KR100883446B1 - Defect diagnostics system and method using acoustic emission - Google Patents

Abstract

A damage detecting system and method using acoustic emission are provided to easily detect whether an inner structure of industrial facility is damaged in real time, effectively remove unnecessary noise from a collected acoustic emission signal and analyze the signal. A damage detection system using acoustic emission includes a plurality of sensing units(10) attached to different points of an inner structure to sense an object signal including an acoustic emission signal and an acoustic vibration signal, an amplifier unit(20) for filtering and amplifying the object signal, a signal processing unit(30) for processing and analyzing the object signal, and an interface unit(40) for indicating the processing result of the signal processing unit and providing a user interface environment.

Description

Defect diagnostics system and method using acoustic emission

The present invention relates to a defect diagnosis system and a method for diagnosing a defect using an acoustic emission signal (AE). More specifically, in order to grasp the abnormality of an internal structure for an industrial facility such as a tube leakage of a boiler, The present invention relates to a defect diagnosis system and a defect diagnosis method for diagnosing a defect, a degree of defect, and a position of a defect of an internal structure by using an acoustic emission signal accompanying plastic deformation of a structure, such as micro deformation.

Recently, with the rapid urbanization, industrialization and specialization of society, the necessity of high-power industrial equipment based on efficient energy distribution is increasing.In response, the construction of energy generating facilities such as boilers and generators, including energy distribution facilities such as water pipes and gas pipes, etc. It is going on.

However, as the number of construction cases of these large industrial facilities increases, there is a growing concern about stability. Especially when defects occur in the internal structure of the equipment that is concealed from the outside, the stability and efficiency are greatly threatened. . Therefore, it is very important to detect and properly deal with defects such as micro deformation and micro cracks of the internal structure for industrial facilities at the early stage.However, applying new damage does not cause any damage to the object. Non-destructive inspection, which can inspect the defects without a defect, is in the spotlight.

In general, nondestructive testing is a general technical act of investigating and judging incompleteness without damaging an object by using physical phenomena of materials. Specific examples are radiographic, ultrasonic, magnetic, and penetrant methods. And electromagnetic induction flaw detection methods. However, most of these non-destructive testing methods are used to inspect defects by applying direct and one-time energy to the object, which makes it difficult to apply to restricted internal structures. There is a limit to the identification of defects only after significant damage has occurred due to a difficult relationship.

On the other hand, recently known `` non-destructive testing method using acoustic emission (AE) '' is an acoustic emission signal propagating along the inside of the object as an elastic wave accompanying the deformation or cracking of the inspection object. AE) is a method of determining defects, which is strictly a non-destructive testing method in the ultrasonic region, but unlike other methods, it is possible to have high sensitivity and continuous inspection, as well as the size and direction of an object's structure or defect. It can be applied even if access is restricted because it is not limited to the back.

However, the acoustic emission signal, which is the analysis target of the non-destructive testing method using the acoustic emission signal, contains various noises ranging from the low frequency band of the audible frequency to the high frequency band of several MHz, which makes it difficult to accurately interpret the results. Etc. There is a problem of inferior reliability and repeatability.

Therefore, in order to diagnose defects on complex structures such as boilers, most conventional non-destructive inspection devices using acoustic emission signals use only acoustic emission signals of a specific frequency band according to the leakage of tubes. As a result, a problem that can only be performed according to the follow-up management.

In addition, the existing non-destructive inspection system using the acoustic emission signal leaks through representative measurement values (for example, alarms and alarms generated above the upper limit of dB) after the occurrence of the alarm and alarm level. As it is impossible to provide accurate information, it is not reliable at the time of maintenance after the power generation stops, resulting in economic loss, and the approximate leakage location can be identified through the alarm value around the sensor. Since it is difficult to accurately estimate the location of cracks and leaks on the structure, there is a big limitation in prompt maintenance response.

Accordingly, the present invention has been made to solve the above problems.

That is, the present invention is a defect diagnosis system and a defect diagnosis method using an acoustic emission signal, it is possible to easily detect in real time the defects of the internal structure for limited access industrial equipment, as well as from the acoustic emission signal collected in the field The purpose of this paper is to present a concrete method to effectively remove and analyze unnecessary noise, and to present a defect diagnosis system and a method for diagnosing defects that can accurately diagnose the defects, the degree of defects, and the location of defects.

In order to achieve the above object, the present invention is a defect diagnosis system for defect diagnosis of the internal structure of the industrial equipment, it is mounted at different positions of the internal structure, each crack from the acoustic wave according to the plastic deformation of the internal structure A plurality of sensing units for detecting a target signal including an acoustic emission signal by generation and an acoustic vibration signal by leakage; An amplifier unit for filtering and amplifying a target signal of the sensing unit; A signal processing unit which processes and analyzes a target signal of the amplifier unit to calculate whether the internal structure is defective, a degree of defect, and a defect position; A defect diagnosis system using an acoustic emission signal including an interface unit for displaying a calculation result of the signal processing unit and providing a user interface environment is provided.

In this case, each of the sensing unit, the waveguide is mounted to the internal structure to propagate the acoustic wave; Low frequency sensing means for collecting the low frequency signal in the 2 to 20 kHz band from the elastic wave; It characterized in that it comprises a high frequency sensing means for collecting the acoustic emission signal of the 20 ~ 500kHz band from the elastic wave, in particular, the low frequency sensing means comprises an accelerometer, the high frequency sensing means comprises an AE sensor, The amplifier unit comprises: a plurality of preamps connected one-to-one with the sensing unit to amplify and filter the target signal first; And a main amplifier for integrally connecting the plurality of preamplifiers and the signal processing unit and performing secondary amplification and filtering on the target signal, wherein the main amplifier comprises: a first broadband filter for low frequency signal processing; And a second wideband filter for high frequency signal processing.

The signal processing unit may further include: a digital signal processing unit for converting the target signal into an adaptive signal having a finite impulse through an input variable and removing noise through adaptive filtering of wavelet transform; Determining whether the defect and the degree of the defect through the analysis of the target signal, and estimating the position of the defect by positioning based on the time delay between the sensing unit, the mounting position of the sensing unit, the detection time of each sensing unit Characterized in that it comprises a calculation unit, wherein the positional determination is obtained by the correction of the estimation result by at least one of the regional expression method and the time difference of arrival method, the correction is the welding part of the internal structure and the resulting acoustic emission It is characterized in that based on the type and characteristics of the internal structure including a change in the speed of the signal.

In addition, the industrial equipment is a once-through boiler, and the sensing unit has a lower furnace, a platen superheater (platen SH), a final superheater (final SH), a cold superheater (Cold RH), and a cutter (ECO). A plurality of adjacent sensing units are arranged in a triangular shape, and the interface unit includes: a display unit which displays a calculation result of the signal processing unit; An alarm unit for emitting a warning sound according to the operation result of the signal processing unit; And a wired or wireless local terminal for mutual communication with the signal processing unit and control of the display unit and the alarm unit, wherein the interface unit further includes a remote controller communicating with the signal processing unit. It is done.

In addition, the present invention is a defect diagnosis method for the defect diagnosis of the internal structure of the industrial equipment, (a) the acoustic wave generated from different positions of the internal structure according to the plastic deformation of the internal structure, the acoustic emission signal due to crack generation And detecting a low frequency signal including an acoustic vibration signal due to leakage and a target signal in an acoustic emission signal band; (b) filtering and amplifying the target signal; (c) converting the target signal into an adaptive signal through an input variable and removing noise through adaptive filtering; (d) The target signal is analyzed to determine whether the defect is present and the degree of the defect, and the defect position is determined based on a time delay between the sensing units, a mounting position for each sensing unit, and a position indication based on a sensing time for each sensing unit. Estimating; (e) providing a method for diagnosing defects using an acoustic emission signal comprising displaying the result of the calculation to a user.

At this time, after the step (e), it characterized in that it further comprises the step of beeping in accordance with the calculation result, after the step (b) before the step (c), the user is the input variable of the step (c) And the adaptive signal of step (c) is a finite impulse signal, the adaptive filtering of step (c) includes wavelet transform, and The positional expression of step d) may include correction of the estimation result according to at least one of the regional expression method and the time difference of arrival method. In addition, the correction is characterized in that based on the type and characteristics of the internal structure, including the welded portion of the internal structure and the resulting speed change of the acoustic emission signal.

The present invention enables the accurate defect diagnosis of the object by greatly improving the objectivity and accuracy of the signal analysis while maintaining the advantages of the acoustic emission signal.

That is, the present invention shows the advantage that the high sensitivity and continuous inspection is possible by using the acoustic emission signal, and can be applied even when the access is restricted regardless of the structure of the inspection object, the size, the direction of the defect, etc. It provides a concrete way to effectively remove and analyze unnecessary noise from the collected acoustic emission signals. It improves the objectivity and accuracy of the signal analysis and adds reliability to the diagnosis results. In particular, the present invention has the effect of more accurate diagnosis of the location of the defect of the internal structure by positioning the defect through a database-based optimization correction process based on the local marking method and the time difference method.

Therefore, the present invention is more effective when applied to large industrial equipment, such as boilers, it is possible to promote the safe process by the inspection and maintenance in real time.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

Prior to the full description, the present invention collects and analyzes a target signal of a specific frequency band from a seismic wave accompanying a plastic deformation such as a micro deformation or a micro crack in an internal structure of a large industrial facility. It is characterized by diagnosing whether there is a defect, the degree of defect, and the location of the defect. At this time, the target signal includes an acoustic emission signal due to crack generation and an acoustic vibration signal due to leakage, which includes an audible signal in a low frequency band and an acoustic emission signal band in a high frequency band, hereinafter collectively referred to as an acoustic emission signal. .

Here, the acoustic emission signal is briefly described. When a stress is applied to a predetermined structure, plastic deformation occurs by moving from the elastic region to the plastic region, and when continuous stress is applied, defects such as micro deformation and micro cracks are generated and grown and destroyed. To this. During this process, the strain energy accumulated therein at each step from the firing zone is released in the form of an elastic wave, propagates along the structure and propagates as an acoustic emission signal which is an elastic wave. At this time, the acoustic emission signal, which is an acoustic wave, is attenuated according to the propagation distance, the type and the type of the medium, and thus is detected with a time difference at different positions from the initial generation point.

1 is a schematic diagram illustrating a frequency range of an acoustic emission signal, and a band that is focused on the present invention corresponds to an industrial acoustic emission signal band of 2 to 1.2 MHz.

On the other hand, the acoustic emission signal includes a longitudinal wave (A-wave, A wave) that vibrates side by side in the direction of wave propagation, and a transverse wave (S-wave, S wave) that vibrates perpendicular to the direction of wave propagation. Depending on the aspect, it is a burst sound emission signal that shows a large energy for a short time. The primary sound emission signal due to the yield, deformation, crack, breakage, phase transformation, etc. of the structure is continuously maintained. As a continuous acoustic emission signal represented by the can be divided into a secondary acoustic emission signal due to friction, leakage of the structure.

At this time, the acoustic emission signal is substantially similar to the white noise since the reflection component and other background noise are mixed in the process of generation and propagation as well as loss and noise due to attenuation, and especially the primary acoustic emission signal. In addition, in addition to the occurrence of defects, it can be caused by background noise exceeding the threshold, which causes considerable difficulty in signal analysis.

Accordingly, the present invention proposes a reliable defect diagnosis method for the internal structure of a large industrial equipment bar, by effectively removing noise from a plurality of target signals collected at different positions of the internal structure to increase the reliability of the signal analysis defect It is characterized by accurately estimating and diagnosing the defect location on the basis of the signal delay of the target signal, the collection position, the time of collection, etc.

Hereinafter, a preferable example of the defect diagnosis system according to the present invention will be described.

2 is a schematic diagram of a defect diagnosis system according to the present invention, which includes a plurality of sensing units 10, an amplifier unit 20, and a signal processing unit 30. ) And an interface unit 40.

First, the plurality of sensing units 10 are parts for collecting target signals of desired frequency bands from different positions of the internal structures for large industrial facilities, respectively, and are mounted at different positions of the internal structures to provide elastic waves according to plastic deformation. A plurality of waveguides 12 to propagate, and a sensor 14 for collecting the target signal of the frequency band from the elastic waves propagating through each waveguide 12.

In this case, the waveguide 12 refers to a general one for propagating elastic waves of microwaves or more, and a plurality of waveguides 12 are mounted in a linear or nonlinear manner along an internal structure, and the sensors 14 respond to low frequency signals in a 2 to 20 kHz band, respectively. low frequency sensing means 16, such as an accelerometer, and high frequency sensing means 18, such as an AE sensor, which responds to high frequency signals in the 20 to 500 kHz band. In this case, an AE sensor may be used as a wide bandwidth model.

3 is a schematic diagram showing the correct mounting position of the sensing units 10 when the sensing object of the defect diagnosis system according to the present invention is, for example, a perfusion boiler of a supercritical pressure one-pass type. .

As is well known, boilers are generators for generating power or industrial steam, circulating systems including drums, furnaces for burning fuel, ventilators for supplying combustion air and releasing combustion gases, and others. It consists of auxiliary equipment including air preheater. The circulation system forms a plate structure of the tube, and according to its role, a superheater (SH) for heating saturated steam and converting it to superheated steam, and a reheater for reheating steam having a lower temperature. RH) and the economizer (ECO) which preheats feedwater using the heat of combustion gas.

In order to detect the defect of the boiler, the sensing unit 10 of the defect diagnosis system according to the present invention has four locations each along a lower furnace, six locations along a platen superheater, and a final superheater (final SH). ) Are installed in 16 to 22 places, including 4 places along the 6), 6 places along the Cold RH, and 2 places along the ECO.Each sensor can detect and locate leaks more efficiently. The arrangement structure by trigonometric method is shown. That is, three adjacent sensing units are arranged in a triangular shape.

At this time, the quantity of use of the sensing unit 10 is a recommended quantity for defect diagnosis. Although the quantity of the low temperature low pressure and small width of the economizer (ECO) is reduced, the sensing unit 10 is additionally used for more accurate defect diagnosis. Of course, it can be mobilized.

Referring back to FIG. 2, the amplifier unit 20 serves to filter and amplify the target signals collected through the plurality of sensing units 10.

The amplifier unit 20 for this purpose is the same number of pre-amps (pre-AMP) 22 that is connected one-to-one correspondence to the rear end of each sensing unit 10, and one pre-amplification amplifier 22 is integrally connected It can be divided into a main amplifier (main-AMP) 24, each pre-processing amplifier 22 is built in a junction box of a predetermined type, one-to-one with the sensor unit 10 through a special cable resistant to high temperatures While connected, it may be connected to the main amplifier 24 via an RG58 cable having oil and radiation resistant characteristics. In this case, each of the pre-processing amplifiers 22 includes a high pass filter (HPF) and a low pass filter (LPF), and a band pass filter (BPF) by a combination thereof. The main amplifier 24 may include a first wideband filter 26 for low frequency signal processing and a second wideband filter 28 for high frequency signal processing to output an analog signal in real time.

Next, the signal processing unit 30 is a part for diagnosing whether there is a defect, a degree of a defect, and a defect position by appropriately processing a real-time target signal transmitted from the amplifier unit 20 to remove noise, and analyzing and calculating the same. And a digital signal processor 32 for signal conversion and noise cancellation, and a calculator 34 for signal analysis and calculation.

At this time, the digital signal processor 32 converts the target signal of the amplifier unit 20 into an adaptive signal such as a real-time digital signal by filtering the target signal with an appropriate variable, while wavelets according to time-scale. Adaptive filtering, such as transform, removes noise. To this end, the digital signal processor 32 may be provided with a digital filter such as a finite impulse response filter (FIRF), and an application having a built-in calculation algorithm may be mounted. In addition, the calculation unit 34 calculates a time delay of the target signal using a cross-correlation method, and uses at least one of a local expression method and a time difference method based on a mounting position of the sensing unit 10 and a detection time point. After the results of the first position estimation are obtained through the optimal calibration process based on the database, the final position of the defect is calculated by performing the second position estimation. The central processing unit (CPU) and Memory may be included and a dedicated application may be mounted. At this time, the memory stores various corrections according to the type and characteristics of the object, such as the welding weight and the velocity weight of the acoustic emission signal, and becomes a substantial database.

On the other hand, general internal structures develop from cracks to cracks due to stress and grow into cracks.In order to correctly diagnose defects, the target signal must be larger than noise in the microcavity generation stage. In the generation and expansion stages, the time difference between the target signals, the speed decay and the change according to the speed and path should be clear, and in the crack generation stage, other external factors should be excluded to estimate the time of the target signals.

Accordingly, the digital signal processing unit 32 of the signal processing unit 30 approximates the target signal of the amplifier unit 20 to an appropriate variable and converts it into an adaptive signal such as a digital signal of a finite impulse, while following the time-scale. The noise is removed by performing adaptive filtering such as wavelet transform. The former signal conversion corresponds to an approximation step for the interpretation of the acoustic emission signal. The latter adaptive filtering can be identified as a sampling process for removing noise. have.

Look at each of these processes.

First, the digital signal processing unit 32 of the signal processing unit 30 according to the present invention converts an analog signal into an adaptive signal suitable for a situation through appropriate parameters. The main variable for this purpose is extracted from the time domain and the frequency domain as shown in FIG. 4, and the criterion for detecting the acoustic emission signal can be summarized as shown in FIG. 6. That is, the main variables for the conversion of the target signal are RMS, Amplitude, Rising Time, Duration time, Ring down count (Count), Event, Energy, Peak Frequency, etc., and their signal definitions are shown in FIG. As a waveform diagram showing the signal definition of the setting environment for the signal processing unit 30 of the defect diagnosis system according to the present invention, once an event is determined, it is not judged as an acoustic emission signal even if the threshold value is exceeded until the end point. The definition includes the definition of the peak component of the signal. At this time, the variable and the setting environment, etc. can be appropriately selected and used by the user through the local terminal 44 of the interface unit 40 described later according to the purpose of use, thereby improving the reliability of the signal analysis. have.

Subsequently, the target signal approximated through the above-described variable and setting environment is removed with noise through adaptive filtering such as wavelet transform.

In this case, the wavelet transform is a signal transform method that appropriately adjusts the window according to time and frequency, and can efficiently remove noise through scaling and shifting. Since the wavelet transform is well known in the art, even a detailed description will be easily understood by those skilled in the art.

Next, the calculating unit 34 of the signal processing unit 30 estimates the defect position based on the time delay between the sensing units 10, the mounting position of the sensing unit 10, and the detection time of each sensing unit 10. The calculation unit checks the time delay between each sensing unit 10 using a cross-correlation, and clearly distinguishes between symmetric and asymmetric waves and estimates the speed through each time delay.

In addition, the calculation unit 34 of the signal processing unit 30 performs calculation according to the positioning method for estimating the defect location. The positioning method used in the present invention is a zone location method and a time difference of arrival method Δt. source location) and at least one primary location step and a second location location step, which is a database-based optimization correction process that reflects the corrections of the database in the estimation results obtained therefrom.

Briefly, the regional marking method of the primary positioning step is not calculated from the defect position of the internal structure as a point but is calculated as the effective measurement sensitivity range of the sensing unit, which is difficult to determine by the time difference method described below. For example, the speed of sound is applied to anisotropic materials different in the propagation direction, welding materials having the same propagation direction, and high damping materials. In addition, the time difference of arrival method obtains a defective position through the detection time of each sensing unit, that is, the time difference of arrival, and if necessary, the position detection is performed by overlapping the signal detection range of each sensing unit and the signal detection order between the sensing units. In addition, the database-based optimization correction process of the second regional marking stage reflects the corrections stored in the database, such as the speed weight of the welded portion, in the first position determination result obtained by the regional marking method and the time difference of arrival method. As a result, the accurate positioning results are estimated.

Referring to the drawings, Figure 7 is a schematic diagram showing the positioning process by the operation unit 34 of the defect diagnosis system according to the present invention, for example, when the defect diagnosis object is a tube t of the boiler, first As shown in (a), the defect positions are estimated by the local marking method and the time difference method, respectively, and the defect positions are reflected by reflecting the reflections of the database, for example, the position of the welding portion P and the resulting speed weighting factors. The final correct position of the defect is calculated by the expression. Therefore, the position determination by the calculation unit 34 proceeds to the local indexing method according to the shape and purpose of the object, the time difference of arrival method, and the database-based optimization correction process according to the type and the characteristic of the object.

Referring back to FIG. 2, the interface unit 40 of the defect diagnosis system according to the present invention displays a calculation result of the signal processing unit 30 and provides an interface environment for user convenience, and displays the calculation result. A display unit 42, such as a monitor, and the like, and a speaker or the like for promptly notifying the user by emitting a warning sound when a defect is detected or the degree of defect is severe when the calculated result is a preset reference value. And an alarm unit 46 provided.

In addition, the interface unit 40 is provided with a wired or wireless local terminal 44 which communicates with the signal processing unit 30 and can control the display unit 42 and the alarm unit 46. Computer, so that the user inputs and changes the appropriate variable and / or setting environment to the digital signal processing unit 32 of the signal processing unit 30 using the local terminal 44, and the position of the calculation unit 34. By selecting and switching the facial expression method, the defect diagnosis system according to the present invention can be used for a purpose suitable for the purpose. In addition, the interface unit 40 may be provided with a separate remote controller 48 that performs some or all of the functions of the local terminal 44, thereby allowing the user to more conveniently control and diagnose the defect diagnosis system. It can be utilized.

On the other hand, the above description is only an example for explaining a preferred aspect of the present invention, there may be various modifications. However, all these modifications should be determined that they belong to the scope of the present invention as long as it satisfies the technical spirit of the present invention, the scope of the present invention will be readily understood by those skilled in the art through the following claims. .

1 is a schematic diagram showing a frequency band of an acoustic emission signal.

2 is a schematic diagram of a defect diagnosis system according to the present invention;

Figure 3 is an installation example for the sensing unit of the defect diagnosis system according to the present invention.

Figure 4 is a schematic diagram showing the type of variables for the signal processing unit of the defect diagnosis system according to the present invention.

5 is a waveform diagram showing the signal definition of a variable for a signal processing unit of a defect diagnosis system according to the present invention;

6 is a waveform diagram showing a signal definition of a setting environment for a signal processing unit of a defect diagnosis system according to the present invention;

Figure 7 is a schematic diagram showing a position determination method of a defect diagnosis system according to the present invention.

<Name of symbols for main parts of drawing>

10: sensing unit 12: waveguide

14,16: low frequency and high frequency sensor 20: amplifier unit

22: preprocessing amplifier 24: main amplifier

30: signal processing unit 32: digital signal processing unit

34: calculator 40: interface unit

42: display unit 44: local terminal

46: alarm unit 48: remote controller

Claims (16)

delete delete delete delete delete Defect diagnosis system for defect diagnosis of internal structure for industrial facilities, A plurality of sensing units mounted at different positions of the internal structure to detect target signals including acoustic emission signals due to crack generation and acoustic vibration signals due to leakage from elastic waves according to plastic deformation of the internal structures; ; An amplifier unit for filtering and amplifying a target signal of the sensing unit; A signal processing unit which processes and analyzes a target signal of the amplifier unit to calculate whether the internal structure is defective, a degree of defect, and a defect position; In the defect diagnosis system using the acoustic emission signal including an interface unit for displaying the calculation result of the signal processing unit, and provides a user interface environment, The signal processing unit comprises: a digital signal processing unit for converting the target signal into an adaptive signal having a finite impulse through an input variable, and removing noise through adaptive filtering of wavelet transform; Determining whether the defect and the degree of the defect through the analysis of the target signal, and estimating the position of the defect by positioning based on the time delay between the sensing unit, the mounting position of the sensing unit, the detection time of each sensing unit A defect diagnosis system using an acoustic emission signal including a calculation unit. The method of claim 6, And the position indication is a defect diagnosis system using an acoustic emission signal obtained by the correction of the estimation result by at least one of a regional expression method and a time difference of arrival method. The method of claim 7, wherein The correction is a defect diagnosis system using the acoustic emission signal based on the type and characteristics of the internal structure, including the welding portion of the internal structure and the resulting speed change of the acoustic emission signal. The method of claim 6, The industrial equipment is a once-through boiler, the sensing unit is 18 ~ throughout the lower furnace, platen superheater (platen SH), final superheater (final SH), cold superheater (Cold RH), cutter (ECO) A defect diagnosis system using sound emission signals in which 26 places are mounted, and three adjacent sensing units are arranged in a triangular shape. delete delete Defect diagnosis method for defect diagnosis of internal structure for industrial facilities, (a) Acoustic emission signal and leakage caused by crack generation as elastic waves generated from different positions of the inner structure according to plastic deformation of the inner structure using a plurality of sensing units mounted at different positions of the inner structure. Detecting a low frequency signal including an acoustic vibration signal and a target signal in an acoustic emission signal band; (b) filtering and amplifying the target signal; (c) converting the target signal into an adaptive signal through an input variable and removing noise through adaptive filtering; (d) Analyzing the target signal to determine whether there is a defect and the degree of the defect of the internal structure, the time delay between the sensing units, the mounting position for each sensing unit, the position indication based on the detection time for each sensing unit Calculating a defect location of the structure; (e) a defect diagnosis method using an acoustic emission signal comprising displaying the calculation result to a user. The method of claim 12, After the step (e), the fault diagnosis method using the sound emission signal further comprising the step of emitting a warning sound according to the calculation result. The method of claim 12, After the step (b) and before the step (c), the user further comprises the step of selecting and specifying the input variable of the step (c) fault diagnosis method using a sound emission signal. The method of claim 12, The adaptive signal of step (c) is a finite impulse signal, the adaptive filtering of step (c) includes wavelet transform, and the positioning of step (d) comprises at least one of a local method and a time difference of arrival method. A defect diagnosis method using an acoustic emission signal including correction of an estimation result according to one. The method of claim 15, And the correction is a defect diagnosis method using an acoustic emission signal based on the type and characteristics of the internal structure, including the welding part of the internal structure and the resulting speed change of the acoustic emission signal.

Images