Sound velocity profile completion method and device based on historical data and machine learning

Abstract

The invention relates to a sound velocity profile completion method based on historical data and machine learning. The method comprises the following steps: obtaining sea area historical average data of a target sea area as first data and historical data collected by instrument equipment as second data; performing EOF analysis on the first data and the second data to obtain an EOF projection coefficient set; taking the second data and the EOF projection coefficient set as training samples for training to obtain neurons representing different reference classification information; obtaining sea area field measured data of a target sea area, and calculating and obtaining a reference neuron with the highest correlation degree with the sea area field measured data; and reconstructing a sound velocity profile according to the reference neurons and the sea area historical average data, and completing sound velocity profile completion of the target sea area. The sound velocity profile reconstruction result established through the method conforms to the disturbance law, the reconstruction result not conforming to the reality is avoided, the characteristics of the small-time-scale sound velocity profile fine structure of the target sea area are included, and the complex disturbance state in the practical situation can be reflected.

Description

technical field [0001] The invention relates to the technical field of sea area analysis, in particular to a sound velocity profile complement method and device based on historical data and machine learning. Background technique [0002] The sound velocity profile is the characteristic of the sound velocity structure in the vertical direction of seawater, which reflects the distribution of hydrological elements of temperature and salinity in the local sea area, and also has a decisive impact on sound propagation and underwater sound channel characteristics. The present invention provides a sound velocity profile complement method based on historical data and machine learning, which can estimate the sound velocity distribution of the whole sea depth when only partial depth sound velocity values ​​are measured, which can be used for marine environment monitoring and sonar systems for reference. [0003] Therefore, mastering the temporal and spatial variation characteristics o...

AI Technical Summary

Problems solved by technology

[0016] 3. Relying on pure mathematical processing such as variance to reconstruct the sound velocity profile, it is easy to produce unrealistic reconstruction results in the case of sudden changes in the sound velocity profile

Its accuracy and reliability are low, and it cannot reflect the real disturbance of the sound velocity profile, and its effect is not as good as that of method 1.

Images