JPH0545347A - Automatic ultrasonic flaw detecting method - Google Patents

Abstract

PURPOSE:To reduce the time necessary for inputting/measuring the shape data before detecting a flaw and to detect the flaw properly without being disturbed by the surface roughness of a object to-be-measured having many curved surfaces or when the shape is measured in the ultrasonic flaw detection. CONSTITUTION:The shape data of a to-be-measured body 1 obtained from a shape measuring device 1 is reproduced to a smooth form by a shape processing part 12 after the interpolation of the data using the spline function is conducted, and temporarily stored by a shape memory part 13. The posture of an ultrasonic probe 5 is obtained by a locus calculating part 14 based on the reproduced shape data so that the position and direction of the probe are maintained with a constant angle and at a constant distance to an entering point of ultrasonic waves on the surface of the to-be-measured body 1. Driving of the probe is controlled by a control device 16.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically and highly accurately performing ultrasonic flaw detection on an object having multiple curved surfaces.

[0002]

2. Description of the Related Art As a method of automatically and highly accurately performing ultrasonic flaw detection on an object having a multi-curved surface using a multi-degree-of-freedom scanner, shape data of an object to be measured is previously input and stored,
Or, it is measured by using a distance sensor and stored, and based on the stored shape data, the direction and position of the ultrasonic probe at the incident point on the surface of the object to be measured are at a constant angle and at a constant distance. The method of driving and controlling the ultrasonic probe so that the ultrasonic probe is held at this point is adopted. Of these, when measuring and storing the shape data of the object to be measured using the distance sensor, the distance meter is scanned and each point (x) of the X axis and the Y axis is scanned.
_i , y _j ) (i, j = 1, 2, 3 ...) And the surface of the object to be measured is measured, and the shape function of the object to be measured z _ij = f (x
A method of obtaining _i , y _j ) is used (for example, JP-A-63-309852 and JP-A-63-30985).
3).

[0003]

In the conventional automatic flaw detection apparatus for inspecting an object having a multi-curved surface, accurate ultrasonic wave flaw detection is performed so that the movement of the probe accurately follows the shape. Attitude control is needed, and therefore
It is necessary to input or measure the data at many points with respect to the shape data that is the basis for determining the movement of the probe, and it takes a lot of time and effort to input the shape data or measure the shape. Furthermore, when inputting the shape data of the measured object using the distance sensor, if the surface of the measured object is rough, or if the distance sensor data contains a lot of disturbance, the shape of the measured object is measured. The function z _ij = f (x _i ,
y _j ) may not be a smooth function. Based on such shape data, the ultrasonic probe is held so that the position and direction of the ultrasonic probe are held at a certain angle and a certain distance at the incident point on the surface of the object to be measured. When scanning is attempted, the scanning trajectory of the ultrasonic probe does not become smooth, so the movement of the ultrasonic probe becomes unstable, and proper flaw detection cannot be performed.

The present invention has been made to solve the above problems, and reduces the time and labor required for inputting or measuring the shape data, and the surface roughness of the object to be measured or disturbance to the distance sensor. The purpose of the present invention is to smoothly drive and control the ultrasonic probe to enable appropriate flaw detection regardless of the influence of.

[0005]

In order to achieve the above object, in the automatic ultrasonic flaw detector of the present invention, the shape data of the object to be measured is previously input and stored as data of a representative point, Interpolation using a spline function is performed from the stored data to reproduce a smooth shape, and based on the reproduced shape data, the position and direction of the ultrasonic probe are set to the surface of the measured object. The ultrasonic probe is automatically driven and controlled so as to hold it at a point at a constant angle and a constant distance at the point of incidence. Further, in order to obtain the shape data of the object to be measured as data of a representative point, it is possible to measure and store the data using a distance sensor.

[0006]

As described above, in the present apparatus, by performing the shape processing using the spline function, the data of a limited number of points representing the shape of the object to be measured are interpolated and smoothed at the same time. Since it is possible to obtain data for a large number of points that can represent a shape that is close to the actual shape, if the ultrasonic probe is scanned based on the interpolated smooth shape data, accurate and stable It will be a flaw detection.

In general, a method of using a Lagrange polynomial in addition to the spline function is well known for the interpolation of discrete data, but this is intended to represent a curve by a single polynomial. When expressing the shape of an object with multiple curved surfaces by interpolation as in the technique, curves and curved surfaces that include various curvatures are generated, and therefore extra vibration such as the Runge phenomenon occurs during interpolation. In some cases, the exact shape cannot be reproduced. On the other hand, in the method using the spline function, since an attempt is made to represent a curve / curved surface by a combination of a plurality of polynomials, unnecessary vibration does not occur,
Since it is possible to represent curves and curved surfaces that include arbitrary curvature,
In this technique, it can be said that it is optimal to perform interpolation using a spline function.

When interpolation using this spline function is performed by fitting to a spline curved surface or by fitting to a spline curve in two directions perpendicular to each other, a smoother shape is reproduced. be able to.

[0009]

EXAMPLES As one example of the present invention, the shape of an object to be measured having a multi-curved surface is previously measured using a distance sensor,
An example of performing ultrasonic flaw detection for detecting a defect inside the object to be measured in water will be described with reference to the drawings.

FIG. 1 shows an outline of the measuring unit of the apparatus in this embodiment, and FIG. 2 shows an outline of the data flow and the whole.
In FIG. 1, the ultrasonic probe 5 has an x-axis, a y-axis, a z-axis, and a θ.
It is attached to a 5-axis scanner consisting of an axis and an α-axis,
Its position and posture are controlled. The distance sensor 2 is attached to the z-axis, and its position is controlled by the movements of the x-axis, the y-axis, and the z-axis. Note that the x-axis,
The y axis and the z axis are orthogonal to each other, the θ axis is attached to the z axis and rotates around the z axis, and the α axis is attached to the θ axis and rotates in a direction perpendicular to the θ axis.

Next, the operation of this apparatus will be described according to the flaw detection procedure. First, the distance sensor 2 is scanned above the device under test 1 installed in the water tank 6, and each point (x
_The distance between the surface of the measured object 1 and the distance sensor 2 at _i , y _j ) is measured, and the shape function z _ij = f (x _i ,
y _j ). FIG. 3 shows how to set the mesh of (x _i , y _j ). In this example d along the x-axis
It is assumed that M points are measured at a pitch of x and N points are measured at a pitch of dy along the y-axis.

As the distance sensor 2, a non-contact type laser range finder, an ultrasonic range finder, or a contact type displacement type can be used, and the distance data thereof is acquired by the shape measuring instrument 11. The data from the shape measuring instrument 11 is interpolated in the shape processing unit 12 by fitting to a spline curved surface or fitting to a spline curve in two directions of x and y, and converted into smooth shape data.

At this time, the pitch in the x-axis direction at the point where ultrasonic flaw detection is desired is dx2, and the pitch in the y-axis direction is dy.
2, the measured M * N data are M * (dx / dx2) in the x-axis direction and N * (dy in the y-axis direction by the interpolation using the spline function in the shape processing unit 12.
/ Dy2) shape data. When performing interpolation using this spline function, generally dx / dx
Since 2 and dy / dy2 are sufficient even if they are 10 or more, the data of the representative points to be initially measured will be 1/100 or less of the points at which ultrasonic flaw detection is finally performed.

The data from the shape processing unit 12 is temporarily stored in the shape storage unit 13 and sent to the trajectory calculation unit 14 as needed. The trajectory calculation unit 14 separates the ultrasonic probe 5 from the shape data stored in the shape storage unit 13 at each point (x _i , y _j , z _ij ) in the normal direction and at a constant distance. The coordinates (x, y, z, θ, α) of the 5-axis scanner for moving from the point to the surface of the DUT 1 are calculated. This is orbital data. From this orbital data, x
The movements of the motors for the axes, y-axis, z-axis, θ-axis, and α-axis are calculated and sent to the control device 16 to drive the 5-axis scanner.

In this way, the ultrasonic probe 5 moves so as to follow the surface of the object to be measured 1 at a constant distance from the normal direction, and ultrasonic flaw detection is possible. In the ultrasonic flaw detection, while the ultrasonic probe 5 is moving so as to follow the surface of the DUT 1 as described above, the ultrasonic flaw detector 5 uses data from the ultrasonic probe 5 at each point where flaw detection is performed. 17 takes in the flaw detection data, and the flaw detection data processing device 18 carries out CRT.
This is done by displaying the flaw detection result above.

Next, the difference between before and after execution of the present flaw detection method will be described below by taking as an example the case of flaw detection in a 100 mm × 100 mm range in a flaw detection sample including a curved surface as shown in FIG. First, the same flaw detection pitch dx2 = dy2 = 0.
Table 1 shows a comparison of the time required for shape measurement using the distance meter before and after the present invention when flaw detection is performed with 1 mm. Before carrying out the present invention, the measurement is performed at a pitch of dx = dy = 0.1 mm, which requires about 40 minutes. On the other hand, after carrying out the present invention, the measurement can be performed at a pitch of dx = dy = 1 mm, and in this case, it takes about 6 minutes. From this, it can be seen that the present invention significantly reduces the time required for shape measurement.

Next, the shape measurement pitch is set to the same dx = d.
When ultrasonic flaw detection was performed with y = 1 mm, the numbers of defects in the samples detected before and after the present invention were compared. The pitch of flaw detection is dx in the embodiment of the present invention.
2 = dy2 = 1 mm, and the ultrasonic flaw detection point is 100 x 10
0 point, and dx2 = dy2 = 0.1 mm after execution, and ultrasonic flaw detection points were 1000 × 1000 points.
The results are shown in Table 2. Twelve defects were detected in the sample before the present invention was carried out, and 35 defects were carried out after the present invention. It can be seen that the present invention significantly improved the accuracy of ultrasonic flaw detection.

In the above embodiments, the case where the ultrasonic probe always points in the direction of the normal to the surface of the object to be measured and is at a constant distance, that is, an example in which the inside of the object to be inspected is flawed is described. It is also possible to detect the surface by generating a surface wave while maintaining a constant angle and distance to the surface.

In the above embodiment, the shape data of the representative point of the object to be measured is measured using the distance sensor, the shape data is created based on this data, and the probe trajectory calculation is performed. For inputting the shape data at the representative point, shape information obtained from CAD or the like at the time of manufacturing or obtaining the object may be used. In that case, since it is sufficient to input only the data of the representative points, the labor and time for inputting the shape data are significantly reduced as compared with the conventional method that requires the shape data at all the flaw detection points.

[0020]

[Table 1]

[0021]

[Table 2]

[0022]

As described above, according to the present invention,
The shape data of the object to be measured is input and stored in advance as representative point data, the shape is reproduced from the stored data by using interpolation using a spline function, and based on the reproduced shape data. , The ultrasonic probe is automatically driven and controlled so that the position and direction of the ultrasonic probe are held at a point at a constant angle and a constant distance at the incident point on the surface of the object to be measured. Therefore, the data of the shape representative points to be input can be reduced.

The data representing the points representing the shape can also be obtained by measurement using a distance sensor. In this case, however, the number of points to be measured can be small, so that the time required for measurement can be reduced. In addition to reducing labor, the data is smoothed at the same time when the interpolation using the spline function is performed. Since data can be obtained and the ultrasonic probe can be driven smoothly by it, highly accurate and stable flaw detection can be performed.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a system measurement unit.

FIG. 2 is a schematic diagram of a data flow and the entire system.

FIG. 3 is a shape measurement diagram using a distance sensor.

FIG. 4 is a flaw detection sample.

[Explanation of symbols]

 1 Object to be Measured 2 Distance Sensor 3 α-axis Motor 4 θ-axis Motor 5 Ultrasonic Probe 6 Water Tank 11 Shape Measuring Instrument 12 Shape Processing Section 13 Shape Storage Section 14 Trajectory Calculation Section 15 Digital Computer 16 Controller 17 Ultrasonic Flaw Detector 18 Flaw detection data processor

Claims (2)

[Claims] 1. An ultrasonic probe for an object to be measured having a multi-curved surface, based on shape data stored in advance in a computer,
At the incident point on the surface of the DUT whose direction and position are,
In the method for automatic flaw detection by scanning while holding at a constant angle and at a constant distance point, the shape of the measured object is input in advance in the computer as the data of the representative point, and the data is spline function. An ultrasonic flaw detection method characterized in that a smooth shape is reproduced by performing interpolation by using, and the direction and position of the ultrasonic probe are determined and scanned based on the reproduced shape data. 2. The ultrasonic flaw detection method according to claim 1, wherein the shape of the object to be measured is measured as data of a representative point using a distance sensor and is stored in a computer.