JPS58103661A - Signal processing unit for flaw detecting data - Google Patents

Abstract

PURPOSE:To permit recording of existing places, length and width of flaws by X-scanning the surface of a plate material at a right angle to the rolling direction, drawing out the flaw information in unit positions, quantizing the same, then processing and storing the same as the flaw information in steel plate units and making logical judgement. CONSTITUTION:A steel plate 1 which is a material to be inspected is conveyed and controlled in an arrow direction. Ultrasonic probes 2A-2D are juxtaposed at equal intervals in the rolling direction of the plate 1, and are so driven as to scan the inside of the plate 1 in the X direction at a right angle to the rolling direction. A flaw detector 3 transmits ultrasonic signals to the plate 1 and extracts flow information from the reflected waves. A quantizing device 6 classifies the information in four stages; absence of flaws, light flaws, about middle flaws and heavy flaws, and applies said information as code information to a signal processing unit 7. The unit 7 adds the information from the unit section in the X direction, and applies the number of the unit sections having flaws as 8 sets of code information to a logical judging device 8. The device 8 has a device for storing the code information to be outputted by the unit 7 for one sheet of the steel plate, decides the existing places, and the length and width of the flaws in the rolling direction of the plate 1 and feeds the same to a recorder 9.

Description

[Detailed Description of the Invention] This invention performs flaw information signal processing with high accuracy in a so-called X-scan ultrasonic automatic flaw detection device that scans an ultrasonic probe perpendicular to the main rolling direction of a plate material such as a thick steel plate. 1. In general, ultrasonic testing of thick steel plates involves two methods: an X-scanning method in which an ultrasonic probe is scanned parallel to the rolling direction of the steel plate, and an There are two scanning methods, the X-scanning method, and each has its advantages and disadvantages.

There is a close relationship between the rolling direction and the form of flaws, and it is generally known that flaws extend in the rolling direction, so it is best to detect the length of flaws using the X scanning method. For K, the X-scanning method is better because it detects flaws with sufficient detection parts.

Conventionally, there has been an automatic flaw detection device using the X-scanning method, as shown in FIG. In FIG. 1, 1xltj is a steel plate to be inspected, and is controlled to be transported and stopped in the direction of the arrow by a transport control mechanism (not shown).

Four (2mm)S(2D)Fi ultrasonic probes are arranged at equal intervals in the rolling direction of the steel plate, and the interior of the steel plate is aligned with the rolling direction by a probe block tracking mechanism device (IIC) (not shown). Scan at right angles, ie in the X direction. Probe here + (2A)
, -(21)) is explained as 4 (however, if some probes are configured as blocks, or if the number of probes is increased, or if the distance between blocks is reduced) Since the flaw detection area covered by one X scan becomes larger,
It has the ability to improve flaw detection efficiency.

(31ti flaw detection device, probe (2A) -= (
2D) controls the ultrasonic signals sent to the copper plate (11), and the probes (2A) and -(2D) detect internal flaws in the copper plate from among the reflected waves received from the steel plate (11), and the wave height is also high. This is to sample and extract flaw information.

(4A) S<4D) The above signals are transmitted and received between the probes (2A), -(2p) and the flaw detection device (3) using the Fi transmission/reception cable. F)) The Fi recording device records the flaw information extracted by the flaw detection device (3) for each probe (2D) while changing the recording speed according to the scanning speed. .

Next, we will explain the operation. A steel plate (1) is fed into the apparatus shown in FIG. 1, and the steel plate is stopped by a detection signal from the top end of the steel plate. Probes (2A) - (2D) contact the steel plate surface through the water film and move to the outside of the steel plate until they detect the top edge and 8th edge of the steel plate as shown in Figure 2 (a) K. do. After this, probe (2A)-(2D)
One flaw detection is completed by scanning perpendicularly to the rolling direction, ie, X scanning, to the N edge side of the Fi steel plate and moving to the N edge end. Next, each of the probes (2A) to (2D) is shifted in the rolling direction by one block toward the bottom side of the steel plate, and the second X-scan flaw detection is completed from the N edge end to the 8 edge end. When the flaw detection for the first part is completed, the flaw-detected part is transported, and the adjacent part is flaw-detected on the second round trip. Repeat these steps until you reach the bottom edge of the copper plate.

Simultaneously with these flaw detection operations, the flaw detection device (3)
A) 5 (2D) generates an ultrasonic signal and transmits the ultrasonic wave inside the steel plate, and extracts information on the internal flaws in the steel plate from the reflected waves that are input in response to the transmitted waves. A recording device (5) that records the magnitude of the reflected wave as analog information.
Output to.

The recording device (5) is composed of, for example, a pen recorder,
If you change the recording speed according to the scanning speed of probes (2A) 5 (2D) and record for each probe (2M) - (2D), the result will be as shown in Figure 2 (1)) 〇Figure 2 (a),
In ('b), (1 MU) - (1111) is the scanning plane of the probe on the steel plate, (1 MU) - (ID) is the probe (2 MU), 5 (2D), and 8. X scanning from the edge end to the N edge end* (1K) -= (IB) corresponds to the X scanning from the N edge end to the 8th edge end of the probe (2M) 5 (2D). (2x)S(2z) is a scratch on the steel plate, and when the probe scans this scratch, the pen recorder output is (3mm) - (3D), and the probe (2z) is the scratch on the steel plate.
(2D).

(4) is a pen recorder device, and (5) is pen recorder paper, which is fed 9 in the direction of the arrow. Pen recorder output waveform (61A) 5 (66D) and scratches (2: [) - (2Z
) is the waveform (61mu1 (62mu), (65B
) is scratched (2X) Vc, waveform (64B) e (65
B) corresponds to the flaw (2Y), and the waveform (66D) corresponds to the flaw (2z).0 In conventional automatic flaw detection, in devices that scan the probe in the X direction, the flaw information in the X scanning direction is directly analog. Since the recording paper was recorded, the presence or absence of scratches could be known, but the length in the rolling direction and the relative position of the scratches detected by each probe could not be determined just by looking at the recording paper. This was developed in order to solve the above drawbacks, and it is a steel plate flaw detection data processing device that processes and records data so that the flaw length and mutually related positions can be directly observed from the recorded flaw information. is intended to provide.

An embodiment of the present invention will be described below with reference to the drawings. Third
In Fig. 1, the same reference numerals as in Fig. 1 represent the same or phased parts. Using a window comparator, etc. to detect the scratch signal without scratches.

Classified into four levels: light, moderate, severe, and 9 severe.

The flaw information in unit categories is expressed by four code information according to the degree of flaw as shown in Table 1, and is converted into a digital signal and output.

Table 1 (7) shows a signal processing device that sequentially inputs and holds the digital flaw information of unit divisions in the X-scanning direction of the steel plate 0) sequentially output from the quantizer (6), and processes a predetermined number of continuous data, for example, 2
The flaw information from five unit categories is added for each name code, and the highest flaw level and the number of flawed unit categories are output as eight code information.

(8) is a logic judgment device, which has a memory device that can store eight code information of the scratches generated by the signal processing device (7) for one steel plate, and the probe (2 people) - (2D). The flaw information obtained by the
) 5 (IH), determine where and how long the flaw is on the steel plate (1), and connect it to the recording device (
0 (9) is a recording device that has the function of outputting to 9), and is a printing device that lays out the flaw location minus 1 length determined by the logic judgment device (8) on the actual steel plate and records it digitally. .

Next, the operation will be explained. A steel plate (1) is fed into the apparatus shown in FIG. 3, and the steel plate is stopped by a detection signal from the top end of the steel plate. The probe block (2M) -= (2D) comes into contact with the surface of the steel plate through the water film and moves to the outside of the steel plate until it detects the trough end and S edge end of the steel plate.

Thereafter, the probe block (2M)-(2D) is moved toward the N edge side of the plate at right angles in the pressure direction, that is, X scanning, to the end of the N edge, and one flaw detection is completed. Next, each of the probe blocks (2M) to (2D) is shifted one block toward the bottom of the copper plate in the rolling direction, and the second X' scan flaw detection is completed from the N edge end to the 8 edge end. When the flaw detection for one bale is completed, the inspected part is transported and 1i
The part in contact with the il should be inspected twice. Repeat these steps.

At the same time as these flaw detection operations, the flaw detection device (31a probe (2)
A) - (2D) generates an ultrasonic signal and transmits the ultrasonic wave inside the steel plate, and extracts information on the internal flaws in the steel plate from the reflected waves that are input in response to the transmitted waves. A quantization circuit (6) uses the magnitude of the reflected wave as analog information.
The zero quantization circuit (6) outputs data to the signal processing device (7), which digitizes each unit section according to the flaw level.
output sequentially. The signal processing device (7) adds a predetermined number of coded flaw information for each unit category, which is input sequentially, to a register of a signal holding circuit (not shown) of a 250 unit category, for each code. The highest level of damage in the 25 credit category.

The number of damaged unit sections is output to the logical judgment device (8) when data for 25 unit sections are collected.

The logic judgment device (8) stores the collected flaw information in 25 unit categories in the storage device (
(not shown), but the structure of the scratch information on the storage device is as follows: each part (one person) of the steel plate (1) - (IH
) and connect it to the steel plate (1) and memorize it so that you can understand the length and position of the flaw on the steel plate (1).

To explain the implementation method step by step, the signal processing device (
The flaw information summarized in 25 unit categories from 7) is composed of flaw information per unit category for each probe (2A) and -(2D) as shown in Figure 4. n) I (
H)-(In) I (B1)-(Bn), (11
)−=(Fn) I(01) , −(On) , (G
1) 5(Gn), (DI) −(Dn), (
Hl) - = (Hn), which is sequentially inputted into the computer memory by the logical judgment device (8). -The input flaw information is stored in the computer memory for each part of the steel plate (1) as shown in Convert to the developed steel plate image corresponding to 0 This is done for one steel plate (1).

Steel plate on computer memory converted to steel plate image (1)
Based on the data for one sheet, the logical judgment device (8) can determine the width and length of the scratch.

The width and length of the wound can be determined by using the computer memory data shown in Figure 5 (Asu * (”) * (”) (IF
F) , (01)..., (A2) I (15)
I (B2) I (F5) , (02)...,
(Mu') * (El)t (”s) s (Fse
By extracting the flaw information in the order of ('6)... and determining the continuity of the flaw, the length of the flaw in the Y direction can be determined. Again (I can’t do it.

(A2), (Mu5), (A4), (A5)
...(B1), (B2), (B3).

(B4) The width of the scratch in the X direction can be determined from K by extracting the scratch information in the order of (B5) and determining the continuity of the scratch.

In the example shown in Fig. 4, the steel plate is divided into 6 in the X direction, but this is determined by the number of probes of the unit division size, the width of the steel plate, etc. Not exclusively.

In the recording device (9), a rectangular shape of the entire surface of a steel plate is laid out on printing paper so that it can be seen at a glance where on the rectangle, and how wide and how long the flaw is. to be recorded. An example of the recording is shown in FIG. 6. FIG. 6 will be explained below. (9') is a recording paper that is printed digitally at high speed and is fed in the direction of the arrow.
(9m) is steel plate information such as the name and specifications of the steel plate that was subjected to ultrasonic flaw detection. (9B) is a spot with no scratches and is recorded and output as (.). (9X) -=(9z) is a damaged area and is recorded and output as bad.

In addition, although the embodiments of the ultrasonic flaw detection apparatus for thick steel plates have been described above, the examples are not limited to ultrasonic flaw detection apparatuses.
This method is applicable to those using #, and those that process signals obtained by X scanning when performing non-destructive inspection of plate-shaped objects other than thick steel plates. Furthermore, in order to record the position and length of the scratches obtained, although the description has been made using a high-speed printing device, it is also possible to display the results on a screen such as a CRT.

As described above, with this invention, the movement of the machine uses the X scanning method, which is advantageous in terms of the flaw detection ability of the probe, while processing the flaw information to determine the length and position of flaws in the Y direction. can be calculated by logical judgment.

[Brief explanation of the drawing]

Figure 1 is a configuration diagram of a general X-scanning ultrasonic automatic flaw detection system, Figure 2 (!L) is a diagram explaining a general flaw detection method, and Figure 2 (b) shows flaws in a general system. Figure 3 shows an example of the output of The figure shows a data diagram in which the scratch data in Figure 4 is developed and connected in memory, and Figure 6 is a diagram showing an example of outputting scratches on recording paper using this method. (2) is the probe, (3)
is flaw detection equipment, (41ti flaw detection data transmission/reception cable, (
6) is a quantization circuit, (7) is a signal processing device, (8) is a logic judgment device, and (9) is a digital recording device. In addition, in the figure -- the code indicates the same or equivalent part 0 Agent Shin Kuzuno - Figure 1 Figure 4 Figure 5

Claims (1)

[Scope of Claims] An automatic flaw detection device that X-scans the surface of a plate material perpendicular to the rolling direction and extracts flaw information regarding a unit part. A quantization device that digitizes the flaw information and outputs it as primary information; and a quantization device that sequentially inputs and holds the above primary information, performs signal processing on a predetermined number of primary information units, and then outputs secondary information about the flaw. A signal processing device, a device that stores the above-mentioned secondary information for each steel plate, and a device that searches the flaw information on the storage device to determine the actual location of the flaw and the length and width of the flaw in the rolling direction. A flaw detection data signal processing device for plate material, comprising a logic judgment device and a recording device for recording the locations of flaws and the length and width of the flaws.