Robust passive underwater acoustic detection method for propeller

Song, Yongxing; Liu, Jingting; Cao, Linlin; Chu, Ning; Wu, Dazhuan

doi:10.1016/j.apacoust.2018.12.024

Cited by 16 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Principal Component Analysis (PCA) is a classic data dimensionality reduction method commonly used for processing, feature extraction, and visualization of high-dimensional data [ 18 ]. After time-frequency analysis, principal component analysis can be performed on the time-frequency distribution matrix of a single component modulated signal to extract its most significant time-frequency information, thereby obtaining the low-frequency characteristic modulation frequency of the modulated signal.…”

Section: Dpca-vgg16 State Recognition Modelmentioning

confidence: 99%

See 1 more Smart Citation

Traction Machine State Recognition Method Based on DPCA Algorithm and Convolution Neural Network

Yang

Pan

et al. 2023

Sensors

View full text Add to dashboard Cite

It is important to improve the identification accuracy of the operating status of elevator traction machines. The distribution difference of the time-frequency signals utilized to identify operating circumstances is modest, making it difficult to extract features from the vibration signals of traction machines under various operating conditions, leading to low recognition accuracy. A novel method for identifying the operating status of traction machines based on signal demodulation method and convolutional neural network (CNN) is proposed. The original vibration time-frequency signals are demodulated by the demodulation method based on time-frequency analysis and principal component analysis (DPCA). Firstly, the signal demodulation method based on principal component analysis is used to extract the modulation features of the experimentally measured vibration signals. Then, The CNN is used for feature vector extraction, and the training model is obtained through multiple iterations to achieve automatic recognition of the running state. The experimental results show that the proposed method can effectively extract feature parameters under different states. The diagnostic accuracy is up to 96.94%, which is about 16.61% higher than conventional methods. It provides a feasible solution for identifying the operating status of elevator traction machines.

show abstract

Section: Dpca-vgg16 State Recognition Modelmentioning

confidence: 99%

“…In addition to VGG16, commonly used CNN models also include AlexNet, ResNet, etc. [ 18 ]. The ResNet network has a deeper structure, consisting of 152 convolutional layers.…”

Section: Experiments and Analysismentioning

confidence: 99%

Traction Machine State Recognition Method Based on DPCA Algorithm and Convolution Neural Network

Yang

Pan

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Underwater acoustic target recognition mainly extracts the revolution rate of the propeller, blade number, low-frequency line spectrum and other features through ship noise DEMON spectrum and LOFAR spectrum and uses them to distinguish target types [1]. However, with the development of noise reduction technology, the noise of modern ships is getting lower and lower, and the amplitude modulation characteristics of propeller noise have not been obvious, which makes it difficult to extract the revolution rate of the propeller and the blade number [2]. Meanwhile, a large amount of low-frequency line spectrum can usually be obtained through the LOFAR spectrum of ship noise, and it is very difficult to distinguish which part of the ship the line spectrum comes from.…”

Section: Introductionmentioning

confidence: 99%

Finite element simulation analysis of vibration transmission characteristics of double shell based on ANSYS

Chen,

Ding,

Liu

et al. 2024

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Ship noise contains a large number of low-frequency line spectrum features. It is difficult to distinguish where they come from and even more difficult to identify the ship’s target types using the line spectrum. Therefore, research on ship targets’ vibration transmission characteristics and radiation noise is of great importance for underwater target recognition. From the perspective of underwater acoustic target recognition, this article establishes a simplified finite element model of the double shell using ANSYS, calculates the displacement response of the double shell under harmonic loads, and conducts modal analysis, harmonic response analysis, and transient dynamic analysis, which verifies the consistency of vibration between the inner shell and outer shell. Marine experimental data show that the low-frequency line spectrum of ship noise is related to the vibration frequency of ship main engines. It may be caused by diesel engine vibration transmitted through ship hull vibration and can be used as an auxiliary feature for underwater acoustic target recognition.

show abstract

“…Underwater research and engineering often involve underwater acoustic operations close to the seafloor, such as seafloor mapping, sub-sea gas leakage monitoring, tracking and navigation of underwater vehicles or divers, detection of submariners, and underwater acoustic communications [1,2,3,4,5,6,7,8,9,10,11,12,13,14]. Acoustic operations benefit from a high signal-to-noise ratio (SNR), which may be compromised by noise in the ocean, e.g., from transit vessel noise and sea surface noise.…”

Section: Introductionmentioning

confidence: 99%

Seafloor noise ensemble from vessel manoeuvre in the central North Sea

White

Roche

2020

Ocean Engineering

View full text Add to dashboard Cite

Anthropogenic noise, e.g., shipping noise, are usually unavoidable and radiate over a certain area, e.g., 10 km depending on the noise source level and acoustic propagation channel, which might affect seabed acoustic engineering as well as species' communication, behaviour, fitness and consequently their survival. Therefore, better understanding of these noise sources allows better prediction of performance of seafloor acoustic research, engineering, and biological environment. Measurements from a hydrophone system mounted on the seafloor in the central North Sea permit comparisons between temporal and spectral seafloor noise as a vessel transits nearby. The measured data indicates that the peak energy in the power spectral density (PSD) of seafloor noise is dominated by that of multiple vessel tonal noise. The tidal level has a strong negative correlation with the seafloor noise variance, and the vessel to hydrophone angle has a modest negative correlation with both the noise variance and kurtosis, particularly within a certain distance (6.6 km at depth 150 m here). As the vessel approaches, the seafloor noise is sensitive to the noise radiated from the vessel manoeuvre dominated at frequencies 400-900 Hz. Further, as the vessel speed increases from 2 knots to 15 knots, the vessel tonal increases the PSD by 10-20 dB at multiple narrow frequency bands (770 and 850 Hz). Results reported here advance the knowledge of seafloor acoustic sensitivity to nearby transit vessels.

show abstract

Robust passive underwater acoustic detection method for propeller

Cited by 16 publications

References 16 publications

Traction Machine State Recognition Method Based on DPCA Algorithm and Convolution Neural Network

Traction Machine State Recognition Method Based on DPCA Algorithm and Convolution Neural Network

Finite element simulation analysis of vibration transmission characteristics of double shell based on ANSYS

Seafloor noise ensemble from vessel manoeuvre in the central North Sea

Contact Info

Product

Resources

About