Construction method of ship radiation noise data set

A technology of noise data and radiation noise, which is applied in the field of ship radiation noise data set construction, can solve the problem of scarcity of ship radiation noise samples, and achieve the effect of overcoming the scarcity of samples

Pending Publication Date: 2021-11-12
HARBIN ENG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of scarcity of samples of ship radiation noise actually collected, and propose a construction method of ship radiation noise data set

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Construction method of ship radiation noise data set
  • Construction method of ship radiation noise data set
  • Construction method of ship radiation noise data set

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0017] Embodiment 1. A method for constructing a ship radiation noise data set described in this embodiment, the method specifically includes the following steps:

[0018] Step 1. Establishing a mathematical model of the actually collected ocean noise data, using the established mathematical model to simulate the simulated ship noise data, and using the set of the actually collected ocean noise data and the simulated ship noise data as the original data set;

[0019] Both the actual ocean noise data and the simulated ship noise data are sound files in the original data set;

[0020] Use the LSTM and CNN network to extract the features of the sound file, and verify the accuracy and missing rate of the obtained audio to detect the accuracy of the original data set; the specific process is:

[0021] (1) The first is audio feature extraction. Feature extraction mainly includes four parts, namely the number of channels, audio sampling size, audio sampling rate and total frame numbe...

specific Embodiment approach 2

[0028] Specific embodiment two, the difference between this embodiment and specific embodiment one is: the mathematical model of the actual mining ocean noise data is:

[0029] S(t)=1+[G(t)]×S x (t)+S l (t) (1)

[0030] In the formula: S(t) is the actual ocean noise data, G(t) is the modulation signal, S x (t) is the time-domain waveform of the continuum component, S l (t) is the time-domain waveform of the line spectrum component.

[0031] Other steps and parameters are the same as those in Embodiment 1.

[0032] According to the ship noise continuum characteristics and line spectrum statistical law, modeling and analysis are carried out respectively.

specific Embodiment approach 3

[0033] Specific embodiment three, this embodiment is different from one of the specific embodiments one or two in that: the described mathematical model is used to simulate the simulated ship noise data, and its specific process is:

[0034] Step 11: spectral peak frequency f 0 Affected by factors such as ship target speed, voyage depth, displacement, propeller size, water pressure, etc.; among them, Ross proposed the spectral peak frequency f 0 The empirical formula for the relationship with the target working condition is:

[0035]

[0036] Among them, B represents the number of propeller blades, σ represents the number of cavitation cells, and P 0 Represents hydrostatic pressure, ρ represents density, "~" represents related to these parameters;

[0037] Step 1 and 2: Assume hydrostatic pressure P 0 constant, the density ρ is:

[0038]

[0039] Among them, v is the target speed, H is the flight depth, J p is the running speed coefficient;

[0040] Step 13: Substi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a construction method of a ship radiation noise data set, belongs to the field of underwater acoustic signal recognition, and solves the problem of scarcity of actually-collected ship radiation noise samples. According to the invention, ship radiation noise meeting actual needs is generated by using a simulation technology according to existing actually-collected ship radiation noise data samples; simulated ship radiation noise and actually-collected ocean noise are used as an original noise data set; an LOFAR spectrum is used for preprocessing; key features are reserved; and finally sample expansion is realized by using a GAN. Therefore, more data sets are obtained to meet requirements of deep learning for large amount of data. The invention can be applied to construction of the ship radiation noise data set.

Description

technical field [0001] The invention belongs to the field of underwater acoustic signal recognition, and in particular relates to a method for constructing a ship radiation noise data set. Background technique [0002] In recent years, deep learning related technologies have achieved rapid application development in image radar, sonar and other application fields, and have made significant contributions in many application fields. In some special fields, such as the field of underwater acoustic signal recognition, due to safety concerns, the training data set may be scarce. In the field of underwater target recognition, due to confidentiality and security reasons, it is difficult to collect and produce data sets. At the same time, there is currently a lack of unified and standardized data sets in the industry. Therefore, the number of actual ship radiation noise samples is scarce. Since deep learning techniques require large datasets to guarantee high accuracy, the lack of ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F30/27G06F17/14G06F16/36G06F16/21G06N3/04G06N3/08G06F119/10
CPCG06F30/27G06N3/084G06F17/14G06F16/367G06F16/212G06F2119/10G06N3/045
Inventor 徐丽钱婧捷王海光申林山娄茹珍闫鑫李悦齐贾我欢张旭张立国
Owner HARBIN ENG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap