Rope inspection device and rope inspection method

Abstract

In a rope inspection device that inspects a rope (100) including a band-shaped CFRP and a coating layer that covers the CFRP, an alternating-current magnetic field is applied to the CFRP, eddy currents (200) generated by the alternating-current magnetic field are detected, a detection signal (B) representing the intensity of the eddy currents and a fiber image representing the distribution of the intensity of the eddy currents are generated based on the detection results in each of a plurality of unit regions in which a respective one end portion is located at a plurality of positions different from each other in the length direction of the CFRP, and it is determined whether or not there is a defect in the unit region based on the comparison result of the detection signal (B) with a threshold value and the result of image recognition for the fiber image (C). Various defects including longitudinal cracks in carbon fibers can be detected.

Description

Technical Field

[0001] This disclosure relates to rope inspection apparatus and rope inspection methods. In particular, this disclosure relates to non-destructive inspection of ropes containing CFRP (carbon fiber reinforced plastic) for detecting defects. Background Technology

[0002] As an inspection device for CFRP ropes, there are known inspection devices that detect defects in CFRP based on eddy currents generated in CFRP when a magnetic field is applied (e.g., Patent Document 1).

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: U.S. Patent Application Publication No. 2019 / 0202666 (paragraphs 0094-0095, Figure 9) Summary of the Invention

[0006] The problem the invention aims to solve

[0007] The inspection device in Patent Document 1 has the problem of being unable to detect longitudinal cracks within carbon fibers.

[0008] The purpose of this disclosure is to provide a rope inspection apparatus and rope inspection method capable of detecting various defects including longitudinal cracks within carbon fibers.

[0009] means for solving problems

[0010] The rope inspection apparatus disclosed herein inspects a rope comprising a strip of CFRP and a coating covering the CFRP, wherein the rope inspection apparatus comprises: a detector that applies an alternating magnetic field to the CFRP and detects eddy currents generated by the alternating magnetic field; and a signal processing device that, based on the detection results in each of a plurality of unit regions having one end at a plurality of different locations along the length direction of the CFRP, generates a detection signal representing the intensity of the eddy currents and a fiber image representing the distribution of the intensity of the eddy currents, and determines whether there are defects in the unit region based on a comparison result of the detection signal with a threshold and an image recognition result of the fiber image.

[0011] The effects of the invention

[0012] According to this disclosure, a fiber image representing the distribution of eddy current intensity is generated, and the presence or absence of defects is determined based on the fiber image. Therefore, various defects, including longitudinal cracks in carbon fibers, can be detected. Attached Figure Description

[0013] Figure 1(a) is a perspective view showing an example of a rope containing CFRP, and (b) is a magnified view showing a portion of the CFRP in (a).

[0014] Figure 2 This is a block diagram illustrating the rope inspection device according to Embodiment 1.

[0015] Figure 3 (a) to (c) are schematic diagrams showing different structural examples of the detector.

[0016] Figure 4 This is a block diagram illustrating an example of the detection element of a detector.

[0017] Figure 5 Figures (a) to (e) are diagrams illustrating the inspection method in Embodiment 1.

[0018] Figure 6 This is a flowchart illustrating the processing steps in the rope inspection device of Embodiment 1.

[0019] Figure 7 This is a block diagram illustrating another example of the detection element of the detector.

[0020] Figure 8 (a) to (d) are diagrams showing the processing used to improve the image quality of fiber images, and (e) is a diagram showing an example of an image obtained as a result of the image quality improvement.

[0021] Figure 9 It shows that it was used Figure 3 The diagram shows the inspection method in the case of detector (c).

[0022] Figure 10 This is a block diagram illustrating the rope inspection device of Embodiment 2.

[0023] Figure 11 (a) is a diagram showing a portion of a reference image generated in Embodiment 2, and (b) is a diagram showing an example of a fiber image.

[0024] Figure 12 This is a flowchart illustrating the steps of the process for obtaining reference images in the rope inspection device of Embodiment 2.

[0025] Figure 13 This is a flowchart illustrating the steps of the inspection process performed in the rope inspection device of Embodiment 2.

[0026] Figure 14 This is a block diagram showing the rope inspection device according to Embodiment 3.

[0027] Figure 15(a) is a graph used to illustrate the skin effect, (b) is a graph showing the relationship between excitation frequency and skin depth, and (c) is a graph showing the relationship between excitation frequency and the generated fiber image.

[0028] Figure 16 This is a flowchart illustrating the processing steps in the rope inspection device of Embodiment 3.

[0029] Figure 17 This is a block diagram illustrating the rope inspection device of Embodiment 4.

[0030] Figure 18 This is a flowchart illustrating the steps of the process for obtaining reference images in the rope inspection device of Embodiment 4.

[0031] Figure 19 This is a flowchart illustrating the steps of the inspection process performed in the rope inspection device of Embodiment 4.

[0032] Figure 20 This is a block diagram illustrating the rope inspection device of Embodiment 5.

[0033] Figure 21 This is a diagram showing the detection element used in Embodiment 5.

[0034] Figure 22 It is a graph showing the relationship between excitation frequency, gap, and the generated fiber image.

[0035] Figure 23 This is a flowchart illustrating the steps of the process for obtaining reference images in the rope inspection device of Embodiment 5.

[0036] Figure 24 This is a flowchart illustrating the steps of the inspection process performed in the rope inspection device of Embodiment 5. Detailed Implementation

[0037] Implementation method 1.

[0038] The rope inspection device and rope inspection method of Embodiment 1 will be described below.

[0039] Figure 1 (a) shows an example of a strip-shaped rope 100 that is the object of inspection in the rope inspection device and rope inspection method of Embodiment 1. Figure 1 In (a), L corresponds to the length direction of rope 100, W corresponds to the width direction of rope 100, and T corresponds to the thickness direction of rope 100. As shown in the figure, the cross-section of rope 100 is rectangular, and one side (W direction) of the rectangle is longer than the other side (T direction). Hereinafter, the surface 101 of rope 100 extending along the L and W directions will be referred to as the main surface.

[0040] Rope 100 comprises CFRP (carbon fiber reinforced plastic) 110 and a non-metallic coating 120 covering the CFRP. CFRP 110 is composed of carbon fiber 114 and resin material 115. The cross-section of CFRP 110 is the same as that of rope 100, being rectangular, with one side (W direction) longer than the other side (T direction). Therefore, CFRP 110 is also strip-shaped. The length direction (L direction), width direction (W direction), and thickness direction (T direction) of CFRP 110 are consistent with the L direction, W direction, and T direction of rope 100.

[0041] Carbon fiber 114 extends along the L direction, such as Figure 1 As shown in (b), it is distributed in resin material 115. The distribution is not necessarily uniform, and sometimes it is non-uniform as shown in the figure.

[0042] In the rope inspection device and rope inspection method of Embodiment 1, the rope 100 that is the object of inspection is, for example, an elevator rope, a bridge rope, or a building component rope. An elevator rope, for example, is used to suspend the elevator car.

[0043] For example, inspections can be conducted during the rope manufacturing process, the installation phase of elevators using ropes, the construction phase of bridges, the construction phase of buildings using structural components, and the operation or maintenance phase of elevators.

[0044] When inspecting an elevator during operation, it can be done, for example, while the ropes are moving along with the elevator car.

[0045] The following explanation assumes that the situation mainly occurs in the operation of elevators.

[0046] Figure 2 The rope inspection device 1 of Embodiment 1 is shown.

[0047] The rope inspection device 1 shown in the figure has a detector 200, a signal source 280 and a signal processing device 300, which is connected to an output device 400.

[0048] During inspection, detector 200 is positioned opposite rope 100, and thus opposite CFRP 110. An alternating magnetic field is applied to CFRP 110, and the intensity of the eddy currents generated by the alternating magnetic field is detected. The detection result is output as an eddy current signal A. Eddy current detection is performed sequentially at multiple different locations along the L direction of rope 100, and thus sequentially at multiple different locations along the L direction of CFRP 110.

[0049] Hereinafter, it is assumed that the detector 200 is set to face the main surface 101 of the rope 100.

[0050] For example, Figure 3 As shown in (a), detector 200 includes a plurality of detection elements 210. The plurality of detection elements 210 are configured to form a one-dimensional array 202, which, during inspection, ... Figure 3 As shown in (a), the length direction of the one-dimensional array 202, i.e. the arrangement direction of the detection elements 210, is consistent with the W direction.

[0051] As a result, multiple detection elements 210 are configured to be opposite to multiple positions that are different from each other in the W direction of CFRP110, detect the eddy current at each opposite position and output the eddy current signal A.

[0052] Since the length direction of the one-dimensional array 202 is aligned with the W direction as described above, the eddy currents of the linear portion extending along the W direction of the CFRP 110 are simultaneously detected by the one-dimensional array 202. This linear portion is referred to as the measurement line or simply the line, and the portion through which the eddy current is detected by each detection element 210 is referred to as the detection point.

[0053] By moving the detector 200, which is composed of a one-dimensional array 202 of detection elements, relative to the rope 100 along the L direction, the portion of the CFRP 110 opposite to the detector 200 changes, and eddy currents with respect to multiple measurement lines are detected sequentially. The multiple measurement lines are located at several different positions along the L direction of the CFRP 110. It is desirable that the spacing between the measurement lines is fixed.

[0054] The eddy current signal A, indicating the detected eddy current, is input to the signal processing device 300.

[0055] When moving detector 200 relative to rope 100, either detector 200 or rope 100 can be moved. In elevator operation, when performing inspections, rope 100 moves with the car; therefore, this movement can also be used for inspection. That is, detector 200 can be fixed, and the portion of rope 100 opposite to detector 200 can change as rope 100 moves. In the following description, "moving detector 200 relative to rope 100" refers to relative movement.

[0056] For example, such as Figure 4 As shown, the detection element 210 includes an excitation coil 220 and a detector 230. The excitation coil 220 generates an alternating magnetic field by flowing an excitation current according to a high-frequency excitation signal H supplied from the signal source 280, and applies this magnetic field to the CFRP 110. In the portion of the CFRP 110 opposite the excitation coil 220 (the detection point), eddy currents flow through the alternating magnetic field, forming a secondary magnetic field. The detector 230 detects the eddy currents by detecting this secondary magnetic field.

[0057] The frequency of the excitation signal H is determined by the relative speed between the rope 100 and the detector 200 and the size of the defect to be detected.

[0058] As an example, suppose the rope inspection device 1 is used in elevator operation and is required to detect a breakage of 0.1 mm. In this case, for a medium-speed elevator moving at 50 m / min, the frequency of the excitation signal H needs to be above 10 kHz, and for a high-speed elevator moving at 1000 m / min, the frequency of the excitation signal H needs to be above 200 kHz.

[0059] As detector 230, it uses components such as detection coil, Hall sensor, magnetoresistive element, fluxgate magnetometer, and superconducting quantum interference device (SQUID) to convert the magnitude of alternating magnetic field into an electrical signal.

[0060] The detection element 210 is mounted on a printed circuit board, for example. If mounted on a printed circuit board, a large number of detection elements can be arranged in a high density.

[0061] The signal processing device 300 detects defects in the CFRP110 based on the eddy currents detected by the detector 200. When a defect is determined to exist, it outputs defect information D indicating the defect determination result.

[0062] The output device 400 will notify the user or external device of the defect information D output from the signal processing device 300.

[0063] The output device 400 can be a device that displays the inspection results to a user, such as a monitor, a flashing display device, or a communication terminal with a display section, or it can be a device that transmits the inspection results to an external device. Here, "user" refers to the person performing the inspection or the person using the inspection results. "External device" refers to other devices not shown in the diagram.

[0064] For example, such as Figure 2 As shown, the signal processing device 300 includes an eddy current signal evaluation unit 302, an image processing unit 304, and a defect determination unit 306.

[0065] The eddy current signal A output from the detector 200 is input to the eddy current signal evaluation unit 302.

[0066] When performing eddy current testing on different measurement lines of the CFRP110, input eddy current signal A.

[0067] Examples of measuring lines are provided by Figure 5 The labels L(j) and L(j+1) of (a) are used to represent it.

[0068] These measuring lines are located at different positions in the L direction, extending from one edge of the CFRP110 in the W direction to the other edge.

[0069] Eddy currents are simultaneously detected at multiple locations (detection points) in the W direction on each measurement line by multiple detection elements 210 constituting a one-dimensional array 202.

[0070] The eddy current signal evaluation unit 302 acquires the eddy current signal A, processes it, and generates and outputs a detection signal B. For example, the eddy current signal evaluation unit 302 can perform amplification, level shifting, filtering, etc., on the eddy current signal A to generate the detection signal B. Furthermore, the eddy current signal evaluation unit 302 can also have a pre-defined correction value for the eddy current signal A (the output of the detector 200). When generating the detection signal B based on the eddy current signal A, the correction value is used to correct the intensity. The eddy current signal A is obtained as a result of detection at the end of the CFRP 110 in the W direction. This prevents false detection of defects caused by deviations in the intensity of the eddy current signal A in the W direction due to the edge effect of the eddy current.

[0071] Figure 5 (b) and (c) show examples of the detection signal B output from the eddy current signal evaluation unit 302. Figure 5 (b) and (c) show the detection signal B generated based on the eddy current signal A obtained by detection on the measurement lines L(j) and L(j+1), respectively.

[0072] As described above, the detection signal B is generated by applying correction and other processing to the eddy current signal A, but it can be said to represent the intensity of the eddy current in the same way as the eddy current signal A.

[0073] The image processing unit 304 acquires the detection signal B output from the eddy current signal evaluation unit 302 and stores it internally to generate an image C.

[0074] Image C represents the state of the carbon fibers in CFRP110, and is therefore referred to as the fiber image in this specification.

[0075] For example, the image processing unit 304 accumulates detection signals B of different predetermined numbers of Nf lines in the L direction, and generates a fiber image C of one frame using the accumulated detection signals B of Nf lines.

[0076] Since a fiber image C of one frame is generated using the detection signal B of the Nf line, the detection signal B of the Nf line is called the detection signal B of one frame.

[0077] exist Figure 5In (a), the numeral Lh denotes the L-direction dimension of adjacent regions formed by Nf lines. In this specification, the adjacent regions formed by Nf lines are referred to as unit regions. A unit region is the area where eddy current detection is performed to generate a detection signal B for one frame and a fiber image C for one frame. One end of the unit region is located at the position where the eddy current detection for generating the detection signal B for one frame and the fiber image C for one frame begins.

[0078] Each measurement line contains multiple detection points at different locations in the W direction. Therefore, a unit region formed by the Nf line contains multiple detection points arranged in the W and L directions, that is, multiple detection points arranged in a matrix.

[0079] The fiber image C is generated repeatedly based on the detection signal B of the Nf line.

[0080] exist Figure 5 In the example shown in (a), adjacent unit regions (whose L-direction dimensions are indicated by Lh) do not overlap with each other and there are no gaps between them.

[0081] In this case, from the start of eddy current detection for generating one frame of fiber image C to the start of eddy current detection for generating the next frame of fiber image C, the distance that detector 200 moves relative to rope 100 is equal to the L-direction dimension Lh of the unit region mentioned above.

[0082] If the moving speed of detector 200 relative to rope 100 is fixed and the period of eddy current detection (generation of eddy current signal A) is fixed, then the L-direction dimension Lh of the unit region mentioned above becomes fixed.

[0083] The image processing unit 304 maps the intensity of the detection signal B obtained from the eddy current signal A onto an imaginary two-dimensional plane, thereby generating a fiber image C representing the distribution of the intensity of the detection signal B, wherein the eddy current signal A is obtained by detection in each unit region.

[0084] As described above, the detection signal B represents the intensity of the eddy current; therefore, each fiber image C can be said to represent the distribution of the intensity of the eddy current obtained by detection within each unit region. However, more precisely, each fiber image C represents the intensity of the eddy current after corrections, etc., have been applied by the eddy current signal evaluation unit 302.

[0085] In the mapping, pixels are defined at positions in the two-dimensional plane corresponding to the aforementioned detection points. The detection points referred to here are the portions of the CFRP110 where eddy currents are detected by each detection element 210. The pixels defined at the positions corresponding to the aforementioned detection points in the fiber image C generated by the mapping represent the intensity of the eddy current detected at that detection point (the intensity of the detection signal B generated based on the eddy current signal obtained from the detection at that point). The intensity of the detection signal B is represented using the color of each pixel. The components of the color referred to here include brightness, hue, and chroma.

[0086] The intensity of eddy currents will be represented by the brightness, or density, of each pixel.

[0087] The resolution of the fiber image in the W direction is determined by the density of the configuration of the detection elements 210 of the one-dimensional array 202, and the number of pixels in the W direction of the fiber image is equal to the number of detection elements 210 of the one-dimensional array 202.

[0088] A column is formed by multiple pixels arranged along the W direction of the fiber image. Each column corresponds to a detection signal B generated from the detection results on each measurement line based on CFRP110.

[0089] The resolution of the fiber image in the L direction is determined by the interval between the measurement lines mentioned above, which is determined by the moving speed of the detector 200 relative to the rope 100 in the L direction and the period of eddy current detection (generation of eddy current signal A).

[0090] The defect determination unit 306 determines whether a defect exists in a given unit area based on the detection signal B generated from the detection results of eddy currents in each unit area and the fiber image C.

[0091] The defects mentioned here include fractures, peeling, and longitudinal cracks.

[0092] For example, the defect determination unit 306 performs threshold determination on the detection signal B and performs image recognition on the fiber image C, and determines whether there is a defect based on the results of the threshold determination and the image recognition.

[0093] In threshold determination, the detection signal B is compared with the threshold. If the detection signal B is greater than the threshold, it is determined that there is a defect.

[0094] In image recognition, for example, if a pattern similar to a predetermined pattern appears, it is determined to be defective.

[0095] Figure 5 (d) shows an example of the relationship between the position in the L direction and the detection signal B within a linear portion W(i) extending in the L direction at a certain position in the W direction of the CFRP110.

[0096] exist Figure 5 In the example shown in (d), at a certain location in the L direction, the detected signal B is larger than the threshold. This is estimated to be caused by a defect. This is because eddy currents tend to concentrate at fractured or peeled ends.

[0097] By performing a threshold determination for the detection signal B, the rise of the aforementioned detection signal B can be detected, thereby enabling the detection of defects.

[0098] In addition, although Figure 5 (d) shows the variation of the detection signal B along a linear portion extending in the L direction, but it is not necessary to generate it. Figure 5 The signal shown in (d). If for all measurement lines... Figure 5 The threshold determination of the detection signal B for each measurement line illustrated in (b) and (c) enables the determination of more than [a certain threshold]. Figure 5 The detection of the threshold portion shown in (d).

[0099] Figure 5 (e) shows an example of fiber image C.

[0100] If the carbon fiber has a uniform density and is free of defects, the detection signal B is fixed. However, in the actual manufacturing process of CFRP110 ropes, such as during the CFRP sheet fabrication, molding, and strip forming processes, the carbon fiber twists and becomes non-uniform in density. Therefore, the detection signal B sometimes varies depending on its position in the W direction.

[0101] For example, when carbon fibers are densely packed, a large number of eddy currents flow, thus increasing the detection signal B and brightening the corresponding portion of the fiber image C. On the other hand, when carbon fiber density is low or longitudinal cracks are present, eddy currents are difficult to flow, thus decreasing the detection signal B and darkening the corresponding portion of the fiber image C.

[0102] As described above, fiber image C reflects the density and longitudinal cracks of the carbon fibers in each section.

[0103] For example, if there are dark stripe-like portions extending along the L direction in fiber image C, it can be estimated that there is a longitudinal crack.

[0104] The above shows one example of a situation where a defect is determined. However, if a pattern similar to the predetermined pattern appears in other cases besides the example above, it can also be determined that a defect exists.

[0105] Regarding the threshold determination for the detection signal B of one frame and the image recognition for the fiber image C, if a defect is found midway through processing, the process can be terminated at this point. That is, if a defect is found midway through the processing of one frame, the subsequent parts can be left unprocessed.

[0106] The defect determination unit 306 notifies the output device 400 of the determination result as defect information D. Upon receiving the notification, the output device 400 notifies the user or external device of the defect information D.

[0107] The above processes are performed sequentially along the entire length of rope 100, resulting in an overall inspection of rope 100.

[0108] That is, as the detector 200 moves relative to the rope 100, it inspects multiple unit areas at multiple different locations at one end of the CFRP 110 in the L direction.

[0109] During the inspection of each unit area, a determination is made as to whether there is a defect in that area based on the detection signal B generated according to the eddy current detected in that unit area and the fiber image C.

[0110] Figure 6 The steps of the processing in the rope inspection device 1 of Embodiment 1 are shown.

[0111] Hereinafter, it is assumed that the detector 200 is fixed and the rope 100 is moved. As the rope 100 moves, eddy currents are detected at multiple different positions (the measuring line at that position) in the L direction of the CFRP 110.

[0112] For example, in the case of inspecting along the entire length of rope 100, the inspection begins with the detector 200 facing one end of rope 100. Figure 6 The processing. When only a portion of the total length of rope 100 is being inspected, the process begins with the detector 200 positioned opposite the beginning of the portion to be inspected. Figure 6 The processing.

[0113] First, in step ST101, an excitation signal H is supplied when the detector 200 is positioned opposite a certain position in the L direction of the rope 100 and thus opposite a certain position (measuring line) in the L direction of the CFRP 110, and an eddy current signal A of the quantity of one line is output. The eddy current signal evaluation unit 302 generates a detection signal B of the quantity of one line based on the eddy current signal A of the quantity of one line.

[0114] The image processing unit 304 acquires the generated detection signal B and stores it internally.

[0115] In step ST102, it is determined whether a detection signal B of 1 frame has been obtained. If the result in step ST102 is "No", then proceed to step ST104; if the result in step ST102 is "Yes", then proceed to step ST111.

[0116] In step ST111, the image processing unit 304 generates a fiber image C of one frame based on the accumulated detection signal B of one frame.

[0117] In step ST112, the defect determination unit 306 determines whether there is a defect based on the threshold determination of the detection signal B and the image recognition of the fiber image C.

[0118] In step ST113, based on the determination of whether there is a defect, if there is no defect, proceed to step ST103; if there is a defect, proceed to step ST114.

[0119] In step ST114, the defect determination unit 306 notifies the output device 400 of the defect information D.

[0120] After step ST114, proceed to step ST103.

[0121] In step ST103, a determination is made as to whether processing should continue. If processing should continue, the process proceeds to step ST104.

[0122] In step ST104, wait for detector 200 to move by one line relative to rope 100, and then return to step ST101.

[0123] In step ST103, if processing should not continue, the process ends. For example, if an inspection is performed along the entire length of rope 100 and processing begins at one end of rope 100, the process ends if the other end of rope 100 is reached.

[0124] Furthermore, if only a portion of the total length of rope 100 is inspected, the process ends when the end of the portion to be inspected is reached.

[0125] Alternatively, steps ST111 to ST114 can be processed through another routine. That is, if "yes" is set in step ST102, the image processing unit 304 and the defect determination unit 306 can perform steps ST111 to ST114, and in parallel, the detector 200 and the eddy current signal evaluation unit 302 can proceed to step ST104.

[0126] As described above, the detector 200 outputs an eddy current signal A, and the signal processing device 300 performs threshold determination for the detection signal B and image recognition for the fiber image C, thereby enabling the detection of defects such as breakage, peeling, and longitudinal cracks.

[0127] In the example above, it is assumed that adjacent unit regions (ranges of length Lh) do not overlap and there are no gaps between them. Alternatively, the unit regions could partially overlap.

[0128] In this case, from the start of eddy current detection for generating one frame of fiber image C to the start of eddy current detection for generating the next frame of fiber image C, the distance that detector 200 moves relative to rope 100 should be shorter than the aforementioned length Lh.

[0129] In summary, based on the detection results of eddy currents in multiple unit regions at multiple different locations along the L direction of the CFRP110, a detection signal B representing the intensity of the eddy currents and a fiber image C representing the distribution of the intensity of the eddy currents are generated. Based on the threshold determination result of the generated detection signal B and the image recognition result of the fiber image C, it is determined whether there is a defect in the unit region.

[0130] Variation Example 1.

[0131] exist Figure 2 In the rope inspection device shown, a common signal source 280 is provided for multiple detection elements 210. Alternatively, each detection element 210 may have its own signal source. For example, each detection element 210 may be configured as follows: Figure 7 It is constructed as shown.

[0132] Figure 7 The detection element 210 shown includes an excitation coil 220, a detector 230, and a signal source 240. A high-frequency current supplied from the signal source 240 flows through the excitation coil 220, generating an alternating magnetic field. Eddy currents are induced by this alternating magnetic field, forming a secondary magnetic field. The detector 230 detects the eddy currents by detecting this secondary magnetic field.

[0133] In the detection element 210 respectively as follows Figure 7 In the case of the configuration shown, it is not necessary Figure 2 The shared signal source 280 is shown.

[0134] Variation Example 2.

[0135] In the example above, image recognition is performed on the fiber image C obtained by mapping the intensity of the detection signal B, thereby determining whether there is a defect.

[0136] Instead, image recognition can also be performed on the following image, which is obtained by performing FFT (High-Speed ​​Fourier Transform) on the fiber image C, filtering to remove high-frequency components, and then performing IFFT (Fast Fourier Transform).

[0137] Such processing can remove noise from the fiber image C, making it easier to observe signal changes caused by the defects to be detected. Furthermore, it also reduces the amount of data.

[0138] The above aspects will be explained in more detail below, based on specific image examples. Figure 8 (a) to (d) show the case where no defects are found in the CRFP110 (the part to be inspected).

[0139] Figure 8 (a) shows an example of a fiber image C obtained by mapping the intensity of the detection signal B (denoted by label C-1). Figure 8 (a) and the following Figure 8 (d) and Figure 8 In (e), the bright areas of the image are those where a large amount of eddy current is flowing. Figure 8 In (a), stripes extending along the L direction are observed, indicating that an image corresponding to carbon fibers has been obtained. However, the overall contrast of the image is low, and the state of the carbon fibers is unclear.

[0140] Figure 8 (b) is through the Figure 8 The image obtained by performing FFT on image (a). Figure 8 In (b), "horizontal frequency" represents the spatial frequency in the L direction, and "vertical frequency" represents the spatial frequency in the W direction. Figure 8 In (b), the center of the longitudinal and transverse directions is the position where the spatial frequency is zero.

[0141] exist Figure 8 In image (b), brightly colored stripes extend longitudinally. The portion of these stripes that is separated from the center of the image corresponds to high-frequency noise components.

[0142] Figure 8 (c) is through the Figure 8 The image obtained by performing BPF (bandpass filtering) on ​​the image in (b). By performing BPF, unwanted noise components (high-frequency components) are removed.

[0143] Figure 8 (d) is through the Figure 8 Image C-1b is obtained by performing IFFT on image (c). Figure 8 Image C-1b of (d) and Figure 8 Compared to image (a), the contrast is higher, allowing for clearer identification of the stripes corresponding to the carbon fibers.

[0144] In this way, by using the image processing unit 304 to perform FFT, BPF and IFFT, noise removal, sharpening of the fiber image C and reduction of data volume can be achieved.

[0145] exist Figure 8 In (e), as an example, fiber image C-2 is shown, which is obtained by generating a detection signal B, generating a fiber image C, and performing signal processing (FFT, BPF, and IFFT processing) on ​​a CFRP110 with longitudinal cracks confirmed after fatigue testing.

[0146] exist Figure 8 In (e), a black band extending laterally near the center is identified. This corresponds to the portion where eddy currents are not flowing, i.e., a longitudinal crack.

[0147] exist Figure 8 Two black dots are visible in the upper center of (e). These are believed to be internal defects that occurred during fatigue testing. Furthermore, these defects could not be visually confirmed. Thus, by utilizing the results of image quality improvement, defects that cannot be visually confirmed can be detected.

[0148] In addition, the images obtained by performing FFT, BPF and IFFT as described above can also be described as fiber images representing the distribution of eddy current intensity in each unit region.

[0149] Variation Example 3.

[0150] exist Figure 2 In the rope inspection device 1, the following is used Figure 3 The detector 200 shown in (a) can also replace... Figure 3 The detector 200 shown in (a) is used Figure 3 The detector 200b is shown in (b). Figure 3 (b) shows the detector 200b together with the rope 100, which is the object of inspection.

[0151] Figure 3 The detector 200b shown in (b) has a single detection element 210 and an actuator 215.

[0152] When detector 200b is set to be opposite rope 100 and thus opposite CFRP 110, detection element 210 is set to be opposite CFRP 110, detects eddy current at the opposite position of CFRP 110 (the detection point at that position) and outputs eddy current signal A.

[0153] When the actuator 215 detects eddy currents, by moving the detection element 210 along the W direction, the detection element 210 can be sequentially positioned opposite the CFRP 110 at multiple different locations along the W direction.

[0154] While moving the detection element 210 along the W direction on each measurement line, the detector 200b also positions the detection element 210 opposite the CFRP 110 at multiple different positions in the W direction, and detects eddy currents at the positions opposite the CFRP 110, and outputs an eddy current signal A.

[0155] The eddy current signal evaluation unit 302 receives the eddy current signal A sequentially output from the detector 200b and outputs the detection signal B.

[0156] The image processing unit 304 accumulates the detection signals B of each measurement line sequentially output from the eddy current signal evaluation unit 302, and generates a fiber image C based on the detection signal B of the Nf line.

[0157] As described above, the process of detecting eddy currents at multiple detection points along each measurement line at different locations in the L direction is performed sequentially (i.e., while moving the detector 200b relative to the rope 100). Thus, the detection results of eddy currents at multiple different locations in the W and L directions (i.e., in the whole of each unit region) for constituting a fiber image C of one frame can be obtained.

[0158] Other than the above, the processing is the same as that described in Implementation 1.

[0159] Variation Example 4.

[0160] It can also replace the detectors 200 and 200b mentioned above. Figure 3 The detector 200c is shown in (c). Figure 3 (c) shows the detector 200c together with the rope 100, which is the object of inspection.

[0161] Figure 3 The detector 200c shown in (c) has a two-dimensional array 204 consisting of multiple detection elements 210.

[0162] The detection elements 210 constituting the two-dimensional array 204 are arranged at multiple different positions in mutually orthogonal directions, namely the U direction and the V direction, and are arranged in a manner that follows the U direction and the V direction.

[0163] During eddy current detection, detector 200c is configured such that the U direction aligns with the L direction and the V direction aligns with the W direction. That is, detector 200c is configured such that multiple detection elements 210 constituting the two-dimensional array 204 are positioned opposite the CFRP 110 at multiple different locations along the L and W directions. In this configuration, the detection elements 210 detect the eddy currents at their respective opposite locations on the CFRP 110 and output an eddy current signal A. In this case, a unit region is formed by the collection of detection points from the multiple detection elements 210 constituting the two-dimensional array 204 that detect eddy currents.

[0164] If using Figure 3 The detector 200c (c) can simultaneously detect eddy currents in the entire unit area and generate a 1-frame eddy current signal A. The eddy current signal evaluation unit 302 generates a 1-frame detection signal B based on the 1-frame eddy current signal A output from the detector 200c. The image processing unit 304 generates a fiber image C based on the 1-frame detection signal B output from the eddy current signal evaluation unit 302.

[0165] Each pixel of the fiber image C corresponds to a detection element 210. The color of each pixel corresponds to the intensity of the eddy current detected by the corresponding detection element 210 (i.e., the intensity of the detection signal B generated based on the eddy current signal A output from the corresponding detection element 210).

[0166] The resolution of the fiber image C in the W and L directions is determined by the spacing of the arrangement of the detection elements 210 in the V and U directions, respectively.

[0167] After the generation of the eddy current signal A used to generate a 1-frame fiber image C is completed, the detector 200c is moved 1 frame relative to the opposite position of the rope 100, and the eddy current detection and image generation are performed in the same manner as described above.

[0168] Here, a 1-frame movement refers to the length Lh of the range corresponding to 1 frame.

[0169] That is, such as Figure 9 As shown, within a certain unit area, after generating the eddy current signal A for generating a fiber image C(n) for one frame, the detector 200c is moved a length Lh relative to the rope 100 to generate the eddy current signal A for generating the fiber image C(n+1) for the next frame. The same process is repeated below.

[0170] Here, as referenced Figure 5As explained in (a), it is assumed that adjacent unit regions do not overlap and have no gaps between them. Alternatively, the unit regions may partially overlap. In such a case, the distance from the position where eddy current detection is performed to generate one frame of fiber image C to the position where eddy current detection is performed to generate the next frame of fiber image C can be shorter than the length Lh described above.

[0171] By using a two-dimensional array of detection elements, inspections can be performed at a much higher speed.

[0172] In addition, it can improve the accuracy of defect detection.

[0173] Variation Example 5.

[0174] Figure 3 The detector 200 shown in (a) and Figure 3 The detector 200c shown in (c) has multiple detection elements 210, each detection element 210 having an excitation coil 220 and a detector 230.

[0175] Alternatively, the detector 200 may be configured such that each of the multiple detection elements 210 constituting the detector 200 has one excitation coil 220, and a single detector 230 is shared between two or more excitation coils 220 arranged adjacent to or consecutively with each other. For example, the detector 200 may be configured such that the multiple detection elements 210 constituting the detector 200 are divided into multiple groups, each consisting of two or more detection elements arranged adjacent to or consecutively with each other. Each group of two or more detection elements 210 may have an excitation coil 220, and they may share a single detector 230. This structure may be used, for example, when it is not possible to make the size of the detector 230 (especially its size in the L and W directions when facing the rope 100) as small as that of the excitation coil 220.

[0176] In this configuration, two or more detection elements 210 belonging to each group operate in a time-sharing manner. That is, excitation current flows sequentially through the excitation coils 220 of the two or more detection elements 210 belonging to each group, and a shared detector 230 sequentially detects the eddy currents induced by the excitation current flowing through each excitation coil 220. The eddy current signal A and the detection signal B, obtained from the detection results of the shared detector 230 when the excitation current flows through each excitation coil 220, can be used as signals representing the intensity of the eddy current at each detection point. In this configuration, the number of detection elements 210 can be considered to be the same as the number of excitation coils, consisting of two or more excitation coils 220 and a shared detector 230.

[0177] Furthermore, in the above structure, eddy currents are detected at slightly different times among the two or more detection elements belonging to each group. Even so, eddy currents can be detected one at a time using each detection element 210 constituting the one-dimensional array 202 or the two-dimensional array 204, thereby obtaining the eddy current detection result for the entire one-dimensional array 202 or the two-dimensional array 204. That is, in the case of the one-dimensional array 202, eddy currents can be detected one at a time using each detection element 210 constituting the one-dimensional array 202, thereby obtaining the eddy current detection result for the entire measurement line opposite to the one-dimensional array 202; in the case of the two-dimensional array 204, eddy currents can be detected one at a time using each detection element 210 constituting the two-dimensional array 204, thereby obtaining the eddy current detection result for the entire unit area opposite to the two-dimensional array 204.

[0178] Implementation method 2.

[0179] Figure 10 The rope inspection device 1b of Embodiment 2 is shown. The overall structure of the rope inspection device 1b of Embodiment 2 is similar to... Figure 2 The rope inspection device 1 shown is the same. The rope inspection device 1b of Embodiment 2 replaces it. Figure 2 The signal processing device 300 includes a signal processing device 300b.

[0180] Signal processing device 300b and Figure 2 The signal processing device 300 shown is largely the same, but instead of the image processing unit 304 and the defect determination unit 306, an image processing unit 304b and a defect determination unit 306b are provided, and a reference image recording unit 312 and a position determination unit 314 are added.

[0181] Image processing unit 304b and defect determination unit 306b and Figure 2 The image processing unit 304 and the defect determination unit 306 are the same, but there are obvious differences according to the following description.

[0182] The rope inspection device 1b of Embodiment 2 operates in either the first mode, i.e., the image acquisition mode, or the second mode, i.e., the inspection execution mode.

[0183] In the first mode, the detector 200 detects eddy currents at multiple different locations in the L direction of the CFRP 110, and the signal processing device 300b generates a reference image E based on the eddy currents detected at the multiple different locations in the L direction of the CFRP 110.

[0184] For example, the detector 200 outputs eddy current signals A from multiple different unit regions located at different positions along the L direction of the CFRP 110. The eddy current signal evaluation unit 302 generates a detection signal B, and the image processing unit 304b generates a single-frame fiber image C. The multiple frames of fiber images C generated based on the detection results in the multiple unit regions are interconnected by the image recording unit 312 to generate a reference image E. The generated reference image E is recorded in the reference image recording unit 312.

[0185] Depend on Figure 10 The dashed lines with arrows indicate the signal flow and image data flow during the above processing.

[0186] For example, the above processing is performed over the entire length of the rope 100 to generate a continuous reference image E over the entire length of the rope 100.

[0187] Information Fa, representing the position of each part, is recorded together with reference image E in reference image recording unit 312. This information Fa indicates the position where eddy current detection was performed (the position where the eddy current signal A used to generate the fiber image C used in generating the aforementioned parts of reference image E was obtained). Here, "position" refers to the position of the rope 100 in the L direction, and more specifically, the position of the CFRP 110 in the L direction. For example, by moving the detector 200 relative to the rope 100 and sequentially acquiring eddy current signals A from one end of the rope 100 or one end of a specific range of the rope (one end of a fixed length portion), the information Fa representing the aforementioned position can be obtained by measuring the moving distance.

[0188] In the second mode, i.e., the inspection execution mode, multiple unit regions at different locations along the L direction of the CFRP110, each with one end, are inspected sequentially. That is, similar to Embodiment 1, the detector 200 detects eddy currents in each unit region and outputs an eddy current signal A, while the signal processing device 300c performs defect detection based on the eddy current signal A.

[0189] Depend on Figure 10 The solid line with arrows shows the signal flow and image data flow during processing in mode 2.

[0190] In the inspection of each unit area, similar to Embodiment 1, the detector 200 outputs an eddy current signal A based on the detection result in that area, the eddy current signal evaluation unit 302 acquires the eddy current signal A and generates a detection signal B, the image processing unit 304b acquires the detection signal B and generates a fiber image C, and the defect determination unit 306b determines whether there is a defect.

[0191] The defect determination unit 306b determines whether a defect exists in a given unit area based on the detection signal B generated from the detection results in each unit area and the fiber image C.

[0192] When the defect determination unit 306b determines that a defect exists, it notifies the location determination unit 314 of the defect information D.

[0193] The position determination unit 314 acquires and outputs defect position information Fb, which indicates the location of the defect. Here, "position" also refers to the position of the rope 100 in the L direction, and thus the position of the CFRP 110 in the L direction. The defect position information Fb is acquired as follows.

[0194] That is, the location determination unit 314 compares the fiber image C at the defect location with the reference image E recorded in the reference image recording unit 312, and determines the portion in the reference image E that matches the fiber image C at the defect location. For example, if the similarity is above a threshold, it is determined to be consistent.

[0195] Here, the fiber image C at the defect location refers to the fiber image C used when a defect is determined to exist, or the fiber image C corresponding to the detection signal B used when a defect is determined to exist.

[0196] The fiber image C corresponding to the detection signal B used when a defect is determined to exist refers to the fiber image C generated based on the detection signal B, and thus refers to the fiber image C generated based on the detection results within the same unit area as the detection signal B.

[0197] Figure 11 (a) shows an example E-1 of a portion of a reference image E recorded in the reference image recording unit 312. Figure 11 (b) shows an example of a 1-frame fiber image C, C-3.

[0198] The number of pixels in the W direction of the reference image is equal to the number of detection elements 210 in the one-dimensional array 202. The number of pixels in the L direction of the reference image is equal to the number of measurement lines along the entire length of the rope 100, and consequently, the entire length of the CFRP 110. A column is formed by multiple pixels arranged along the W direction of the reference image. Each column corresponds to the intensity of the eddy current detected on each measurement line of the CFRP 110 (the detection signal B generated by detection on each measurement line).

[0199] The position determination unit 314 extracts the portion of the same size as the fiber image C from the reference image E (in... Figure 11(An example is shown in the dashed box Ee in (a)). A comparison is made. While shifting the positions of the extracted portions, a search is conducted for portions that match the fiber image C. For example, while shifting the portions column by column, a search is conducted for matching portions. Whether they match is determined based on the similarity between pixels at corresponding positions, the similarity of the shape of the stripes corresponding to the carbon fibers, the similarity of contrast, etc.

[0200] The location determination unit 314 obtains information Fa indicating the location of the part that is determined to be consistent as defect location information Fb, and transmits the defect location information Fb to the defect judgment unit 306b.

[0201] The defect determination unit 306b notifies the output device 400 of the defect information D and the defect location information Fb. Upon receiving the notification, the output device 400 notifies the user or external device of the defect information D and the defect location information Fb.

[0202] Figure 12 and Figure 13 The steps of the processing in the rope inspection device 1b of Embodiment 2 are shown. Hereinafter, the processing in the first mode, i.e., acquiring the reference image, will be performed first, followed by the processing in the second mode, i.e., inspection.

[0203] Figure 12 The processing in mode 1 is shown.

[0204] Here, it is assumed that the reference image E is generated continuously along the entire length of the rope 100. In this case, with the detector 200 facing one end of the rope 100, the process begins... Figure 12 The processing.

[0205] exist Figure 12 In the process, steps ST101, ST102, ST111, ST103, and ST104 are processed and Figure 6 The steps ST101, ST102, ST111, ST103 and ST104 are the same.

[0206] In step ST211, the image processing unit 304b supplies one frame of fiber image C to the reference image recording unit 312. The supplied fiber image C is recorded as a part of the reference image E.

[0207] When the initial frame of fiber image C is supplied to the reference image recording unit 312, the reference image E is formed using only the supplied fiber image C. When a new fiber image C is supplied after the reference image E has already been formed, the newly supplied fiber image C is appended to the end of the already formed reference image E, thus extending the reference image E to a longer length.

[0208] After step ST211, proceed to step ST103.

[0209] In step ST103, it is determined whether processing should continue. If processing should continue, proceed to step ST104; otherwise, the processing ends.

[0210] Here, assuming the reference image E is generated along the entire length of the rope 100, the decision to continue processing is made by checking whether the detector 200 is in a state opposite to the other end of the rope 100.

[0211] Figure 13 The processing in mode 2 is shown.

[0212] exist Figure 13 In the process, steps ST101, ST102, ST111, ST112, ST113, ST103, and ST104 are... Figure 6 The steps ST101, ST102, ST111, ST112, ST113, ST103 and ST104 are the same.

[0213] In step ST121, the position determination unit 314 determines the portion in the reference image E that corresponds to the fiber image C at the defect location.

[0214] In step ST122, the position determination unit 314 obtains the position information Fa of the consistent part as the defect position information Fb, and notifies the defect judgment unit 306b of the defect position information Fb.

[0215] After step ST122, proceed to step ST114b.

[0216] In step ST114b, the defect determination unit 306b notifies the output device 400 of the defect information D and the defect location information Fb.

[0217] As described above, in Embodiment 2, a reference image recording unit 312 and a position determination unit 314 are included. In the first mode, a reference image E is generated and stored, and in the second mode, a portion of the reference image E that matches the generated fiber image C is determined. Therefore, defect position information Fb can be obtained. Furthermore, defect position information Fb that accurately represents the defect position can be obtained.

[0218] Since the location of the defect can be determined, there is no need to pre-mark the rope 100 or use a separate location detector.

[0219] In the example above, the image processing unit 304b generates one frame of fiber image C, and each time one frame of fiber image C is generated, it is supplied to the reference image recording unit 312 and linked with the reference image E.

[0220] Alternatively, each time the eddy current signal evaluation unit 302 generates a detection signal B for one line, that detection signal B is supplied to the reference image recording unit 312, accumulating the detection signals B of each line supplied sequentially, thereby generating a reference image E. That is, in the first mode, instead of using the image processing unit 304b to generate a single frame of fiber image C, the detection signals B of each line generated by the eddy current signal evaluation unit 302 are supplied to the reference image recording unit 312, and the reference image E is generated using the detection signals B of multiple lines supplied sequentially. In this case, when the initial detection signal B for one line is supplied, the reference image E is generated using only the supplied detection signal B; when a new detection signal B is supplied after the reference image E has already been formed, the newly supplied detection signal B is connected to the end of the already formed reference image E.

[0221] The threshold used by the defect determination unit 306b can also be adjusted based on the magnitude of the detection signal B used to generate the reference image E. For example, the larger the detection signal B used to generate the reference image E, the larger the threshold can be. In this way, even if the distance (gap) between the rope 100 and the detector 200, and thus between the CFRP 110 and the detector 200, changes, the accuracy of defect detection can be maintained.

[0222] Implementation method 3.

[0223] Figure 14 The rope inspection device 1c of embodiment 3 is shown. Figure 14 The overall structure of the rope inspection device 1c shown is similar to Figure 2 The rope inspection device 1 shown is the same. However, instead of Figure 2 The detector 200c, signal source 280c and signal processing device 300c are provided.

[0224] Signal processing device 300c has the same characteristics as Figure 2 The signal processing device 300 has the same structure. However, instead of the eddy current signal evaluation unit 302, the image processing unit 304, and the defect determination unit 306, it includes an eddy current signal evaluation unit 302c, an image processing unit 304c, and a defect determination unit 306c.

[0225] Eddy current signal evaluation unit 302c, image processing unit 304c, and defect determination unit 306c and Figure 2 The eddy current signal evaluation unit 302, image processing unit 304 and defect determination unit 306 are the same, but there are obvious differences according to the following description.

[0226] As detector 200c, using reference Figure 3 The detector described in (c) is a detector comprising a two-dimensional array 204 with detection elements. Therefore, the detector 200c is capable of simultaneously detecting eddy currents in the entirety of each unit region, and can obtain an eddy current signal A representing the intensity of the detected eddy current in one frame from the detector 200c.

[0227] Even when the detector 200c is the structure described in Variation 5 of Embodiment 1, eddy current detection can be performed in the whole of each unit area by detecting eddy current one by one by the multiple detection elements 210 constituting the two-dimensional array 204, and an eddy current signal A representing the intensity of the detected eddy current can be obtained from the detector 200c in one frame.

[0228] Signal source 280c and Figure 2 Signal source 280 also outputs a high-frequency excitation signal. However, signal source 280c differs in that it can change the frequency of the output excitation signal.

[0229] As a signal source 280c, it can use a frequency-variable current source or voltage source, or a function generator.

[0230] For example, signal source 280c sequentially selects and outputs multiple excitation signals H1 to HM with different frequencies, such as the first frequency f1 to the Mth (M is an integer greater than 2) frequency fM. Hereinafter, let M be 4.

[0231] When an excitation signal Hm with frequency fm (m is 1, 2, 3 or 4) is supplied from signal source 280c, detector 200c applies an AC magnetic field of frequency fm to CFRP110 and detects the eddy current when the AC magnetic field is applied.

[0232] The signal source 280c can supply an excitation signal Hm of a selected frequency from multiple frequencies. Therefore, the detector 200c can be said to be able to apply an alternating magnetic field of a selected frequency from multiple frequencies.

[0233] The signal source 280c supplies the reference signal Im (1, 2, 3 or 4) of the selected frequency at each time point to the eddy current signal evaluation unit 302c.

[0234] For example, when the detector 200c is opposite to the CFRP110 in a unit area, the signal source 280c sequentially selects frequencies f1 to f4 and outputs an excitation signal Hm at the selected frequency fm.

[0235] During the period of outputting excitation signals Hm at various frequencies fm, eddy currents are detected in the unit area opposite to detector 200c, and eddy current signals Am of 1 frame are output.

[0236] The signal processing device 300c generates a detection signal Bm and a fiber image Cm based on a one-frame eddy current signal Am output from the detector 200c.

[0237] When the above series of processes ends within the selected frequency period, the next frequency is selected, and the same series of processes are performed.

[0238] After completing the above series of processes for all frequencies, the detector 200c is moved by 1 frame relative to the cable 100.

[0239] The operation of the signal processing device 300c when a frequency fm is selected will be explained in more detail below. An excitation signal Hm at the selected frequency fm is supplied from the signal source 280c to the detector 200c, and a reference signal Im at that frequency fm is sent to the eddy current signal evaluation unit 302c. A one-frame eddy current signal Am output from the detector 200c is sent to the eddy current signal evaluation unit 302c.

[0240] The eddy current signal evaluation unit 302c receives the eddy current signal Am, extracts the frequency components that are consistent with the reference signal Im, and generates a detection signal Bm based on the extracted frequency components. The detection signal Bm generated by the eddy current signal evaluation unit 302c is sent to the image processing unit 304c and the defect determination unit 306c.

[0241] The image processing unit 304c generates a fiber image Cm based on a detection signal Bm of 1 frame, and sends the fiber image Cm to the defect determination unit 306c.

[0242] The above processing is performed on frequencies f1 to f4 respectively. As a result, the image processing unit 304c generates images C1 to C4 corresponding to frequencies f1 to f4 respectively.

[0243] The defect determination unit 306c performs threshold determination for detection signals B1 to B4 and image recognition for fiber images C1 to C4. Based on the results of the threshold determination and image recognition, it determines whether a defect exists. For example, if any of the detection signals B1 to B4 is greater than the threshold, a defect can also be determined. Furthermore, if a pattern similar to a predetermined pattern is found in the image recognition of any of the fiber images C1 to C4, a defect can also be determined.

[0244] If a defect is found during the threshold determination and image recognition for each unit area, the processing can be terminated at that point.

[0245] For example, if threshold determination and image recognition are performed sequentially for multiple frequencies, and the result of threshold determination or image recognition for any frequency is that a defect exists, then processing for that unit region can be terminated at that time point, without processing other frequencies.

[0246] The defect determination unit 306c notifies the output device 400 of the determination result of whether a defect exists, as defect information D. Upon receiving this notification, the output device 400 notifies the user or external equipment of the defect information D. The significance of changing the frequency of the excitation signal as described above will be explained below.

[0247] When an alternating current flows in a conductor, the range of current flow is limited to a region closer to the surface as the frequency of the alternating current increases. This is due to the skin effect. The depth Ds of the range in which most of the current flows is called the skin depth.

[0248] Figure 15 (a) shows the skin depth Ds in CFRP110. The skin depth Ds is represented by equation (1).

[0249] In equation (1), f is the frequency of the current, μ r σ is the relative permeability of CFRP110, μ0 is the permeability of free space, and σ is the electrical conductivity of CFRP110.

[0250] Figure 15 (b) is μ r =1, μ0 = 1.257 × 10 -6 The relationship between skin depth Ds and frequency f is calculated by setting the conductivity σ in the T and W directions of CFRP110 to be the same as 100 S / m.

[0251] A smaller skin depth Ds results in a narrower depth range that can be inspected using eddy currents, but at the same time, more precise inspection results can be obtained within that range. Furthermore, the thinner the object being inspected, the higher the frequency required.

[0252] For example, with a CFRP110 thickness of 1 mm, to define the range of eddy current flow (skin depth) from one surface to 0.5 mm and from another surface to 0.5 mm, the excitation signal frequency can be set to 10 GHz. When obtaining more precise inspection results limited to the surface area, an excitation signal with a higher frequency than 10 GHz can be used.

[0253] Figure 15 (c) is a graph showing the relationship between the frequency of the excitation signal (excitation frequency) and the fiber image C.

[0254] When the excitation frequency f is low, eddy currents flow throughout the entire range in the T direction, thus obtaining a fiber image C that reflects the state of CFRP110 up to the deeper parts.

[0255] On the other hand, the higher the excitation frequency f, the more accurately an image reflecting only the shallower portion of the CFRP110 is obtained. In this image, the contrast is higher for the shallower portions compared to the case with a lower excitation frequency f. Conversely, while a lower excitation frequency f yields an image reflecting the state up to the deeper portions, the contrast is low, and even if defects exist in the shallower portions, they may sometimes be undetectable due to the influence of the deeper portions.

[0256] The higher the excitation frequency f, the shallower the area being evaluated is limited to, i.e., closer to the surface; conversely, the lower the excitation frequency f, the deeper the area being evaluated expands. Figure 15 In (c), “surface portion” means the part close to the surface, and “whole” means including a wider range.

[0257] As described above, when a defect exists at a specific depth from the surface 111 of CFRP110, the overlap of signals at other depths can be prevented by adjusting the excitation frequency f, thus improving the accuracy of defect detection. Specifically, by using excitation signals of different frequencies to detect eddy currents through detector 200c, fiber images Cm of CFRP110 corresponding to different depths can be generated, enabling precise determination of the presence or absence of defects regardless of their depth.

[0258] Figure 16 The steps of the processing in the rope inspection device 1c of Embodiment 3 are shown.

[0259] For example, when inspecting along the entire length of rope 100, with detector 200c facing one end of rope 100, the inspection begins... Figure 16 The processing. When only a portion of the total length of rope 100 is being inspected, the process begins with detector 200c positioned opposite the beginning of the portion to be inspected. Figure 16 The processing.

[0260] In step ST301, multiple frequencies are set, namely the first frequency to the Mth frequency f1, f2, ..., fM. Here, it is assumed that the relationship is f1 < f2 < ..., fM. Furthermore, it is assumed that M is 4.

[0261] In step ST302, signal source 280c selects one frequency fm from a plurality of frequencies. For example, the frequencies are selected in ascending order. In this case, the first frequency f1 is selected first.

[0262] The signal source 280c supplies the excitation signal Hm of the selected frequency fm to the detector 200c, and outputs the reference signal Im of the selected frequency fm.

[0263] In step ST303, an excitation signal Hm at a selected frequency fm is used within a unit area to acquire a detection signal Bm of 1 frame and generate a fiber image Cm.

[0264] That is, the detector 200c outputs a 1-frame eddy current signal Am, the eddy current signal evaluation unit 302c acquires the 1-frame eddy current signal Am and generates a 1-frame detection signal Bm, and the image processing unit 304c acquires the 1-frame detection signal Bm and generates a fiber image Cm.

[0265] In step ST304, the defect determination unit 306c performs threshold determination for the detection signal Bm and image recognition for the fiber image Cm, and determines whether there is a defect based on the results.

[0266] In step ST305, if the determination result in step ST304 is "no", then proceed to step ST306; if it is "yes", then proceed to step ST308.

[0267] In step ST306, a determination is made as to whether all frequencies have been selected. If the result is "No", the process returns to step ST302. If the result is "Yes", the process proceeds to step ST309.

[0268] If the result in step ST306 is "No" and the process returns to step ST302, then the next frequency is selected. The next frequency refers to the frequency following the frequency selected in the preceding step ST302.

[0269] In step ST308, the defect determination unit 306c notifies the output device 400 of the defect information D.

[0270] In step ST309, a decision is made as to whether to continue processing. If no further processing is required, the processing ends.

[0271] For example, if an inspection is performed on the entire length of rope 100, and processing begins at one end of rope 100, processing ends upon reaching the other end of rope 100. Furthermore, if only a portion of the total length of rope 100 is inspected, processing ends upon reaching the end of that inspected portion.

[0272] If processing should continue in step ST309, proceed to step ST310.

[0273] In step ST310, wait for detector 200c to move 1 frame relative to cable 100, then return to step ST302. If the process returns to step ST302 via step ST310, select the initial frequency f1.

[0274] The same effect as in Embodiment 1 is obtained in Embodiment 3. In addition, Embodiment 3 has the following additional effects.

[0275] That is, by setting different frequencies in the signal source 280c and switching frequencies, more detailed information related to defects at different depths can be obtained, thereby improving the accuracy of defect detection.

[0276] In implementation method 3, using Figure 3 The detector 200c shown in (c) is a two-dimensional array 204 with detection elements. Alternatively, a detector 200c can also be used. Figure 3 The detector 200 shown in (a) with a one-dimensional array 202 of detection elements can also be used Figure 3 The detector 200b shown in (b) has a single detection element and actuator.

[0277] In use Figure 3 In the case of detector 200 shown in (a), the frequency can also be selected sequentially during the period when the one-dimensional array 202 is opposite to a measurement line.

[0278] The actions in this situation are as follows.

[0279] That is, during the period when the one-dimensional array 202 is opposite to a measurement line, multiple frequencies f1 to fM are selected sequentially, thereby generating eddy current signals A1 to AM and detection signals B1 to BM for a single line corresponding to the multiple frequencies f1 to fM respectively.

[0280] While changing the opposing position of the detector 200 relative to the rope 100 in the L direction, the above processing is repeated Nf times for each measurement line, namely, generating eddy current signals A1-AM and detection signals B1-BM. Thus, one frame of eddy current signals A1-AM and detection signals B1-BM are generated sequentially. Based on the generated one frame of detection signals B1-BM, one frame of fiber images C1-CM are generated.

[0281] Alternatively, the following processing can be performed for frequencies from the 1st to the Mth: when a frequency fm is selected (m is any one of 1 to M), by changing the opposing position of the detector 200 relative to the rope 100 in the L direction, a 1-frame eddy current signal Am and a detection signal Bm for that frequency are generated, and a 1-frame fiber image Cm is generated based on the generated 1-frame detection signal Bm.

[0282] The actions in this situation are as follows.

[0283] That is, with one frequency fm selected, while changing the opposing position of detector 200 relative to rope 100 in the L direction, the generation of a linear eddy current signal Am and a detection signal Bm is repeated Nf times. This generates one frame of eddy current signal Am and detection signal Bm. A one-frame fiber image Cm is generated based on the generated one-frame detection signal Bm. Then, the opposing position of detector 200 relative to rope 100 in the L direction is returned to its original position. This process is repeated M times (thus obtaining M fiber images C1 to CM).

[0284] In use Figure 3 In the case of detector 200b shown in (b), the frequency can also be selected sequentially during the period when the detection element 210 is opposite to a detection point.

[0285] The actions in this situation are as follows.

[0286] That is, during the period when the detection element 210 is opposite to a detection point, multiple frequencies f1 to fM are selected sequentially, thereby generating eddy current signals A1 to AM and detection signals B1 to BM corresponding to one detection point, respectively, for the multiple frequencies f1 to fM.

[0287] While changing the opposing positions of the detection element 210 relative to the rope 100 in the W and L directions, the aforementioned processing is performed on all detection points constituting one frame, namely, the generation of eddy current signals A1-AM and detection signals B1-BM. Thus, one frame's worth of eddy current signals A1-AM and one frame's worth of detection signals B1-BM are generated. Then, one frame's worth of fiber images C1-CM are generated based on the generated one frame's worth of detection signals B1-BM.

[0288] Alternatively, the following processing can be performed for frequencies from the 1st to the Mth: when a frequency fm is selected, by changing the opposing position of the detection element 210 relative to the rope 100 in the W and L directions, a 1-frame eddy current signal Am and a detection signal Bm for that frequency are generated, and a 1-frame fiber image Cm is generated based on the generated 1-frame detection signal Bm.

[0289] The actions in this situation are as follows.

[0290] That is, with one frequency fm selected, the opposing positions (detection points) of the detection element 210 relative to the rope 100 in the W and L directions are changed, and the eddy current signal Am and detection signal Bm of each detection point constituting one frame are generated and obtained. Thus, one frame's worth of eddy current signal Am and one frame's worth of detection signal Bm are generated. One frame's worth of fiber image Cm is generated based on the generated one frame's worth of detection signal Bm. Then, the opposing positions of the detection element 210 relative to the rope 100 in the W and L directions are returned to their original positions. The above process is repeated M times (thus obtaining M fiber images C1 to CM).

[0291] As described above, it can also be used as a detector. Figure 3 Any of the detectors 200, 200b, and 200c shown in (a) to (c). In summary, the following structure is sufficient: Signal source 280c sequentially selects multiple different frequencies f1 to fM, and uses the selected frequency fm (m is any one of 1 to M) for inspection. During the inspection using each frequency fm, the excitation signal Hm of that frequency fm is supplied to detector 200c (or 200a or 200b). Detector 200c (or 200a or 200b) outputs eddy current signal Am by applying an AC magnetic field of that frequency fm to CFRP110. Signal processing device 300c generates detection signal Bm and fiber image Cm based on the eddy current signal Am output by detector 200c (or 200a or 200b). The generated detection signal Bm and fiber image Cm are used to determine whether there is a defect.

[0292] Implementation method 4.

[0293] Figure 17 The rope inspection device 1d according to Embodiment 4 is shown. The overall structure of the rope inspection device 1d according to Embodiment 4 is similar to... Figure 14 The rope inspection device 1c shown is the same. The rope inspection device 1d of Embodiment 4 includes a signal processing device 300d to replace... Figure 14 The signal processing device 300c.

[0294] Signal processing device 300d and Figure 14 The signal processing device 300c shown is generally the same, but instead of the image processing unit 304c and the defect determination unit 306c, it is provided with an image processing unit 304d and a defect determination unit 306d, and it is also provided with a reference image recording unit 312d and a position determination unit 314d.

[0295] Image processing unit 304d and defect determination unit 306d and Figure 14The image processing unit 304c and the defect determination unit 306c are the same, but there are obvious differences according to the following description.

[0296] Referring to the image recording unit 312d and the position determining unit 314d and Figure 10 The reference image recording unit 312 and the position determination unit 314 are the same, but there are obvious differences according to the following description.

[0297] As described in Embodiment 3, the signal source 280c can change the frequency of the excitation signal H. Hereinafter, excitation signals H1 to H4 with frequencies f1 to f4 are output sequentially.

[0298] The rope inspection device 1d of Embodiment 4 is similar to the rope inspection device 1b of Embodiment 2, and operates in either the first mode, i.e., the image acquisition mode, or the second mode, i.e., the inspection execution mode.

[0299] Depend on Figure 17 The dashed lines with arrows indicate the signal and data flow during processing in Mode 1.

[0300] In Mode 1, reference images E1 to E4 for multiple frequencies f1 to f4 are generated. For example, multiple frequencies f1 to f4 are selected sequentially, and reference images Em for each frequency fm are generated. The reference image Em for each frequency fm is generated based on the intensity of eddy currents detected at multiple different locations in the L direction of the CFRP110 when that frequency fm is selected.

[0301] For example, when the detector 200c is positioned opposite the CFRP110 within a unit area, the signal source 280c sequentially selects frequencies f1 to f4 and outputs an excitation signal Hm at the selected frequency fm.

[0302] During the period when the excitation signal Hm of each frequency fm is output, the detector 200c detects eddy currents in the opposite unit area and outputs a frame of eddy current signal Am. The eddy current signal evaluation unit 302c acquires the frame of eddy current signal Am and generates the detection signal Bm. The image processing unit 304d acquires the frame of detection signal Bm and generates the fiber image Cm.

[0303] When the above series of processes ends within the selected frequency period, the next frequency is selected, and the same series of processes are performed.

[0304] After completing the above series of processes for all frequencies, the detector 200c is moved by 1 frame relative to the cable 100.

[0305] By sequentially performing the above processing on multiple unit regions, multiple fiber images Cm of various frequencies fm for multiple frequencies f1 to f4 are generated. These multiple fiber images Cm are then interconnected in the reference image recording unit 312d to generate a reference image Em. The above processing is then performed on multiple frequencies f1 to f4 to generate reference images E1 to E4 for multiple frequencies f1 to f4. The generated reference images E1 to E4 are recorded in the reference image recording unit 312d.

[0306] For example, the above processing is performed over the entire length of the rope 100 to generate a continuous reference image Em over the entire length of the rope 100.

[0307] Information Fa, representing the position of each part, is recorded together with the reference image Em in the reference image recording unit 312d. Information Fa indicates the position where eddy current detection was performed to generate the fiber image C used in generating the aforementioned parts of each reference image Em (the position of the eddy current signal Am used to generate the fiber image Cm was obtained).

[0308] Depend on Figure 17 The solid line with arrows shows the signal and data flow during processing in mode 2.

[0309] In mode 2, each of the multiple frequencies was used for the check.

[0310] During the inspection using various frequencies, the signal source 280c supplies the excitation signal Hm of that frequency to the detector 200c. The detector 200c outputs an eddy current signal Am by applying an alternating magnetic field of that frequency. The signal processing device 300d generates a detection signal Bm and a fiber image Cm based on the eddy current signal Am output by the detector 200c.

[0311] The operation of the eddy current signal evaluation unit 302c, image processing unit 304d, and defect determination unit 306d in the second mode is similar to... Figure 14 The operation of the eddy current signal evaluation unit 302c, the image processing unit 304c, and the defect determination unit 306c is the same.

[0312] However, the defect determination unit 306d sends the defect information D to the location determination unit 314d. The defect determination unit 306d also receives defect location information Fb from the location determination unit 314d and outputs the defect location information Fb together with the defect information D.

[0313] The location determination unit 314d acquires and outputs defect location information Fb, which represents the location of the defect. The defect location information Fb is acquired as follows.

[0314] That is, the position determination unit 314d selects a reference image Em from the reference images E1 to E4 recorded in the reference image recording unit 312d, and determines the portion of the selected reference image Em that matches the fiber image Cm at the defect location. This reference image Em is generated using the same frequency fm as the fiber image Cm used when a defect is determined to exist, or the detection signal Bm used when a defect is determined to exist. For example, if the similarity is above a threshold, it is determined to be consistent.

[0315] Here, the fiber image Cm at the defect location refers to the fiber image Cm used when a defect is determined to exist, or the fiber image Cm corresponding to the detection signal Bm used when a defect is determined to exist.

[0316] The fiber image Cm corresponding to the detection signal Bm used when a defect is determined to exist refers to the fiber image Cm generated based on the detection signal Bm, and further refers to the fiber image Cm generated based on the detection results within the same unit area as the detection signal Bm.

[0317] The reference image Em generated using the same frequency fm as the fiber image Cm used when a defect is determined to exist or the detection signal Bm used when a defect is determined to exist refers to the reference image Em generated based on the eddy current signal Am obtained when the same frequency as the frequency selected when the eddy current signal Am used to generate the fiber image Cm or the detection signal Bm is obtained.

[0318] Referring to Implementation Method 2 Figure 11 The processing described in (a) and (b) is similarly performed to determine the consistent parts.

[0319] The location determination unit 314d obtains information Fa indicating the location of the part determined to be consistent as defect location information Fb, and transmits the defect location information Fb to the defect judgment unit 306d.

[0320] The defect determination unit 306d notifies the output device 400 of the defect information D and the defect location information Fb. Upon receiving the notification, the output device 400 notifies the user or external device of the defect information D and the defect location information Fb.

[0321] Figure 18 and Figure 19 The steps of the processing in the rope inspection device 1d of Embodiment 4 are shown. Hereinafter, the processing in the first mode, namely, the acquisition of the reference image, will be performed first, and then the processing in the second mode, namely, the execution of the inspection, will be performed.

[0322] Figure 18 The processing in mode 1 is shown.

[0323] Here, we assume the generation of continuous reference images along the entire length of the rope 100. In this case, with the detector 200c facing one end of the rope 100, the process begins... Figure 18 The processing.

[0324] exist Figure 18 In the process, steps ST301, ST302, ST303, ST306, ST309, and ST310 are... Figure 16 The steps are the same for ST301, ST302, ST303, ST306, ST309 and ST310.

[0325] In step ST211d, the image processing unit 304d supplies one frame of fiber image Cm to the reference image recording unit 312d. The supplied fiber image Cm is recorded as a part of the reference image Em.

[0326] In the reference image recording unit 312d, regarding each frequency fm, when the initial one-frame fiber image Cm is supplied, the reference image Em is formed only using the supplied fiber image Cm. When a new fiber image Cm is supplied after the reference image Em has already been formed, the newly supplied fiber image Cm is appended to the end of the already formed reference image Em, thus extending the reference image Em to a longer length.

[0327] After step ST211d, proceed to step ST306.

[0328] Figure 19 The processing in mode 2 is shown.

[0329] exist Figure 19 In the process, steps ST301, ST302, ST303, ST304, ST305, ST306, ST309, and ST310 are... Figure 16 The steps for ST301, ST302, ST303, ST304, ST305, ST306, ST309 and ST310 are the same.

[0330] In step ST121d, the position determination unit 314d determines the portion in the reference image Em that corresponds to the fiber image Cm at the defect location.

[0331] In step ST122d, the position determination unit 314d obtains the position information Fa of the consistent part as the defect position information Fb, and notifies the defect judgment unit 306d of the defect position information Fb.

[0332] After step ST122d, proceed to step ST308d.

[0333] In step ST308d, the defect determination unit 306d notifies the output device 400 of the defect information D and the defect location information Fb.

[0334] In Implementation 4, in addition to the same effects as in Implementation 2, the following effects are also present.

[0335] That is, in the first mode, multiple reference images E1 to E4 are generated and saved using multiple different frequencies f1 to f4. In the second mode, when a defect is determined to exist, the portion of the reference image Em generated using the same frequency fm as the fiber image Cm or the detection signal Bm used for the determination is determined to be consistent with the fiber image Cm, and the defect location information Fb is obtained. Therefore, the accuracy of the defect location information Fb can be improved.

[0336] Similar to embodiment 3, embodiment 4 also uses... Figure 3 The detector shown in (c) is a two-dimensional array 204 with detection elements. Alternatively, it can also be used Figure 3 The detector 200 shown in (a) can also be used Figure 3 The detector 200b is shown in (b). Using... Figure 3 The detector 200 shown in (a) and its use Figure 3 The generation of eddy current signal, detection signal and fiber image in the case of detector 200b shown in (b) can also be performed in the same way as the operation described in the variation of embodiment 3.

[0337] Similar to that described in Embodiment 2, in Embodiment 4, the threshold used by the defect determination unit 306d can also be adjusted based on the magnitude of the detection signal Bm used to generate the reference image Em. In this way, even if the distance (gap) between the rope 100 and the detector 200c, and further the distance (gap) between the CFRP 110 and the detector 200c, changes, the accuracy of defect detection can be maintained.

[0338] Implementation method 5.

[0339] Figure 20 The rope inspection device 1e of Embodiment 5 is shown. The overall structure of the rope inspection device 1e of Embodiment 5 is similar to... Figure 17 The rope inspection device shown is the same as 1d. However, it differs in the following aspects.

[0340] First, a detector 200e, a signal source 280e, and a signal processing device 300e are installed instead of the detector 200c, the signal source 280c, and the signal processing device 300c.

[0341] Detector 200e, like detector 200c, has a two-dimensional array of detection elements 210e. However, the detection elements 210e are respectively, for example, as... Figure 21 As shown, it has multiple sub-elements, such as sub-elements 1 to 3, 212a, 212b, and 212c.

[0342] When detector 200e is set to be opposite rope 100 and thus opposite CFRP 110, each sub-element is set to be opposite CFRP 110 to detect eddy currents at the opposite position of CFRP 110.

[0343] Each sub-element has an excitation coil and a detector. Specifically, sub-element 212a has an excitation coil 220a and a detector 230a, sub-element 212b has an excitation coil 220b and a detector 230b, and sub-element 212c has an excitation coil 220c and a detector 230c.

[0344] When detector 200e is positioned opposite rope 100, multiple sub-elements 212a-212c are configured to be at different distances from rope 100. Figure 21 The diagram shows that the distances (gap) ga, gb, and gc between the first sub-element 212a to the third sub-element 212c and the rope 100 are all different. Since the gaps ga, gb, and gc between them and the rope 100 are all different, the gaps between them and the CFRP 110 are also all different.

[0345] Sub-elements 212a to 212c are formed on different layers 250a, 250b, and 250c of the printed circuit board. The sub-elements 212a, 212b, and 212c within each detection element are arranged in the T direction. In this case, it can be said that the sub-elements 212a to 212c are stacked on top of each other.

[0346] Sub-elements 212a, 212b, and 212c are arranged sequentially from the side closest to the rope 100. For example, sub-elements 212a is arranged at the position closest to the rope 100, sub-elements 212b is arranged at the second closest position, and sub-elements 212c is arranged at the farthest position.

[0347] Excitation coils 220a, 220b, and 220c are connected to signal source 280e via independent wiring, and detectors 230a, 230b, and 230c are connected to signal processing device 300e via independent wiring.

[0348] Similar to signal source 280c in Embodiment 3, signal source 280e sequentially outputs excitation signals H1 to HM at multiple frequencies, such as the first frequency f1 to the Mth frequency fM. Hereinafter, M is assumed to be 4. Signal source 280e can also select multiple sub-elements 212a to 212c and supply excitation signal Hm to the selected sub-elements.

[0349] The signal processing apparatus 300e includes an eddy current signal evaluation unit 302e, an image processing unit 304e, a defect determination unit 306e, a reference image recording unit 312e, and a position determination unit 314e. These are respectively connected to… Figure 17 The eddy current signal evaluation unit 302c, image processing unit 304d, defect determination unit 306d, reference image recording unit 312d, and position determination unit 314d are the same, but there are obvious differences according to the following description.

[0350] The rope inspection device 1e of embodiment 5 also operates in either the first mode, i.e., the image acquisition mode, or the second mode, i.e., the inspection execution mode.

[0351] Depend on Figure 20 The dashed lines with arrows indicate the signal and data flow during processing in Mode 1.

[0352] In the first mode, reference images E(m, q) are generated for combinations of multiple frequencies f1 to f4 and multiple sub-elements 212a to 212c, respectively.

[0353] In the first mode, all sub-elements 212a, 212b, and 212c are used and selected sequentially. From here on, selection is performed in the order closest to rope 100, i.e., in the order of sub-elements 212a, 212b, and 212c.

[0354] Signal source 280e sequentially selects frequencies from 1 to 4, f1 to f4, and supplies the excitation signal Hm (m = 1, 2, 3 or 4) at the selected frequency fm to the selected sub-element 212q.

[0355] For example, when the detector 200e is opposite to the CFRP110 in a unit area, the signal source 280e sequentially selects frequencies f1 to f4, and sequentially selects multiple sub-elements 212a, 212b, and 212c during the selected frequency period. The signal source 280e supplies the excitation signal Hm of the selected frequency fm to the selected sub-elements 212q (q is a, b, or c).

[0356] The signal source 280e also supplies the reference signal Im (m is 1, 2, 3 or 4) of the selected frequency to the eddy current signal evaluation unit 302e.

[0357] When an excitation signal Hm of each frequency fm is supplied, the detector 200e detects eddy currents in the opposite unit region through the selected sub-element 212q and outputs an eddy current signal A of 1 frame.

[0358] Let A(m, q) represent the eddy current signal A obtained when an excitation signal Hm at frequency fm (m is 1, 2, 3 or 4) is supplied to sub-element 212q (q is a, b or c).

[0359] The same applies to the detection signal B, fiber image C, and reference image E.

[0360] As a result of selecting multiple sub-elements sequentially during the selected frequency fm, eddy current signals A(m,a), A(m,b), and A(m,c) based on sub-elements 212a, 212b, and 212c are sequentially output from the detection element 210e.

[0361] As a result of sequentially selecting multiple frequencies f1 to f4, the detection element 210e sequentially outputs eddy current signals A(1,a), A(2,a), A(3,a), A(4,a) based on sub-element 212a, eddy current signals A(1,b), A(2,b), A(3,b), A(4,b) based on sub-element 212b, and eddy current signals A(1,c), A(2,c), A(3,c), A(4,c) based on sub-element 212c.

[0362] The eddy current signal evaluation unit 302e outputs a detection signal B(m,q) of one frame based on the eddy current signal A(m,q) output from the sub-element 212q when each frequency fm is selected and each sub-element 212q is selected.

[0363] When the frequency fm is selected, a reference signal Im of frequency fm is supplied. Therefore, the eddy current signal evaluation unit 302e extracts the frequency fm component in the eddy current signal A(m,q) based on the reference signal Im and generates a detection signal B(m,q).

[0364] The above processing is performed on the combination of multiple frequencies f1 to f4 and multiple sub-elements 212a to 212c respectively.

[0365] The image processing unit 304e generates a fiber image C(m,q) based on a detection signal B(m,q) of one frame for each combination of multiple frequencies f1 to f4 and multiple sub-elements 212a to 212c.

[0366] That is, based on the detection signal B(m,q) corresponding to the eddy current signal A(m,q) output from the sub-element 212q at the selected m-th frequency fm, the fiber image C(m,q) is generated.

[0367] Since the selected frequency and the selected sub-element combination are 12, 12 fiber images C(1,a) to C(4,c) are generated.

[0368] When the above series of processes is completed for all 12 combinations of frequency and sub-elements, the detector 200e is moved by 1 frame relative to the cable 100.

[0369] By performing the above processing sequentially in multiple unit regions, fiber images C(1,a) to C(4,c) of multiple frames are generated for each of the 12 combinations of frequency and sub-element.

[0370] The reference image unit 312e records reference images E(1,a) to E(4,c).

[0371] Corresponding to the image processing unit 304e generating 12 fiber images C(1,a) to C(4,c) for each unit area, 12 reference images E(1,a) to E(4,c) are stored in the reference image recording unit 312e.

[0372] The fiber images C(1,a) to C(4,c) generated by the image processing unit 304e are respectively linked to the end of the reference images E(1,a) to E(4,c) in the reference image recording unit 312e as part of the reference images E(1,a) to E(4,c).

[0373] When initially generating each of the fiber images C(m,q) to C(4,c), the fiber image is recorded as a reference image E(m,q).

[0374] As described above, in the first mode, reference images E(m, q) are generated for multiple combinations of multiple sub-elements 212a to 212c and multiple frequencies f1 to f4, respectively. The reference image E(m, q) for each combination is generated based on eddy current signals A(m, q) obtained at multiple different locations in the L direction of CFRP110 when the frequency fm constituting the combination and the sub-elements 212q are selected.

[0375] For example, the above processing is performed over the entire length of the rope 100 to generate 12 consecutive reference images E(1,a) to E(4,c) over the entire length of the rope 100.

[0376] Depend on Figure 20 The solid line with arrows shows the signal and data flow during processing in mode 2.

[0377] In the second mode, only one of the sub-elements 212a, 212b, and 212c from the first to the third sub-elements is used, such as the first sub-element 212a.

[0378] Signal source 280e sequentially selects frequencies from 1 to 4, f1 to f4, and supplies the excitation signal Hm (m = 1, 2, 3 or 4) at the selected frequency fm to one of the sub-elements 212a.

[0379] The signal source 280e also supplies the reference signal Im (m is 1, 2, 3 or 4) of the selected frequency to the eddy current signal evaluation unit 302e.

[0380] When an excitation signal Hm of each frequency fm is supplied, the sub-element 212a detects the eddy current caused by the excitation signal Hm and outputs the eddy current signal A(m,a).

[0381] As a result of sequentially selecting multiple frequencies f1 to f4, eddy current signals A(1,a), A(2,a), A(3,a), and A(4,a) based on sub-element 212a are sequentially output from the detection element 210e.

[0382] The eddy current signal evaluation unit 302e outputs a detection signal B(m,a) based on a frame of eddy current signal A(m,a) obtained from the detection results of eddy currents in each unit region when each frequency fm is selected.

[0383] When the frequency fm is selected, a reference signal Im of frequency fm is supplied. Therefore, the eddy current signal evaluation unit 302e extracts the frequency fm component in the eddy current signal A(m,a) based on the reference signal Im and generates a detection signal B(m,a).

[0384] The above processing is performed on multiple frequencies f1 to f4 respectively.

[0385] The image processing unit 304e generates a fiber image C(m,a) based on a detection signal B(m,a) of one frame for each fm among multiple frequencies f1 to f4.

[0386] That is, based on the detection signal B(m,a) corresponding to the eddy current signal A(m,a) output from the sub-element 212a at the selected m-th frequency fm, the fiber image C(m,a) is generated.

[0387] The selected frequencies fm are 4, thus generating 4 fiber images C(1,a) to C(4,a).

[0388] The position determination unit 314e determines, for each C(m,a) (m is 1, 2, 3 or 4) in the fiber images C(1,a) to C(4,a), the portion J(m,p) that is consistent with the fiber image C(m,a) in the reference images E(m,a), E(m,b) and E(m,c) generated using the same frequency fm as the fiber image C(m,a).

[0389] For example, it is also possible to extract the part J(m,a) with the highest similarity to the fiber image C(m,a) from the reference image E(m,a), extract the part J(m,b) with the highest similarity to the fiber image C(m,a) from the reference image E(m,b), extract the part J(m,c) with the highest similarity to the fiber image C(m,a) from the reference image E(m,c), and extract the part J(m,a), J(m,b), and J(m,c) with the highest similarity as the consistent part J(m,p).

[0390] In these processes, a similarity score above a certain threshold can also be used as an additional condition.

[0391] The position determination unit 314e obtains the consistent part J(m, p) as a comparison image.

[0392] The position determination unit 314e also obtains position information Fa about the part J(m, p) that is determined to be consistent, as position information about the comparison image.

[0393] The obtained comparison image J(m, p) and location information Fa are sent to the defect determination unit 306e.

[0394] The defect determination unit 306e determines whether there is a defect based on the threshold determination result for the detection signal B(m,a), the image recognition result for the fiber image C(m,a), and the comparison result between the fiber image C(m,a) and the comparison image J(m,p), and outputs the determination result as defect information D.

[0395] Alternatively, if the defect determination unit 306e determines that a defect exists in any one of the following: threshold determination for the detection signal B(m,a), image recognition for the fiber image C(m,a), and comparison result between the fiber image C(m,a) and the comparison image J(m,p), then it comprehensively determines that a defect exists.

[0396] In the comparison between the fiber image C(m, a) and the comparison image J(m, p), the corresponding pixel values ​​can be compared, as well as the results of the recognition of each image.

[0397] When the defect determination unit 306e determines that a defect exists, it obtains the position information Fa of the comparison image J(m, p) used in the defect determination as the defect position information Fb, and outputs the defect position information Fb.

[0398] The comparison image J(m, p) used in defect determination mentioned here is a comparison image obtained as a fiber image C(m, a) used when a defect is determined to exist, or a fiber image C(m, a) corresponding to the detection signal B(m, a) used when a defect is determined to exist.

[0399] If a defect is found during the threshold determination and image recognition for each unit area, the processing can be terminated at that point.

[0400] For example, threshold determination and image recognition are performed sequentially for multiple frequencies and multiple sub-elements. As a result of threshold determination or image recognition for any frequency or any sub-element, if a defect is determined to exist, the processing for that unit region can also be terminated at that time point.

[0401] As described above, in the second mode, the signal source 280e sequentially selects multiple frequencies f1 to f4 and supplies the excitation signal Hm of the selected frequency fm to one of the multiple sub-elements 212a to 212c.

[0402] When selecting each frequency fm, the signal processing device 300e generates a detection signal B(m, a) and a fiber image C(m, a). The signal processing device 300e obtains the portion of the multiple reference images E(m, a) to E(m, c) generated using the same frequency as the generated fiber image C(m, a) that corresponds to the generated fiber image C(m, a), and uses it as a comparison image J(m, p). Based on the result of threshold determination for the detection signal B(m, a), the result of image recognition for the fiber image C(m, a), and the comparison result between the fiber image C(m, a) and the comparison image J(m, p), it determines whether there is a defect.

[0403] The defect determination unit 306e notifies the output device 400 of the defect information D and defect location information Fb, indicating whether a defect exists. Upon receiving this notification, the output device 400 notifies the user or external device of the defect information D and defect location information Fb.

[0404] The following explains the significance of using a detection element having multiple sub-elements arranged at different distances from the rope 100, as described above.

[0405] When performing an inspection while the detector 200e is fixed and the rope 100 is moved, the rope 100 sometimes vibrates, causing the distance g between the rope 100 and the detector 200e to change. When the gap g changes, the eddy current signal A changes, and the fiber image C obtained from the eddy current signal A changes. Furthermore, when the frequency of the excitation signal changes, the fiber image C also changes.

[0406] Figure 22 The diagram conceptually illustrates the changes in fiber image C caused by variations in gap g and frequency of the excitation signal (excitation frequency).

[0407] exist Figure 22 In the diagram, the vertical axis represents the gap g, and the horizontal axis represents the excitation frequency.

[0408] For reference Figure 15 As illustrated in (c), the higher the excitation frequency, the more the fiber image C reflects the state of the portion of CFRP110 closer to surface 111.

[0409] Furthermore, as the gap g increases further, the eddy current signal A decreases further, the shape of the strip-shaped portion corresponding to the carbon fiber in the fiber image C changes, and the contrast decreases.

[0410] In the case of comparison with each fiber image C, the more similar the reference image is to the portion of the reference image E that corresponds to the fiber image C (the image portion obtained according to the same portion of CFRP110) and originates from the eddy current signal A obtained according to the detection result in the sub-element, the higher the similarity is, wherein the sub-element is used to generate the fiber image C and the gap (ga, gb or gc) is closer to the gap g when the eddy current is detected.

[0411] Therefore, in this embodiment, all reference images E(m,a), E(m,b), and E(m,c) with different gaps are compared with the fiber image C(m,a). The image portion with the highest similarity is determined as the image portion that matches the fiber image C(m,a). Furthermore, the comparison result with the image portion with the highest similarity is also used to determine whether a defect exists. In this way, robust defect detection can be achieved even if the gap changes during inspection.

[0412] Figure 23 and Figure 24 The steps of the processing in the rope inspection device 1e of Embodiment 5 are shown. Hereinafter, the processing in the first mode, namely, the acquisition of the reference image, will be performed first, followed by the processing in the second mode, namely, the execution of the inspection.

[0413] Figure 23 The processing in mode 1 is shown.

[0414] Here, we assume the case where a continuous reference image is generated along the entire length of the rope 100. In this case, with the detector 200e facing one end of the rope 100, the process begins... Figure 23 The processing.

[0415] exist Figure 23In the process, steps ST301, ST302, ST306, ST309, and ST310 are... Figure 18 ST301, ST302, ST306, ST309 and ST310 are the same or similar.

[0416] In step ST501, any sub-element 212q (q is a, b, or c) from a plurality of sub-elements is selected. If the processing of step ST501 is performed after the processing of step ST302, the initial sub-element, i.e., the sub-element 212a closest to the rope 100, is selected.

[0417] In step ST303e, within a unit area, an excitation signal Hm at a selected frequency fm and a selected sub-element 212q are used to acquire a detection signal B(m,q) of one frame and generate a fiber image C(m,q).

[0418] That is, the detector 200e outputs a 1-frame eddy current signal A(m,q), the eddy current signal evaluation unit 302e acquires the 1-frame eddy current signal A(m,q) and generates a 1-frame detection signal B(m,q), and the image processing unit 304e acquires the 1-frame detection signal B(m,q) and generates a fiber image C(m,q).

[0419] In step ST211e, the image processing unit 304e supplies one frame of fiber image C(m, q) to the reference image recording unit 312e. The supplied fiber image C(m, q) is recorded as a part of the reference image E(m, q).

[0420] For each combination of multiple frequencies f1 to f4 and multiple sub-elements 212a to 212c, when the initial single-frame fiber image C(m, q) is supplied to the reference image recording unit 312e, the reference image E(m, q) is formed only from the supplied fiber image C(m, q). When a new fiber image C(m, q) is supplied after the reference image E(m, q) has already been formed, the newly supplied fiber image C(m, q) is appended to the end of the already formed reference image E(m, q), thus extending the reference image E(m, q) to a longer length.

[0421] After step ST211e, proceed to step ST502.

[0422] In step ST502, it is determined whether all sub-components have been selected.

[0423] If the answer is "No", then return to step ST501.

[0424] If the result in step ST502 is "No" and the process returns to step ST501, select the next sub-element.

[0425] If the answer is "yes" in step ST502, then proceed to step ST306.

[0426] Step ST306 determines whether all frequencies have been selected. If the result is "No", the process returns to step ST302; if the result is "Yes", the process proceeds to step ST309.

[0427] If the result in step ST306 is "No" and the process returns to step ST302, select the next frequency.

[0428] Figure 24 The processing in mode 2 is shown.

[0429] exist Figure 24 In the process, steps ST301, ST302, ST305, ST306, ST309, and ST310 are... Figure 19 The steps are the same for ST301, ST302, ST305, ST306, ST309 and ST310.

[0430] also, Figure 24 Step ST303f and Figure 23 The steps are similar to those of ST303e.

[0431] In step ST303f, within a unit area, using the excitation signal Hm at the selected frequency fm and the selected sub-element 212a, a detection signal B(m, a) of 1 frame is acquired and a fiber image C(m, q) is generated.

[0432] That is, detector 200e outputs a 1-frame eddy current signal A(m,a), eddy current signal evaluation unit 302e acquires the 1-frame eddy current signal A(m,a) and generates a 1-frame detection signal B(m,a), and image processing unit 304e acquires the 1-frame detection signal B(m,a) and generates fiber image C(m,a).

[0433] In step ST511, the position determination unit 314e determines, for the fiber image C(m,a), a portion J(m,p) in reference images E(m,a), E(m,b), and E(m,c) generated using the same frequency fm as the fiber image C(m,a) that is consistent with the fiber image C(m,a).

[0434] In step ST512, the position determination unit 314e outputs the portion J(m, p) determined to be consistent in step ST511 as a comparison image. The position determination unit 314e also obtains position information Fa about the portion J(m, p) determined to be consistent as position information about the comparison image J(m, p) and outputs it.

[0435] In step ST304f, the defect determination unit 306e determines whether there is a defect based on the threshold determination result for the detection signal B(m,a), the image recognition result for the fiber image C(m,a), and the comparison result between the fiber image C(m,a) and the comparison image J(m,p).

[0436] In step ST305, if the determination result in step ST304f is "no", then proceed to step ST306; if it is "yes", then proceed to step ST308e.

[0437] In step ST308e, the defect determination unit 306e notifies the output device 400 of the defect information D and the defect location information Fb.

[0438] In Implementation 5, in addition to the same effects as in Implementation 4, the following effects are also present.

[0439] That is, the detection element 210 has multiple sub-elements 212a to 212c arranged at different distances from the rope 100. By recording multiple reference images corresponding to multiple different gaps ga, gb, gc, robust defect detection can be achieved for changes in gaps during inspection.

[0440] In the above-described embodiment 5, the threshold used in the threshold determination for the detection signal B can also be adjusted based on the estimated value of the gap when the eddy current signal A is used to generate the detection signal B.

[0441] For example, a threshold is determined in advance based on the magnitude of the detection signal B(m,q) used when generating each reference image E(m,q), and the determined threshold is recorded in association with the reference image E(m,q) generated based on the detection signal B(m,q).

[0442] During the inspection, a reference image E(m,p) containing an image portion J(m,p) determined to be consistent with the fiber image C(m,a) can also be determined. A threshold recorded in association with the determined reference image E(m,p) is used as a threshold for comparison with the detection signal B(m,a) used when generating the fiber image C(m,a).

[0443] In other words, when comparing the detection signal B(m,a) with the threshold, a threshold recorded in association with reference image E(m,p) among multiple reference images E(m,a), E(m,b), and E(m,c) can also be used. This reference image E(m,p) includes a fiber image C(m,a) generated based on the detection results within the same unit area as the aforementioned detection signal B(m,a), that is, a portion that is consistent with the fiber image C(m,a) generated based on the detection results within the same unit area as the aforementioned detection signal B(m,a).

[0444] Each reference image corresponds to an arbitrary sub-element, and each sub-element has a defined gap. Therefore, if a threshold recorded in association with a reference image is used, a threshold suitable for the gap of the sub-element corresponding to that reference image is used.

[0445] By performing the above processing, it is possible to prevent false detection of defects caused by variations in the detection signal B resulting from changes in the gaps during inspection.

[0446] The above explains the relationship between the distance (gap) between the rope 100 and the detector 200e or sub-element, and the threshold for the fiber image or detection signal. However, the same explanation also applies to the relationship between the distance (gap) between the CFRP 110 and the detector 200e or sub-element, and the threshold for the fiber image or detection signal.

[0447] In embodiment 5 described above, multiple sub-elements are selected sequentially during a period in which multiple frequencies are selected respectively. Alternatively, it can be configured such that multiple frequencies are selected sequentially during a period in which multiple sub-elements are selected respectively.

[0448] In the above-described embodiment 5, the following was used: Figure 3 The detector 200e shown in (c) is a two-dimensional array 204 with detection elements. Alternatively, a detector with... Figure 3 The detector 200 shown in (a) similarly has a one-dimensional array of detection elements and each detection element is connected to... Figure 21 The detection element 210e shown also includes a detector with multiple sub-elements (denoted by reference numeral 200ea), and can also be used with... Figure 3 The detector 200b shown in (b) similarly has a single detection element and actuator, and this single detection element is... Figure 21 The detection element 210e shown also includes a detector with multiple sub-elements (denoted by reference numeral 200eb).

[0449] In use and Figure 3In the case of detector 200ea, which is also equipped with a single-element array of detection elements, as shown in (a), the generation of eddy current signal, detection signal and fiber image in the first mode can also be performed as follows. Hereinafter, the case of having 4 frequencies and 3 sub-elements will be described.

[0450] That is, the frequency can be selected sequentially and the sub-elements can be selected sequentially during the period when the one-dimensional array 202 is opposite to a measurement line.

[0451] The actions in this situation are as follows.

[0452] That is, during the period when the one-dimensional array 202 is aligned with a measurement line, multiple frequencies f1 to f4 are selected sequentially, and sub-elements 212a to 212c are selected sequentially. This generates eddy current signals A(1,a) to A(4,c) and detection signals B(1,a) to B(4,c) corresponding to the combinations of the multiple frequencies f1 to f4 and the multiple sub-elements 212a to 212c, respectively, for one line. While changing the alignment position of the detector 200ea relative to the rope 100 in the L direction, the above process is repeated Nf times for each measurement line, i.e., the generation of eddy current signals A(1,a) to A(4,c) and detection signals B(1,a) to B(4,c). Thus, one frame of eddy current signals A(1,a) to A(4,c) and one frame of detection signals B(1,a) to B(4,c) are generated sequentially. Then, fiber images C(1,a) to C(4,c) of one frame are generated based on the generated detection signals B(1,a) to B(4,c).

[0453] Alternatively, the following processing can be performed on the above combinations: when one frequency fm and one sub-element 212q (q is a, b or c) are selected, by changing the opposing position of detector 200ea relative to rope 100 in the L direction, one frame of eddy current signal A (m, q) and detection signal B (m, q) are generated, and one frame of fiber image C (m, q) is generated based on the generated one frame of detection signal B (m, q).

[0454] The actions in this situation are as follows.

[0455] That is, with one frequency fm and one sub-element 212q (q being a, b, or c) selected, while changing the position of detector 200ea relative to rope 100 in the L direction, the process of generating one line's eddy current signal A(m, q) and detection signal B(m, q) is repeated Nf times. This generates one frame's worth of eddy current signal A(m, q) and detection signal B(m, q) sequentially. One frame's worth of fiber image C(m, q) is generated based on the generated one frame's worth of detection signal B(m, q). Then, the position of detector 200ea relative to rope 100 in the L direction is returned to its original position. This process is repeated 12 times (thus obtaining 12 fiber images C(1, a) to C(4, c)).

[0456] In use and Figure 3 In the case of detector 200b shown in (b), which also has a single detection element and actuator, the generation of eddy current signal, detection signal and fiber image in the first mode can also be performed as follows.

[0457] That is, the frequency can be selected sequentially and the sub-elements can be selected sequentially during the period when the detection element 210 is opposite to a detection point.

[0458] Alternatively, the following process can be repeated: with one frequency fm selected and one sub-element 212q (q is a, b or c) selected, by changing the opposing position of the detection element 210eb relative to the rope 100 in the W and L directions, a frame of eddy current signal A(m,q) and detection signal B(m,q) can be generated.

[0459] In the rope inspection device 1e of embodiment 5, regarding the use and Figure 3 The detector 200 shown in (a) also has the case of a detector 200ea with a one-dimensional array of detection elements and is used in conjunction with Figure 3 The operation of generating eddy current signals, detection signals and fiber images in the second mode of the detector 200b shown in (b) is similar to the operation described in the variation of embodiment 3, where the detector 200b also has a single detection element and actuator.

[0460] In embodiment 5, a detection element having multiple sub-elements is used, and the frequency of the excitation signal is switched. Alternatively, a detection element having multiple sub-elements can be used, but the frequency of the excitation signal is not switched.

[0461] The operation in this case is summarized below. Hereinafter, as described above, each detection element is assumed to contain 3 sub-elements.

[0462] That is, in the first mode, reference images E(a) to E(c) are generated for each of the multiple sub-elements 212a to 212c. The reference image E(q) for each sub-element 212q (q is a, b or c) is generated based on the eddy current signal A(q) obtained at multiple different positions in the L direction of CFRP110 when the sub-element 212q is selected.

[0463] In the second mode, one of the multiple sub-elements 212a to 212c is selected, for example, sub-element 212a. The signal processing device 300e generates a detection signal B(a) and a fiber image C(a). The portion of multiple reference images E(a) to E(c) that is consistent with the generated fiber image C(a) is obtained as a comparison image J. Based on the result of threshold determination for the detection signal B(a), the result of image recognition for the fiber image C(a), and the comparison result between the fiber image C(a) and the comparison image J, it is determined whether there is a defect.

[0464] The structure described above, which uses a detection element with multiple sub-elements and does not switch the frequency of the excitation signal, can also achieve robust defect detection in response to changes in the gap during inspection.

[0465] Various modifications can be made to embodiments 1 to 5 described above, in addition to the above.

[0466] For example, the variations described for Embodiment 1 can also be applied to Embodiments 2 through 5. For example, in Embodiment 1, it was explained that image recognition can also be performed on images obtained by performing FFT, BPF, and IFFT on the fiber image C. The same variations can also be performed on Embodiments 2 through 5.

[0467] Furthermore, regarding Embodiment 1, it is explained that the eddy current signal evaluation unit 302 can also generate a detection signal B by amplifying, adjusting, level shifting, and filtering the eddy current signal A, and can also generate a detection signal B by correcting the eddy current signal A obtained from the detection results at the W-direction end of the CFRP110. The same modifications can also be made to Embodiments 2 to 5.

[0468] In embodiments 2, 4, and 5, the generation of reference image E in the first mode and the inspection in the second mode are described separately. However, the generation of reference image E in the first mode and the inspection in the second mode can also be performed in parallel. If the generation and inspection of reference image E are performed in parallel, it is effective for identifying initial defects and preventing false detections of defects caused by updates to reference image E.

[0469] In Embodiment 5, a comparison image extracted from the reference image is also used to determine whether there is a defect. However, in Embodiments 2 and 4, the reference image can also be used to determine whether there is a defect. That is, the presence or absence of a defect can be determined based on the comparison results of the portion of one or more reference images that matches the fiber image with the fiber image.

[0470] In the first mode of embodiments 2, 4, and 5, it is assumed that reference images are generated over the entire length of the rope. Instead, reference images may be generated only for a portion of the rope, i.e., a specific range. For example, reference images may be generated separately for a fixed length portion centered at a specified position set at fixed intervals every other rope.

[0471] In this case, Figure 12 , Figure 18 and Figure 23 The process begins when the detector is positioned opposite one end of the aforementioned fixed-length portion and ends when the detector is positioned opposite the other end of the aforementioned fixed-length portion.

[0472] The signal processing apparatus in embodiments 1 to 5 is implemented by a processing circuit. The processing circuit can be dedicated hardware or a control circuit of a CPU (Central Processing Unit) that has a memory and executes the program stored in the memory. The memory can be, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, magnetic disk, optical disk, etc.

[0473] The rope inspection device has been described above. Rope inspection methods can be implemented using the rope inspection device described above.

[0474] Explanation of reference numerals in the attached figures

[0475] 1, 1b, 1c, 1d, 1e Rope inspection device; 100 Rope; 110 CFRP; 114 Carbon fiber; 115 Resin material; 120 Coating; 200, 200c, 200e Detector; 202 One-dimensional array; 204 Two-dimensional array; 210 Detection element; 212a, 212b, 212c Sub-element; 215 Actuator; 220, 220a, 220b, 220c Excitation coil; 230, 230a, 230b, 230c Detector; 240 Signal source; 250a, 250b, 250c Layer; 280, 280c, 280e Signal source; 300, 300b, 300c, 300d, 300e Signal processing device; 302, 302c, 302e Eddy current signal evaluation unit, 304, 304b, 304c, 304d, 304e; image processing unit, 306, 306b, 306c, 306d, 306e; defect determination unit, 312, 312d, 312e; reference image recording unit, 314, 314d, 314e; position determination unit, 400; output device.

Claims

1. A rope inspection device for inspecting a rope comprising a strip of CFRP and a coating covering the CFRP, wherein, The rope inspection device has: A detector that applies an alternating magnetic field to the CFRP and detects the eddy currents generated by the alternating magnetic field; as well as A signal processing device generates a detection signal representing the intensity of eddy currents and a fiber image representing the distribution of eddy current intensity in each of multiple unit regions located at multiple different positions along the length direction of the CFRP, based on the detection results of each unit region. Based on a comparison of the detection signal with a threshold and the result of image recognition of the fiber image, the device determines whether a defect exists in the unit region. In the first mode, the signal processing device generates a reference image based on the intensity of eddy currents detected at multiple different locations along the length of the CFRP. In the second mode, the signal processing device performs the following processing: The presence or absence of defects is determined based on the detection signal generated according to the detection results in each of the plurality of unit regions and the fiber image. When a defect is determined, the portion of the reference image that matches the fiber image used when the defect was determined, or the portion that matches the fiber image generated based on the detection results of the same unit area as the detection signal used when the defect was determined, is identified, and information indicating the location of the matching portion is output as information indicating the location of the defect.

2. The rope inspection device according to claim 1, wherein, In the first mode, the signal processing device repeatedly performs the process of generating fiber images representing the intensity of eddy currents detected in each of the plurality of unit regions, thereby sequentially generating a plurality of frames of fiber images, and generating the reference image by linking the generated plurality of frames of fiber images together.

3. The rope inspection device according to claim 1, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

4. The rope inspection device according to claim 2, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

5. The rope inspection device according to claim 1, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

6. The rope inspection device according to claim 2, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

7. The rope inspection device according to claim 3, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

8. The rope inspection device according to claim 4, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

9. The rope inspection device according to claim 1, wherein, The detector has at least one detection element. The at least one detection element has the following features: An excitation coil that applies the alternating magnetic field to the CFRP; as well as The detector detects the secondary magnetic field of the eddy currents generated in the CFRP, thereby detecting the eddy currents that generate the secondary magnetic field.

10. The rope inspection device according to any one of claims 1 to 9, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

11. The rope inspection device according to any one of claims 1 to 9, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

12. The rope inspection device according to claim 10, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

13. A rope inspection device for inspecting a rope comprising a strip of CFRP and a coating covering the CFRP, wherein, The rope inspection device has: A detector that applies an alternating magnetic field to the CFRP and detects the eddy currents generated by the alternating magnetic field; as well as A signal processing device generates a detection signal representing the intensity of eddy currents and a fiber image representing the distribution of eddy current intensity in each of multiple unit regions located at multiple different positions along the length direction of the CFRP, based on the detection results of each unit region. Based on a comparison of the detection signal with a threshold and the result of image recognition of the fiber image, the device determines whether a defect exists in the unit region. The detector can apply an alternating magnetic field of a selected frequency from a plurality of frequencies as the alternating magnetic field. In mode 1, the signal processing device performs the following processing: Generate reference images for each of the plurality of frequencies. A reference image for a given frequency is generated based on the intensity of eddy currents detected at multiple different locations along the length of the CFRP when each frequency is selected. In mode 2, The rope inspection device performs inspections using each of the multiple frequencies. During the checks using various frequencies, The detector detects the eddy currents when an alternating magnetic field of that frequency is applied. The signal processing device performs the following processing: The detection signal and the fiber image are generated based on the eddy currents detected by the detector. The presence or absence of defects is determined based on the detection signal and the fiber image. If a defect is determined to exist, a portion of the reference image generated using the same frequency as the fiber image used in the determination or the detection signal used in the determination, that corresponds to the fiber image or that corresponds to the fiber image generated based on the detection result within the same unit area as the detection signal, is identified. The output information representing the location of the consistent portion is used as information representing the location of the defect.

14. The rope inspection device according to claim 13, wherein, In the first mode, the signal processing device repeatedly performs the process of generating fiber images representing the intensity of eddy currents detected in each of the plurality of unit regions, thereby sequentially generating a plurality of frames of fiber images, and generating the reference image by linking the generated plurality of frames of fiber images together.

15. The rope inspection device according to claim 13, wherein, The rope inspection device also includes a signal source that supplies an excitation signal to the detector to generate an alternating magnetic field of the selected frequency. The signal source supplies a reference signal to the signal processing device at the same frequency as the excitation signal supplied to the detector. The signal processing device uses the reference signal to extract the components in the detector output that have the same frequency as the reference signal, and generates the detection signal.

16. The rope inspection device according to claim 13, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

17. The rope inspection device according to claim 14, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

18. The rope inspection device according to claim 13, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

19. The rope inspection device according to claim 14, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

20. The rope inspection device according to claim 15, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

21. The rope inspection device according to claim 16, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

22. The rope inspection device according to claim 17, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

23. The rope inspection device according to claim 13, wherein, The detector has at least one detection element. The at least one detection element has the following features: An excitation coil that applies the alternating magnetic field to the CFRP; as well as The detector detects the secondary magnetic field of the eddy currents generated in the CFRP, thereby detecting the eddy currents that generate the secondary magnetic field.

24. The rope inspection device according to any one of claims 13 to 23, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

25. The rope inspection device according to any one of claims 13 to 23, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

26. The rope inspection device according to claim 24, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

27. A rope inspection device for inspecting a rope comprising a strip of CFRP and a coating covering the CFRP, wherein, The rope inspection device has: A detector that applies an alternating magnetic field to the CFRP and detects the eddy currents generated by the alternating magnetic field; as well as A signal processing device generates a detection signal representing the intensity of eddy currents and a fiber image representing the distribution of eddy current intensity in each of multiple unit regions located at multiple different positions along the length direction of the CFRP, based on the detection results of each unit region. Based on a comparison of the detection signal with a threshold and the result of image recognition of the fiber image, the device determines whether a defect exists in the unit region. The detector has at least one detection element. The at least one detection element has multiple sub-elements. The plurality of sub-elements are configured such that, when the detector is positioned opposite the rope, they are opposite the CFRP and at different distances from the CFRP, and each sub-element, when selected, detects eddy currents at its opposite position to the CFRP. In mode 1, the signal processing device performs the following processing: Generate reference images for each of the plurality of sub-elements. A reference image for each sub-element is generated based on the intensity of the eddy currents detected at multiple different locations along the length of the CFRP when each sub-element is selected. In mode 2, Select one of the plurality of sub-elements. The signal processing device performs the following processing: Based on the detection results of eddy currents in the selected sub-element, the detection signal and the fiber image are generated. The portion of the multiple reference images that matches the generated fiber image is obtained as a comparison image. Based on the comparison results of the detection signal and the threshold, the image recognition results of the fiber image, and the comparison results of the fiber image and the comparison image, it is determined whether there is a defect.

28. The rope inspection device according to claim 27, wherein, In the first mode, the signal processing device repeatedly performs the process of generating fiber images representing the intensity of eddy currents detected in each of the plurality of unit regions, thereby sequentially generating a plurality of frames of fiber images, and generating the reference image by linking the generated plurality of frames of fiber images together.

29. The rope inspection device according to claim 27, wherein, In the second mode, If the signal processing device determines that a defect exists based on the comparison result of the detection signal and the threshold, the image recognition result of the fiber image, and the comparison result of the fiber image and the comparison image, it outputs information indicating the location of the defect in the comparison image.

30. The rope inspection device according to claim 27, wherein, The signal processing device adjusts the threshold used for comparison with the detection signal in the second mode based on an estimated distance between the sub-element and the CFRP.

31. The rope inspection device according to claim 29, wherein, The signal processing device adjusts the threshold used for comparison with the detection signal in the second mode based on an estimated distance between the sub-element and the CFRP.

32. The rope inspection device according to claim 27, wherein, The signal processing device performs the following processing: In the first mode, a threshold is determined based on the intensity of the eddy current used to generate each of the plurality of reference images, and the determined threshold is recorded in association with the reference images generated based on the intensity of the eddy current. In the second mode, in the comparison of the detection signal with the threshold, a threshold recorded in association with a reference image that includes a portion consistent with the fiber image generated using the detection signal is used.

33. The rope inspection device according to claim 29, wherein, The signal processing device performs the following processing: In the first mode, a threshold is determined based on the intensity of the eddy current used to generate each of the plurality of reference images, and the determined threshold is recorded in association with the reference images generated based on the intensity of the eddy current. In the second mode, in the comparison of the detection signal with the threshold, a threshold recorded in association with a reference image that includes a portion consistent with the fiber image generated using the detection signal is used.

34. The rope inspection device according to claim 27, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

35. The rope inspection device according to claim 28, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

36. The rope inspection device according to any one of claims 27 to 35, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

37. The rope inspection device according to any one of claims 27 to 35, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

38. The rope inspection device according to claim 36, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

39. The rope inspection device according to any one of claims 27 to 35, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

40. The rope inspection device according to claim 36, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

41. The rope inspection device according to claim 37, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

42. The rope inspection device according to claim 38, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

43. A rope inspection device for inspecting a rope comprising a strip of CFRP and a coating covering the CFRP, wherein, The rope inspection device has: A detector that applies an alternating magnetic field to the CFRP and detects the eddy currents generated by the alternating magnetic field; as well as A signal processing device generates a detection signal representing the intensity of eddy currents and a fiber image representing the distribution of eddy current intensity in each of multiple unit regions located at multiple different positions along the length direction of the CFRP, based on the detection results of each unit region. Based on a comparison of the detection signal with a threshold and the result of image recognition of the fiber image, the device determines whether a defect exists in the unit region. The detector can apply an alternating magnetic field of a selected frequency from a plurality of frequencies as the alternating magnetic field. The detector has at least one detection element. The at least one detection element has multiple sub-elements. The plurality of sub-elements are configured such that, when the detector is positioned opposite the rope, they are opposite the CFRP and at different distances from the CFRP, and each of the plurality of sub-elements, when selected, detects eddy currents at the position opposite the CFRP. In mode 1, the signal processing device performs the following processing: Generate reference images for each of the multiple combinations of the plurality of sub-elements and the plurality of frequencies. A reference image for the CFRP is generated based on the intensity of eddy currents detected at multiple different locations along the length of the CFRP when the frequencies and sub-elements constituting each combination are selected. In mode 2, Select one of the plurality of sub-elements, and then sequentially select the plurality of frequencies. The signal processing device performs the following processing: Based on the detection results of eddy currents in the selected sub-elements at each selected frequency, the detection signal and the fiber image are generated. A portion of the multiple reference images generated using the same frequency as the generated fiber image that matches the generated fiber image is obtained and used as a comparison image. Based on the comparison results of the detection signal and the threshold, the image recognition results of the fiber image, and the comparison results of the fiber image and the comparison image, it is determined whether there is a defect.

44. The rope inspection device according to claim 43, wherein, In the first mode, the signal processing device repeatedly performs the process of generating fiber images representing the intensity of eddy currents detected in each of the plurality of unit regions, thereby sequentially generating a plurality of frames of fiber images, and generating the reference image by linking the generated plurality of frames of fiber images together.

45. The rope inspection device according to claim 43, wherein, In the second mode, If the signal processing device determines that a defect exists based on the comparison result of the detection signal and the threshold, the image recognition result of the fiber image, and the comparison result of the fiber image and the comparison image, it outputs information indicating the location of the defect in the comparison image.

46. ​​The rope inspection device according to claim 43, wherein, The signal processing device adjusts the threshold used for comparison with the detection signal in the second mode based on an estimated distance between the sub-element and the CFRP.

47. The rope inspection device according to claim 45, wherein, The signal processing device adjusts the threshold used for comparison with the detection signal in the second mode based on an estimated distance between the sub-element and the CFRP.

48. The rope inspection device according to claim 43, wherein, The signal processing device performs the following processing: In the first mode, a threshold is determined based on the intensity of the eddy current used to generate each of the plurality of reference images, and the determined threshold is recorded in association with the reference images generated based on the intensity of the eddy current. In the second mode, in the comparison of the detection signal with the threshold, a threshold recorded in association with a reference image that includes a portion consistent with the fiber image generated using the detection signal is used.

49. The rope inspection device according to claim 45, wherein, The signal processing device performs the following processing: In the first mode, a threshold is determined based on the intensity of the eddy current used to generate each of the plurality of reference images, and the determined threshold is recorded in association with the reference images generated based on the intensity of the eddy current. In the second mode, in the comparison of the detection signal with the threshold, a threshold recorded in association with a reference image that includes a portion consistent with the fiber image generated using the detection signal is used.

50. The rope inspection device according to claim 43, wherein, The rope inspection device also includes a signal source that supplies an excitation signal to the detector to generate an alternating magnetic field of the selected frequency. The signal source supplies a reference signal to the signal processing device at the same frequency as the excitation signal supplied to the detector. The signal processing device uses the reference signal to extract the components in the detector output that have the same frequency as the reference signal, and generates the detection signal.

51. The rope inspection device according to claim 43, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

52. The rope inspection device according to claim 44, wherein, In the first mode, the signal processing device generates the reference image based on the intensity of the eddy current detected in the CFRP in the entire length direction of the rope or within a specific range.

53. The rope inspection device according to any one of claims 43 to 52, wherein, The detector is configured to face the surface extending in both the length and width directions of the rope. The signal processing device corrects the output of the detector representing the detection result at the end of the CFRP in the width direction, and generates the detection signal.

54. The rope inspection device according to any one of claims 43 to 52, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

55. The rope inspection device according to claim 53, wherein, The fiber image generated based on the detection results in each of the plurality of unit regions consists of pixels representing the intensity of eddy currents detected at detection points arranged in a matrix within the unit region.

56. The rope inspection device according to any one of claims 43 to 52, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

57. The rope inspection device according to claim 53, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

58. The rope inspection device according to claim 54, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

59. The rope inspection device according to claim 55, wherein, The eddy current detection is performed at multiple detection points, which are located at different positions along the length of the CFRP and are respectively along measurement lines extending along the width of the CFRP. Each unit region is composed of a predetermined number of measuring lines.

60. A rope inspection method comprising a strip of CFRP and a coating covering the CFRP, wherein, An alternating magnetic field is applied to the CFRP, and the eddy currents generated by the alternating magnetic field are detected. Based on the detection results in multiple unit regions located at different positions along the length of the CFRP at one end of each unit region, a detection signal representing the intensity of eddy currents and a fiber image representing the distribution of eddy current intensity are generated. Based on the comparison result of the detection signal with a threshold and the image recognition result of the fiber image, it is determined whether there is a defect in the unit region. In the first mode, a reference image is generated based on the intensity of eddy currents detected at multiple different locations along the length direction of the CFRP. In mode 2, The presence or absence of defects is determined based on the detection signal generated according to the detection results in each of the plurality of unit regions and the fiber image. When a defect is determined, the portion of the reference image that matches the fiber image used when the defect was determined, or the portion that matches the fiber image generated based on the detection results of the same unit area as the detection signal used when the defect was determined, is identified, and information indicating the location of the matching portion is output as information indicating the location of the defect.

Images