Typhoon orientation estimation method, system and terminal based on single vector hydrophone

Abstract

The invention discloses a typhoon orientation estimation method, system and terminal based on a single vector hydrophone, and the method comprises the steps: carrying out the strict inspection of sound pressure and vibration velocity data in three directions collected by the single vector hydrophone, and guaranteeing the data quality; preprocessing the data, including filtering to remove noise interference, normalizing the filtered data, calculating the sound intensity in each direction in a certain azimuth angle and pitch angle search range, constructing a sound intensity matrix, and finding the azimuth angle and pitch angle corresponding to the maximum sound intensity as real reference values; and dividing the processed data into a training set and a test set in proportion, constructing a multilayer perceptron model, performing model optimization by testing different learning rates and small batch sizes, and selecting parameters with minimum loss to train an optimal model. And finally, predicting the test set by using the trained model to obtain estimated values of the azimuth angle and the pitch angle of the typhoon. According to the invention, the accuracy of typhoon azimuth estimation is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ocean observation, and particularly to a typhoon azimuth estimation method, system, terminal and computer-readable storage medium based on a single vector hydrophone. Background Art

[0002] As a natural disaster with extremely strong destructiveness, the accurate monitoring of typhoons is crucial for disaster prevention and mitigation. Although traditional monitoring means such as satellite remote sensing, meteorological radar and buoy observation are widely used, they have significant defects: satellite remote sensing is affected by cloud cover and sea surface reflection, resulting in low data accuracy and high latency; meteorological radar has a limited coverage range (only hundreds of kilometers), is easily interfered by terrain, and the equipment deployment cost is high; although the acoustic monitoring technology based on traditional hydrophones can capture typhoon low-frequency noise, it only relies on a single parameter of sound pressure and lacks the fusion of particle velocity information, resulting in insufficient azimuth estimation accuracy, and traditional algorithms rely on complex signal processing with low computational efficiency, making it difficult to meet the real-time requirements. The single vector hydrophone has become an important tool in the field of ocean monitoring because it can synchronously obtain sound pressure and three-dimensional particle velocity information, and its multi-parameter characteristics provide a new idea for typhoon azimuth estimation.

[0003] However, at present, the typhoon azimuth estimation technology based on single vector hydrophones is mostly limited to traditional physical models and simple sound intensity calculations, and does not fully utilize deep learning technology to explore the potential correlations of multi-parameter data, resulting in a single data dimension, poor algorithm real-time performance, and inability to adapt to the monitoring requirements of typhoon dynamic changes.

[0004] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0005] The main purpose of the present invention is to provide a typhoon azimuth estimation method, system, terminal and computer-readable storage medium based on a single vector hydrophone, aiming to solve the problems in the existing technology that the typhoon azimuth estimation based on a single vector hydrophone does not fully utilize deep learning technology to explore the potential correlations of multi-parameter data, resulting in a single data dimension, poor algorithm real-time performance, and inability to adapt to the monitoring requirements of typhoon dynamic changes.

[0006] To achieve the above object, the present invention provides a typhoon azimuth estimation method based on a single vector hydrophone, and the typhoon azimuth estimation method based on a single vector hydrophone includes the following steps:

[0007] Arrange the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data through the target single vector hydrophone;

[0008] Perform data inspection, filtering processing and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data;

[0009] Within the search range of the preset azimuth, calculate the true azimuth of the typhoon according to the target sound pressure data and the target particle velocity data;

[0010] Use the target sound pressure data and the target particle velocity data as sample data, and use the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels to construct a data set according to the sample data and the labels;

[0011] Construct a typhoon azimuth estimation model, and use the data set to train and evaluate the typhoon azimuth estimation model to obtain a target model;

[0012] Collect the sound pressure data to be predicted and the particle velocity data to be predicted through the single vector hydrophone, and input the sound pressure data to be predicted and the particle velocity data to be predicted into the target model to obtain the estimated azimuth of the typhoon.

[0013] Optionally, in the typhoon azimuth estimation method based on a single vector hydrophone, wherein the target single vector hydrophone is arranged at a preset depth below the sea surface, and the sound pressure data and the particle velocity data are collected through the target single vector hydrophone, specifically including:

[0014] Select a single vector hydrophone with sensitivity, frequency response range, working depth and directivity meeting the preset standards as the target single vector hydrophone;

[0015] Arrange the target single vector hydrophone at a preset depth below the sea surface, and collect the sound pressure data and the particle velocity data in three orthogonal directions through the target single vector hydrophone.

[0016] Optionally, in the typhoon azimuth estimation method based on a single vector hydrophone, wherein the data check, filtering process and normalization process are performed on the sound pressure data and the particle velocity data to obtain the target sound pressure data and the target particle velocity data, specifically including:

[0017] Check whether the sound pressure data and the particle velocity data are vectors and whether the vector lengths are consistent. If not, it is determined that the sound pressure data and the particle velocity data are collected incorrectly and an exception is thrown;

[0018] If so, use a fourth-order Butterworth filter to perform a filtering process on the sound pressure data and the particle velocity data to obtain filtered data, and map the filtered data to a preset range for normalization processing to obtain the target sound pressure data and the target particle velocity data.

[0019] Optionally, in the typhoon azimuth estimation method based on a single vector hydrophone, wherein the true azimuth includes the true azimuth angle and the true pitch angle;

[0020] Within the search range at a preset azimuth, based on the target sound pressure data and the target particle velocity data, calculate the true azimuth of the typhoon, specifically including:

[0021] Within the search range of the first preset azimuth angle and the second preset pitch angle, based on the target sound pressure data and the target particle velocity data, calculate multiple sound intensities in each direction;

[0022] Construct a sound intensity matrix based on the multiple sound intensities, and determine the true azimuth angle and the true pitch angle of the typhoon according to the sound intensity matrix.

[0023] Optionally, for the typhoon azimuth estimation method based on a single vector hydrophone, wherein, constructing the data set according to the sample data and the label specifically includes:

[0024] Construct a data set according to the sample data and the label corresponding to the sample data, and divide the data set into a training set and a test set according to a preset ratio;

[0025] The training set is used for training and parameter adjustment of the model, and the test set is used for evaluating the performance of the model.

[0026] Optionally, for the typhoon azimuth estimation method based on a single vector hydrophone, wherein, constructing the typhoon azimuth estimation model, training and evaluating the typhoon azimuth estimation model using the data set to obtain a target model, specifically including:

[0027] Construct a multi-layer perceptron model, the multi-layer perceptron model includes an input layer, a hidden layer and a regression output layer, the hidden layer contains multiple neurons, and each neuron processes the data through weighted summation and a non-linear activation function;

[0028] Increase the number of hidden layers and the number of neurons to obtain a typhoon azimuth estimation model, train the typhoon azimuth estimation model using the training set, and calculate the loss of the typhoon azimuth estimation model on the training set;

[0029] Select the parameter combination with the minimum loss as the optimal parameter, and continue to train the typhoon azimuth estimation model using the optimal parameter until the typhoon azimuth estimation model converges or reaches a preset number of training rounds to obtain a trained model;

[0030] Evaluate the trained model using the test set to obtain a target model.

[0031] Optionally, for the typhoon azimuth estimation method based on a single vector hydrophone, when inputting the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the target model to obtain the estimated azimuth of the typhoon, it specifically includes:

[0032] Input the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the input layer of the target model for transformation to obtain a feature vector;

[0033] Input the feature vector into the hidden layer of the target model for feature extraction and summation to obtain a fused feature, and map the fused feature through a non-linear activation function to obtain a target feature;

[0034] Input the target feature into the regression output layer of the target model for prediction, and output the estimated azimuth of the typhoon.

[0035] In addition, to achieve the above object, the present invention also provides a typhoon azimuth estimation system based on a single vector hydrophone, where the typhoon azimuth estimation system based on a single vector hydrophone includes:

[0036] A data acquisition module, configured to arrange a target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data through the target single vector hydrophone;

[0037] A data preprocessing module, configured to perform data inspection, filtering processing, and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data;

[0038] A true azimuth calculation module, configured to calculate the true azimuth of the typhoon within a search range of a preset azimuth according to the target sound pressure data and the target particle velocity data;

[0039] A data set construction module, configured to use the target sound pressure data and the target particle velocity data as sample data, use the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels, and construct a data set according to the sample data and the labels;

[0040] A model training and evaluation module, configured to construct a typhoon azimuth estimation model, and use the data set to train and evaluate the typhoon azimuth estimation model to obtain a target model;

[0041] A typhoon azimuth estimation module, configured to collect to-be-predicted sound pressure data and to-be-predicted particle velocity data through the single vector hydrophone, and input the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the target model to obtain the estimated azimuth of the typhoon.

[0042] In addition, to achieve the above object, the present invention further provides a terminal, wherein the terminal includes: a memory, a processor, and a typhoon azimuth estimation program based on a single vector hydrophone stored on the memory and executable on the processor. When the typhoon azimuth estimation program based on the single vector hydrophone is executed by the processor, the steps of the typhoon azimuth estimation method based on the single vector hydrophone as described above are implemented.

[0043] In addition, to achieve the above object, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a typhoon azimuth estimation program based on a single vector hydrophone. When the typhoon azimuth estimation program based on the single vector hydrophone is executed by a processor, the steps of the typhoon azimuth estimation method based on the single vector hydrophone as described above are implemented.

[0044] In the present invention, strict checks are performed on the sound pressure and three-direction vibration velocity data collected by the single vector hydrophone to ensure data quality. Then, the data is preprocessed, including filtering to remove noise interference, and then the filtered data is normalized. Next, the sound intensity in each direction is calculated within a certain azimuth angle and pitch angle search range to construct a sound intensity matrix, and the azimuth angle and pitch angle corresponding to the maximum sound intensity are found as the true reference values. The processed data is divided into a training set and a test set according to a certain proportion, a multi-layer perceptron model is constructed, and the model is optimized by experimenting with different learning rates and mini-batch sizes, and the parameters with the minimum loss are selected to train the best model. Finally, the trained model is used to predict the test set to obtain the estimated values of the typhoon azimuth angle and pitch angle. The present invention effectively improves the accuracy of typhoon azimuth estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flowchart of a preferred embodiment of the typhoon azimuth estimation method based on a single vector hydrophone of the present invention;

[0046] Figure 2 is a schematic diagram of the data obtained after filtering in the typhoon azimuth estimation method based on a single vector hydrophone of the present invention;

[0047] Figure 3 is a schematic diagram of the change of sound intensity with azimuth angle in the typhoon azimuth estimation method based on a single vector hydrophone of the present invention;

[0048] Figure 4 is a loss curve graph of model training in the typhoon azimuth estimation method based on a single vector hydrophone of the present invention;

[0049] Figure 5 is a structural diagram of a preferred embodiment of the typhoon azimuth estimation system based on a single vector hydrophone of the present invention;

[0050] Figure 6Structural diagram of a preferred embodiment of the terminal of the present invention. Detailed implementation manners

[0051] The present application provides a typhoon azimuth estimation method, system and terminal based on a single vector hydrophone. To make the objectives, technical solutions and effects of the present application clearer and more definite, the following further describes the present application in detail with reference to the accompanying drawings and by way of examples. 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.

[0052] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as generally understood by those of ordinary skill in the field to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted in an idealized or overly formal sense unless specifically defined as here.

[0053] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, such descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first", "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions appears to be contradictory or unable to be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0054] The typhoon azimuth estimation method based on a single vector hydrophone according to a preferred embodiment of the present invention, as Figure 1 shown, the typhoon azimuth estimation method based on a single vector hydrophone includes the following steps:

[0055] Step S10: Arrange the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data through the target single vector hydrophone.

[0056] Specifically, select a single vector hydrophone with sensitivity, frequency response range, working depth and directivity meeting the preset standards as the target single vector hydrophone; arrange the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data in three orthogonal directions through the target single vector hydrophone.

[0057] It is understandable that when selecting a suitable single vector hydrophone, its performance indicators such as sensitivity, frequency response range, working depth, and directivity need to be considered. For typhoon monitoring, a vector hydrophone with a relatively wide low-frequency response range (such as 0.1 Hz - 1 kHz) is generally selected as the target single vector hydrophone (because the ocean ambient noise generated by typhoons is mainly concentrated in the low-frequency band). The target single vector hydrophone is deployed at a preset depth below the sea surface, such as 10 - 50 meters, to avoid the interference of direct impact of surface waves, and at the same time ensure that the underwater acoustic signal propagation characteristics triggered by typhoons can be effectively captured. The installation of the single vector hydrophone should ensure its stability and reliability to avoid affecting the accuracy of signal acquisition due to changes in water flow or the ocean environment.

[0058] It should be noted that in addition to collecting data based on single vector hydrophones, in other implementation manners of this application, combinations of other types of sensors can also be considered for replacement. For example, multiple distributed hydrophone arrays can be used, and similar acoustic information can be obtained through signal fusion processing. The distributed hydrophone array can cover a larger area and obtain richer spatial information, and the estimation of typhoon azimuth and pitch angle may be more accurate. However, this solution requires more complex signal synchronization and fusion technologies, and the cost is relatively high.

[0059] The target single vector hydrophone is used to collect sound pressure data and particle velocity data. The single vector hydrophone continuously collects acoustic signals within time t, and simultaneously measures the sound pressure signal p(t) and the particle velocity signals v x (t), v y (t), v z (t). The sound pressure signal p(t) reflects the pressure fluctuations in the medium, and the particle velocity signals v x (t), v y (t), v z (t) respectively represent the motion velocities of the medium particles in three orthogonal directions, and these three directions are usually the horizontal x-axis, y-axis, and vertical z-axis. These signals provide multi-dimensional data for subsequent analysis.

[0060] In addition, in addition to estimating by combining the above acoustic data, in other implementation manners of this application, other types of data such as optical and meteorological data can also be fused. For example, the optical images of satellites are used to obtain the macroscopic position information of typhoons, and then the acoustic data of single vector hydrophones are combined for refined azimuth and pitch angle estimation. Further, this application can also perform azimuth estimation without relying on the characteristics of sound pressure and particle velocity data. Instead, by performing time-frequency analysis on the acoustic signals, spectral features or time-frequency distribution features are extracted, and these features are used for model training and azimuth and pitch angle estimation, and the purpose of typhoon azimuth estimation can also be achieved.

[0061] Step S20: Perform data inspection, filtering, and normalization on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data.

[0062] In this embodiment, check whether the sound pressure data and the particle velocity data are vectors and whether the vector lengths are the same. If not, it is determined that the sound pressure data and the particle velocity data are collected incorrectly and an error is thrown.

[0063] It can be understood that before data processing, the input sound pressure and velocity data are checked to ensure that the data are vectors and have the same length. This technical feature can avoid calculation errors caused by incorrect data formats or inconsistent lengths, improving the stability and reliability of the system.

[0064] If so, use a fourth-order Butterworth filter to filter the sound pressure data and the particle velocity data to obtain filtered data, and map the filtered data to a preset range for normalization to obtain target sound pressure data and target particle velocity data (as Figure 2 shown).

[0065] In this embodiment, before data processing, a fourth-order Butterworth filter is used to filter the data. The Butterworth filter has flat passband and stopband characteristics, which can effectively remove noise interference and make the data smoother. Through filtering, the quality of the data can be improved, reducing the impact of noise on subsequent sound intensity calculation and model training, thereby improving the accuracy of estimation. Finally, the filtered data is normalized by mapping the data to a specific range. Normalization can eliminate the dimensional differences between data in different dimensions, making the model easier to converge and improving the training efficiency and generalization ability of the model.

[0066] It is understandable that in this application, in addition to the fourth-order Butterworth filter, other types of filters can also be used, such as Chebyshev filters or Elliptic filters. Chebyshev filters have steeper attenuation characteristics in the passband or stopband and can more effectively suppress noise at specific frequencies; Elliptic filters have faster attenuation rates in both the passband and stopband but may introduce certain phase distortions. Different types of filters vary in performance such as passband ripple and stopband attenuation and may be more suitable for certain specific ocean environments or typhoon signal characteristics. Depending on different noise characteristics and application scenarios, changing the filter type may bring better signal processing effects. Adaptive filtering technology is adopted to dynamically adjust the filter parameters according to the real-time signal noise level and characteristics to adapt to the noise characteristics in different stages of typhoons or different ocean environments, making the signal preprocessing more flexible and intelligent.

[0067] Step S30: Within the search range of the preset azimuth, calculate the true azimuth of the typhoon based on the target sound pressure data and the target particle velocity data.

[0068] In this embodiment, the true azimuth includes the true azimuth angle and the true pitch angle; the process of calculating the true azimuth of the typhoon based on the target sound pressure data and the target particle velocity data is as follows: within the search range of the first preset azimuth angle and the second preset pitch angle, calculate multiple sound intensities in each direction based on the target sound pressure data and the target particle velocity data; construct a sound intensity matrix based on the multiple sound intensities, and determine the true azimuth angle and the true pitch angle of the typhoon according to the sound intensity matrix.

[0069] It is understandable that within a certain search range of azimuth angle and pitch angle, calculate the sound intensities in each direction and construct a sound intensity matrix. By comparing the magnitudes of the sound intensities in different directions, the approximate direction of the typhoon can be preliminarily determined (as Figure 3 shown). The accuracy of sound intensity calculation directly affects the label accuracy of subsequent model training, thereby affecting the accuracy of the entire estimation system.

[0070] Furthermore, the principle of sound intensity calculation is as follows: collect data at a certain sampling frequency f s for a period of time T, and obtain discrete time series data p[n], v x [n], v y [n], v z [n], where n = 0, 1,..., n - 1, N, and N is the total number of sampling points, N = f s T. Calculate the instantaneous sound intensity components I in the x, y, and z directions x[n], I y [n] and I y [n]:

[0071] I x [n] = p[n]·v x [n]

[0072] I y [n] = p[n]·v y [n]

[0073] I z [n] = p[n]·v z [n];

[0074] Then calculate the amplitude I of the average sound intensity: According to the target azimuth intensity formula calculate the target azimuth intensity, where the reference sound intensity I0 = 10 -12 W / m 2 . Establish a spherical coordinate system with the three-dimensional vector hydrophone as the center. The azimuth angle usually ranges from 0 to 2π, and the elevation angle θ ranges from 0 to π. Assume that azimuth estimation is performed on the horizontal plane. First, fix the elevation angle θ = π / 2 (i.e., only consider the horizontal direction). At the azimuth angle search at a certain angular interval (such as ). For each value, calculate the target azimuth intensity in the corresponding direction according to the above steps Search for the maximum value, and the corresponding azimuth angle is the azimuth of the typhoon on the horizontal plane.

[0075] Step S40: Use the target sound pressure data and the target particle velocity data as sample data, and use the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels to construct a data set according to the sample data and the labels.

[0076] Specifically, construct a data set according to the sample data and the labels corresponding to the sample data, and divide the data set into a training set and a test set according to a preset ratio; the training set is used for training and parameter adjustment of the model, and the test set is used for evaluating the performance of the model.

[0077] Divide the preprocessed target sound pressure data and target particle velocity data into a training set and a test set according to a set training set ratio (e.g., 7:3). The training set is mainly used to train the model and adjust the model's parameters (such as the weights of the neural network). By using the data in the training set, the model learns features and patterns for prediction or classification, and the test set is used to evaluate the performance of the final model. The results of the test set represent the performance of the model in actual applications. The data in the test set is neither used for training nor for parameter tuning, but for finally evaluating the accuracy and performance of the model. By the performance on the test set, the generalization ability of the model to unknown data can be verified to ensure that the model can also perform well on unknown data. Reasonable data set division can avoid overfitting and underfitting problems of the model and improve the generalization ability of the model.

[0078] It can be understood that in addition to setting the training set and the test set, this application can further set a validation set to adjust the model hyperparameters and evaluate the model performance. During the training process, the validation set is used to evaluate the performance of the model on unseen data to prevent overfitting of the model. Through the performance of the validation set, the best model parameters can be selected to further optimize the model.

[0079] Step S50: Construct a typhoon azimuth estimation model, and use the data set to train and evaluate the typhoon azimuth estimation model to obtain a target model.

[0080] Select to construct a Multilayer Perceptron (MLP) model as the typhoon azimuth estimation model. The multilayer perceptron model consists of an input layer, a hidden layer, and an output layer. The input layer receives the preprocessed data, and each input node corresponds to a feature. The hidden layer contains multiple neurons, and each neuron processes the input through weighted summation and a non-linear activation function to extract the features in the data. The output layer outputs the predicted values of the typhoon azimuth angle and pitch angle. Through the backpropagation algorithm, the model can adjust the weights and biases between neurons according to the prediction error to improve the prediction accuracy of the model.

[0081] Specifically, construct a multilayer perceptron model, which includes an input layer, a hidden layer, and a regression output layer. The hidden layer contains multiple neurons, and each neuron processes the data through weighted summation and a non-linear activation function.

[0082] In this embodiment, select to construct a multilayer perceptron model as the typhoon azimuth estimation model. The multilayer perceptron model has strong non-linear fitting ability and can capture complex features in the data. The multilayer perceptron model includes an input layer, multiple hidden layers, and a regression output layer. The input layer receives the preprocessed data, the hidden layers perform feature extraction and non-linear transformation, and the regression output layer outputs the predicted values of the typhoon azimuth angle and pitch angle.

[0083] It is understood that this application can choose to construct different machine learning models instead of the multilayer perceptron based on actual circumstances, such as support vector machine models, decision tree models, random forest models, etc. The support vector machine model has good generalization capabilities and is suitable for processing high-dimensional data; the decision tree model and random forest model can handle nonlinear problems and have strong interpretability.

[0084] Furthermore, by increasing the number of hidden layers and the number of neurons, a typhoon direction estimation model is obtained. By increasing the number of hidden layers and the number of neurons, the expressive power of the model can be improved, thereby improving the accuracy of the estimation.

[0085] The typhoon direction estimation model is trained using the training set, and the loss of the typhoon direction estimation model on the training set is calculated. In each experiment, the model is trained using the training set, and the loss of the model on the training set is calculated. By comparing the loss values under different parameter combinations, the parameter combination with the smallest loss is selected as the optimal parameter combination (such as Figure 4 shown).

[0086] Experiment with different learning rates and mini-batch sizes, and select the best parameters to train the model. The learning rate controls the step size of the model's weight updates during training, and the mini-batch size determines the number of data samples used in each training session. By properly adjusting these parameters, the model can converge faster, improving training efficiency and prediction accuracy. By experimenting with different learning rates and mini-batch sizes, the multi-layer perceptron model is tuned, and the parameter combination with the smallest loss is selected for model training. This process fully taps the potential of the model and improves the model's adaptability and prediction accuracy for complex ocean acoustic data. In addition to experimenting with different learning rates and mini-batch sizes, this application can also adopt an adaptive learning rate strategy, such as the adaptive learning rate mechanism provided by optimizers such as AdaGrad, RMSProp, or Adam. These optimizers can automatically adjust the learning rate based on the update of the parameters to improve the training efficiency and stability of the model.

[0087] Furthermore, the parameter combination with the smallest loss is selected as the optimal parameter, and the typhoon direction estimation model is continued to be trained using the optimal parameter. During the training process, the model continuously adjusts the weights and biases to minimize the error between the predicted value and the true value until the typhoon direction estimation model converges or reaches a preset number of training rounds to obtain a trained model; finally, the trained model is evaluated using the test set to obtain a target model.

[0088] Step S60: Collect the sound pressure data to be predicted and the particle velocity data to be predicted through the single vector hydrophone, and input the sound pressure data to be predicted and the particle velocity data to be predicted into the target model to obtain the estimated azimuth of the typhoon.

[0089] Specifically, input the sound pressure data to be predicted and the particle velocity data to be predicted into the input layer of the target model for transformation to obtain a feature vector; input the feature vector into the hidden layer of the target model for feature extraction and summation to obtain a fused feature, map the fused feature through a non-linear activation function to obtain a target feature; input the target feature into the regression output layer of the target model for prediction, and output the estimated azimuth of the typhoon.

[0090] It can be seen that the present invention provides a more accurate label for model training by calculating the sound intensity matrix and finding the azimuth angle and elevation angle corresponding to the maximum sound intensity as the true reference value, and then directly outputs the estimated values of the typhoon azimuth angle and elevation angle in the regression output layer of the MLP model. This design enables the model to be trained and optimized more directly for the estimation tasks of azimuth and elevation angle. The present invention systematically integrates multiple steps such as data inspection, filtering, normalization, sound intensity calculation, dataset division, model construction and tuning, etc., to form a complete and coherent single vector hydrophone typhoon azimuth and elevation angle estimation process. The design and connection of the entire process ensure that each step works together to provide reliable support for accurate estimation.

[0091] Furthermore, as Figure 5 shown, based on the above typhoon azimuth estimation method based on a single vector hydrophone, the present invention also correspondingly provides a typhoon azimuth estimation system based on a single vector hydrophone, wherein the typhoon azimuth estimation system based on a single vector hydrophone includes:

[0092] A data acquisition module 51, configured to arrange the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data through the target single vector hydrophone;

[0093] A data preprocessing module 52, configured to perform data inspection, filtering processing and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data;

[0094] A true azimuth calculation module 53, configured to calculate the true azimuth of the typhoon according to the target sound pressure data and the target particle velocity data within the search range of the preset azimuth;

[0095] The dataset construction module 54 is used to take the target sound pressure data and the target particle velocity data as sample data, take the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels, and construct a dataset according to the sample data and the labels;

[0096] The model training and evaluation module 55 is used to construct a typhoon azimuth estimation model, train and evaluate the typhoon azimuth estimation model using the dataset, and obtain a target model;

[0097] The typhoon azimuth estimation module 56 is used to collect the sound pressure data to be predicted and the particle velocity data to be predicted through the single vector hydrophone, input the sound pressure data to be predicted and the particle velocity data to be predicted into the target model, and obtain the estimated azimuth of the typhoon.

[0098] Further, as Figure 6 shown, based on the above typhoon azimuth estimation method and system based on a single vector hydrophone, the present invention also correspondingly provides a terminal, and the terminal includes a processor 10, a memory 20, and a display 30. Figure 6 Only some components of the terminal are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0099] The memory 20 may be an internal storage unit of the terminal in some embodiments, such as the hard disk or memory of the terminal. The memory 20 may also be an external storage device of the terminal in other embodiments, such as a plug-in hard disk equipped on the terminal, a SmartMedia Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 20 may also include both the internal storage unit of the terminal and the external storage device. The memory 20 is used to store application software installed on the terminal and various types of data, such as the program code for installing the terminal. The memory 20 may also be used to temporarily store data that has been output or will be output. In one embodiment, a typhoon azimuth estimation program 40 based on a single vector hydrophone is stored on the memory 20, and the typhoon azimuth estimation program 40 based on a single vector hydrophone can be executed by the processor 10, so as to implement the typhoon azimuth estimation method based on a single vector hydrophone in the present application.

[0100] The processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chips in some embodiments, and is used to run the program code stored in the memory 20 or process data, such as executing the typhoon azimuth estimation method based on a single vector hydrophone, etc.

[0101] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.

[0102] In one embodiment, when the processor 10 executes the typhoon direction estimation program 40 based on the single vector hydrophone in the memory 20, the following steps are implemented:

[0103] placing a target single vector hydrophone at a preset depth below the sea surface, and collecting sound pressure data and particle velocity data through the target single vector hydrophone;

[0104] Performing data inspection, filtering, and normalization on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data;

[0105] Calculating the true direction of the typhoon based on the target sound pressure data and the target particle velocity data within a preset direction search range;

[0106] Taking the target sound pressure data and the target particle velocity data as sample data, taking the true positions corresponding to the target sound pressure data and the target particle velocity data as labels, and constructing a data set based on the sample data and the labels;

[0107] Constructing a typhoon direction estimation model, and using the data set to train and evaluate the typhoon direction estimation model to obtain a target model;

[0108] The single vector hydrophone is used to collect the sound pressure data and the particle velocity data to be predicted, and the sound pressure data and the particle velocity data to be predicted are input into the target model to obtain the estimated direction of the typhoon.

[0109] The step of placing a target single vector hydrophone at a preset depth below the sea surface and collecting sound pressure data and particle velocity data through the target single vector hydrophone specifically includes:

[0110] A single-vector hydrophone whose sensitivity, frequency response range, working depth and directivity meet the preset standards is selected as the target single-vector water;

[0111] The target single vector hydrophone is arranged at a preset depth below the sea surface, and sound pressure data and particle velocity data in three orthogonal directions are collected by the target single vector hydrophone.

[0112] Among them, the data checking, filtering, and normalization processing of the sound pressure data and the particle velocity data to obtain the target sound pressure data and the target particle velocity data specifically include:

[0113] Check whether the sound pressure data and the particle velocity data are vectors and whether the vector lengths are the same. If not, it is determined that the acquisition of the sound pressure data and the particle velocity data is incorrect and an exception is thrown.

[0114] If so, use a fourth-order Butterworth filter to filter the sound pressure data and the particle velocity data to obtain the filtered data, and map the filtered data to a preset range for normalization processing to obtain the target sound pressure data and the target particle velocity data.

[0115] Among them, the true azimuth includes the true azimuth angle and the true elevation angle;

[0116] Calculating the true azimuth of the typhoon according to the target sound pressure data and the target particle velocity data within the search range of the preset azimuth specifically includes:

[0117] Within the search range of the first preset azimuth angle and the second preset elevation angle, calculate multiple sound intensities in each direction according to the target sound pressure data and the target particle velocity data;

[0118] Construct a sound intensity matrix according to the multiple sound intensities, and determine the true azimuth angle and the true elevation angle of the typhoon according to the sound intensity matrix.

[0119] Among them, constructing the dataset according to the sample data and the label specifically includes:

[0120] Construct a dataset according to the sample data and the label corresponding to the sample data, and divide the dataset into a training set and a test set according to a preset ratio;

[0121] The training set is used for training and parameter adjustment of the model, and the test set is used for evaluating the performance of the model.

[0122] Among them, constructing the typhoon azimuth estimation model, training and evaluating the typhoon azimuth estimation model using the dataset to obtain the target model specifically includes:

[0123] Construct a multi-layer perceptron model. The multi-layer perceptron model includes an input layer, a hidden layer, and a regression output layer. The hidden layer contains multiple neurons, and each neuron processes the data through weighted summation and a non-linear activation function;

[0124] Increase the number of hidden layers and the number of neurons to obtain the typhoon azimuth estimation model.

[0125] Train the typhoon azimuth estimation model using the training set, and calculate the loss of the typhoon azimuth estimation model on the training set;

[0126] Select the parameter combination with the minimum loss as the optimal parameter, and continue to train the typhoon azimuth estimation model using the optimal parameter until the typhoon azimuth estimation model converges or reaches the preset number of training epochs to obtain the trained model;

[0127] Evaluate the trained model using the test set to obtain the target model.

[0128] Among them, inputting the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the target model to obtain the estimated azimuth of the typhoon specifically includes:

[0129] Input the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the input layer of the target model for transformation to obtain a feature vector;

[0130] Input the feature vector into the hidden layer of the target model for feature extraction and summation to obtain a fused feature, and map the fused feature through a non-linear activation function to obtain a target feature;

[0131] Input the target feature into the regression output layer of the target model for prediction, and output the estimated azimuth of the typhoon.

[0132] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a typhoon azimuth estimation program based on a single vector hydrophone, and when the typhoon azimuth estimation program based on a single vector hydrophone is executed by a processor, the steps of the typhoon azimuth estimation method based on a single vector hydrophone as described above are implemented.

[0133] In summary, the present invention proposes a typhoon azimuth estimation method, system and terminal based on a single vector hydrophone. The method includes: strictly checking the sound pressure and three-direction velocity data collected by the single vector hydrophone to ensure data quality. Then preprocess the data, including filtering to remove noise interference, and then perform normalization processing on the filtered data. Then calculate the sound intensity in each direction within a certain azimuth angle and pitch angle search range, construct a sound intensity matrix, and find the azimuth angle and pitch angle corresponding to the maximum sound intensity as the true reference value. Divide the processed data into a training set and a test set according to a ratio, construct a multi-layer perceptron model, and optimize the model by experimenting with different learning rates and mini-batch sizes. Select the optimal model trained with the parameter with the minimum loss. Finally, use the trained model to predict the test set to obtain the estimated values of the typhoon azimuth angle and pitch angle. The present invention effectively improves the accuracy of typhoon azimuth estimation.

[0134] It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or terminal. Without more limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or terminal including that element.

[0135] Of course, those of ordinary skill in the art can understand that all or part of the processes of implementing 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 embodiments provided by the present invention can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various 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.

[0136] It should be understood that the application of the present invention is not limited to the above examples. For those of ordinary skill in the art, improvements or transformations can be made according to the above description. All such improvements and transformations should fall within the protection scope of the appended claims of the present invention.

Claims

1. A typhoon azimuth estimation method based on a single vector hydrophone, characterized in that, The described typhoon azimuth estimation method based on a single vector hydrophone includes: Deploy the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data through the target single vector hydrophone; Perform data inspection, filtering processing, and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data; Within the search range of the preset azimuth, calculate the true azimuth of the typhoon according to the target sound pressure data and the target particle velocity data; Use the target sound pressure data and the target particle velocity data as sample data, and use the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels to construct a data set according to the sample data and the labels; Construct a typhoon azimuth estimation model, use the data set to train and evaluate the typhoon azimuth estimation model to obtain a target model; Collect the sound pressure data to be predicted and the particle velocity data to be predicted through the single vector hydrophone, and input the sound pressure data to be predicted and the particle velocity data to be predicted into the target model to obtain the estimated azimuth of the typhoon.

2. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 1, characterized in that The step of deploying the target single vector hydrophone at a preset depth below the sea surface and collecting sound pressure data and particle velocity data through the target single vector hydrophone specifically includes: Select a single vector hydrophone with sensitivity, frequency response range, working depth, and directivity meeting the preset standards as the target single vector hydrophone; Deploy the target single vector hydrophone at a preset depth below the sea surface, and collect sound pressure data and particle velocity data in three orthogonal directions through the target single vector hydrophone.

3. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 1, characterized in that The step of performing data inspection, filtering processing, and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data specifically includes: Check whether the sound pressure data and the particle velocity data are vectors and whether the vector lengths are the same. If not, it is determined that the collection of the sound pressure data and the particle velocity data is incorrect and an exception is thrown; If so, use a fourth-order Butterworth filter to perform filtering processing on the sound pressure data and the particle velocity data to obtain filtered data, and perform normalization processing on the filtered data by mapping it to a preset range to obtain target sound pressure data and target particle velocity data.

4. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 1, characterized in that The true azimuth includes a true azimuth angle and a true pitch angle; The step of calculating the true azimuth of the typhoon according to the target sound pressure data and the target particle velocity data within the search range of the preset azimuth specifically includes: Within the search range of the first preset azimuth angle and the second preset pitch angle, calculate multiple sound intensities in each direction according to the target sound pressure data and the target particle velocity data; Construct a sound intensity matrix based on the multiple sound intensities, and determine the true azimuth angle and the true pitch angle of the typhoon according to the sound intensity matrix.

5. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 1, wherein The step of constructing the data set according to the sample data and the labels specifically includes: Construct a data set according to the sample data and the labels corresponding to the sample data, and divide the data set into a training set and a test set according to a preset ratio; The training set is used for the training and parameter adjustment of the model, and the test set is used to evaluate the performance of the model.

6. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 5, characterized in that, To construct the typhoon azimuth estimation model, use the dataset to train and evaluate the typhoon azimuth estimation model to obtain the target model, which specifically includes: Construct a multi-layer perceptron model. The multi-layer perceptron model includes an input layer, a hidden layer, and a regression output layer. The hidden layer contains multiple neurons, and each neuron processes data through weighted summation and a non-linear activation function. Increase the number of hidden layers and the number of neurons to obtain the typhoon azimuth estimation model. Use the training set to train the typhoon azimuth estimation model and calculate the loss of the typhoon azimuth estimation model on the training set. Select the parameter combination with the minimum loss as the optimal parameter, and continue to train the typhoon azimuth estimation model using the optimal parameter until the typhoon azimuth estimation model converges or reaches the preset number of training epochs to obtain the trained model. Use the test set to evaluate the trained model to obtain the target model.

7. The typhoon azimuth estimation method based on a single vector hydrophone according to claim 6, characterized in that To input the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the target model to obtain the estimated azimuth of the typhoon, specifically includes: Input the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the input layer of the target model for transformation to obtain a feature vector. Input the feature vector into the hidden layer of the target model for feature extraction and summation to obtain a fused feature, and map the fused feature through a non-linear activation function to obtain a target feature. Input the target feature into the regression output layer of the target model for prediction and output the estimated azimuth of the typhoon.

8. A typhoon azimuth estimation system based on a single vector hydrophone, characterized in that, The typhoon azimuth estimation system based on a single vector hydrophone includes: A data acquisition module for arranging the target single vector hydrophone at a preset depth below the sea surface and collecting sound pressure data and particle velocity data through the target single vector hydrophone. A data preprocessing module for performing data inspection, filtering, and normalization processing on the sound pressure data and the particle velocity data to obtain target sound pressure data and target particle velocity data. A true azimuth calculation module for calculating the true azimuth of the typhoon within the search range of the preset azimuth according to the target sound pressure data and the target particle velocity data. A dataset construction module for using the target sound pressure data and the target particle velocity data as sample data, using the true azimuth corresponding to the target sound pressure data and the target particle velocity data as labels, and constructing a dataset according to the sample data and the labels. A model training and evaluation module for constructing a typhoon azimuth estimation model and using the dataset to train and evaluate the typhoon azimuth estimation model to obtain the target model. A typhoon azimuth estimation module for collecting the to-be-predicted sound pressure data and the to-be-predicted particle velocity data through the single vector hydrophone, and inputting the to-be-predicted sound pressure data and the to-be-predicted particle velocity data into the target model to obtain the estimated azimuth of the typhoon.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and a typhoon azimuth estimation program based on a single vector hydrophone stored on the memory and executable on the processor. When the typhoon azimuth estimation program based on the single vector hydrophone is executed by the processor, the steps of the typhoon azimuth estimation method based on the single vector hydrophone according to any one of claims 1-7 are implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a typhoon azimuth estimation program based on a single vector hydrophone. When the typhoon azimuth estimation program based on the single vector hydrophone is executed by a processor, the steps of the typhoon azimuth estimation method based on the single vector hydrophone according to any one of claims 1-7 are implemented.

Images