Underwater Acoustic Research Trends with Machine Learning: Passive SONAR Applications

양해상,이근화,추영민,김국현 한국해양공학회 2020 韓國海洋工學會誌 Vol.34 No.3

Underwater acoustics, which is the domain that addresses phenomena related to the generation, propagation, and reception of sound waves in water, has been applied mainly in the research on the use of sound navigation and ranging (SONAR) systems for underwater communication, target detection, investigation of marine resources and environment mapping, and measurement and analysis of sound sources in water. The main objective of remote sensing based on underwater acoustics is to indirectly acquire information on underwater targets of interest using acoustic data. Meanwhile, highly advanced data-driven machine-learning techniques are being used in various ways in the processes of acquiring information from acoustic data. The related theoretical background is introduced in the first part of this paper (Yang et al., 2020). This paper reviews machine-learning applications in passive SONAR signal-processing tasks including target detection/identification and localization.

Underwater Acoustic Research Trends with Machine Learning: General Background

양해상,이근화,추영민,김국현 한국해양공학회 2020 韓國海洋工學會誌 Vol.34 No.2

Underwater acoustics that is the study of the phenomenon of underwater wave propagation and its interaction with boundaries, has mainly been applied to the fields of underwater communication, target detection, marine resources, marine environment, and underwater sound sources. Based on the scientific and engineering understanding of acoustic signals/data, recent studies combining traditional and data-driven machine learning methods have shown continuous progress. Machine learning, represented by deep learning, has shown unprecedented success in a variety of fields, owing to big data, graphical processor unit computing, and advances in algorithms. Although machine learning has not yet been implemented in every single field of underwater acoustics, it will be used more actively in the future in line with the ongoing development and overwhelming achievements of this method. To understand the research trends of machine learning applications in underwater acoustics, the general theoretical background of several related machine learning techniques is introduced in this paper.

다공성 입자 매질에서 고주파 영역 음향 측정 자료와 Kramers-Krönig 관계식의 비교

양해상,이근화,성우제,Yang, Hae-Sang,Lee, Keun-Hwa,Seong, Woo-Jae 한국음향학회 2011 韓國音響學會誌 Vol.30 No.7

The necessary and sufficient condition for causality of a physical system can be expressed as Kramers-Kr$\ddot{o}$nig (K-K) relation. K-K relation for acoustic wave is a Hilbert transforms pair between dispersion equations of phase speed and attenuation. In this study, we quantitatively compare the acoustic measurements in water-saturated glass beads for the frequency ranges from 400 kHz to 1.1 MHz with the predictions of differential form of K-K relation obtained by Waters et al. For media with attenuation obeying an arbitrary frequency power law, acoustic measurements show good agreements with the predictions of Kramers-Kr$\ddot{o}$nig relation. 물리현상의 인과성에 대한 필요충분조건은 Kramers-Kr$\ddot{o}$nig (K-K) 관계식으로 표현된다. 음파에 대한 Kramers-Kr$\ddot{o}$nig 관계식은 음파의 위상속도 분산식과 감쇠계수 분산식 사이의 힐버트 변환 쌍으로 나타난다. 본 연구에서는 400 kHz-1.1 MHz의 고주파 영역에서 물이 찬 다공성 유리구슬 매질에서 측정된 p파 음속 및 감쇠계수를 Waters 등에 의해 얻어진 미분 형태의 Kramers-Kr$\ddot{o}$nig 관계식과 정량적으로 비교했다. 감쇠계수는 주파수의 거듭제곱형태를 따르며, 이때 실험값은 Kramers-Kr$\ddot{o}$nig 관계식과 비교적 정확히 일치한다.

Underwater Acoustic Research Trends with Machine Learning: Active SONAR Applications

양해상,변성훈,이근화,추영민,김국현 한국해양공학회 2020 韓國海洋工學會誌 Vol.34 No.4

Underwater acoustics, which is the study of phenomena related to sound waves in water, has been applied mainly in research on the use of sound navigation and range (SONAR) systems for communication, target detection, investigation of marine resources and environments, and noise measurement and analysis. The main objective of underwater acoustic remote sensing is to obtain information on a target object indirectly by using acoustic data. Presently, various types of machine learning techniques are being widely used to extract information from acoustic data. The machine learning techniques typically used in underwater acoustics and their applications in passive SONAR systems were reviewed in the first two parts of this work (Yang et al., 2020a; Yang et al., 2020b). As a follow-up, this paper reviews machine learning applications in SONAR signal processing with a focus on active target detection and classification.

물로 포화된 이봉 입도 분포 유리 구슬에서 P 파 고주파 특성

양해상(Haesang, Yang),이근화(Keunhwa Lee),성우제(Woojae Seong) 한국소음진동공학회 2017 한국소음진동공학회 학술대회논문집 Vol.2017 No.4

흡음재 물성치 변화에 따른 소음기 음향성능 연구

이용범,양해상,Lee, Yongbeom,Yang, Haesang 한국음향학회 2021 韓國音響學會誌 Vol.40 No.4

In this study, the acoustic performance of a dissipative silencer used in the ship with excellent performance compared to its size was predicted and analyzed using a numerical analysis method to reduce the pipe noise. To this end, the performance of the single expansion chamber-shaped silencer was verified using experimental and numerical analysis methods. The acoustic performance of the silencer was expressed using the Transmission Loss (TL), an indicator of its own performance, and the result was derived using the two-load method, which measured by changing the impedance at the end of the pipe. For the numerical analysis method, a general-purpose finite element analysis program was used, and the Delany-Bazley-Miki model with the flow resistivity of the sound absorbing material as an input parameter was applied. Finally, we compared the experimental and simulated results for each of the acoustic performances of the single expansion type and the dissipative silencer to confirm the consistency of the results, and predicted and analyzed the simulation results for four cases according to the properties of the sound absorbing material.

모드 매칭법을 이용한 다층 다공성 탄성 흠음재가 채워진 원통형 소음기의 음향투과손실 계산

이종무,양해상,성우제 한국음향학회 2022 韓國音響學會誌 Vol.41 No.4

This paper deals with the process of obtaining sound transmission loss of a cylindrical silencer lined with multi-layered poroelastic sound absorbing materials. The Biot model and the Johnson-Champoux-Allard- Lafarge (JCAL) model were used to deal with waves propagating in multi-layered poroelastic materials. The boundary conditions required for analysis of the silencer were obtained and the numerical process of finding modes was explained. A numerical experiment was conducted on the 2-layered silencer using the modes and the transmission loss converged with the first 12 modes. Finally, the mode matching method proposed in this research was validated by being compared with the results calculated from Finite Element Method (FEM) about different kinds of sound absorbing materials. 본 논문은 다층 다공성 흡음재가 채워진 원통형 소음기의 음향투과손실을 구하는 과정을 다루었다. 다층다공성 흡음재 내부에서 전파되는 파동을 다루기 위해 Biot모델과 Johnson-Champoux-Allard-Lafarge(JCAL) 모델을 이용했다. 소음기 해석에 필요한 경계조건들을 얻었고 그것들을 토대로 수치적으로 모드를 구하는 과정을 설명했다. 얻은 모드들을 이용하여 2층 소음기에 대해 수치적인 실험을 진행했으며 처음 12개의 모드만으로도 음향투과손실이 수렴함을 보였다. 마지막으로 흡음재의 종류를 바꿔가면서 음향투과손실을 계산했고 이를 유한요소법을 이용한결과와 비교함으로써 본 연구에서 제시한 모드 매칭법의 유효성을 검증했다.

희소 베이지안 학습 기법을 적용한 초생 프로펠러 캐비테이션 위치추정 연구

최하민,양해상,이석규,성우제 한국음향학회 2023 韓國音響學會誌 Vol.42 No.6

초생 프로펠러 캐비테이션에 의한 소음은 광대역 신호를 방출하는 적은 수의 음원으로 가정된다. 전통적인캐비테이션 위치 추정 방법은 정확도가 낮으며, 해상도가 낮아 인접한 음원을 구분할 수 없다는 한계점이 있다. 희소베이지안 학습 기법은 희소성을 가지는 신호에 대해 고해상도의 복원 성능을 보이는 기법으로, 전통적인 캐비테이션위치추정 방법에 비해 고해상도로 위치를 추정하는 특성을 가진다. 본 논문에서는 희소 베이지안 학습 기법을 적용한초생 프로펠러 캐비테이션 위치 추정 기법을 제안하고 실제 모형선 실험 결과를 통해 제안한 기법이 정확도 및 해상도측면에서 기존 방식보다 뛰어남을 보였다. Noise originating from incipient propeller cavitation is assumed to come from a limited number of sources emitting a broadband signal. Conventional methods for cavitation localization have limitations because they cannot distinguish adjacent sound sources effectively due to low accuracy and resolution. On the other hand, sparse Bayesian learning technique demonstrates high-resolution restoration performance for sparse signals and offers greater resolution compared to conventional cavitation localization methods. In this paper, an incipient propeller cavitation localization method using sparse Bayesian learning is proposed and shown to be superior to the conventional method in terms of accuracy and resolution through experimental data from a model ship.

동해 해역에서 양상태 잔향음 통계적 특징 분석

염수현,윤승현,양해상,성우제 한국음향학회 2022 韓國音響學會誌 Vol.41 No.4

In this study, the reverberation of a bistatic sonar operated in southeastern coast in the East Sea in July 2020 was analyzed. The reverberation sensor data were collected through an LFM sound source towed by a research vessel and a horizontal line array receiver 1 km to 5 km away from it. The reverberation sensor data was analyzed by various methods including geo-plot after signal processing. Through this, it was confirmed that the angle reflected from the sound source through the scatterer to the receiver has a dominant influence on the distribution of the reverberation sound, and the probability distribution characteristics of bistatic sonar reverberation varies for each beam. In addition, parametric factors of K distribution and Rayleigh distribution were estimated from the sample through moment method estimation. Using the Kolmogorov-Smirnov test at the confidence level of 0.05, the distribution probability of the data was analyzed. As a result, it could be observed that the reverberation follows a Rayleigh probability distribution, and it could be estimated that this was the effect of a low reverberation to noise ratio. 최근 단상태 소나의 표적탐지에 대한 한계로 인해 양상태 소나의 사용이 늘어나고 있다. 또한 소나 시스템이고해상도의 성능으로 발전하면서 잔향음의 확률분포가 다양한 형태로 나타나 이에 대한 연구가 활발히 이루어 지고있다. 본 연구에서는 2020년 7월, 동해 천해환경에서 수행된 양상태 소나의 잔향음을 분석하였다. 잔향음 센서 데이터는 Linear Frequency Modulated(LFM) 음원을 예인하는 연구선과 이로부터 1 km~ 5 km 떨어진 수평 선배열 수신기를 통해 수집되었으며 빔 형성 및 프리엠프 게인(Preamp-gain) 보상과정 등의 신호처리를 거친 후 Geo-plot을 비롯한다양한 방법으로 분석되었다. 이를 통해 음원에서 산란체를 거쳐 수신기로 반향되는 각도가 잔향음의 분포에 지배적인 영향을 미친다는 점과 빔별로 잔향음 확률분포가 달라진다는 양상태 소나 잔향음의 특징을 확인할 수 있었다. 또한, 모멘트 추정 기법을 통해 샘플로부터 K 분포 및 레일리 분포의 모수 인자를 추정하였으며 Kolmogorov-Smirnov test(K-S test) 기법을 이용하여 유의수준 0.05에서 데이터가 어느 확률분포에 일치하는 지를 분석하였다. 결과적으로잔향음이 레일리 확률분포를 따른다는 것을 관찰할 수 있었으며 이는 낮은 Reveration to Noise Ratio(RNR)의 영향임을 추정할 수 있었다.

머신러닝을 활용한 프로펠러 날개끝 캐비테이션 소음원 위치 추정 연구

홍성훈 ( Sunghoon Hong ),양해상 ( Haesang Yang ),성우제 ( Woojae Seong ) 한국국방기술학회 2022 한국국방기술학회 논문지 Vol.4 No.1

프로펠러에서 발생한 캐비테이션 소음은 선박 수중방사소음의 주요 원인으로 특히 해군 함정의 경우 잠수함에 의한 피탐가능성과 밀접한 관련이 있다. 따라서 수중방사소음을 최소화하기 위해 설계 단계부터 프로펠러의 요구성능을 추정하고 공동터널에서 모형 프로펠러 시험을 수행하고 있다. 기존에 수행된 연구에서 프로펠러 상부 선체에 센서를 매립하여 수중음향 분야에서 음원의 위치 추정 또는 환경인자 역산에 사용되는 정합장처리 기법을 적용해 캐비테이션의 위치를 추정하였는데 실선에 적용하기 위해서는 매립 과정 및 비용 등을 고려하여 매립되는 센서 수량을 최소화하는 것이 필요하다. 따라서 본 연구에서는 기존의 연구에서 낮은 정확도를 나타낸 3개의 음향 센서를 이용한 캐비테이션 위치 추정기법에 최근 다양한 분야에서 활용되고 있는 순방향 신경망 학습 기법을 적용하여 위치 추정 성능을 향상 하였다. The Cavitation noise generated from propellers is a major cause of underwater radiated noise in ships, and especially in the case of naval ships, it is closely related to the possibility of detection by submarines. Therefore, from the design stage, the required performance of the propeller is estimated, and a model propeller test is performed in a common tunnel. In the previous study, the location of cavitation was estimated by embedding the sensor in the upper hull of the propeller and applying the matching field processing technique used for estimating the source localization in the underwater acoustic field or inverting environmental factors. In order to apply the cavitation localization technique using the acoustic sensor to real ships, it is necessary to minimize the number of sensors to be installed on the hull in consideration of the process and cost. Therefore, in this study, the localization performance is improved by applying the forward neural network learning technique used in various fields to the cavitation localization technique using three acoustic sensors, which showed low accuracy in previous studies.