Method, device, equipment and medium for predicting complex sound field distribution in deep sea environment

Abstract

The present application relates to a method, device, equipment and medium for predicting the complex sound field distribution in a deep-sea environment. The method includes: constructing a sound propagation loss data set in a deep-sea environment as a training set; performing Gaussian filtering smoothing to obtain the global structure information and local detail information of the sound propagation loss in a deep-sea environment; constructing a double-branch U-Net neural network model; preprocessing the source depth, source frequency and sound speed profile in the training set after Gaussian smoothing; inputting the training set preprocessed with the source depth, source frequency and sound speed profile into the model for iterative training, and inputting the data of the new unknown source depth, unknown source frequency and unknown sound speed profile into the trained model after the same preprocessing steps, and predicting and outputting the corresponding sound propagation loss distribution in a deep-sea environment. Using this method can achieve efficient and high-precision prediction of the sound propagation loss distribution in a complex sound field in a deep-sea environment.

Description

Technical Field

[0001] This application relates to the technical field of signal processing, and particularly to a method, device, equipment and medium for predicting complex sound field distribution in a deep-sea environment. Background Art

[0002] Sound waves are currently the only form of radiation that can propagate over long distances in seawater, and are also the main means of underwater long-distance detection and information transmission. In a deep-sea environment, sound waves often undergo multi-path reflection, refraction and scattering due to factors such as undulating seabed topography, dynamic changes in the seawater sound speed profile, and sea surface fluctuations, forming obvious and complex interference fringes. For the simulation and prediction of such complex acoustic phenomena, traditional numerical methods usually rely on solving complex partial differential equations, with a long calculation process and huge resource consumption, making it difficult to meet the requirements of real-time deep-sea long-distance applications.

[0003] Neural network models, with their powerful non-linear fitting ability, provide a feasible way for quickly predicting complex acoustic phenomena such as interference fringes and multi-path propagation. By training on the sound field distribution under various sound source conditions and environmental parameters, the model can significantly shorten the calculation time during the inference stage and better adapt to the dynamic changes of the ocean environment. However, when existing neural network models are applied to the prediction task of the sound propagation loss distribution in a complex deep-sea long-distance sound field, due to the poor accuracy of extracting the characteristics of the sound propagation loss distribution, there is still a problem of low prediction accuracy of the sound propagation loss distribution. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, equipment and medium for predicting complex sound field distribution in a deep-sea environment, so as to achieve efficient and high-precision prediction of the sound propagation loss distribution in a complex deep-sea sound field.

[0005] A method for predicting complex sound field distribution in a deep-sea environment, the method includes:

[0006] Generating a sound propagation loss dataset in a deep-sea environment based on different sound source depths, sound source frequencies and sound speed profiles, and using this dataset as a training set;

[0007] Using a Gaussian filter to smooth the sound propagation loss in the training set in a deep-sea environment to obtain the global structure information and local detail information of the sound propagation loss in a deep-sea environment;

[0008] Construct a dual-branch U-Net neural network model for predicting sound propagation loss in the deep-sea environment. This model consists of a global feature branch and a local detail branch, both of which are in the U-Net structure. The model learns the global structural information of the input data through the global feature branch, extracts the local detail information of the input data through the local detail branch. Finally, by fusing the information of different scales learned by the two branches, it outputs the prediction result of the sound propagation loss in the deep-sea environment that takes into account both the global structural information and the local detail information.

[0009] Adopt the sound source space encoding algorithm to preprocess the sound source depth, sound source frequency, and sound speed profile in the training set after Gaussian smoothing. The preprocessed sound source depth, sound source frequency, and sound speed profile are encoded into a matrix with the same size as the predicted sound field.

[0010] Input the training set preprocessed with the sound source depth, sound source frequency, and sound speed profile into the dual-branch U-Net neural network model for iterative training until a trained dual-branch U-Net neural network model is obtained. Then, after passing the new unknown sound source depth, unknown sound source frequency, and unknown sound speed profile data through the same preprocessing steps, input them into the trained dual-branch U-Net neural network model to predict and output the corresponding sound propagation loss distribution in the deep-sea environment.

[0011] In one embodiment, generate a sound propagation loss dataset in the deep-sea environment based on different sound source depths, sound source frequencies, and sound speed profiles, including:

[0012] Take different sound source depths, sound source frequencies, sound speed profiles, terrain settings, and geoacoustic parameter settings as input parameters, and use the traditional numerical model Bellhop based on ray theory to run and generate a sound propagation loss dataset in the deep-sea environment.

[0013] In one embodiment, use the traditional numerical model Bellhop based on ray theory to run and generate a sound propagation loss dataset in the deep-sea environment, including:

[0014] Bellhop uses ray theory to simulate the propagation process of sound waves in the ocean environment and generates a sound propagation loss dataset in the deep-sea environment, including eigen sound rays and underwater sound propagation loss. Among them, the underwater sound propagation loss is calculated from the sound pressure at a certain receiver position relative to the reference signal at the sound source and is specifically expressed as:

[0015] ;

[0016] ;

[0017] where represents the number of eigen - acoustic rays contributing to the sound field at a specific receiving position, is the sound pressure caused by the th eigen - acoustic ray, and , and represent the horizontal and vertical (i.e., depth) direction coordinates of the receiving position, respectively.

[0018] In one embodiment, a sound - source spatial - encoding algorithm is adopted to pre - process the sound - source depth, sound - source frequency, and sound - speed profile in the Gaussian - smoothed training set, including:

[0019] Discretize the research area into a zero matrix , and the matrix element represents the relative position of the position at the th row and th column of the zero matrix within the area ; among them, the horizontal distance and vertical distance of the research area are and , and represent the number of rows and columns of the matrix, respectively;

[0020] For the known sound source , for the set underwater sound - source depth , search for the vertical - direction index closest to in the depth range , expressed as:

[0021] ;

[0022] Among them, ;

[0023] Normalize the sound - source depth , sound - source frequency and sound - speed profile respectively;

[0024] According to the characteristics of cylindrical coordinates, set the sound source to only appear in the first column of the zero matrix. Therefore, encode the normalized sound - source depth, sound - source frequency, and sound - speed profile into the first column of the zero matrix respectively, and corresponding to the vertical - direction index , obtain the matrix after encoding the sound - source depth, the matrix after encoding the sound - source frequency, and the matrix after encoding the sound - speed profile.

[0025] In one of the embodiments, the sound source depth , the sound source frequency and the sound speed profile are respectively normalized, including:

[0026] Within the corresponding defined ranges, the sound source depth and the sound source frequency are respectively normalized to obtain the normalized sound source depth and the normalized sound source frequency , which are respectively expressed as:

[0027] ;

[0028] ;

[0029] wherein, , and respectively represent the minimum and maximum values of the sound source depth; , and respectively represent the minimum and maximum values of the sound source frequency;

[0030] Each sound speed profile sample in the sound speed profiles is normalized to obtain the normalized sound speed profile sample , which is expressed as:

[0031] ;

[0032] wherein, represents the serial number of the sound speed profile sample, is the total number of sound speed profiles, and respectively represent the minimum and maximum values of the sound speed profile.

[0033] In one of the embodiments, the normalized sound source depth, sound source frequency, and sound speed profile are respectively encoded in the first column of the zero matrix , and corresponding to the vertical direction index , to obtain the matrix after encoding the sound source depth , the matrix after encoding the sound source frequency and the matrix after encoding the sound speed profile , which are expressed as:

[0034] ;

[0035] ;

[0036] 。

[0037] In one of the embodiments, the total loss function of the dual-branch U-Net neural network model is expressed as:

[0038] ;

[0039] ;

[0040] ;

[0041] wherein, represents the loss function of the global feature branch, represents the loss function of the local detail branch, and both represent weights used to adjust the proportion of the loss of the global feature branch and the loss of the local detail branch, represents the global structural information learned by the global feature branch, represents the label value of the global structural information, represents the local detail information extracted by the local detail branch, represents the label value of the local detail information.

[0042] A complex sound field distribution prediction device in a deep-sea environment, the device includes:

[0043] A dataset generation module, configured to generate a sound propagation loss dataset in a deep-sea environment based on different sound source depths, sound source frequencies, and sound speed profiles, and use this dataset as a training set;

[0044] A Gaussian smoothing module, configured to use a Gaussian filter to smooth the sound propagation loss in the deep-sea environment in the training set to obtain the global structural information and local detail information of the sound propagation loss in the deep-sea environment;

[0045] A model construction module, configured to construct a dual-branch U-Net neural network model for predicting the sound propagation loss in a deep-sea environment. This model consists of a global feature branch and a local detail branch, both of which are in the U-Net structure; the model learns the global structural information of the input data through the global feature branch, extracts the local detail information of the input data through the local detail branch, and finally, by fusing the information of different scales learned by the two branches, outputs a prediction result of the sound propagation loss in the deep-sea environment that takes into account both the global structural information and the local detail information;

[0046] A data preprocessing module, which is used to preprocess the sound source depth, sound source frequency, and sound speed profile in the training set after Gaussian smoothing by using a sound source spatial encoding algorithm. The preprocessed sound source depth, sound source frequency, and sound speed profile are encoded into a matrix with the same size as the predicted sound field.

[0047] A model training and prediction module, which is used to input the training set preprocessed by the sound source depth, sound source frequency, and sound speed profile into a dual-branch U-Net neural network model for iterative training until a trained dual-branch U-Net neural network model is obtained. After the new unknown sound source depth, unknown sound source frequency, and unknown sound speed profile data go through the same preprocessing steps, they are input into the trained dual-branch U-Net neural network model to predict and output the sound propagation loss distribution in the corresponding deep-sea environment.

[0048] A computer device, including a memory and a processor. The memory stores a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of the above-mentioned method for predicting the complex sound field distribution in the deep-sea environment.

[0049] A computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-mentioned method for predicting the complex sound field distribution in the deep-sea environment.

[0050] The above-mentioned method, device, equipment, and medium for predicting the complex sound field distribution in the deep-sea environment have the following beneficial effects:

[0051] 1. Using a Gaussian filter to decompose the sound propagation loss in the deep-sea environment into global structure information and local detail information improves the model's learning ability for complex interference fringe patterns, so that it can reduce noise while retaining edges and details as much as possible.

[0052] 2. Constructing a dual-branch U-Net neural network model for predicting the sound propagation loss in the deep-sea environment. This model fuses the global structure information and local detail information learned by the global feature branch and the local detail branch respectively, and finally outputs the prediction result of the sound propagation loss in the deep-sea environment that takes into account both global structure information and local detail information, realizing the accurate prediction of the sound propagation loss distribution in the deep-sea environment.

[0053] 3. Introduce a sound source spatial encoding algorithm to encode the sound source depth, sound source frequency, and sound speed profile into a matrix with the same size as the predicted sound field. The encoded matrix contains the sound source depth, sound source frequency, sound speed profile, and sound source spatial position information. Moreover, the encoded matrices can be stacked or combined in a specific manner to form a multi-channel input with rich information, ensuring the consistency of the data format with the input of the dual-branch U-Net neural network model. Meanwhile, it enables the model to fully learn the relevant features of sound propagation loss in the deep-sea environment, improving the accuracy of predicting the distribution of sound propagation loss in the deep-sea environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic flowchart of a method for predicting the complex sound field distribution in the deep-sea environment in one embodiment;

[0055] Figure 2 It is a schematic diagram for generating a sound propagation loss data set in the deep-sea environment in one embodiment;

[0056] Figure 3 It is a schematic diagram of the architecture of a dual-branch U-Net neural network model in one embodiment;

[0057] Figure 4 It is a schematic diagram for model training and prediction in one embodiment;

[0058] Figure 5 It is an internal structure diagram of a computer device in one embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0060] In one embodiment, as Figure 1 shown, a method for predicting the complex sound field distribution in the deep-sea environment is provided, including the following steps:

[0061] Step S1: Generate a sound propagation loss data set in the deep-sea environment based on different sound source depths, sound source frequencies, and sound speed profiles, and use this data set as a training set.

[0062] Among them, due to the lack of underwater sound field observation data, it is difficult to support the effective training of the neural network. The present application uses the results of traditional numerical calculations as training label values. In the simulation of underwater sound fields and acoustic channels, Bellhop is a commonly used numerical calculation model. Select the Bellhop model to generate a sound propagation loss data set in the deep-sea environment, and use this data set as a training set for model training, as Figure 2As shown in the figure, the process of generating the sound propagation loss dataset in the deep sea environment is as follows: different sound source depths, sound source frequencies, sound velocity profiles, terrain settings, and geoacoustic parameter settings are used as input parameters, and the traditional numerical model Bellhop based on ray theory is used to generate the sound propagation loss dataset in the deep sea environment. Among them, the sound velocity profile comes from the ocean temperature, salinity, and depth dataset, the terrain settings come from the terrain dataset, and the geoacoustic parameter settings come from the geoacoustic dataset.

[0063] Specifically, Bellhop uses ray theory to simulate the propagation of sound waves in the ocean environment and generates a data set of sound propagation losses in the deep sea environment, including intrinsic sound rays and underwater sound propagation losses. Among them, the underwater sound propagation loss reflects the degree of energy attenuation during the sound wave propagation process. By sound pressure A reference signal at a receiver location relative to the source To calculate, specifically expressed as:

[0064] ;

[0065] ;

[0066] in, represents the number of eigenvalues ​​that contribute to the sound field at a specific receiving position, By The complex sound pressure caused by the eigenvalue sound ray describes the The intrinsic sound rays are at the receiving position The intensity and phase information at , and Respectively represent the horizontal and vertical (ie, depth) coordinates of the receiving position.

[0067] Step S2, using a Gaussian filter to smooth the sound propagation loss in the deep sea environment in the training set, to obtain global structural information and local detail information of the sound propagation loss in the deep sea environment.

[0068] Among them, Gaussian filtering is a smoothing technology widely used in image processing and signal processing. It smoothes data by using a filter constructed using a Gaussian function to reduce noise while retaining edges and details as much as possible. Its mathematical expression is:

[0069] ;

[0070] in, is the spatial coordinate; is the standard deviation, which is used to control the width of the Gaussian function and determines the smoothness of the filter. Through Gaussian smoothing, the global features of the low-frequency part can be better preserved, while the high-frequency part focuses on complex interference fringes.

[0071] Step S3: Construct a dual-branch U-Net neural network model for predicting sound propagation loss in the deep-sea environment. This model consists of a global feature branch and a local detail branch, both of which are in the U-Net structure. The model learns the global structure information of the input data through the global feature branch, extracts the local detail information of the input data through the local detail branch, and finally, by fusing the information of different scales learned by the two branches, outputs the prediction result of sound propagation loss in the deep-sea environment that takes into account both global structure information and local detail information.

[0072] Among them, the architecture of the dual-branch U-Net neural network model is as Figure 3 shown. The U-Net in the architecture adopts a symmetric autoencoder structure, and extracts and reconstructs image features by gradually compressing and restoring the spatial resolution of the image.

[0073] Furthermore, the total loss function of the dual-branch U-Net neural network model is expressed as:

[0074] ;

[0075] ;

[0076] ;

[0077] Among them, represents the loss function of the global feature branch, represents the loss function of the local detail branch, and both represent weights, which are used to adjust the proportion of the loss of the global feature branch and the loss of the local detail branch. represents the global structure information learned by the global feature branch, represents the label value of the global structure information, represents the local detail information extracted by the local detail branch, represents the label value of the local detail information.

[0078] Step S4: Adopt the sound source space encoding algorithm to preprocess the sound source depth, sound source frequency and sound speed profile in the training set after Gaussian smoothing. The preprocessed sound source depth, sound source frequency and sound speed profile are encoded into a matrix with the same size as the predicted sound field.

[0079] In step S4, this application considers that using the sound source depth, sound source frequency, and sound speed profile as the direct inputs of the neural network model has limitations. This input method fails to leverage the advantages of the convolutional kernel in spatial feature extraction. To overcome this limitation, this application further introduces a sound source spatial encoding algorithm. This algorithm encodes the sound source depth, sound source frequency, and sound speed profile into a matrix with the same size as the predicted sound field, enabling the encoded matrix to contain the sound source depth, sound source frequency, sound speed profile, and sound source spatial position information simultaneously. Thus, it can fully utilize the advantages of the convolutional kernel in the dual-branch U-Net neural network model for spatial feature extraction, improving the accuracy of feature extraction related to sound propagation loss in the deep-sea environment and further enhancing the accuracy of model prediction. The specific steps of this sound source spatial encoding algorithm are as follows:

[0080] (1) Regional discretization: Discretize the research region into a zero matrix . The matrix element represents the relative position of the position at the -th row and -th column in the zero matrix within the region . Among them, the horizontal distance and vertical distance of the research region are and respectively. and represent the number of rows and columns of the matrix respectively.

[0081] (2) Sound source depth positioning: Given the sound source , for the set underwater sound source depth , search for the vertical direction index closest to within the depth range , which is expressed as:

[0082] ;

[0083] where .

[0084] (3) Normalization processing of sound source depth, frequency, and sound speed profile: Perform normalization processing on the sound source depth , sound source frequency , and sound speed profile respectively. Specifically, within the corresponding defined ranges, perform normalization processing on the sound source depth and sound source frequency respectively to obtain the normalized sound source depth and normalized sound source frequency , which are expressed as:

[0085] ;

[0086] ;

[0087] Among them, , and respectively represent the minimum and maximum values of the sound source depth; , and respectively represent the minimum and maximum values of the sound source frequency.

[0088] Normalize each sound speed profile sample in the sound speed profiles to obtain the normalized sound speed profile sample , which is expressed as:

[0089] ;

[0090] Among them, represents the serial number of the sound speed profile sample, is the total number of sound speed profiles, and respectively represent the minimum and maximum values of the sound speed profile.

[0091] (4) Data embedding: According to the characteristics of cylindrical coordinates, set the sound source to only appear in the first column of the zero matrix. Therefore, encode the normalized sound source depth, sound source frequency, and sound speed profile into the first column of the zero matrix respectively, and correspond to the vertical direction index to obtain the matrix after encoding the sound source depth, the matrix after encoding the sound source frequency, and the matrix after encoding the sound speed profile, which are respectively expressed as:

[0092] ;

[0093] ;

[0094] .

[0095] Through the above preprocessing process, the encoded , and Stacked or combined in a specific way to form a multi-channel input with rich information, ensuring the consistency of the data format with the input of the dual-branch U-Net neural network model. At the same time, the model can fully learn the relevant features of sound propagation loss in the deep-sea environment, improving the accuracy of predicting the sound propagation loss distribution in the deep-sea environment.

[0096] Step S5: Input the preprocessed training set of sound source depth, sound source frequency, and sound speed profile into the dual-branch U-Net neural network model for iterative training until a trained dual-branch U-Net neural network model is obtained. Then, after passing the new unknown sound source depth, unknown sound source frequency, and unknown sound speed profile data through the same preprocessing steps, input them into the trained dual-branch U-Net neural network model to predict and output the corresponding sound propagation loss distribution in the deep-sea environment.

[0097] Specifically, the model training and prediction process is as Figure 4 shown, including:

[0098] Training stage: First, initialize the weights and biases of the dual-branch U-Net neural network model. Then, input the preprocessed training set of sound source depth, sound source frequency, and sound speed profile into the dual-branch U-Net neural network model. Through the forward propagation process, the input data successively passes through multiple convolutional layers, batch normalization layers, and non-linear activation functions to gradually extract the global structure and local details, generating a preliminary prediction result of the sound propagation loss distribution in the deep-sea environment. During this process, the skip connections in the U-Net structure are used to fuse features at different levels to ensure that high-resolution details are retained. Subsequently, calculate the error between the prediction result and the true label, and use weighting to obtain the total loss function to quantify the error.

[0099] Then, use the backpropagation algorithm to adjust the weights and biases of the network according to the gradient information of the loss function. The Adam optimizer is used during the optimization process to dynamically adjust the learning rate, accelerate convergence, and avoid falling into local optima. The entire training process is repeated in multiple iteration cycles, and the training data is batch-processed in each iteration cycle to improve the computational efficiency.

[0100] Prediction stage: After training is completed, save the model parameters of the trained dual-branch U-Net neural network model. Then, after passing the new unknown sound source depth, unknown sound source frequency, and unknown sound speed profile data through the same preprocessing steps, input them into the model. Through forward propagation, the model generates the final prediction result of the sound propagation loss distribution in the deep-sea environment. The entire process is implemented in an end-to-end manner, without additional artificial feature extraction steps, ensuring the consistency between the training and testing stages and the efficiency of the model.

[0101] In summary, a method for predicting the complex sound field distribution in the deep - sea environment proposed in this application uses the pre - processed sound source depth, sound source frequency, and sound speed profile as inputs, and realizes the end - to - end prediction of the sound propagation loss distribution in the complex sound field in the deep - sea environment through a dual - branch U - Net neural network model. Compared with the traditional numerical model, the method proposed in this application can significantly reduce the computational complexity and cost while maintaining the prediction accuracy.

[0102] In one embodiment, a device for predicting the complex sound field distribution in the deep - sea environment is provided, including:

[0103] A data set generation module, configured to generate a sound propagation loss data set in the deep - sea environment based on different sound source depths, sound source frequencies, and sound speed profiles, and use this data set as the training set;

[0104] A Gaussian smoothing module, configured to use a Gaussian filter to smooth the sound propagation loss in the deep - sea environment in the training set to obtain the global structure information and local detail information of the sound propagation loss in the deep - sea environment;

[0105] A model construction module, configured to construct a dual - branch U - Net neural network model for predicting the sound propagation loss in the deep - sea environment. This model consists of a global feature branch and a local detail branch, both of which are in the U - Net structure. The model learns the global structure information of the input data through the global feature branch, extracts the local detail information of the input data through the local detail branch. Finally, by fusing the information of different scales learned by the two branches, it outputs the prediction result of the sound propagation loss in the deep - sea environment that takes into account both the global structure information and the local detail information;

[0106] A data pre - processing module, configured to use a sound source space encoding algorithm to pre - process the sound source depth, sound source frequency, and sound speed profile in the training set after Gaussian smoothing. The pre - processed sound source depth, sound source frequency, and sound speed profile are encoded into a matrix with the same size as the predicted sound field;

[0107] A model training and prediction module, configured to input the training set pre - processed with the sound source depth, sound source frequency, and sound speed profile into the dual - branch U - Net neural network model for iterative training until a trained dual - branch U - Net neural network model is obtained. Then, after passing the new unknown sound source depth, unknown sound source frequency, and unknown sound speed profile data through the same pre - processing steps, input them into the trained dual - branch U - Net neural network model to predict and output the corresponding sound propagation loss distribution in the deep - sea environment.

[0108] For the specific limitations of the device for predicting the complex sound field distribution in the deep - sea environment, reference can be made to the limitations of the method for predicting the complex sound field distribution in the deep - sea environment described above, which will not be elaborated here. Each module in the above - mentioned device for predicting the complex sound field distribution in the deep - sea environment can be implemented in whole or in part by software, hardware, or a combination thereof. The above - mentioned modules can be embedded in the processor of the computer device in hardware form or be independent of it, or be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above - mentioned modules.

[0109] In one embodiment, a computer device is provided. This computer device can be a terminal, and its internal structure diagram can be as Figure 5 shown. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non - volatile storage medium and an internal memory. The non - volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non - volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for predicting the complex sound field distribution in the deep - sea environment. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad set on the outer shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0110] Those skilled in the art can understand that Figure 5 the structure shown in

[0111] is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0112] In one embodiment, a computer - readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above - mentioned method for predicting the complex sound field distribution in the deep - sea environment.

[0113] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0114] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0115] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application should be subject to the appended claims.

Claims

1. A method for predicting complex sound field distribution in a deep sea environment, characterized in that: The method comprises: Generate a sound propagation loss dataset in a deep-sea environment based on different sound source depths, sound source frequencies, and sound velocity profiles, and use the dataset as a training set; Using a Gaussian filter to smooth the sound propagation loss in the deep sea environment in the training set, so as to obtain global structural information and local detail information of the sound propagation loss in the deep sea environment; A dual-branch U-Net neural network model for predicting sound propagation loss in deep-sea environments is constructed. The model consists of a global feature branch and a local detail branch, both of which are U-Net structures. The model learns the global structural information of the input data through the global feature branch, and extracts the local detail information of the input data through the local detail branch. Finally, by fusing the information of different scales learned by the two branches, the prediction result of sound propagation loss in deep-sea environments that takes into account both global structural information and local detail information is output. The sound source space coding algorithm is used to preprocess the sound source depth, sound source frequency and sound velocity profile in the training set after Gaussian smoothing. The preprocessed sound source depth, sound source frequency and sound velocity profile are encoded into a matrix with the same size as the predicted sound field. The training set after preprocessing of the sound source depth, sound source frequency and sound speed profile is input into the dual-branch U-Net neural network model for iterative training until a trained dual-branch U-Net neural network model is obtained. The new unknown sound source depth, unknown sound source frequency and unknown sound speed profile data are subjected to the same preprocessing steps and then input into the trained dual-branch U-Net neural network model to predict the sound propagation loss distribution in the corresponding deep-sea environment.

2. The method according to claim 1, characterized in that Generate a deep-sea sound propagation loss dataset based on different sound source depths, sound source frequencies, and sound velocity profiles, including: Different sound source depths, sound source frequencies, sound velocity profiles, terrain settings and geoacoustic parameter settings are used as input parameters, and the traditional numerical model Bellhop based on ray theory is used to generate a data set of sound propagation loss in deep sea environment.

3. The method according to claim 2, characterized in that Using the traditional numerical model Bellhop based on ray theory, a data set of sound propagation loss in deep sea environment is generated, including: Bellhop uses ray theory to simulate the propagation of sound waves in the ocean environment and generates a data set of sound propagation losses in deep-sea environments, including intrinsic sound rays and underwater sound propagation losses. By sound pressure A reference signal at a receiver location relative to the source To calculate, specifically expressed as: ； ； in, represents the number of eigenvalues ​​that contribute to the sound field at a specific receiving position, By The sound pressure caused by the intrinsic sound rays, and , and Respectively represent the horizontal and vertical coordinates of the receiving position.

4. The method according to claim 1, characterized in that: The sound source spatial coding algorithm is used to preprocess the sound source depth, sound source frequency and sound velocity profile in the training set after Gaussian smoothing, including: The study area Discretize into a zero matrix , the matrix elements represents the zero matrix Line The position of the column in the area The relative position within the study area The horizontal and vertical distances are and , and Respectively represent the number of rows and columns of the matrix; Known sound source , for the set underwater sound source depth , in the depth range Search and The closest vertical index , expressed as: ； in, ; Depth to sound source , sound source frequency Sonic velocity profile Normalization is performed separately; According to the characteristics of cylindrical coordinates, the sound source is set to appear only in the first column of the zero matrix. Therefore, the normalized sound source depth, sound source frequency and sound velocity profile are encoded in the zero matrix respectively. The first column of the vertical index , respectively get the matrix after the sound source depth encoding , the matrix after the sound source frequency encoding The matrix after encoding the sound velocity profile .

5. The method according to claim 4, characterized in that Depth to sound source , sound source frequency Sonic velocity profile Normalization is performed separately, including: Within the corresponding definition range, the sound source depth The frequency of the sound source Perform normalization to obtain the normalized sound source depth and the normalized sound source frequency , respectively expressed as: ； ； in, , and Respectively represent the minimum and maximum values ​​of the sound source depth; , and Respectively represent the minimum and maximum values ​​of the sound source frequency; Will Each sound velocity profile sample in the sound velocity profile Perform normalization processing to obtain the normalized sound velocity profile sample , expressed as: ； in, Indicates the serial number of the sound velocity profile sample, is the total number of sound velocity profiles, and They represent the minimum and maximum values ​​of the sound velocity profile respectively.

6. The method according to claim 5, characterized in that The normalized sound source depth, sound source frequency and sound velocity profile are encoded in the zero matrix respectively. The first column of the vertical index , respectively get the matrix after the sound source depth encoding , the matrix after the sound source frequency encoding The matrix after encoding the sound velocity profile , expressed as: ； ； 。 7. The method according to claim 1, characterized in that The total loss function of the dual-branch U-Net neural network model It is expressed as: ； ； ； in, represents the loss function of the global feature branch, represents the loss function of the local detail branch, and Both represent weights, which are used to adjust the proportion of global feature branch loss and local detail branch loss. Represents the global structural information learned by the global feature branch, The tag value representing the global structure information, Represents the local detail information extracted by the local detail branch, The label value representing the local detail information.

8. A device for predicting complex sound field distribution in a deep sea environment, characterized in that: The device comprises: A data set generation module is used to generate a data set of sound propagation loss in a deep-sea environment based on different sound source depths, sound source frequencies and sound speed profiles, and use the data set as a training set; A Gaussian smoothing module is used to use a Gaussian filter to smooth the sound propagation loss in the deep sea environment in the training set, so as to obtain global structural information and local detail information of the sound propagation loss in the deep sea environment; A model building module is used to build a dual-branch U-Net neural network model for predicting sound propagation loss in deep-sea environments. The model consists of a global feature branch and a local detail branch, both of which are U-Net structures. The model learns the global structural information of the input data through the global feature branch, and extracts the local detail information of the input data through the local detail branch. Finally, by fusing the information of different scales learned by the two branches, the prediction result of sound propagation loss in deep-sea environments that takes into account both global structural information and local detail information is output; A data preprocessing module is used to preprocess the sound source depth, sound source frequency and sound speed profile in the training set after Gaussian smoothing by using a sound source space coding algorithm, and the preprocessed sound source depth, sound source frequency and sound speed profile are encoded into a matrix with the same size as the predicted sound field; The model training and prediction module is used to input the training set of sound source depth, sound source frequency and sound speed profile after preprocessing into the double-branch U-Net neural network model for iterative training until a trained double-branch U-Net neural network model is obtained, and the new unknown sound source depth, unknown sound source frequency and unknown sound speed profile data are subjected to the same preprocessing steps and then input into the trained double-branch U-Net neural network model to predict the sound propagation loss distribution in the corresponding deep-sea environment.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images