JP7332821B1 - WELD INSPECTION METHOD AND INSPECTION DEVICE - Google Patents

Abstract

Provided are an inspection method and an inspection apparatus for a welded portion, which can determine an inspection range for an actual welded portion shape in a short time, simply, and accurately without depending on the skill and experience of an inspector. A method for inspecting a welded portion for inspecting a welded portion of a tubular component using an inspection device includes a first step of measuring a three-dimensional shape of the tubular component after welding and acquiring three-dimensional shape data of the tubular component; a second step of obtaining three-dimensional CAD data including three-dimensional position information regarding the design shape of the tubular part and the welding position of the weld; and data processing to associate the three-dimensional CAD data with the three-dimensional shape data. and a third step of determining an inspection range based on the three-dimensional position information regarding the construction position of the weld, and inspecting the weld of the tubular part by the inspection device based on the determined inspection range and a fourth step of inspecting to obtain inspection data.

Description

TECHNICAL FIELD The present invention relates to an inspection method and an inspection apparatus for welded portions.

Ultrasonic flaw detection (hereinafter abbreviated as UT) is a typical non-destructive inspection method used to detect defects that exist inside cast steel products and welds. etc., are used for various purposes. UT basically measures the time (propagation time) and the signal intensity at that time are measured, and the position and size of the defect are evaluated based on the measured time and signal intensity.

Defect evaluation methods include a method that uses peak signals (echoes) that appear in waveforms called A-scopes, and a method that generates flaw detection images from multiple A-scopes that shift the transmission and reception positions and timings, and identify defects from the images. There are methods of evaluation, and these have become the workhorses of UT. A phased array (PA) method is a representative method for evaluating defects from images, and has already been used in various industrial fields. Also, in recent years, a new method called a full matrix capture (FMC) method has attracted attention and its application is expanding.

The PA method and the FMC method use an array sensor containing a plurality of elements for transmitting and receiving ultrasonic waves. A piezoelectric element that converts a voltage into force or converts an applied force into a voltage using a piezoelectric effect is usually used as the element. The number, size, and arrangement of the elements vary depending on the application. is used. The dimension and pitch of the elements in the arrangement direction are generally 1 mm or less.

In the PA method, by controlling the phase of the ultrasonic waves oscillated from each element of the array sensor, it is possible to scan the ultrasonic beam, which is a composite wave, in any direction and to change the focal position. Technology. The FMC method is a technology that obtains a high-definition image by individually recording all waveforms corresponding to the combination of transmission and reception of each element of the array sensor and applying waveform synthesis processing corresponding to the position of the transmission and reception elements on the software. . The FMC method is described, for example, in JP-A-2019-158876.

In inspection methods using these array sensors, the test surface does not necessarily have to be flat. Even if the flaw detection surface has an arbitrary shape, these techniques can be applied in principle if the coordinates of each element are known. In this case, the PA method is not practical because the phase control is complicated, but the FMC method can easily generate an image by using coordinates for calculation.

For this reason, in recent years, a flexible array sensor, which is called a flexible array sensor, has been put into practical use and is used in an increasing number of cases together with the FMC method. By using a flexible array sensor, for example, the sensor can be placed along the surface of a component with a curved surface, such as a pipe, and brought into direct contact with the surface of the component. It is possible.

Here, as a typical example of tubular parts, an example of a bogie frame of a railway vehicle will be given. The bogie frame of a railway vehicle generally has a structure in which long tubular members (several meters in length and several tens of centimeters in diameter) called lateral beams and long members called side beams are welded in orthogonal directions. It has a structure in which plate materials called brackets are welded at multiple points to fix the equipment to it. Since these welds play an important role in supporting heavy parts, high soundness is required from the viewpoint of safety, and ultrasonic inspections are often performed before shipment from the factory. The welded part of the cross beam can be inspected by ultrasonic waves from both the inner surface and the outer surface of the cross beam. If they are brought into contact with each other, it becomes possible to easily inspect the cross beam welded portion.

At this time, an automatic scanning device (manipulator) is often used that can automatically and mechanically scan the flexible array sensor in the pipe axial direction while keeping the flexible array sensor in contact with the inner surface of the lateral beam. A liquid couplant is usually interposed between the sensor and the contact surface so that air that suppresses the propagation of ultrasonic waves does not enter. Glycerin paste or water is often used as the contact medium.

A manipulator generally has multiple rotation axes, each of which is connected to a servomotor and a reduction gear. By combining these, the position (coordinates) and attitude (angle) of the sensor attached to the tip of the arm in three-dimensional space. is controlled. Since these controls are performed through a computer, if the diameter and length of the horizontal beam pipe and the position of the weld to be inspected are input, the sensor can be automatically pressed and scanned. Since the position of the welded part is usually described in the design drawing, numerical values are entered into the computer by comparing the actual product with the drawing. Numerical input at this time is manually performed by the inspector.

By the way, in recent years, railcars are often designed using CAD (Computer Aided Design) instead of paper drawings. The CAD data generally indicates the welding instruction position (position to be welded) along with the structural information of each part, and the parts are welded by an automatic welder or a welding operator based on this information. After welding, the welded portion is inspected for weld defects, such as blowholes and poor penetration, using the above-described ultrasonic device and method.

However, since the actual weld position generally differs from the welding instruction position on the drawing, the inspection range often does not simply match the welding instruction range on the drawing. The reason for this is that the actual welding area extends from several millimeters to several tens of millimeters around the groove. Furthermore, defects such as blowholes, poor penetration, cracks, etc., which are subject to inspection, occur in the welded part and the area near the welded part called the heat affected zone. The range is wider than the position. However, since the inspection range is currently determined by the inspector, the accuracy of the inspection is often affected by the skill and experience of the inspector.

Patent Literature 1 discloses a method of measuring a weld repaired portion of a reactor pressure vessel before and after repair with a three-dimensional measuring device, converting the obtained point cloud data into CAD data, and comparing the data before and after repair. . According to this technology, it is possible to grasp the actual shape after repair including irregularities due to grinding or welding beads.

JP 2013-145202 A

However, since the construction instruction range and the device operating range when actually performing repairs are instructed to the repair device according to the construction drawing obtained by the repair work support system, the determination of the construction range depends on the skill of the operator of the repair device. affected by experience. In addition, the inspection method after repair is not disclosed in Patent Document 1. Therefore, when inspection is performed after repair, the determination of the inspection range, like the determination of the construction range, depends on the skill and experience of the operator of the inspection device. affected.

SUMMARY OF THE INVENTION The present invention provides a weld inspection method and inspection apparatus capable of determining an inspection range for an actual weld shape in a short time, simply, and accurately without depending on the skill and experience of an inspector. for the purpose.

In order to solve the above-mentioned problems, one of the representative weld inspection methods of the present invention is a weld inspection method for inspecting a weld of a tubular component using an inspection apparatus, comprising:
a first step of measuring a three-dimensional shape of the tubular part after welding, and acquiring three-dimensional shape data of the tubular part;
a second step of acquiring three-dimensional CAD data including three-dimensional position information regarding the design shape of the tubular part and the construction position of the weld;
a third step of performing data processing so as to associate the three-dimensional CAD data with the three-dimensional shape data, and determining an inspection range based on three-dimensional position information regarding the welding position;
and a fourth step of inspecting the welded portion of the tubular part by the inspecting device based on the determined inspection range and obtaining inspection data.

According to the present invention, there is provided a method and apparatus for inspecting a welded portion that can easily and accurately determine the inspection range for the actual shape of the welded portion in a short period of time without depending on the skill and experience of the inspector. can provide.
Problems, configurations, and effects other than those described above will be clarified by the following description of the embodiments.

FIG. 1 is a plan view showing the structure of a railway bogie frame, which is an object to be inspected in an embodiment of the present invention. FIG. 2 is a view of the horizontal beam viewed in the axial direction, and shows a state in which the flexible array sensor is installed along the circumferential direction on the inner surface of the horizontal beam. FIG. 3 is a perspective view of a flexible array sensor. FIG. 4 is a diagram showing a processing flow for inspecting railroad bogie frame lateral beam welds using a flexible array sensor. FIG. 5 is a diagram schematically showing the actual shape point cloud data of the horizontal beam pipe with broken lines. FIG. 6 is a diagram showing the shape based on the extracted three-dimensional CAD data of the horizontal beam. FIG. 7 is a diagram showing visualized three-dimensional CAD data relating to the sensor section of the inspection apparatus. FIG. 8 is a diagram showing a state in which the three-dimensional CAD data of the cross beam, the CAD data of the sensor unit of the inspection device, and the actual shape point cloud data are superimposed on each other in the virtual space. FIG. 9 is a diagram showing an example of a format in which digital data is stored, visualized by a display unit, for example.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing the structure of a railroad bogie frame, which is an object to be inspected in this embodiment. A bogie frame 1 has a structure in which horizontal beams (also called horizontal beam pipes) 2, which are two long tubular members (tubular parts), and two side beams 3, which are also long members, are combined in orthogonal directions as a skeleton. there is The cross beam 2 and the side beam 3 are welded together. Furthermore, a top plate 4 for fixing various fittings and a brake device mounting seat 5 for mounting a brake device are welded to the cross beam 2 . In general, many other members are welded to the bogie frame 1, but they are omitted here.

The cross beam 2 is a tubular elongated member generally having a diameter of several tens of centimeters and a length of several meters. In this embodiment, flaw detection is performed from the inner surface of the horizontal beam using a flexible array sensor. As described above, since there is no obstacle on the inner surface of the horizontal beam during flaw detection, flaw detection can be performed efficiently.

FIG. 2 shows a state in which the flexible array sensor 200 is installed on the inner surface of the horizontal beam 2 along the circumferential direction. FIG. 2 is a diagram of the horizontal beam 2 viewed from the axial direction, and shows an example in which a welded portion 201 between the horizontal beam 2 and the side beam 3 is an object to be inspected. The flexible array sensor 200 incorporates a plurality of piezoelectric elements 202 inside, and ultrasonic waves are transmitted from each piezoelectric element 202 at desired timings. Adjacent element surfaces of the plurality of piezoelectric elements 202 are connected to each other so as to be tiltable, so that the plurality of element surfaces can face the inner surface 204 so as to follow the curvature of the inner surface 204 of the horizontal beam 2.

As the type of array sensor, a composite type is preferable. Composite-type sensors are made by filling the gaps between multiple piezoelectric materials cut into a lattice shape with epoxy resin and solidifying it into a block. It can be made to remain. Both sides of the block are plated for electrical connections and cut into shapes such as squares, rectangles, or circles to fit the type of sensor to be incorporated.

These multiple piezoelectric elements 202 are individually controlled by voltage signals sent from the transmitting/receiving unit 205, and each element deforms and vibrates at a timing corresponding to the voltage signal, and the material in contact with the element is heated. It is possible to vibrate and, as a result, generate and transmit ultrasonic waves. In the receiving process, the process is the reverse of the transmitting process. The displacement of the element due to the ultrasonic waves is converted into a voltage signal, and the signal is sent to the transmitting/receiving unit 205, whereby the received waveform is recorded in the recording/signal processing unit 206. It is recorded, subjected to appropriate signal processing, and the result is displayed on the display unit 207 . Although omitted in FIG. 2, since the transmission/reception processing is performed for each element of the piezoelectric element 202, the received waveform is also recorded for each element.

Appropriate signal processing here means, for example, recording received waveforms corresponding to the combination of all elements using the above-mentioned full matrix capture method, and applying those received waveforms to total focusing method (TFM method), aperture synthesis method, etc. It is the process of visualizing with Through these processes, an ultrasonic flaw detection image corresponding to a tomographic image on a plane parallel to the element arrangement direction of the flexible array sensor 200 is obtained, and the image is displayed on the display unit 207 . If a recent computer is used, these processes are performed in a sufficiently short time, so the image is immediately updated according to the position of the flexible array sensor 200 and displayed on the display unit 207 .

Of course, the received waveform may be visualized by other visualization techniques. Also, instead of individually recording the received waveform of each element, it may be recorded by a phased array method according to an appropriate delay time pattern, and a visualized image may be displayed.

FIG. 3 illustrates an inspection system including flexible array sensor 200 . In the flexible array sensor 200, guide pieces 203a and 203b having the same curvature as that of the lateral beam 2 are provided on both sides in the width direction of the element surface 303, and have an easily attachable/detachable structure. Details of the structure will be described later. Although not shown, the surface of the element surface 303 is provided with a thin protective material made of resin or gel.

The flexible array sensor 200 is urged from the back side of the element surface 303 by a pressing mechanism 304 containing a spring or the like, and is placed in close contact with the inner surface 204 (FIG. 2) of the horizontal beam 2 . The flexible array sensor 200 is connected to a moving mechanism 305 via a pressing mechanism 304, and the moving mechanism 305 moves an extension arm 306, which also serves as a moving track, according to a control signal from the recording/signal processing unit 206. It is possible to scan the flexible array sensor 200 in the axial direction of the horizontal beam 2 .

The flexible array sensor 200, the pressing mechanism 304, the moving mechanism 305, and the extension arm 306 constitute a sensor unit, and the transmitting/receiving unit 205, recording/signal processing unit 206, and display unit 207 constitute a signal control unit. An inspection device is configured by the sensor section and the signal control section.

The guide 203 may be an integral type or may be divided into a plurality of sections. In this embodiment, the guide 203 includes a pair of guide pieces 203a, 203b. Guide pieces 203a and 203b show cross sections perpendicular to the direction in which the elements 202 are arranged. Grooves 401a (only one of the grooves is shown) are machined in the guide pieces 203a and 203b so as to face each other. are inserted and sandwiched, and the curved shape of the flexible array sensor 200 is maintained.

At this time, with the flexible array sensor 200 assembled, the outer surface of the guide 203 projects slightly outward from the element surface 303, so that when the guide 203 contacts the inner surface 204 of the lateral beam 2, the element The surface 303 has a structure that does not directly contact the inner surface 204 . It is desirable to secure a gap of approximately 1 mm to several mm between the element surface 303 and the inner surface 204 .

As shown in FIG. 3, the guide piece 203a is provided with contact medium supply holes 301a, 301b and 301c for supplying the contact medium to the gap between the element surface 303 and the inner surface 204. As shown in FIG. The couplant supply holes 301a, 301b and 301c communicate with the couplant supply tube 302 via the internal passage (not shown) of the guide piece 203a.

The couplant supplied from the outside through the couplant supply tube 302 using air pressure or the like is supplied from the couplant supply holes 301a, 301b and 301c. The couplant supplied to the gap between the element surface 303 and the inner surface 204 is held from both sides by the guide pieces 203a and 203b in contact with the inner surface 204 and remains there.

Here, three couplant supply holes are arranged at representative positions, but the number and positions of the couplant supply holes are applied according to the diameter of the horizontal beam pipe, the dimensions of the flexible array sensor 200, and the like. It is preferable to select . Glycerin paste or water is used as the couplant, and substances generally used for ultrasonic flaw detection can be used.

The material of the guide 203 is desirably a flexible resin, and particularly desirably an engineering plastic made of polyoxymethylene, which has excellent abrasion resistance. However, the material of the guide 203 does not necessarily have to be resin, and the effects of the present embodiment can be similarly obtained by using metal or other materials, so that may be used.

The pressing mechanism 304 may be rotatable with respect to the moving mechanism 305 to freely change the orientation of the flexible array sensor 200 with respect to the extension arm 306 . For example, in FIG. 3, the arrangement direction of the piezoelectric elements 202 of the flexible array sensor 200 is drawn so as to be orthogonal to the movement track 306. In this state, by aligning the movement track 306 with the axial direction of the horizontal beam pipe, the flexible array Circumferential flaw detection is performed while pressing the sensor 200 against the inner surface of the transverse beam pipe, and inspection is performed while moving the flaw detection position in the axial direction of the transverse beam pipe.

This makes it possible to efficiently inspect the welded portion of the long transverse beam pipe. Further, by rotating the flexible array sensor 200 together with the movement track 306 around the axis of the movement track 306 at an arbitrary position in the axial direction in the horizontal beam pipe, the flexible array sensor 200 performs circumferential flaw detection of the horizontal beam pipe. Furthermore, inspection can be performed while moving the flaw detection position in the circumferential direction of the transverse beam pipe. Although not shown in the figure, the rotation around the axis may be performed manually, or the movement track 306 may be connected to a rotating device and rotated electronically.

An inspection method for inspecting a welded portion of a railroad bogie frame lateral beam using an inspection apparatus including the flexible array sensor 200 of the present embodiment will be specifically described below. An example using a flexible array sensor as an ultrasonic sensor will be described below, but an ultrasonic sensor other than the flexible array sensor may be used. Further, a sensor for non-destructive inspection other than an ultrasonic sensor, for example, a sensor for eddy current flaw detection, can be applied in exactly the same manner. Also, the object to be inspected may be a member other than the horizontal beam of the railway bogie frame, and the same method can be applied.

FIG. 4 and FIG. 5 show a processing flow and processing details when inspecting a railroad bogie frame lateral beam weld using the flexible array sensor 200 as a representative example. The processing of steps 3 to 10 in FIG. 4 is preferably executed autonomously by the recording/signal processing unit 206, but may be executed by an external personal computer or the like connected to the recording/signal processing unit 206. An example executed by the recording/signal processing unit 206 is shown below.

(Step 1)
A flexible array sensor 200 is manually installed at a predetermined position inside the cross beam 2 . The predetermined position varies depending on the product to be inspected and the inspection apparatus, but here it is set at the initial position when the flexible array sensor 200 is moved for scanning.

(Step 2: First step)
Furthermore, using a three-dimensional measuring machine, three-dimensional shape data of the horizontal beam 2 and the flexible array sensor 200 is obtained with the flexible array sensor 200 installed at a predetermined position (measurement position) of the horizontal beam 2, and this is recorded and signaled. Stored in the processing unit 206 . A three-dimensional measuring machine is a device that converts the shape of a three-dimensional object into three-dimensional data. There is a contact type that acquires coordinates while actually touching the object with a sensor, and a non-contact type that acquires the three-dimensional shape without touching the object. There is a contact type. The contact method is a method of pressing a probe against an object, measuring the position of the contact point, and obtaining the coordinates, and generally it takes time to perform the measurement. On the other hand, the non-contact type is a method of acquiring a three-dimensional shape by analyzing the time difference and irradiation angle in which a light beam such as a laser is applied to the object and reflected, and is a light (lattice pattern) projection that measures by projecting a stripe pattern. There are methods such as a laser beam cutting method that scans an object with a slit laser beam. It is preferable to use a non-contact three-dimensional measuring machine for a large-sized structure such as a railway bogie frame, which is the object of inspection in this embodiment.

(Step 3)
The recording/signal processing unit 206 converts the three-dimensional shape data obtained in step 2 into a point group in three-dimensional space coordinates (having three-dimensional position coordinates for each point). However, when using a three-dimensional measuring machine whose output from the three-dimensional measuring machine in step 2 is a point cloud from the beginning, step 3 is executed at the same time as step 2. Through the above steps, the three-dimensional actual shape point cloud data (three-dimensional shape data of the tubular part) 500 in the state where the inspection device is installed is obtained.

FIG. 5 is a diagram schematically showing the actual shape point cloud data 500 of the transverse beam pipe with broken lines. In addition to the point cloud data 501 of the horizontal beam 2 corresponding to the point cloud data of the horizontal beam 2, the point cloud data 502 of the parts welded to the horizontal beam 2 and the point cloud data 504 of the inspection device are also shown. Also shown is point cloud data 503 of the uneven part. However, each point in the point group has only simple coordinate information, and does not have information as to which part the point belongs to.

Furthermore, FIG. 5 also shows point cloud data 505 of the extension arm inside the horizontal beam pipe and point cloud data 506 of the sensor attached to the tip thereof. Measurement may not be possible. In that case, the point cloud data of only a part of the main body of an inspection device capable of three-dimensional measurement may be used.

(Step 4: Second step and third step)
Next, the recording/signal processing unit 206 acquires the three-dimensional CAD data used for the design of the transverse beam pipe stored in advance, aligns them with the actual shape point cloud data 500, and superimposes them in the virtual space of the computer. . Either acquisition of the actual shape point cloud data 500 (first step) or acquisition of the three-dimensional CAD data 601 (second step) may be performed first.

FIG. 6 is a drawing showing the shape based on the extracted three-dimensional CAD data of the horizontal beam 2. As shown in FIG. The three-dimensional CAD data 601 of the horizontal beam 2 includes three-dimensional positional information on the shape of the horizontal beam 2 such as diameter and length, and three-dimensional positional information of a planned welding position 602 with the parts to be welded to the horizontal beam 2. is indicated by the axial length extent and the circumferential angular extent.

Alignment (association) between the three-dimensional CAD data 601 and the actual shape point cloud data 500 is performed, for example, by using alignment markers (design data for alignment markers) 603 included in the three-dimensional CAD data 601 and real shape point cloud data. This is performed by performing data processing so that the point cloud data (marker shape data) 507 of the alignment marker included in the data 500 overlaps. The alignment markers actually attached to the cross beams 2 have a unique uneven shape that does not impair the performance of the product. Can be easily extracted. Further, for alignment, in addition to the alignment marker 603 and the point cloud data 507 of the alignment marker, a characteristic shaped portion that can replace the marker, such as the end of a horizontal beam, may be used. .

(Step 5)
Furthermore, the recording/signal processing unit 206 aligns the CAD data of the inspection device with the actual shape point cloud data 500 and superimposes them in the virtual space of the computer. Here, FIG. 7 shows three-dimensional CAD data 701 relating to the sensor section of the inspection device, and the three-dimensional CAD data 701 of the inspection device includes three-dimensional position information relating to the shape of the inspection device. FIG. 7A shows three-dimensional CAD data 701a of the inspection device with the extension arm 306 contracted, and FIG. 7B shows three-dimensional CAD data 701b of the inspection device with the extension arm 306 extended. Although shown, since the portions other than the extension arm 306 have a common shape, alignment is performed using the CAD data of the inspection device other than the extension arm 306 .

Since the 3D CAD data 601 of the cross beam 2 is also aligned and superimposed on the actual shape point cloud data 500 in step 4, at this point, the 3D CAD data 601 of the cross beam 2, the 3D CAD data 701 of the inspection device, and the actual shape point cloud data 500 are superimposed in the virtual space (see FIG. 8).

(Step 6)
Next, the recording/signal processing unit 206 calculates the difference Δ between the three-dimensional CAD data 601 of the horizontal beam 2 and the actual shape point cloud data 500 . A difference Δ is obtained for each point of the actual shape point cloud data 500 . Specifically, it is obtained by calculating the distance between each point of the actual shape point cloud data 500 and the closest plane constituting the three-dimensional CAD data 601 of the cross beam 2 . From the obtained difference Δ, the distribution of the spatial deformation amount due to the weld bead, welding distortion, etc. can be grasped, and therefore the range in which the difference Δ is larger than a predetermined value can be determined as the welding range.

Since the three-dimensional CAD data 601 usually includes shape information such as a plane, a cylindrical surface, and a spherical surface, the distance between these local geometric shapes and the corresponding real shape point cloud data 500 is calculated. Alternatively, the three-dimensional CAD data 601 is first converted into a format called an STL (Stereolithography) format, which expresses a three-dimensional solid shape by a collection of triangles, and each point of the point group and the triangular plane (triangle plane containing the area) may be calculated. The range where the calculated distance is greater than the threshold can be determined as the welding range. The STL format is a general-purpose file format adopted by many CAD software.

(Step 7)
Next, the recording/signal processing unit 206 extracts a region 801 in which the difference Δ obtained in step 6 exceeds a predetermined value, and further extracts a three-dimensional welding planned position (welding execution position) 602 attached to the three-dimensional CAD data 601 . The actual inspection range 802 is determined by matching with the position information. This is because if there is a difference Δ exceeding a predetermined value in the actual shape point cloud data 500, it can be assumed to be a welded portion. However, even in a region where the difference Δ exceeds a predetermined value, if the distance from the planned welding position 602, which is attached to the three-dimensional CAD data 601 and whose three-dimensional position information is known, exceeds the specified distance, Since it can be determined that the difference Δ is not due to welding, it is excluded from the inspection range candidates. That is, a range in which the difference Δ exceeds a predetermined value and is within a specified distance from the welding target position 602 is determined as a welding range, and is set as an inspection range 802 to be inspected. Thereby, the scanning range RS of the flexible array sensor 200 is determined, and the welding range can be accurately scanned.

(Step 8)
The inspection range 802 obtained in step 7 is stored in the signal control section (recording/signal processing section 206) of the inspection apparatus. However, when the inspection range 802 is determined by an external personal computer or the like, it is input to the recording/signal processing section 206 . The input may be performed manually by the operator, or the data may be automatically transferred on the system. According to the inspection range 802, the recording/signal processing unit 206 calculates the scanning range RS of the sensor unit in the virtual space based on the three-dimensional CAD data 701 of the inspection apparatus with the inspection apparatus as a reference.

(Step 9)
Based on the inspection range (calculated scanning range RS) set in step 8, the sensor unit of the inspection device is moved in the axial direction within the horizontal beam 2 in accordance with the control signal from the recording/signal processing unit 206 for flaw detection inspection. (fourth step), and the inspection data D (received waveform) obtained within the inspection range is recorded in the memory in the recording/signal processing unit 206 as a digital value.

(Step 10)
The three-dimensional CAD data 601 of the cross beam 2, the three-dimensional CAD data 701 of the inspection device, the actual shape point cloud data 500, and the inspection data D obtained so far are all stored as a set in the recording/signal processing unit 206. , can be visualized via the display unit 207 as needed. By doing this for each weld on each product, the traceability of each product is preserved.

FIG. 9 shows an example of a format in which these digital data are stored, visualized by the display unit 207, for example. On the display screen 901 of the display unit 207, an image 906 displayed based on the three-dimensional CAD data of the cross beam 2 is displayed together with the welding part 902, and a flaw detection analysis image 907 is displayed. An area 904 surrounding internal defects 903a, 903b, and 903c is shown in a flaw analysis image 907 as an example. Information 905 (inspection object name, inspection lot number, weld part name to be inspected, inspection date, inspector code, inspection result, etc.) input in association with the digital data is also displayed on the display screen 901. . These data and information are stored in the recording/signal processing unit 206 while being associated with each other.

As described above, the processing of the present embodiment makes it possible to determine the inspection range for the actual weld shape easily and accurately in a short time without depending on the skill and experience of the inspector. It is possible to provide the inspection method and inspection apparatus of

1: bogie frame, 2: lateral beam, 3: side beam, 4: upper plate, 5: brake device mounting seat, 200: flexible array sensor, 201: welding part, 202: piezoelectric element, 203: guide, 203a, 203b: Guide piece 204: Inner surface of horizontal beam 205: Transmission/reception unit 206: Recording/signal processing unit 207: Display unit 301a, 301b, 301c: Contact material supply hole 302: Contact material supply tube 303: Element surface 304 : Pressing mechanism, 305: Moving mechanism, 306: Extension arm, 500: Actual shape point cloud data, 501: Horizontal beam point cloud data, 502: Part point cloud data, 503: Point cloud data of irregularities due to weld beads, 504: point cloud data of inspection device, 505: point cloud data of extension arm, 506: point cloud data of sensor, 507: point cloud data of alignment marker, 601: three-dimensional CAD data, 602: planned welding position, 603: Alignment markers, 701, 701a, 701b: CAD data of inspection device, 801: area, 802: inspection range

Claims (10)

A weld inspection method for inspecting a weld of a tubular part using an inspection device, comprising:
a first step of measuring a three-dimensional shape of the tubular part after welding, and acquiring three-dimensional shape data of the tubular part;
a second step of acquiring three-dimensional CAD data including three-dimensional position information regarding the design shape of the tubular part and the construction position of the weld;
a third step of performing data processing so as to associate the three-dimensional CAD data with the three-dimensional shape data, and determining an inspection range based on three-dimensional position information regarding the welding position;
and a fourth step of inspecting the welded portion of the tubular part by the inspection device based on the determined inspection range to obtain inspection data. In the method for inspecting a weld according to claim 1,
wherein the three-dimensional CAD data includes design data for alignment markers in the design shape of the tubular part;
extracting marker shape data corresponding to the alignment marker of the tubular part from the three-dimensional shape data;
In the third step, the three-dimensional CAD data and the three-dimensional shape data are associated so that the design data of the positioning marker of the three-dimensional CAD data and the marker shape data of the three-dimensional shape data overlap. A method for inspecting a weld, characterized by: In the method for inspecting a weld according to claim 1,
A method of inspecting a welded portion, wherein the three-dimensional shape data is point cloud data each including three-dimensional position information. In the method for inspecting a weld according to claim 3,
In the third step, the shape based on the three-dimensional CAD data and the shape based on the point cloud data are superimposed to obtain a difference in shape, and the difference exceeds a predetermined value and the welded portion A method for inspecting a welded portion, wherein a range within a specified distance from the execution position of is determined as the inspection range. In the method for inspecting a weld according to claim 3,
In the third step, the three-dimensional CAD data is converted into STL format data, and the distance between each point forming the point cloud data and a plane including each triangular region forming the STL format data is calculated. and determining, as the inspection range, a range in which the distance exceeds a threshold value and is within a specified distance from a position where the weld is to be performed. In the method for inspecting a weld according to claim 1,
The three-dimensional CAD data includes three-dimensional position information regarding the design shape of the inspection device,
A method of inspecting a welded portion, wherein the three-dimensional shape data includes shape data of the inspection device installed at a measurement position of the tubular part. In the method for inspecting a weld according to claim 1,
A method for inspecting a welded portion, wherein the inspection data, the three-dimensional CAD data, and the three-dimensional shape data are stored in association with each other. In the method for inspecting a weld according to claim 1,
A method for inspecting a welded portion, wherein the tubular part is a cross beam of a bogie frame of a railway vehicle. In the welded portion inspection apparatus used in the welded portion inspection method according to claim 1,
a signal control unit that executes the first to third steps;
and a sensor unit that performs the fourth step. In the weld inspection device according to claim 9,
The weld inspection apparatus, wherein the sensor unit includes a nondestructive inspection sensor and a moving mechanism for scanning the nondestructive inspection sensor over the inspection range on the inner surface of the tubular part.

Images