8 results about "Wave separation" patented technology

This application provides a wavefield decomposition method, apparatus, device, and medium based on wavefront phase direction, belonging to the field of geophysical exploration technology. The method includes: determining a wavefield decomposition operator based on the original wavefield and Thomsen parameters; the wavefield decomposition operator is a first-order pseudo-Helmholtz decomposition operator based on the wavefront phase direction; determining the wavenumber domain wavefield obtained after transforming the original wavefield to the wavenumber domain; performing computational processing based on the wavenumber domain wavefield and wavenumber, followed by inverse Fourier transform to obtain multiple intermediate quantities constituting an auxiliary wavefield; determining the auxiliary wavefield based on the multiple intermediate quantities and Thomsen parameters through an auxiliary wavefield expression; and performing wavefield decomposition based on anisotropic wavefield decomposition formula to achieve P-wave and S-wave separation, obtaining the P-wave and S-wave wavefields. This method not only achieves P-wave and S-wave vector wavefield separation and reduces imaging crosstalk noise, but also reduces computational load, balancing wavefield decomposition effect and computational efficiency.

The present application relates to a kind of diffracted wave separation methods based on multi-feature reuse hollow convolutional neural network.Based on hollow convolution and multi-feature reuse module, neural network is built, training data set is constructed by synthesizing data and actual data, neural network is trained, L1-multiple scale structural similarity index hybrid function is used to measure the distance between network output and label in training to obtain optimal network model;Seismic common offset data or poststack data are input into optimal network model, and separated diffracted wave data are obtained;According to seismic migration velocity file, the diffracted wave data separated by multi-feature reuse hollow convolutional neural network are migrated and imaged.This method can effectively remove steep dip reflection while preserving diffracted wave energy, and can solve the problems of diffracted energy loss and reflection residue in the process of diffracted wave separation.For the case where diffracted wave and reflection event cross, the separated diffracted wave is more complete.

The embodiment of the application provides a diffraction wave separation method, device, equipment, storage medium and program product. The method comprises the following steps: performing conversion processing on seismic data to obtain a seismic trace set; performing multi-scale and multi-angle wave field decomposition on the seismic trace set to obtain a wave field component set; the wave field component set comprises mixed wave components containing diffraction waves and reflection waves; performing modal decomposition on the mixed wave components in the wave field component set to obtain a plurality of modal components; and determining separated diffraction wave data based on the plurality of modal components.

The application discloses a kind of mixing light module and passive optical network system, mixing light module includes shell assembly, light emitting receiving component being arranged in shell assembly, light emitting receiving component is equipped with optical port and lens, and is integrated with emitting unit and receiving unit.Emitting unit includes multiple laser assembly and wave combining component, and multiple optical signals are output from optical port after wave combining.Receiving unit includes wave separating optical separating component, first detection component and second detection component, wave separating optical separating component receives mixed frequency optical signal input from optical port, first, receive waveband and emission waveband are separated, then the optical signal of receive waveband is divided into first light beam and second light beam according to optical power, and first light beam and second light beam are respectively incident to two detection components in opposite directions.The application solves the technical problem that multiple system PON uplink signal cannot be independently separated by wavelength selective filter due to wavelength overlap by receiving mode of first wave separation and then light separation.

The application discloses a kind of marine mesoscale vortex and near-inertial internal wave energy exchange analysis method, belong to the technical field of ocean dynamics, including the following steps: S1, determining target mesoscale vortex area, obtaining high-resolution three-dimensional flow field data and preprocessing to obtain standardized three-dimensional flow field data;S2, by frequency domain filtering separation processing extracts near-inertial flow component and background flow component;S3, by sliding time average processing, obtain Reynolds stress component;S4, by spatial derivative calculation and strain rate derivation processing, obtain horizontal direction strain rate component;S5, substitute energy exchange equation, obtain energy exchange rate;S6, combined with seawater density, complete depth and time integration processing, obtain net energy exchange and realize energy transfer direction discrimination.The above method is used, realize the high-precision quantitative analysis of marine mesoscale vortex and near-inertial internal wave energy exchange direction and intensity and accurate discrimination of transfer direction.

The application discloses a MIMO-SAR robust beam forming echo separation method based on LASSO regression, which comprises the following steps: S1, a MIMO-SAR transmitting waveform is designed, and a pitch multi-channel system is used to collect echo signals; S2, a pitch dimension space snapshot is constructed, and a sample covariance matrix is estimated; S3, a space domain search is performed based on a LASSO regression model, and the angle of arrival of each signal component is roughly estimated; S4, the width of a beam main lobe and a null is set, and a constraint matrix is designed according to the estimated angle of arrival information of the expected and interference signals; S5, a LASSO regression model is constructed based on the constraint matrix, and an optimal weight vector is solved; and S6, the space snapshot in the step S2 is processed by using the optimal weight vector, the interference signals in the echo are suppressed, and the final echo separation is realized. The method can still apply a non-distortion response to the expected signal direction under the condition of terrain undulation, can sufficiently suppress the interference components, and has high robustness and wide application prospect.

This invention discloses a weak signal separation imaging method based on reflected wave separation, relating to the field of petroleum and natural gas geophysics. The method includes: creating a synthetic record; tracing the sequence stratigraphic framework; rock physical analysis; well de-reservoirization; three-dimensional fine wave impedance modeling based on sequence stratigraphic framework constraints; obtaining the three-dimensional spatially varied reflection coefficient volume of the lithologic interface; extracting the well-side wavelet; obtaining the minimum half-wavelength transit time of seismic waves at the well-side lithologic interface; obtaining the matching factor; obtaining the dominant frequency of the spatial variation at the lithologic interface; obtaining the three-dimensional spatially varied wavelet; obtaining the three-dimensional forward model; and recovering the weak signal from the de-reservoir. This method not only establishes a three-dimensional forward model of the lithologic interface but also simultaneously performs forward modeling of the model using the Ricker wavelet in porous viscoelastic media. Finally, the forward modeling results are separated from the actual seismic data using wavefield separation.

The application discloses a VSP transmission converted wave PS reverse-time migration imaging method based on a traveling wave separation, which introduces a traveling wave separation technology under an elastic wave reverse-time migration framework, effectively restrains wave types and propagation directions in a wave field continuation process, constructs imaging conditions consistent with transmission PS converted wave propagation paths and vector polarization characteristics, realizes effective focusing of transmission PS converted wave energy while ensuring imaging polarity stability, effectively distinguishes transmission wave and reflection wave imaging results, suppresses imaging false images caused by non-target wave fields, and improves stability and reliability of transmission wave imaging; and the method should have good engineering realizability and scalability, and can be seamlessly embedded into an existing process.