934 results about "Seismic exploration" patented technology

The invention discloses a coal seam passive source seismic survey system and method based on deep learning, and relates to the technical field of coal mining safety monitoring and early warning, and the method comprises the following steps: S1, collecting the passive source seismic data of a coal seam area; s2, deep fusion features are extracted; s3, constructing a semi-supervised self-attention prediction network model; s4, extracting coal seam structure prediction parameters; s5, the corrected coal seam structure parameters are extracted; s6, constructing a fast seismic wave forward model, and calculating and outputting an initial model of the coal seam underground structure; s7, constructing a Transform optimization network, and extracting a depth feature vector; and S8, constructing a coal seam structure inversion model, and training iteration by using the improved batch depth Bayesian optimization algorithm. The method overcomes the limitation of manual analysis and slow response in a traditional coal seam seismic survey method, and provides an efficient and accurate solution for safe production of a coal mine.

The invention belongs to the technical field of seismic exploration, and discloses a pre-stack gather optimization method based on multi-dimensional constraint dynamic time warping, which remarkably improves the space continuity and stability of a time shift field. According to the method, the time shift field which is smoother and more continuous in a three-dimensional space can be generated, and the common abnormal jump and unreasonable jitter of the time shift field in a conventional DTW method are effectively inhibited. This better conforms to the gradient characteristics of the geologic structure. Mismatching and processing illusion are effectively avoided, the adaptability of a complex structure area is improved, the global optimization capacity of the DTW algorithm is enhanced through multi-dimensional constraint, and the phenomenon that different horizons or effective waves are wrongly matched with interference waves due to local waveform complexity or noise interference is reduced. Therefore, common processing illusions such as event dislocation and local distortion in a conventional DTW method can be remarkably reduced.

The invention belongs to the technical field of seismic exploration, and discloses a robust full-waveform inversion method, system and device based on generative modeling, and a medium, and the method comprises the steps: obtaining seismic observation data; the method comprises the following steps: starting from random Gaussian noise, establishing a model space comprising a plurality of initial velocity models through an unconditional score model, and determining a global optimal initial velocity model of the model space by adopting a strategy search algorithm; de-noising is carried out on the global optimal initial velocity model based on back diffusion, seismological observation data are introduced as observation constraints, the velocity model is updated by calculating the mismatch gradient of forward modeling data of the current velocity model and the seismological observation data, and an implicit condition sample is obtained; performing forward diffusion processing on the implicit condition sample to obtain a speed model for a subsequent annealing time step; and when the annealing process reaches a preset condition, outputting a final speed model. According to the method, the robustness and accuracy of full-waveform inversion are improved, and the bottleneck problem of traditional full-waveform inversion is solved.

The invention discloses a geological modeling method for fusing remote sensing and seismic data based on a multi-mode auto-encoder, and the method comprises the steps: collecting a remote sensing image covering a target area and seismic exploration data, and converting the attribute of the seismic exploration data into a resolution consistent with the remote sensing image; respectively extracting remote sensing image features and seismic attribute features by using a parallel double-branch coding structure of a multi-mode encoder; a fusion module is arranged in the middle layer of the remote sensing image encoder and the seismic attribute encoder, fusion is carried out in the fusion module, remote sensing images and seismic data are reconstructed by using a decoder based on fusion features, and difference data between reconstructed data and the remote sensing images and the seismic data are calculated based on a combined loss function introducing regular terms. Training a multi-modal encoder by using the difference data to obtain an encoding feature and a difference heat map; and on the basis of the coding features and the difference heat map, key geological structure region identification and visual display are carried out, and key geological structures comprise faults and lithology.

The invention discloses a seismic surface wave suppression method based on EWT-Curvelet, and belongs to the technical field of seismic exploration data processing. The method comprises the following steps: firstly, performing empirical wavelet transform (EWT) on post-stack seismic data channel by channel, extracting a frequency localized modal component through adaptive spectrum segmentation, and constructing a surface wave noise model; then converting the model to a curvelet domain, performing physical constraint filtering at a specific scale and in a nearly horizontal direction, and combining pixel connectivity analysis to suppress a continuous noise region; and finally, reconstructing and optimizing noise through inverse transformation and removing the noise from the original data. The method has the advantages that the modal aliasing problem is solved through channel-by-channel adaptive spectrum segmentation, a surface wave energy dense region is accurately locked by curvelet domain scale-direction combined filtering, isolated effective signals are protected in combination with connectivity analysis, high-precision suppression of surface waves is efficiently achieved in a non-iterative mode, meanwhile, the structural integrity of reflected waves is kept, and high-precision suppression of the surface waves is achieved. The seismic section signal-to-noise ratio and the geological interpretation precision are remarkably improved, and the method is suitable for land seismic exploration data processing.

The invention discloses a three-dimensional tomography method and system for seismic exploration data migration, and relates to the technical field of geological exploration, and the three-dimensional tomography method comprises the following steps: laying a layered heterogeneous sensing array of adaptive terrain coupling in an exploration target area, the layered heterogeneous sensing array comprises a shallow fiber grating sensing layer, a middle MEMS sensing layer and a deep distributed fiber sensing layer, sensors are all provided with terrain adaptive coupling assemblies, and sensing shafts all point to the radial direction of an underground medium. According to the seismic exploration three-dimensional tomography imaging method, an iterative imaging mechanism of real-time geological constraint feedback is constructed, the limitation of traditional static geological constraint one-time importing is broken through, the imaging process can dynamically fuse newly added local geological data in the exploration process, the imaging model and the actual geological conditions are continuously adapted and optimized, and the imaging efficiency is improved. The integrating degree of the imaging result and the actual geological condition is enhanced, and the high requirement of deep complex area exploration for the imaging accuracy is met.

The invention provides a three-dimensional geological structure modeling method, device and system based on drilling information. The method comprises the following steps: S1, collecting drilling data, seismic exploration data and remote sensing image data of a target area; s2, constructing a neural network model composed of a simple full-connection module, a complex convolution residual module and a hierarchical integration module which are connected in sequence; s3, training the initial three-dimensional geologic model by adopting a joint loss function; and S4, in the network training process, a plurality of geological scene models are generated, and each scene model corresponds to a group of different network parameter combinations. According to the three-dimensional geologic structure modeling method, device and system based on drilling information, a spatial receptive field for feature extraction is expanded through a three-dimensional cavity convolution residual block of a complex convolution residual module according to a symmetric expansion rate, input features and deep convolution features are added, processing is carried out through a hierarchical integration module, and the three-dimensional geologic structure modeling method, device and system based on drilling information are obtained. And geologic structure correlation characteristics across drill holes can be effectively captured.

The invention provides a shale reservoir fracture prediction method and system based on three-dimensional full-waveform inversion and a medium, and relates to the technical field of data analysis, and the method comprises the steps: arranging a multi-channel seismic exploration instrument in a shale reservoir region, and collecting full-waveform seismic data, including longitudinal waves, transverse waves and reflected waves; performing three-dimensional full-waveform inversion based on the logging data, the geological structure map and the full-waveform seismic data to generate a shale reservoir attribute model; a multi-scale inversion strategy is constructed, fracture sensitivity analysis and fracture form dynamic prediction are carried out, and reservoir fracture distribution characteristics and reservoir fracture form characteristics are obtained; discrete fracture point connection and fracture network reconstruction are carried out, a reservoir three-dimensional fracture network model is constructed, and fracture prediction visualization display is carried out. The technical problems that in the prior art, only inference based on limited data can be provided, comprehensive capture of a complex fracture network is lacked, distribution and form of fractures cannot be comprehensively and accurately described, and reservoir analysis and mining decisions are affected are solved.

The invention is suitable for the technical field of coal seam seismic exploration, and particularly relates to a coal seam seismic exploration acquisition method and system, and the method comprises the steps: delimiting an exploration region where a coal seam earthquake needs to be carried out, dividing the exploration region into a mined region and an unmined region, obtaining a daily mining task in the mined region, and positioning a blasting point location; the unmined area is divided into a plurality of levels, the deployment position of an optical fiber sensor is positioned, the linear distance between the deployment position and a blasting point position is calculated, mapping between the linear distance and the deployment density is established, look-up tables are constructed, each level corresponds to one look-up table, and each look-up table corresponds to a blasting point position; the lookup table is composed of a linear distance item and a deployment density item. According to the method, the coal seam earthquake can be quickly acquired by generating the snapshot, so that the speed and the precision of coal seam seismic exploration can be greatly improved, and the safety management and the resource utilization of an exploration area can be greatly optimized.

The invention discloses a seismic wave velocity-resistivity joint inversion method based on a time dimension, belongs to the technical field of geophysical exploration, and can solve the problem that the dynamic change of a geological structure cannot be determined in the prior art. The method comprises the following steps: constructing a seismic wave velocity distribution model according to seismic exploration data of a target area, and constructing a resistivity distribution model according to electromagnetic exploration data of the target area; determining an objective function of seismic wave velocity-resistivity joint inversion according to the seismic wave velocity distribution model and the resistivity distribution model; updating and iterating the seismic wave velocity distribution model and the resistivity distribution model according to the seismic exploration data and the electromagnetic exploration data of the plurality of periods until the objective function is minimized; and according to the corresponding seismic wave velocity distribution model and the resistivity distribution model when the target function is minimized, respectively determining seismic wave velocity inversion results and resistivity inversion results of the target area in multiple periods. The method is used for inverting the seismic wave velocity and the resistivity.

The invention discloses a mining-while-digging earthquake advanced detection system based on distributed optical fiber acoustic sensing. Relates to the technical field of seismic exploration. On the basis of a distributed optical fiber acoustic sensing technology, mining-while-digging seismic detection of a working face is realized by utilizing vibration generated by cutting a coal rock stratum by mining machinery. In the production process, the detector collects vibration data of mining machinery and transmits the vibration data to the server through a network; optical cables are arranged on two sides of a coal face upper and lower crossheading or two sides of a driving face roadway; the demodulator emits pulse laser and transmits the pulse laser to the working face optical cable through the industrial looped network or the independent fiber core, vibration of measuring points at equal intervals of the optical cable is measured, and a measuring data packet is returned to the demodulator through the industrial looped network or the independent fiber core; the demodulator demodulates vibration waveform data and uploads the vibration waveform data to the server; server data processing software carries out noise reduction processing on the waveform, then stores and images the waveform, carries out real-time intelligent interpretation, and realizes detection of a geological structure and an abnormal area in front of a working face.

The invention relates to the technical field of seismic exploration, and provides a three-dimensional seismic exploration method and system based on multi-source geological data fusion. The method comprises the following steps: acquiring geological data through a detection device, constructing a first tensor based on multi-source geological data, extracting principal components of the first tensor in a preset dimension, and performing fusion to generate a fusion tensor; generating global features based on tensor slices and hidden states in the fusion tensor, and screening first features with authenticity from the global features through a generative adversarial network; performing three-dimensional modeling based on the first feature and the seismic wave field data to generate a parameter cube; a Bayesian network is constructed based on the parameter cube and drilling data, and stratum evolution information is obtained through prediction by fusing a long-term and short-term memory network so as to dynamically generate a geologic model. A tensor is constructed based on multi-source geological data to extract global features, three-dimensional modeling is carried out in combination with seismic wave field data to obtain a parameter cube, then drilling data and stratum evolution information are fused to dynamically generate a geological model, and the prediction accuracy is improved.

The invention belongs to the technical field of seismic exploration and artificial intelligence, and relates to a physical perception collaborative fusion-based full waveform inversion neural network construction method, which comprises the steps of required model data construction, full waveform inversion neural network construction, loss function construction, network pre-training, and network fine tuning and application. Physical feature perception and collaborative fusion are carried out through a U-Net network and an adaptive residual learning module, so that the problem that traditional full-waveform inversion depends on synthetic data training and is not matched with an actually measured data field is solved. The designed pre-training strategy not only can get rid of dependence on a large-scale synthetic data set, but also can effectively relieve the domain offset problem through fine adjustment of measured data, thereby improving the generalization ability of the model. And the adaptive residual learning module enables the neural network to adaptively and dynamically optimize feature expression, and improves inversion precision through collaborative fusion of a physical perception mechanism. According to the method, the physical consistency and imaging reliability of speed model prediction can be remarkably improved while the labor cost is reduced.

The invention relates to the technical field of geological exploration, and particularly discloses a seismic exploration well depth analysis method based on a complex surface environment, which is used for solving the problems of blind well depth setting and low efficiency of a traditional artificial experience method. Comprising the steps of analyzing geological conditions of a complex earth surface and a shallow surface layer, constructing a middle-deep layer geological model, designing a partition well depth test, collecting multi-gradient well depth test data, performing automatic seismic reflection data analysis based on deep learning, performing intelligent well depth optimization inversion, performing field verification on the optimized well depth, constructing a three-dimensional data body and performing geological application analysis. According to the method, the well depth gradient and the landform recognition parameters are dynamically optimized for different landform areas such as a bedrock exposed area, a farmland section, a weathered object area and a gully area through a dynamic model fed back in real time, and automatic adaptive adjustment of the seismic exploration well depth is achieved.

The invention discloses a seismic image super-resolution reconstruction method based on a double-branch multiple attention network, and belongs to the technical field of seismic exploration, and the method comprises the steps: constructing a data sample library for model training; adopting a super-resolution backbone network and a gradient branch auxiliary network to construct a double-branch multiple attention network model; and training the double-branch multiple attention network model by adopting the data sample library and the mixed loss function to obtain a network model for realizing seismic image super-resolution reconstruction and high-frequency texture information reconstruction, and performing reconstruction based on the trained network model to obtain a seismic image subjected to super-resolution reconstruction. According to the method, a gradient branch is introduced as an auxiliary network, so that the super-resolution reconstruction effect is enhanced. Meanwhile, a mixed loss function is adopted to train the network, and a loss function for constraining high-frequency information is particularly introduced, so that the common structural deformation problem in a super-resolution task is effectively solved, the generation of artifacts is prevented, and a new technical approach is provided for high-precision seismic interpretation.

One or more embodiments of the invention provide a Scholte wave frequency dispersion curve adaptive inversion method and a related device, and the method comprises the steps: obtaining and processing the original seismic data of a target seabed, and obtaining a Scholte wave frequency dispersion curve; constructing an initial shear wave velocity model through linear inversion based on a Schlte wave frequency dispersion curve; and performing fitting optimization on the Schlte wave frequency dispersion curve based on the initial shear wave velocity model to obtain a two-dimensional shear wave velocity profile. According to the technical scheme, the problem that a traditional linear inversion method depends on an initial model is solved, the calculation efficiency is remarkably improved, the problem of local extremum is avoided, and an automatic high-precision inversion solution suitable for batch data processing is provided for submarine seismic exploration.

The invention discloses a seismic data acquisition device based on a ground 3C-DAS optical cable and an acquisition method thereof, is applied to the field of seismic exploration, and aims to solve the problems of high layout cost, too long construction time and low operation efficiency of an existing ground three-component detector. According to the method, a 3C-DAS armored optical cable is buried under a shallow earth surface in a parallel or orthogonal mode, a ground multichannel low-frequency DAS modulation-demodulation instrument collects phase data of Rayleigh scattering light reflected back from the 3C-DAS armored optical cable and carries out modulation-demodulation processing, and high-density distributed three-component seismic data distributed along the 3C-DAS armored optical cable is obtained; according to the system, high-density distributed three-component ground seismic data along the armored optical cable can be rapidly and efficiently acquired at low cost; the invention also provides a processing method for the collected high-density distributed three-component ground seismic data, an analysis method for a shallow surface velocity structure with high spatial resolution, and an inversion imaging method for longitudinal waves, transverse waves and surface waves.

The invention provides a method for identifying a middle-level boundary of a geological envelope based on artificial intelligence. The method comprises the following steps: determining seismic exploration data and logging data of a target work area; extracting seismic reflection features from the seismic exploration data based on the trained first neural network model, and extracting logging sequence features from the logging data by using the trained second neural network model; according to the seismic reflection characteristics and the logging sequence characteristics, geological characteristic representation of the target work area is generated; based on formation contact relation constraint and lithologic sequence constraint which are constructed for a target work area, according to geological characteristic representation, determining implicit topological structure representation of a representation geological envelope body in the target work area; and processing the implicit topological structure representation based on the trained third neural network model to predict the horizon boundary of the geological envelope in the target work area. According to the invention, the prediction result of the horizon boundary of the geological envelope body better conforms to the actual geological law.

The invention relates to the technical field of seismic exploration, in particular to a seismic offset angle gather generation method and device, equipment and a medium. The method comprises the following steps: acquiring pre-stack seismic data of a work area after preprocessing; performing pre-stack time migration on the pre-stack seismic data, and performing grouping accumulation according to inclination angles in the pre-stack time migration process to generate an inclination angle gather; according to the inclination angle gather, picking up Fresnel zone boundaries of reflection events of all imaging points according to preset intervals; stable-phase offset is carried out by using the picked Fresnel zone boundary, and an offset angle gather of stable-phase control is generated according to grouping accumulation of incident angles in the offset process. The problems that an existing method for generating an angle gather through pre-stack depth migration is difficult in interval velocity modeling and high in calculation cost are solved.

The invention provides a seismic exploration path planning method, system and device and a medium, and relates to the technical field of seismic exploration assistance, and the method comprises the steps: a laser radar module obtains laser radar data of a target region; the fusion correction module obtains current attitude information and current positioning information of the unmanned aerial vehicle, and determines target three-dimensional point cloud data according to the current attitude information, the current positioning information and the laser radar data; the data processing module determines earth surface types and terrain gradients respectively corresponding to different sub-regions in the target region through a machine learning algorithm based on the target three-dimensional point cloud data; constructing a three-dimensional trafficability map of the target area based on the earth surface type and the terrain gradient of each sub-area; and the path planning module performs path planning by adopting a preset path planning algorithm based on the three-dimensional trafficability map to obtain a target exploration path, and the target exploration path is a transportation path of the target equipment. According to the invention, the planning precision and the planning efficiency are improved.

The invention discloses a scattered wave seismic acquisition and imaging method based on a full random observation system, which belongs to the technical field of seismic exploration and comprises a quasi-Monte Carlo random observation array design step, a scattered wave intelligent identification and enhancement step, a sparse wave field reconstruction step and a multi-target optimization imaging step. The limited equidistant property is ensured through the random observation array design based on a low-difference sequence, intelligent separation and enhancement of scattered waves are realized by using a physical constraint neural network, a sparse reconstruction problem is solved by using an alternating direction multiplier algorithm, a multi-target optimization imaging framework is constructed to realize high-resolution imaging, and high-resolution imaging is realized. The four core steps form a closed-loop cooperative system, sparse reconstruction quality evaluation feedback optimizes observation array design, imaging quality measurement feedback adjusts scattered wave identification parameters, and cooperative interaction is achieved.

The invention relates to a DAS-VSP seismic data noise suppression method based on a double-domain generative adversarial network, and belongs to the field of seismic exploration data denoising and deep learning. The method comprises the following steps: constructing a double-domain generative adversarial network, determining an optimal hyper-parameter combination by Bayesian optimization, adding actual noise to a pure seismic signal obtained by a forward modeling method to construct a complete training set, training the double-domain generative adversarial network, and testing the double-domain generative adversarial network. According to the method, DAS-VSP data which is low in signal-to-noise ratio and contains various kinds of complex noise can be effectively processed, the denoised seismic signals are clearer in structure, better in continuity, higher in signal-to-noise ratio and more thorough in noise suppression, effective signals are reserved to the maximum extent, high-quality basic data are provided for subsequent seismic data processing and explanation, and the method is suitable for large-scale popularization and application. The method meets the high-precision requirement of current seismic exploration, and has a wide application prospect in the field of oil and gas resource exploration and development.

The invention discloses a surface modeling method based on high-speed top interface reflection time, and relates to a geophysical exploration method, in particular to a surface modeling method based on high-speed top interface reflection time, which comprises the following steps of: acquiring early seismic exploration acquisition data of a research area, picking up cannon first arrival time and calculating a uniform reference surface static correction value; processing the original single-shot record based on the uniform reference surface static correction value to obtain the first arrival time of the large shot and the minimum offset reflection time after static correction; calculating the weathered layer velocity of the shot point, and obtaining the weathered layer velocities of all detection points; within the minimum offset range of each shot point; the high-speed top interface elevation of each shot point is calculated, and a surface layer model is established according to the weathered layer speeds of all the detection points, the weathered layer thickness of each shot point and the weathered layer thickness parameters of each detection point. According to the method, the technical problems of high exploration cost and low model precision in near-surface modeling of the area are effectively solved.

The present invention belongs to the field of seismic exploration technology and discloses a system, method, device, and medium for online migration of seismic exploration project data. The system comprises a data file server and a seismic data migration client. The data file server is configured to respond to requests from the seismic data migration client for data directory structure information, database data, and disk file data, convert the data into corresponding byte streams, and transmit them to the seismic data migration client. The seismic data migration client is configured to receive the corresponding byte streams sent by the data file server and complete data migration for the seismic exploration project. The present invention is applicable to migrating data for seismic exploration projects and solves the problem of data non-interoperability and migration difficulties among multiple projects during seismic exploration.

The invention discloses a surface wave suppression method and device based on a self-supervised learning deep convolutional neural network, and relates to the technical field of seismic exploration and seismic signal processing, the surface wave suppression method based on the self-supervised learning deep convolutional neural network mainly comprises the steps: obtaining a training set and a test set according to a shot seismic record; constructing a surface wave suppression model according to hyperlinks in the deep convolutional neural network and the residual network; and training the surface wave suppression model by using the loss function and the optimizer according to the training set to obtain a trained surface wave suppression model, and performing surface wave suppression on the test set to obtain a result after surface wave suppression. By implementing the surface wave suppression method and equipment based on the self-supervised learning deep convolutional neural network provided by the invention, surface wave suppression can be better carried out, effective background signals are reserved, and dependence on prior knowledge is reduced.

The invention discloses a frequency-dependent fluid mobility inversion method and device, belongs to the technical field of seismic exploration, and solves the problem that a fluid mobility result extracted by an existing time-frequency analysis method is difficult to directly reflect real physical distribution of a reservoir. The method comprises the following steps: acquiring seismic data including multiple seismic post-stack records and relative wave impedance; the number of frequency bands needing to participate in inversion and the frequency range corresponding to each frequency band are set, continuous wavelet transform is carried out on each seismic post-stack record, and a continuous wavelet transform result is inversely transformed to a time domain to obtain seismic frequency division data needing to participate in inversion; calculating a fluid mobility coefficient of the reservoir according to a logging curve obtained by logging, and interpolating the fluid mobility coefficient into a low-frequency model; and based on the relative wave impedance and the L2 regularization constraint, establishing a fluid mobility inversion equation, substituting the seismic frequency division data, the low-frequency model and the relative wave impedance into the established fluid mobility inversion equation, and solving a mobility inversion result recorded after each seismic stack by using gradient descent.

The invention relates to the technical field of oil-gas exploration, in particular to a micro-amplitude structure seismic exploration detection method and system based on pre-stack depth migration, and the method comprises the steps: generating an equivalent speed model set, calculating speed uncertainty data, probabilistically detecting a micro-amplitude structure, and fusing and displaying a result. Compared with the prior art that an absolutely correct speed model is blindly pursued, but the influence of model errors on tiny structure recognition cannot be overcome finally, the method has the advantages that the speed model errors are converted into valuable detection signals from interference needing to be minimized; by actively generating multiple sets of global equivalent speed models and analyzing the uncertainty of the speed models, the most unstable areas in the modeling process become direct evidences for indicating the existence of the micro-amplitude structure, normal form transfer from error elimination to error utilization is realized, and the detection capability of the micro-amplitude structure is remarkably improved.

The invention belongs to the technical field of seismic exploration, and particularly relates to a coal mine underground hydraulic hammering type controllable seismic source vehicle device. The device comprises a crawler, a rotating platform and a hydraulic hammering system, the rotating platform, a four-column type lifting hydraulic support installed on the rotating platform and a hydraulic support in the hydraulic hammering system are hydraulically controlled, and the hydraulic hammering system can conduct 360-degree horizontal rotation, height stepless adjustment and horizontal stretching and retracting. And meanwhile, the crawler-type vehicle body and remote control operation are adopted, so that the crawler-type underground coal mine hammering vehicle is suitable for a complex walking environment in an underground coal mine, and the construction efficiency and the safety are improved.

The invention belongs to the technical field of seismic exploration and artificial intelligence, and relates to a seismic data high-resolution reconstruction and random noise suppression method, which comprises the following steps of: performing feature extraction and reconstruction on low-resolution noisy seismic data through a classic generator to generate a high-resolution denoised seismic image; and discriminant learning is carried out on real seismic data and a generation result through a quantum enhanced double-path discriminator, and the reconstruction capability of the generator is continuously optimized. According to the quantum enhanced double-path structure designed by the invention, the modeling capability of the model on complex seismic signals can be improved, the reconstructed image is finer, the structure is clearer, the robustness of the network on random noise can be enhanced, and more effective noise suppression is realized. According to the invention, the residual learning mechanism enables the feature interaction between the layers to be more sufficient, and facilitates the transmission and fusion of the seismic features at different layers, thereby improving the overall reconstruction quality. According to the method, the suppression effect on random noise can be remarkably improved while high-resolution reconstruction of seismic data is realized.

The invention relates to the technical field of seismic exploration equipment, in particular to an optical fiber signal transmission seismic cable for seismic exploration. A power line, a twisted-pair shielding signal line and an optical fiber beam tube are wrapped outside the high-pressure plastic composite hose in a cabling mode, and the power line, the twisted-pair shielding signal line and the optical fiber beam tube jointly form a cable core structure. An outer layer non-woven fabric is wrapped outside the cable core structure, a polyurethane inner sheath is extruded outside the outer layer non-woven fabric, an armor layer is wrapped outside the polyurethane inner sheath, and a polymer polyethylene sheath is extruded outside the armor layer. According to the high-pressure plastic composite hose, through the design of the double reinforcing layers of the Kevlar aramid fibers and the polyester fibers, the strength and bursting pressure of the central air pipe are remarkably improved, and the high-pressure plastic composite hose can better adapt to the complex working environment; the cable core structure adopts composite optical fiber analog signal transmission to replace a traditional digital signal seismic cable, so that the quality and efficiency of signal transmission are improved; and the armor layer adopts a double-layer steel wire armor structure, so that the overall mechanical strength and tensile property of the cable are improved.