• The Journal of the Acoustical Society of Korea
• /
• v.40 no.4
• /
• pp.337-346
• /
• 2021

Detection of abnormal signal generally can be done by using features of normal signals as main information because of data imbalance. This paper propose an efficient method for abnormal signal detection using parallel AutoEncoder (AE) which can use features of abnormal signals as well. The proposed Parallel AE (PAE) is composed of a normal and an abnormal reconstructors having identical AE structure and train features of normal and abnormal signals, respectively. The PAE can effectively solve the imbalanced data problem by sequentially training normal and abnormal data. For further detection performance improvement, additional binary classifier can be added to the PAE. Through experiments using public acoustic data, we obtain that the proposed PAE shows Area Under Curve (AUC) improvement of minimum 22 % at the expenses of training time increased by 1.31 ~ 1.61 times to the single AE. Furthermore, the PAE shows 93 % AUC improvement in detecting abnormal underwater acoustic signal when pre-trained PAE is transferred to train open underwater acoustic data.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.33 no.2
• /
• pp.198-204
• /
• 2013

Acoustic emission(AE) signal was detected during charge and discharge of lithium ion battery to investigate relationships among cumulative count, discharge capacity, and microdamages. AE signal was received during accelerated charge/discharge cycle test of a coin-type commercial battery. A number of AE signals were successfully detected during charge and discharge, respectively. With increasing number of cycle, discharge capacity was decreased and AE cumulative count was observed to increase. Microstructural observation of the decomposed battery after cycle test revealed mechanical damages such as interface delamination and microcracking of the electrodes. These damages were attributed to sources of the detected AE signals. Based on a linear correlation between discharge capacity and cumulative count, feasibility of AE technique for evaluation of battery degradation was suggested.

• Journal of the Korean Society for Railway
• /
• v.4 no.3
• /
• pp.79-86
• /
• 2001

The signal pattern recognition method by acoustic emission signal is applied to detect and classify the defects of a journal bearing in a power plant. AE signals of main defects such as overheating, wear and corrosion are obtained from a small scale model. To detect and classify the defects, AE signal pattern recognition program is developed. As the classification methods, the wavelet transformation analysis, the frequency domain analysis and time domain analysis are used. Among three analyses, the wavelet transformation analysis is most effective to detect and classify the defects of the journal bearing..

• Journal of the Korean Wood Science and Technology
• /
• v.22 no.4
• /
• pp.43-50
• /
• 1994

본 연구는 침엽수재를 사용한 손가락결합부재의 강도 및 감쇠특성을 측정함으로서 문창틀재로의 사용가능성을 시험하기 위하여 수행하였다. 손가락결합부 시험편의 형태는 손가락부분의 길이와 경사도에 따라서 다섯가지로 제작하였다. 하중은 4개의 하중단계를 갖는 양방향휨으로 가하였으며 각각의 하중단계는 3개의 하중주기로 구성되었다. 손가락결합부의 파괴특성을 분석하기 위하여 감쇠시험 과정에 AE신호를 측정하였다. MOE는 시험편의 형태와 뚜렷한 관련을 갖고 있지 않으나, MOR은 손가락부분의 경사도가 증가될수록 감소되는 뚜렷한 관계를 나타내었다. 감쇠비는 하중단계가 증가될수록 감소되었으나 파괴직전의 단계에서는 증가되는 경향을 나타내었다. 하중방향에 대하여 수직한 방향의 손가락결합부재가 수평방향부재보다 더 높은 강도를 나타내었다. 60dB 이상의 AE신호는 목재 또는 접착층의 파괴에 의하여 발생되는 것으로 분석되었다. 완전한 파괴가 발생하는 경우에는 100dB 이상의 AE신호가 발생하였다.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.11
• /
• pp.125-129
• /
• 2004

For the diagnosis of compressor abnormal condition in air-conditioner, AE signal which is derived from wear condition, compressed air and assembly error is analyzed experimentally. The burst and continuous type AE signal occurred by metal contact and compressed air and AE raw signal of compressors were directly acquired in production line. After extracting samples according to waveforms, Early Life Test(ELT) is conducted and classified to normal and abnormal waveform. The efficient parameters of waveform pattern are investigated in time and frequency domain and the diagnosis algorithm of air-conditioner by Neural Network estimation is suggested.

• Korean Journal of Materials Research
• /
• v.5 no.1
• /
• pp.12-21
• /
• 1995

The study was performed to study the effects of strain rate on acoustics emission( AE) during bulging test in corrosive environmentsynthetic sea water. The strain rates used were in the range $4 \times 10^{-6}S^{-1}$ to $1 \times 10^{-4} \times S^{-1}$ and the parameters used to evaluate AE signal characteristics were AE hit and amplitude. It can be observed that the cumulative AE hit and average amplitude during fracture process increase highly at decreasing strain rates while the equivalent fracture strain and the crack length of circumferencial direction become decrease. The peak point of AE signal characteristic parameters approach to the first half of test. When the average amplitude per unit equivalent fracture strain was above 20dB, it was definitly observed stress corrosion cracking phenomena. Additional, we knew that the AE test had the possibility to evaluate SCC susceptibility with various strain rates.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.5
• /
• pp.529-535
• /
• 2011

In this study, evaluation of acoustic emission signals characteristics for the post weld heat treated (PWHT) multi-pass weldment and weldment was dealt. Charpy standard specimens were taken from the lowest, middle and highest regions of the weld block. Pre-crack was made using the repeated load. Four point bend and AE tests were conducted simultaneously. Regardless of the specimens, AE signals were absent within elastic region and produced in the process of plastic deformation. AE signals for all specimens were not emitted after the maximum load. Value of signal strength for the all PWHT specimens was lower than that of the weldment. Besides, relations of plastic deformation zone size and accumulated AE counts for the PWHT specimens were more simple compared with the weldment. In case of the PWHT specimen, particles on the fractured surface decreased prominently compared with the weldment due.to PWHT. From these results, it can be concluded that PWHT was effective in reducing the AE sources for the weldment.

• Composites Research
• /
• v.24 no.6
• /
• pp.49-55
• /
• 2011

The use of advanced composite materials in main structures of military and civil aircraft has been increased rapidly because of their considerable metals in high specific strength and stiffness. However, the mechanical properties of composite materials may severely degrade in the presence of damage. Especially, the high-velocity impact such as a hailstorm, and a small piece of tire or stone during high taxing, can cause considerable damage to the structures and sub-system in spite of a very small mass. However, it is not easy to detect the damage in composite plates using a single sensor or any conventional methods. In this paper, the PVDF sensors and AE sensors were used for monitoring high-velocity impact damage initiation and propagation in composite laminates. The WT(wavelet transform) is used to decompose the sensor signals. In the PVDF sensor and AE sensor signal analysis, amounts of high-frequency signals are increased when the impact energy is increased. PVDF sensor and AE sensor signal appeared similar results. This study shows how various sensing techniques can be used to characterize high-velocity impact damage of advanced composite laminates.

• Journal of the Korean Institute of Gas
• /
• v.7 no.2 s.19
• /
• pp.48-53
• /
• 2003

This study is to look at the effect for deformation of carbon steel for high-presure pipe, on the AE signals produced by tensile test. Acoustic emission(AE) has been widely used in various fields because of its extreme sensitivity, dynamic detection ability and location of growing defects. We investigated a relationship failure mode and AE signals by tensile test, From the tensile test, we could divide into four ranges of the failure modes of elastic range, yield range, plastic range before $\sigma$u, plastic range after $\sigma$u. And failure behaviors of elastic range, yield range, plastic range before $\sigma$u, plastic range after $\sigma$u could be evaluated in tensile test by AE counts, accumulation counts and time frequency analysis. It is expected to be basic data that can protect a risk according to tensile test and bending of pipe material for pressure vessel, as a real time test of AE.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.20 no.6
• /
• pp.490-500
• /
• 2000

A new technique for the source location of acoustic emission (AE) in plates whose thichness are close to or thinner than the wavelength has been studied by introducing wavelet transform de-noising technique. The detected AE signals were pre-processed using wavelet transform to be decomposed into the low-frequency, high-amplitude flexural components and the high-frequency, low-amplitude extensional components. If the wavelet transform de-noising was employed, we could successfully filter out the extensional wave component, one of the critical errors of source location in plates by arrival time difference method. The accuracy of source location appeared to be significantly improved and independent of the setting of gain and threshold, plate thickness, sensor-to-sensor distance, and the relative position of source to sensors. Since the method utilizes the flexural component of relatively high amplitude, it could be applied to very large, thin-walled structures in practice.