Marine sound velocity profile classification method and device based on machine learning

Abstract

The invention relates to an ocean sound velocity profile classification method based on machine learning, and the method comprises the following steps: obtaining sound velocity original data of a target classification region, carrying out the preprocessing of the sound velocity original data, interpolating the sound velocity data to a standard depth point, and obtaining a to-be-classified sample set; converting the to-be-classified sample set into a sound velocity anomaly matrix; performing principal component analysis on the sound velocity anomaly matrix to obtain a projection coefficient set; and performing sound velocity profile classification on the target classification region according to the projection coefficient set. According to the method, space grids do not need to be preset in classification, and standard deep processing grids are adopted in the vertical direction, so that deviation of a classification result due to artificial grid setting is avoided; the amplitude of the disturbance mode of the whole section is taken as an input vector to reflect the disturbance characteristics of the whole section, and the amplitude value of the mode is taken as the characteristic quantity to be extracted more simply; according to the invention, the classification of the ocean sound velocity profile of the target classification area can be rapidly and effectively realized.

Description

technical field [0001] The invention relates to the technical field of ocean measurement, in particular to a method and device for classifying ocean sound velocity profiles based on machine learning. Background technique [0002] The speed of sound profile is the distribution of the speed of sound in the water body in the vertical direction, and it is an important environmental parameter that affects the propagation of ocean sound. The mastery of the speed of sound profile also directly affects the application efficiency of the underwater sonar system. Since the distribution of the sound velocity profile is actually a feature of the water state, it has continuously changing temporal and spatial characteristics in time and space, so it can be divided into different categories according to the characteristics. The classification of sound speed profiles has many practical application values: according to the temporal and spatial classification characteristics of sound speed pro...

AI Technical Summary

Problems solved by technology

There is no uniform standard for this advanced space and depth grid division, which may lead to different classification results for different people

[0018] (2) Classification often relies heavily on the gradient and other properties of part of the water layer, without overall processing of the sound velocity profile; feature extraction such as layered gradients or marginal spectral features requires additional workload

[0019] (3) The artificial intelligence algorithm adopted is relatively professional or complex, and there are many parameters that need to be set. It is difficult for general engineering and technical personnel to operate if they are not involved in these professional directions, and the method is difficult to get started.

Images