KR100485450B1 - Ultrasonic testing apparatus and control method therefor - Google Patents

Abstract

The present invention relates to an ultrasonic flaw test apparatus and a control method thereof. The ultrasonic flaw detector according to the present invention includes a memory for converting a signal obtained from an A1-type standard test piece (STB-A1) into a digital signal and storing it as a reference signal, and a predetermined pulse train for driving the ultrasonic probe under the control of a processor. A transmission driver for generating a signal, an ultrasonic probe for generating an ultrasonic signal according to the pulse train from the transmission driver, receiving an echo signal and outputting the echo signal to a receiver, and removing the noise from the echo signal from the ultrasonic probe to reflect the signal And a digital signal corresponding to the reflected signal from the analog and digital converter, and the reference from the memory. Perform a deconvolution calculation with the digital signal corresponding to the signal And a processor for acquiring a deconvolution waveform and determining whether a test object is defective from a shape of the deconvolution waveform.

Description

Ultrasonic Testing Machine and Control Method {ULTRASONIC TESTING APPARATUS AND CONTROL METHOD THEREFOR}

The present invention relates to an apparatus and method for ultrasonic flaw detection, and more particularly, to an apparatus and method for detecting defect signals using modeling-based deconvolution.

In general, rectangular ultrasonic flaw detection tests are widely used for defect detection of welds. However, in real industrial field, the signals obtained through the square ultrasonic test are not only defects but also signals by geometric reflectors such as the corners of the specimen, the counter bore and the weld root. Since both and defect signals are obtained together, there are many difficulties in signal analysis for quantitative evaluation of detection defects. Identifying these complex signals is essential for improving the reliability of weld ultrasonic tests.

Conventional methods of identifying such signals include preliminary methods of recognizing general information about the geometry, and considering the instructions appearing at that location as geometric reflection signals, and how the echo changes as the transducer moves when the instructions appear. There is a method of determining whether or not the relevant indication from the echo dynamics pattern. In particular, the changing shape of the echo has a problem that the results can vary greatly depending on the subjective experience and knowledge of the inspector, the working environment or the equipment settings.

In addition, when using an automatic ultrasonic flaw detector, the geometric shape and the defect can be separated from the B-scan or the C-scan indicating the shape of the defect. However, this method does not only require expensive ultrasonic flaw detection equipment, but also has a disadvantage in that it is difficult to distinguish the defects only when the inspection is completed and the geometry is complicated.

In the previous study, the inventors of the present application first identified that the signals reflected at the crack tip and corners, the counter bore and the weld root would have their own unique patterns, since they would depend on their reflecting mechanisms in order to identify the weld ultrasonic rectangular flaw signals. We have proposed a TIFD (Technique for Identification of Flaw signal using Deconvolution), which is an algorithm to find these patterns more easily and objectively.

The TIFD technique is a technique for identifying defects using a similar function after taking a deconvolution between two signals, a reference signal and a target signal. However, the existing TIFD technique required many reference signals obtained under various conditions and conditions for signal identification. Therefore, there are some limitations to the practical application.

Basic concepts and details of existing TIFDs are described in detail in the following documents.

Journal of the Korean Society for Nondestructive Testing, Vol. 22, No. 4, pp.422-429, 2002 (Identification of Crack and Geometric Reflection Signals Using Deconvolution in Ultrasonic Rectangular Scanning of Weldments)

Briefly describing the existing TIFD, if f (t) and g (t) are reference signals and target signals, the deconvolution type h (t) is expressed by Equation 1 below.

here, Is the symbol for deconvolution.

1 is a diagram illustrating a basic concept of TIFD.

As shown in FIG. 1, when the reference signal and the target signal have similarities, the deconvolution result has a sharp and impulse form, and when it is not, it has a wide spread form. In this way, it is possible to determine the type of defect from the deconvolution type.

However, an important limitation is found in the existing TIFD, which requires a large number of reference signals to apply to each target signal.

An object of the present invention is to propose an improved TIFD for determining a defect using a single reference signal.

According to the present invention for achieving the above object, the ultrasonic flaw detection apparatus, the memory for converting the signal obtained from the A1 type standard testing block (Standard Testing Block-A1) (STB-A1) into a digital signal to store as a reference signal; A transmitter for generating a predetermined pulse train for driving the ultrasonic probe under the control of a processor, an ultrasonic probe for generating an ultrasonic signal according to the pulse train from the transmitter, receiving an echo signal and outputting the echo signal to a receiver, The receiver for removing noise from the echo signal from the probe to generate a reflection signal, an analog to digital converter for converting the reflected signal from the receiver into a digital signal, and outputting the digital signal; and the reflection from the analog to digital converter A digital signal corresponding to a signal and a digital signal corresponding to the reference signal from the memory And a processor for performing a deconvolution calculation with a signal to obtain a deconvolution waveform, and determining whether a test object is defective from the shape of the deconvolution waveform.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, if it is determined that the detailed description of the related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

The present invention proposes a TIFD using ultrasonic flaw test modeling that improves previous studies without acquiring various reference signals. An improved method is to adopt the signal reflected from the curved part of the STB-A1 block as a reference signal and apply this reference signal to all target signals.

In the present invention, four kinds of target signals are applied to prove the performance of the improved technique. Geometric reflectors were applied to counterbore and edge, defect tip and spherical pores. Waveforms and deconvolution shapes in the time domain are predicted using ultrasonic test models. Experimental results show that the deconvolutional shape of the geometrically reflected signal is impulsive, and the deconvolutional shape of the signal reflected from the defect has a form with two poles. That is, the deconvolution shape obtained by the modeling-based TIFD may be a criterion for defect signal identification.

2 shows the concept of an improved TIFD according to the present invention.

As shown, the deconvolution pattern is obtained by deconvolving the four target signals with the reference signal, respectively. Here, the target signals used are obtained by ultrasonic test modeling, and are signals obtained from a counter bore, a corner, a crack, and a spherical void. The signals obtained at the counterbore and the edge are geometric reflection signals, and the signals obtained at the defect tip and the spherical pores are defect signals. The reference signal represents a signal reflected from the curved portion of STB-Al.

As mentioned above, the deconvolution shape of the geometric reflection signal and the reference signal will have an impulse shape, and the deconvolution shape of the defect signal and the reference signal will have two poles.

3 is a view for explaining the correction of the ultrasonic probe using the STB-A1 block. As shown in FIG. 3, the signal reflected from the STB-A1 block is used to calibrate the ultrasonic transducer in an actual industrial site. In the improved TIFD technique, the signal reflected from the curved portion of the STB-A1 block is used as the reference signal.

4 is a diagram illustrating a waveform obtained by experimentally calculating a signal reflected from a curved portion of an STB-A1 block. The present invention uses this signal to determine the system efficiency factor required for square ultrasonic flaw test modeling. This system efficiency factor is one of the important factors for predicting ultrasonic flaw detection signals using ultrasonic measurement models.

As mentioned above, ultrasonic flaw detection signals reflected from geometric reflectors and various defects can be calculated using an ultrasonic measurement model. This measurement model contains four elements.

1) System efficiency factor

2) the sound field of the transducer

3) Scattered sound field in defect

4) receiving model of receiving probe

In the previous study, the inventors of the present application (Hak-Jun Kim and Sung-Jin Song) developed an ultrasonic measurement model based on the multiple Gaussian beam model and the Kirchhoff approach to predict pulse echo ultrasonic flaw detection signals. In the present invention, two types of geometric reflectors (counter bores, corners) and defects (circular defects and spherical pores) are used as target signals. The ultrasonic measurement models calculated to predict the ultrasonic flaw detection signal are as follows.

Equation 2 represents an ultrasonic wave reflected from a counterbore, and Equation 3 represents an ultrasonic wave reflected from a corner, and Equation 4 represents an ultrasonic wave scattered from a circular defect. Equation (5) represents the ultrasonic waves received by being scattered in the spherical pores.

In the above equations, w is each frequency, S is the transducer area, β (w) is the system efficiency factor, and A _n and B _n are coefficients that determine the respective Gaussian range height and width. And ρ ₁ and ρ ₂ are the density of the wedge and the specimen, c ₁ and c ₂ ^s are the longitudinal wave velocity in the wedge and the transverse wave velocity in the specimen, respectively, and A (w) is the far-field sound field representing the scattering sound field at the defect. Indicates scattering amplitude. Coefficients not described herein are described in detail in the following documents.

1) Journal of the Korean Society for Nondestructive Testing, Vol. 22, No. 4, pp. 422-429, 2002 (Identification of Cracked and Geometrically Reflected Signals Using Deconvolution in Ultrasonic Rectangular Scanning of Weldments)

2) schmerr, L. W., Materials Evaluation, 58, 882-888 (2000)

5A to 5B show waveforms calculated using Equations 2 to 5, respectively. FIG. 5A shows a signal reflected from a counterbore, FIG. 5B shows a signal reflected from a corner, FIG. 5C shows a signal reflected from a circular defect, and FIG. 5D shows a signal reflected from a spherical pore. . Here, since the defect tip signal of FIG. 5C cannot be accurately calculated using the Kirchhoff approach, the first signal of the circular defect signal group is calculated.

6A to 6D show waveforms of deconvolution forms of four reflected signals and a reference signal according to the present invention. FIG. 6A shows the deconvolution result of the counterbore reflection signal and the reference signal, FIG. 6B shows the deconvolution result of the edge reflection signal and the reference signal, and FIG. 6C shows the deconstruction of the circular defect and the reference signal. 6 shows a deconvolution result of the spherical pore and the reference signal.

As shown, the deconvolution result of the counterbore reflection signal and the reference signal shows an impulse shape in the positive direction, and the deconvolution result of the edge reflection signal and the reference signal impulses in the negative direction. It is showing form. On the other hand, it can be seen that the scattered signal from the circular defect and the spherical pores and the deconvolution result of the reference signal show the shape of the anode.

From these results, it can be said that the deconvolution shape is changed by the characteristics of scattered signals. It can be seen that the impulse shape is shown in the geometric reflection signal (counterbore and edge reflection signal), and the shape of the anode is shown by the scattered signal in small defects (defect tip and spherical pore). That is, it is possible to determine whether there is a defect from the deconvolution result of the target signal and the reference signal.

<Verification by experiment>

Ultrasonic flaw test modeling was used to experimentally verify the improved TIFD technique. In this experiment, a 5 MHz transducer (Panametrics 45 ° transverse wave) is used.

7A to 7D briefly illustrate experimental apparatuses for acquiring an ultrasonic signal. FIG. 7A illustrates an experimental apparatus for acquiring a counterbore reflection signal, FIG. 7B illustrates an experimental apparatus for acquiring a corner reflection signal, FIG. 7C illustrates an experimental apparatus for acquiring a circular defect reflected signal, and FIG. 7D. Denotes an experimental apparatus for obtaining a spherical pore reflection signal.

8A to 8D illustrate reflection signals obtained from experiments and deconvolution results of the reference signal. FIG. 8A illustrates a waveform obtained by deconvolution of the counterbore reflection signal and the reference signal obtained in FIG. 7A, and FIG. 8B illustrates a waveform obtained by deconvolution of the edge reflection signal and the reference signal obtained in FIG. 7B. 8C illustrates a waveform obtained by deconvolution of the circular defect reflection signal and the reference signal obtained in FIG. 7C, and FIG. 8D illustrates a waveform obtained by deconvolution of the spherical pore reflection signal and the reference signal obtained in FIG. 7D. Indicates.

As shown, it can be seen that the impulse shape is shown in the geometric reflection signal (counterbore and edge reflection signal), and the shape of the anode is shown by the signal scattered in the small defects (defect tip and spherical pores). That is, it is possible to determine whether there is a defect from the deconvolution result of the target signal and the reference signal.

That is, when comparing FIG. 6 and FIG. 8, it can be seen that the deconvolution shape found in the modeling and the deconvolution shape obtained by the experiment match.

Looking at the configuration and operation of the present invention based on the above theoretical basis and experimental results as follows.

9 is a block diagram of an ultrasonic flaw detector according to an exemplary embodiment of the present invention.

As shown in the drawing, the ultrasonic flaw testing apparatus includes a processor 900, a ROM 901, a RAM 902, a key input unit 903, a display unit 904, a transmission driver 905, Ultrasonic transducer 906, receiving unit 907, amplifying unit 908 and the analog-digital converter 909 is configured to include.

Referring to FIG. 9, first, the processor 900 controls the overall operation of the ultrasonic flaw detector. In particular, the present invention deconvolves the reflection signal received by the ultrasonic probe 906 and the previously stored reference signal to determine whether the test object is defective. In particular, the reference signal is a signal reflected from the curved surface of the STB-A1 block as described above. The ROM 901 stores program data and initial service data for controlling the overall operation of the ultrasonic testing apparatus, and the RAM 902 stores temporary data generated according to the operation of the processor 900. In particular, the memory (ROM or RAM) stores digital data corresponding to the reference signal according to the present invention.

The key input unit 903 includes a plurality of operation keys, and transmits a key signal according to a user's operation to the processor 900. The display unit 904 displays the overall state of the ultrasonic flaw detector and input user input information, etc. on a display window (eg, a liquid crystal display window, a CRT) under the control of the processor 900. According to the present invention, the deconvolution result of the target signal and the reference signal and defects are displayed under the control of the processor 900.

The transmission driver 905 generates a predetermined pulse train for driving the ultrasonic transducer 906 under the control of the processor 900. The ultrasonic transducer 906 is driven by the pulse train from the transmission driver 905 to generate an ultrasonic signal, receives an echo signal from the test object, and outputs the echo signal to the receiver 907. The receiver 907 removes noise from the echo signal and outputs a reflected signal. The amplifier 908 amplifies and outputs the reflected signal from the receiver 907 to a predetermined size. The analog to digital converter 909 converts the analog signal from the amplifier 908 into a digital signal and outputs the digital signal to the processor 900.

Then, the processor 900 performs deconvolution by performing deconvolution calculation with the digital data corresponding to the reflected signal from the analog-digital converter 909 and the digital data corresponding to the previously stored reference signal. Acquire the waveform and the test object for defects. The waveform and defects thus obtained are output to the display unit 904 and displayed on the display window. Here, the processor 900 may be implemented as a digital signal processor (DSP) capable of high speed digital processing.

10 shows a control procedure of the ultrasonic testing apparatus according to an embodiment of the present invention.

Referring to FIG. 10, the processor 900 first checks whether an ultrasonic transducer driving key according to a user's operation is input in step 1001. If the transducer driving key is not input, the processor 900 proceeds to step 1023 to perform a corresponding mode. If the transducer driving key is input, the processor 900 proceeds to step 1003 and the transmission driver 905. The ultrasonic transducer 906 is driven by outputting a control signal. At this time, the ultrasonic transducer 906 generates ultrasonic waves and simultaneously outputs the echo signals to the receiver 907 when an echo signal is received. Then, the receiver 907 removes noise from the echo signal to output a reflected signal, the amplifier 908 amplifies the reflected signal to a predetermined size, and the A / D converter 909 amplifies the amplified signal. The reflected signal is converted into a digital signal and provided to the processor 900.

That is, the processor 900 checks whether a signal is received from the A / D converter 909. That is, it is checked whether the reflected signal according to the generation of the ultrasonic wave is received. When the reflected signal is received, the processor 900 proceeds to step 1007 and stores the reflected signal (digital signal) received from the A / D converter 900 in a predetermined area of the memory. In operation 1009, the processor 900 reads the reference signal and the reflected signal stored in advance from the memory, and in step 1011 performs deconvolution with the two signals to obtain a deconvolution waveform. As described above, the reference signal is a signal obtained from the STB-A1 block.

After obtaining the deconvolution waveform, the processor 900 checks the peak value of the waveform in step 1013, and checks whether the waveform has an impulse shape in step 1015. If it has an impulsive shape, the processor 900 proceeds to step 1017 to determine the reflected signal as a geometrically reflected signal (for example, a signal reflected from a counterbore or corner) and then proceeds to step 1021. If the waveform has a bipolar shape, the processor 900 proceeds to step 1019 and determines that the reflected signal is a signal reflected from a defect (for example, a signal reflected from a circular defect or a spherical pore) and then proceeds to step 1021. Proceed.

After determining whether the test object is defective as described above, the processor 900 displays the test result on the display window in step 1021. That is, the obtained waveform and the determination result (defect) are displayed on the display window and then terminated.

As described above, the present invention proposes an improved TIFD using ultrasonic flaw test modeling that applies one well-defined reference signal obtained from the STB-A1 block. It can be seen that the deconvolution shapes of the four kinds of target signals (counterbore, edge, circular defect, spherical pore) are examined in detail using square ultrasonic flaw modeling and its properties are determined by the respective scattering action. . The signals (counterbore, edge) simply reflected from the geometric reflector show impulse shape, and the signals scattered from small defects (defect tip, spherical pore) have the shape of anode. Verified. Therefore, by using this type of deconvolution, it is possible to clearly identify the geometric reflection signal and the signal scattered from the defect.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by those equivalent to the scope of the claims.

As described above, the present invention can exclude the subjective judgment of the operator because the defect can be identified using the modeling, which is an objective expertise in the ultrasonic flaw detection test for finding cracks in the welded part. In addition, because the results can be obtained quickly, there is an advantage in that real-time defect identification is possible.

1 is a view showing a basic concept of TIFD.

2 illustrates the concept of an improved TIFD in accordance with the present invention.

3 is a view for explaining the correction of the ultrasonic probe using the STB-A1 block.

4 is a view showing an experimentally calculated waveform of the signal reflected from the curved portion of the STB-A1 block.

5A to 5B show waveforms of a calculated counterbore reflection signal, a calculated corner reflection signal, a calculated circular defect reflection signal, and a calculated spherical pore reflection signal.

6A to 6D are waveform diagrams illustrating deconvolution forms of four reflected signals and a reference signal according to the present invention.

7a to 7d schematically show experimental apparatus for acquiring an ultrasonic signal.

8A to 8D are diagrams showing deconvolution results of reflection signals and reference signals obtained from experiments.

9 is a block diagram showing an ultrasonic flaw tester according to an embodiment of the present invention.

10 is a view showing a control procedure of the ultrasonic testing apparatus according to an embodiment of the present invention.

Claims (6)

In the ultrasonic flaw testing apparatus, A memory for converting the signal acquired from the A1 type standard test piece (STB-A1) into a digital signal and storing it as a reference signal; A transmission driver for generating a predetermined pulse train for driving the ultrasonic probe under the control of the processor; An ultrasonic probe generating an ultrasonic signal according to the pulse train from the transmission driver, receiving an echo signal and outputting the echo signal to a receiver; The receiver which removes noise from the echo signal from the ultrasonic probe to generate a reflected signal; An analog-digital converter converting the reflected signal from the receiver into a digital signal and outputting the digital signal; A deconvolution waveform is obtained by performing deconvolution calculation with a digital signal corresponding to the reflected signal from the analog-to-digital converter and a digital signal corresponding to the reference signal from the memory to obtain a deconvolution waveform. It is configured to include the processor for determining whether or not the defect of the test object from the form of, The processor determines that the reflected signal is a geometrically reflected signal when the deconvolution waveform has an impulse shape, and determines that the reflected signal is reflected from a defect when the deconvolution waveform has an anode shape. Characterized in that the device. The method of claim 1, And an amplifier for amplifying the signal from the receiver to a predetermined size and outputting the signal to the analog-digital converter. delete In the ultrasonic flaw test apparatus including a memory for converting the signal obtained from the A1 type standard test piece (STB-A1) into a digital signal and storing it as a reference signal, the method for determining whether there is a defect from the reflected signal, Generating a predetermined pulse train for driving the ultrasonic transducer, Generating an ultrasonic signal according to the pulse train and receiving an echo signal from a test object; Generating a reflected signal by removing noise from the echo signal; Converting the reflected signal into a digital signal; Obtaining a deconvolution waveform by performing deconvolution calculation with the digital signal corresponding to the reflected signal and the digital signal corresponding to the reference signal from the memory; Determining whether or not the defect of the test object from the shape of the deconvolution waveform, In the determining process, when the deconvolution waveform has an impulse shape, the reflection signal is determined as a geometric reflection signal, and when the deconvolution waveform has an anode shape, the reflection signal is determined by a signal reflected from a defect. How to feature. delete The method of claim 4, wherein And displaying the deconvolution waveform and the defect on the display window.