20 results about "Acoustical oceanography" patented technology

The invention relates to a wind noise modeling method and system based on machine learning and application, and belongs to the field of marine acoustics and environmental noise modeling, and the method comprises the steps of data preprocessing and feature extraction, hybrid network model building, hybrid network model training and verification. According to the method, a hybrid network model of multiple linear regression and a multi-layer perceptron is constructed based on a known physical mechanism of wind noise and two main noise generation mechanisms of surface turbulence and bubble oscillation, a linear relation and a non-linear relation are modeled respectively, outputs of the two models are integrated through a weighted fusion strategy, and a multi-layer perceptron model is constructed. Smooth transition modeling from a low-wind-speed linear relation to a high-wind-speed nonlinear relation is achieved.

The invention discloses a quantitative evaluation model and system for underwater sonar detection efficiency by mesoscale vortexes, and belongs to the technical field of marine acoustic detection. The method comprises the following steps: acquiring marine environment data; identifying vortex types (cold vortex / warm vortex / mixed vortex), positions, radiuses R and amplitudes A based on a velocity gradient method of optimized parameters and a deep learning model; utilizing an ROMS numerical mode to simulate a vortex three-dimensional thermohaline structure; generating a sound velocity disturbance coefficient in combination with a Wilson sound velocity model and the vortex amplitude A; calculating propagation loss PL (z) through a Bellhop sound field model; an acoustic detection efficiency index ADEI is constructed; and outputting the detection probability matrix. According to the method, the technical bottleneck that a traditional model cannot dynamically couple the vortex acoustic effect and the detection probability is solved, full-process automatic evaluation is realized, and a sonar deployment strategy is remarkably optimized.

The invention relates to the technical field of underwater towing, in particular to an underwater towed body towing performance design method, which comprises the following steps of: performing static dynamic analysis on a towed body; carrying out dragging dynamics analysis on the towed body; constructing a towed body structure model, and performing hydrodynamic simulation calculation; establishing a dynamic coupling numerical simulation model of the towing system; and carrying out dragging attitude simulation, and obtaining an optimal design scheme. According to the invention, key indexes such as trim and rolling of the towed body in the towing system are ensured to meet the stability standard, so that the urgent requirements of marine acoustic detection on high precision and high reliability are met.

The invention belongs to the field of marine acoustics, and discloses a wind speed-based multi-stage modeling regression marine environmental noise model modeling method, which comprises the following steps of: firstly, acquiring and matching marine noise and wind speed data, and extracting a noise power spectrum; fitting the noise spectrum into gamma distribution, and extracting shape and scale parameters as physical statistical characteristics; in the first stage, Gaussian process regression is utilized to establish mapping from wind speed to distribution parameters; in the second stage, the distribution parameters and the wind speed serve as input, a mixed kernel support vector regression model fusing a Wenz empirical physical kernel and an RBF data kernel is adopted, and a full-band noise spectrum is predicted. According to the method, a complex prediction problem is decomposed through a framework of'physical statistical feature guidance ', physical priori and data-driven learning are effectively fused, prediction precision, generalization ability and robustness of the model under the condition of data scarcity are remarkably improved, and a high-precision noise prediction tool is provided for applications such as ocean wind field monitoring and the like.

This invention relates to an adaptive correction method for multi-beam backscattering based on statistical change point detection, belonging to the field of marine acoustic detection and seabed sediment information processing technology. The method includes acquiring multi-beam data and constructing a proxy signal; adaptively grouping the proxy signal sequence using a bisection change point detection algorithm; statistically processing the backscattering intensity data within each group according to the beam incident angle; smoothing the average angle response curve using a smoothing filter; unsupervised partitioning based on curvature and clustering; and constructing a piecewise angle response reference model, using this model to calculate the correction value. This invention effectively weakens the central highlighting, edge attenuation, and striped structure caused by angle response effects, reduces the block effect or mosaic phenomenon common in fixed window methods, and effectively weakens the systematic correlation between backscattering intensity and incident angle, providing a more stable and reliable data foundation for subsequent seabed sediment classification, seabed geomorphological identification, and related marine survey applications.

This invention relates to the field of underwater acoustic positioning and topographic detection in marine environments, providing a rapid calculation method and system for the incident angle and propagation time of a direct sound ray between two points. The method includes: dividing seawater into layers based on a sound velocity profile; adaptively dividing the incident angle interval according to a preset comprehensive error threshold; calculating the horizontal displacement and propagation time of the sound ray corresponding to different incident angles; and constructing an interpolation coefficient table through third-order polynomial fitting; combining the piecewise displacement of the water depth intervals where the sound ray originates and terminates, obtaining the third-order polynomial fitting coefficients of the horizontal displacement with respect to the incident angle from the polynomial interpolation coefficient table, constructing a cubic equation with respect to the incident angle, solving it using the Cardano method, and iteratively obtaining the incident angle correction value according to a preset angle error threshold; calculating the propagation time; and finally outputting the incident angle and propagation time of the sound ray. This method replaces layer-by-layer calculation with interpolation modeling, achieving high accuracy, rapid convergence, and coefficient table reuse.

This invention relates to the field of marine acoustic modeling and underwater acoustic environment forecasting, and in particular to an integrated method for dynamic extraction of Kuroshio frontal features and sound field modeling based on publicly available data. Through the integrated design of dynamic encryption of publicly available data, frontal feature quantization and PINNs modeling, this invention greatly reduces the sound field forecasting error of Kuroshio frontal regions and reduces the cost of CTD measured data.

This invention relates to the field of marine acoustic detection technology and discloses a dynamic suppression algorithm for real-time elimination of the influence of tidal motion on sonar signals. The method collects sonar signal sequences and tidal motion parameter sequences for the current sea area and divides the sonar signal into multiple time windows. Within each window, the phase offset is calculated based on the tidal parameters and converted into time-domain compensation coefficients through an established dynamic coupling model, performing preliminary phase correction on the sonar signal. Spectral analysis is performed on the initially suppressed signal to extract residual interference features, and the parameter weights of the dynamic coupling model are adjusted accordingly to generate an optimized model. This optimized model is applied to the processing of the next time window, forming an iterative suppression closed-loop process. This method can adaptively track the dynamic changes of tidal motion, effectively weakening its periodic modulation interference on sonar signals and improving the accuracy of sonar detection data in strong tidal areas such as nearshore and estuaries.

The application belongs to the field of ocean acoustics, and discloses a marine environmental noise modeling method based on multi-stage modeling regression of wind speed, which firstly collects and matches marine noise and wind speed data, and extracts noise power spectrum; then fits the noise spectrum into a gamma distribution, and extracts shape and scale parameters as physical statistical characteristics; further, a two-stage prediction model is constructed: the first stage uses Gaussian process regression to establish a mapping from wind speed to distribution parameters; the second stage uses a hybrid kernel support vector regression model that fuses Wenz empirical physical kernel and RBF data kernel to predict the noise spectrum in the full frequency band. Through the framework of "physical statistical characteristic guidance", the application decomposes complex prediction problems, effectively fuses physical priori and data-driven learning, significantly improves prediction accuracy, generalization ability and model robustness under data scarcity conditions, and provides a high-precision noise prediction tool for marine wind field monitoring and other applications.

The invention discloses an in-situ acoustic parameter calculation method for a detection device incompletely penetrating into a sedimentary layer, and belongs to the field of marine acoustic characteristic detection. Comprising the steps that after consistency calibration is conducted on sedimentary layer sound wave data according to an underwater reference sound field, corresponding envelope curves are obtained through envelope fitting, and travel time and amplitude of a sedimentary layer sound field are further obtained; calculating to obtain travel time corresponding to each receiving point of the detection device, connecting lines according to the sequence of the receiving points to obtain travel time curves under different scanning sound velocities, and performing least square matching on the travel time curves and the travel time of the sound field of the sedimentary layer to obtain the sound velocity of the sedimentary layer; calculating to obtain the amplitude corresponding to each receiving point, and connecting according to the sequence of the receiving points to obtain an amplitude curve under different scanning sound attenuation; and performing least square matching on all the amplitude curves and the sound field amplitude of the sedimentary layer to obtain the attenuation coefficient of the sedimentary layer. The method disclosed by the invention effectively makes up the blank of related research when the measurement condition that the stratum is hard or the detection device is incompletely penetrated is encountered in the actual construction process.

The application discloses a kind of ice layer parameter inversion system and method based on photoacoustic effect inversion, it is related to marine acoustic technology field, realize non-contact, high efficiency, end to end, high accuracy, physically interpretable detection sea ice parameter, solve the problems of poor real-time in prior art, artificial dependence, complex calculation, the system includes: laser excitation module, for emitting intensity modulated laser to the measured area of sea ice surface;Signal acquisition module is used to receive the target sound wave associated with the time-frequency characteristics of laser;The target sound wave is generated by the photoacoustic effect of the measured area irradiated by laser;Data processing module is used to restore the ice layer parameter of the measured area based on the target sound wave according to the preset inversion process.

This invention belongs to the field of marine acoustic detection and remote sensing technology, and discloses a method and system for inverting the abundance of deep-sea mixed sediments based on bottom echo data. The method involves preprocessing the bottom echo waveforms of multibeam bathymetry data to obtain a bottom echo sequence; extracting features from the bottom echo sequence waveforms and concatenating these features to the corresponding bottom echo waveforms to form a new waveform sequence; fusing the enhanced waveform sequences to construct an observation matrix Y; decomposing the observation matrix Y into an endmember waveform matrix H′ and an abundance distribution matrix W′, and introducing a minimum volume constraint during the decomposition process; adding a spectral reconstruction error term based on dynamic time warping to the loss function; iteratively optimizing H′ and W′ using gradient descent or multiplicative update rules until convergence; and obtaining the endmember waveform H′, which represents the acoustic response waveform of typical sediment types, and the abundance matrix W′, which reflects the mixing ratio of each endmember at different sampling points, demonstrating high classification accuracy.

The invention belongs to the technical field of crossing of marine acoustics and machine vision, and particularly relates to a sound field region segmentation method based on improved ResNet-34, which comprises the following steps of: 1, performing sound propagation loss image data set and sound field simulation setting; 2, designing a segmentation network and improving the structure; constructing a segmentation backbone network based on ResNet-34, extracting multi-layer semantic and fine-grained texture features by adopting residual connection and convolution stacking, and outputting a multi-level feature map required by a segmentation head as input of a subsequent improvement module; a deformable convolution self-adaptive sampling sound linear density mutation region is introduced into the Stage3, multi-scale feature fusion is carried out across to improve the sensing capability of convergence / shadow region structures with different sizes, and a global statistical feature module is added behind the Stage4 to carry out remarking and enhancement on multi-scale statistics, so that local boundaries and global distribution are considered; and step 3, evaluating standard definition based on a result of gray calculation. According to the invention, the accuracy and generalization ability of the segmentation result in different sea areas, different seasons and various sound channel structures are improved.

The application relates to the technical field of data acquisition, in particular to a measurement system and a measurement method of marine physical acoustics parameters. The measurement system comprises an upper computer module, a sensor module, an interface module, a waveform transceiving module and a communication control module. Compared with the prior art, the measurement system can realize double excitation of sinusoidal and pulse wave signals, can simultaneously measure various marine acoustics parameters based on control instructions, can synchronously monitor marine environmental physical parameters such as the salinity and temperature of seawater, the pressure and temperature of seabed sediments and acoustics parameters such as the sound velocity and sound attenuation, and can correct the results, meanwhile, the measurement system supports external power amplification and linear sweep of the emitted signal, the multifunctionality and practicability of the marine physical acoustics parameter measurement system are improved, the seamless switching operation of multiple frequency bands and multiple sensors is realized, portability is realized, and the function integration of in-situ measurement and laboratory measurement is realized.

The invention belongs to the technical field of ocean acoustics, and particularly relates to an ocean sound channel feature recognition method based on improved Faster R-CNN, which comprises the following steps: step 1, sound velocity data preprocessing and imaging; step 2, an improved Faster R-CNN algorithm adopts a Res-Net 101 network to extract image features of the input image, and an image feature map is obtained; step 3, selecting candidate regions through a region suggestion network; step 4, performing detection frame screening by a non-maximum suppression method; the method comprises the following steps: introducing a Soft-NMS strategy into an improved Faster R-CNN algorithm, and screening out a detection frame with relatively high confidence coefficient; 5, after the sizes of the candidate regions are improved to be uniform, the sizes of the candidate regions output by the RPN network are fixed, and the candidate regions are directly input to a full connection layer for classification and regression prediction; and processing the vectors of the full connection layer through a Soft-max classifier, and selecting the type corresponding to the maximum probability as the classification of the target. According to the method, the recognition accuracy and the space-time generalization capability are improved, and the sound channel feature recognition capability in a complex marine environment is enhanced.

A two-dimensional low-frequency shallow sea sound field calculation method based on a physical information neural network relates to the technical field of marine acoustic calculation and artificial intelligence, and comprises the following steps: step 1, obtaining shallow sea area sound velocity profile data; 2, carrying out data preprocessing on the time sequence sound velocity profile data; 3, designing a physical information loss function; 4, constructing a dual-network parallel PINN physical information neural network model; and step 5, using the trained dual-network PINN model to calculate the modal function and the corresponding horizontal wave number of the shallow sea area, embedding the calculation result into the coupling frame, and calculating the low-frequency shallow sea sound field with the sound velocity changing horizontally. According to the method, a loss function is designed by using a modal equation in a normal wave, corresponding boundary conditions and the like, so that the physical interpretability of the neural network is enhanced, and the black box problem of deep learning is solved.

This invention discloses an optical hydrophone structure based on Fresnel zone plate sensitization, belonging to the field of underwater acoustic signal detection technology. It includes a single-mode optical fiber, an elastic cylinder, a Fresnel zone plate, and a metal cavity. The single-mode optical fiber is uniformly and tightly wound around the elastic cylinder, which is located inside the metal cavity. One end of the metal cavity has an acoustic wave incident aperture. The Fresnel zone plate is fixed at the acoustic wave incident aperture of the metal cavity and has several concentric acoustic transmission rings and blocking rings. This invention employs the above-mentioned optical hydrophone structure based on Fresnel zone plate sensitization, which is simple in structure, has few components, and low cost. It possesses good water pressure resistance and electromagnetic interference resistance, making it suitable for complex underwater environments such as marine acoustic exploration and underwater weak target monitoring.

The application provides a method and device for determining a floating photovoltaic platform offshore construction period. The method collects noise data generated by various types of construction machinery in the operating state in real time in the construction sea area of the floating photovoltaic platform; inputs the noise data into a pre-constructed marine sound propagation model, combines the environmental parameters of the construction sea area, and calculates the predicted noise influence value of at least one preset focus point in the construction sea area; after obtaining the corresponding time-division environmental protection noise limit value of the construction sea area, based on the comparison result of the predicted noise influence value and the environmental protection noise limit value of the corresponding period, a compliant construction period table is generated. The method combines real-time noise monitoring with marine acoustic model prediction to automatically generate a construction period table that meets environmental protection regulations, and intelligently and dynamically plans the high-noise operation time of the floating photovoltaic platform offshore.

The invention discloses a miniaturized underwater target real-time monitoring system based on software and hardware collaborative optimization and data interaction, and relates to the field of marine acoustic monitoring, and the system comprises a detection device, a fusion device and an identification device. The detection device comprises a rack, an acoustic array and a data acquisition module, the rack is used for supporting each component; the acoustic array is used for collecting underwater acoustic signals in real time; the data acquisition module is used for converting underwater acoustic signals into digital signals and extracting audio features; lightweight deep learning models are deployed in the fusion device and the recognition device, and the fusion device is used for dynamically fusing digital signals and audio features by adopting the lightweight deep learning models based on professional domain knowledge; and the recognition device is used for carrying out target recognition by adopting a lightweight deep learning model according to the fusion features so as to carry out real-time monitoring on the underwater target. According to the invention, the size of the underwater target monitoring system is reduced, and the real-time performance and accuracy of underwater target detection are improved.

The invention belongs to the technical field of marine acoustic detection and remote sensing, and discloses a deep sea mixed sediment abundance inversion method and system based on water body bottom echo data, and the method comprises the steps: carrying out the preprocessing of a multi-beam water body data bottom echo waveform, and obtaining a bottom echo sequence; performing feature extraction on the waveform of the bottom echo sequence, and joining the features to the corresponding waveform of the bottom echo to form a new waveform sequence; fusing the enhanced waveform sequence, and constructing an observation matrix Y; decomposing the observation matrix Y into an end member waveform matrix H'and an abundance distribution matrix W ', and introducing a minimum volume constraint in the decomposition process; adding a wave spectrum reconstruction error term based on dynamic time warping into the loss function; iteratively optimizing H'and W 'through gradient descent or multiplication updating rules until convergence; the obtained end member waveform H'represents the acoustic response waveform of a typical substrate type, the abundance matrix W 'reflects the mixing proportion of each end member in different sampling points, and the classification precision is high.