from the later expression (Expression 6) and the true nodal displacement vector {U _e } are substituted into the strain displacement relational expression (Expression 7) described later, and the element number e The strain {ε ^e } is calculated.

Then, the maximum principal strain ε _e of the element number _e is calculated from the strain {ε ^e } and the postscript (Equation 8). Furthermore, the maximum principal strain epsilon _e of the element number _e, and three optional peripheral element around said and calculated element number in a similar manner (e) (a, b, c) strain each maximum principal of epsilon Difference processing is performed from _{a 1} , ε _b , and ε _c to obtain a differential strain Δε of element number (e). This differential strain Δε is calculated by substituting ε _e , ε _a , ε _b , and ε _c into the following (Equation 9).

Here, the peripheral elements (a, b, c) for performing the differential processing of the central element (e) are examined. FIG. 4 is a diagram for examining peripheral elements. 4A shows a case where an element sharing one side of the central element (e) is a peripheral element (a, b, c), and FIG. 4B shows one vertex of the central element (e). The case where the three elements sharing the symbol are the peripheral elements (a, b, c) is shown.
In FIG. 4, the crack (C ₁ ) passes through the element number (c, e, a) in FIG. 4 (a), and passes only the element number (e) in FIG. 4 (b). The crack (C ₂ ) passes through the element numbers (a, e, b) in FIG. 4 (a) and passes through the element numbers (a, e) in FIG. 4 (b). Therefore, in both the cracks (C ₁ ) and (C ₂ ), the differential strain Δε of the element number (e) is large in the case of FIG. 4B (see (Formula 9)). Here, as will be described later, the larger the differential strain Δε is, the easier it is to detect a crack occurrence location, and therefore, as shown in FIG. 4B, three elements sharing one vertex of the central element (e) As the element (a, b, c), the differential strain Δε of the element number (e) is calculated from the above (Formula 9).

  Then, the differential strain Δε is calculated for all the discrete elements, and the elements having the differential strain Δε larger than a preset threshold value are extracted using the crack detection device (23). This threshold value is set in advance as a value of the differential strain at which cracks of the detected object occur. Therefore, a crack has occurred in an element having a differential strain Δε larger than the threshold value. This threshold value can be determined according to the type of object to be detected, the width of cracks, and the like. Then, the crack detection device (23) and the image processing device (21) are connected, and as shown in FIG. 5, an element having differential distortion exceeding a threshold value based on the signal from the crack detection device (23). Is black and other elements are colored white and displayed on the image processing device (21), the crack occurrence location in the measurement range (10) can be visualized. The image processing device (21), the calculation device (22), and the crack detection device (23) can also be configured as integrated hardware.

Next, the shape and arrangement of elements will be examined. 6A and 6B are diagrams for examining the element shape. FIG. 6A shows a case where the element shape is an equilateral triangle, and FIG. 6B shows a case where the element shape is a right-angled isosceles triangle.
As shown in FIG. 6, the crack (C) has a width dimension of δ and an angle with respect to the base line (BL) in the x-axis direction of θ (0 ° to 360 °). And let the length of the side shown to the figure of each element (3) be (L). In this case, the relational expressions of δ, θ, L and the maximum principal strain ε _e of the element number (e) are (Expression 10) to (Expression 12) described later. Here, (Expression 10) is a relational expression when the shape of the element (3) is an equilateral triangle, and (Expression 11) and (Expression 12) are relational expressions when the shape of the element (3) is a right isosceles triangle. Show.

Figure 7 is a graph showing the relationship between θ and epsilon _e. In FIG. 7, a curve (I) shows a change when the shape of the element (3) is an equilateral triangle, and a curve (II) shows a change when the shape of the element (3) is a right isosceles triangle. Here, δ and L are arbitrary constant values. As shown by curve (I) in FIG. 7, when the element shape is an equilateral triangle, the maximum principal strain ε _e of the element number (e) is constant at any angle θ (0 ° ≦ θ ≦ 360 °). It is a value that periodically changes within the range. Big Accordingly, the shape of the element (3) is an equilateral triangle, the minimum crack width to be detected [delta] (e.g., 0.05 mm) by the smallest maximum threshold the principal strain epsilon _e in, than the threshold The differential strain Δε of the value can be determined almost stably, and the sensitivity of crack detection is also stable.

When the element shape is a right-angled isosceles triangle, the maximum principal strain ε _e has a significantly different value depending on the angle θ. That is, as shown by curve (II) in FIG. 7, the maximum principal strain ε _e is obtained when the crack (C) passes through the bottom (0 ° ≦ θ ≦ 110 °, 160 ° ≦ θ ≦ 290 °, 340 °). ≦ θ ≦ 360 °), a small value, and when the crack (C) passes through two sides sandwiching a right angle (110 ° ≦ θ ≦ 160 °, 290 ° ≦ θ ≦ 340 °), a large value.
Thus, by using the property that the value of the maximum principal strain ε _e increases when the crack (C) passes through two sides sandwiching the right angle of the element (3), the direction of the crack is grasped. Can do. This will be described in detail.

FIG. 8 is a diagram for examining the directionality of cracks. (A) and (b) of FIG. 8 are the same nodes (2 ′ ₁ , 2 ′ ₂ , 2 ′ ₃ , 2 ′ ₄ ) and cracks (C ₃ ) in the same direction, both of which are elements (3, 3 The shape of ') is a right-angled isosceles triangle, but Fig. 8 (a) shows that when the element (3) is composed of nodes (2' ₁ , 2 ' ₂ , 2' ₃ ), Fig. 8 (b) The case where the element (3 ′) is configured by the nodes (2 ′ ₁ , 2 ′ ₂ , 2 ′ ₄ ) is shown.
In the case of FIG. 8A, since the crack (C ₃ ) passes through the bottom of the number (3), the maximum principal strain ε _e becomes small as described above. On the other hand, in the case of FIG. 8B, since the crack (C ₃ ) passes through two sides sandwiching the right angle of the element (3 ′), the maximum principal strain ε _e increases. Thus, when the shape of the element (3, 3 ′) is a right-angled isosceles triangle, the arrangement of the element (3, 3 ′) is changed with respect to the crack (C ₃ ), and the differential strain Δε is large. if the change, notice that the crack (C ₃₎ is passing through two sides sandwiching a right angle, it is possible to grasp the direction of crack (C _3).

  According to the crack detection system and method of the present invention, by attaching a mark to the object to be detected, outside of large-scale concrete structures such as tunnels and bridges, small-scale structures such as pipes, microchips, etc. Even in the case of microscopic objects, it is possible to easily detect the occurrence of cracks. Cracks are generated not only by directly applying a load, but also by strain and pressure changes due to temperature changes, and the present invention is applied in various technical fields and is very useful.

It is a figure for demonstrating a crack detection system. It is a figure which shows the state which discretized the measurement range into the some element. It is an enlarged view which shows one arbitrary element for demonstrating the calculation method. It is a figure for examining a peripheral element. It is a figure which shows the state which visualized the crack of the measurement range. It is a figure for examining element shape. It is a graph which shows the relationship between (theta) and (epsilon) _e . It is a figure for examining the directionality of a crack.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... To-be-detected object, 10 ... Measuring object 2 ... Mark, 2 '... Node 3 ... Element C ... Crack e ... Generalized element number i, j, k ... Generalized node number 20 ... Imaging device 21 ... Image processing device 22 ... Arithmetic device 23 ... Crack detection device

Claims (8)

  An imaging device that captures a plurality of target points provided in the measurement range on the surface of the detected object, an image processing and arithmetic unit that measures the target points captured by the imaging device and obtains coordinates of each target point, and each target point Distortion of the surface of the detected object caused by an arbitrary load applied to the detected object by connecting at least three points to form one element (virtual region) to discretize the measurement range into a collection of a plurality of elements. And calculating the maximum principal strain of each element from the coordinates of each gauge point before and after applying the load obtained by the image processing and computing device, and differentially processing each maximum principal strain to calculate the differential strain of each element. A crack detection system comprising: a calculation means for calculating; and a crack detection means for detecting an element having a differential strain larger than an arbitrary threshold as a crack occurrence location of a detected object.   Each element has a triangular shape, and the arithmetic means compares the maximum principal strain of an arbitrary central element with the maximum principal strains of three peripheral elements sharing one vertex of the central element and performs differential processing. The crack detection system according to claim 1.   Each of the elements is composed of a right isosceles triangle having substantially the same shape, and the crack detection means detects the directionality of the crack based on the differential strain obtained from each element in a plurality of arrays. The crack detection system according to claim 1 or 2.   3. The crack detection system according to claim 1, wherein each of the elements is formed of an equilateral triangle having substantially the same shape.   A step of imaging a plurality of target points existing in the measurement range of the surface of the detected object, a step of measuring the captured target points to obtain coordinates of each target point, and connecting at least three points of each target point Forming a single element (virtual region) to discretize the measurement range into a collection of a plurality of elements (virtual regions), and distortion of the surface of the detected object caused by an arbitrary load applied to the detected object Calculating the maximum principal strain of each element from the coordinates of each gauge point before and after applying a load to the detected object; calculating the differential strain of each element by differentially processing each of these maximum principal strains; And a step of comparing the differential strain with a preset threshold value and detecting an element having a differential strain larger than the threshold value as a crack occurrence location of the detected object. Method.   The elements are discretized in a triangular shape, and the differential processing is performed by comparing the maximum principal strain of an arbitrary central element and the maximum principal strains of three peripheral elements sharing one vertex of the central element. Item 6. The crack detection method according to Item 5.   7. Each element is formed into a right isosceles triangle having substantially the same shape, each element is made into a plurality of different arrays, and the directionality of the crack is detected from the differential strain based on each array. The crack detection method as described in.   The crack detection method according to claim 5 or 6, wherein each element is formed in a regular triangle having substantially the same shape.

Images