13 results about "Seismic noise" patented technology

The DAS seismic noise suppression method based on the conditional latent diffusion model belongs to the fields of machine learning and seismic data processing. This invention utilizes a latent signal representation module encoder to compress high-dimensional seismic data into a low-dimensional latent space, improving computational efficiency while preserving signal characteristics. In the inference phase, noise components are estimated through a noise prediction network, and clean latent variable estimation and state transitions are iteratively performed according to a deterministic sampling strategy. The decoder then reconstructs the denoised seismic data. In the training phase, self-supervised pre-training of the latent signal representation module, supervised training of the noise prediction network, and joint fine-tuning of all network parameters are performed sequentially, using KL divergence loss, reconstruction loss, and mean square error loss to simultaneously optimize network weights. This invention significantly reduces the computational overhead of the diffusion model while suppressing various complex noises.

The invention belongs to the technical field of seismic signal processing, and provides a seismic noise de-noising processing method and system based on transfer learning, and the method comprises the steps: optimizing an initial de-noising convolutional neural network, and obtaining an optimized de-noising convolutional neural network; inputting target noisy seismic data to be subjected to intelligent denoising into the optimized denoising convolutional neural network for segmentation processing, splicing processing and data edge filling processing in sequence, and predicting corresponding target denoising seismic data; and based on the plurality of denoising effect evaluation indexes, evaluating the denoising effect of the target denoised seismic data from an image angle and a numerical angle, and generating a corresponding evaluation result. Through the intelligent denoising processing method, the data edge can be filled, the overweight splicing illusion can be effectively avoided, and the precision of the intelligent denoising result is finally improved.

The invention discloses a method and system for reconstructing and purifying a noise cross-correlation function of seismic background noise, and the method comprises the following steps: collecting an original noise signal, carrying out the non-uniform downsampling of the original noise signal through employing the sparse characteristic of the noise cross-correlation function of the seismic background noise in an effective frequency band, and combining with a compressed sensing theory, and obtaining a non-uniform downsampling signal; obtaining under-sampling data; reconstructing a full-band noise cross-correlation function from the undersampled data with high precision by adopting a fast iterative shrinkage threshold algorithm to obtain a reconstructed noise cross-correlation function; performing frequency-wave number transformation on the reconstructed noise cross-correlation function, and realizing three-component separation by using an FK filtering mask to obtain a separated FK domain effective signal; and carrying out inverse transformation on the separated FK domain effective signal, and finally outputting a purified noise cross-correlation function.

This invention relates to a desert earthquake noise suppression method based on a multi-scale attention interaction network, belonging to the fields of machine learning and seismic image processing. Addressing the problem of low-frequency noise suppression in desert earthquake images, this invention proposes a multi-scale attention interaction network. This network first downsamples the input seismic data, then adaptively learns the complex features of the multi-scale seismic data using dual-branch convolutional layers with different kernel sizes. It then employs multiple attention mechanisms, including permutation attention and coordinate attention, to integrate and interact with multi-scale features across channel, spatial, and coordinate dimensions to obtain continuous seismic signals. This method uses a hybrid loss function of mean squared error and mean cosine similarity, combining the multi-scale strategy with the attention mechanism to improve the denoising effect of desert earthquake images. Compared with single-scale denoising convolutional neural networks, this invention significantly suppresses desert noise and outperforms single-scale denoising convolutional neural networks in restoring the continuity of the phase axis.

Embodiments of a system and methods for utilizing ambient seismic noise to implement a modality based on the geometric phase for sending the effect of environmental conditions on ground properties are disclosed. The geometric phase is a global measure of the geometry of the seismic field which can be reconstructed via seismic interferometry. The change in geometric phase is the defined as the rotation angle of the complex state vector.

The utility model discloses an anti-seismic noise-reduction type casting supporting structure for a numerical control operation platform. The anti-seismic noise-reduction type casting supporting structure comprises a bottom plate, a damping mechanism on the bottom plate and a limiting mechanism used for fixing a casting. Mounting holes are formed in the four corners of the bottom plate, long-strip-shaped holes are formed in the four corners of the inner end of the bottom plate, damping mechanisms are fixedly connected to the four corners of the upper surface of the bottom plate, lifting cylinders are slidably connected to the upper ends of the inner sides of cylinders in the damping mechanisms, and sound insulation plates are fixedly connected to the inner side faces of the lifting cylinders. The upper and lower ends of the springs in the cylinder and the sound insulation plate are fixedly connected to the bottom surface of the first connecting plate and the top surface of the bottom plate respectively; a shock pad is fixedly connected between the first connecting plate and the second connecting plate, and a limiting mechanism is arranged at the top of the second connecting plate. According to the anti-seismic noise reduction type casting supporting structure for the numerical control operation platform, anti-seismic noise reduction can be conducted on castings, and the situation that when the numerical control operation platform operates, the use conditions of the castings are affected is avoided.

This invention discloses a method and system for reconstructing and purifying the noise cross-correlation function of seismic background noise, comprising the following steps: acquiring the original noise signal; utilizing the sparsity characteristics of the noise cross-correlation function of seismic background noise in the effective frequency band, and combining compressed sensing theory to perform non-uniform downsampling on the original noise signal to obtain undersampled data; employing a fast iterative shrinking threshold algorithm to reconstruct the full-band noise cross-correlation function from the undersampled data with high precision, obtaining the reconstructed noise cross-correlation function; performing frequency-wavenumber transformation on the reconstructed noise cross-correlation function, and using an FK filter mask to achieve three-component separation, obtaining the separated effective signal in the FK domain; performing an inverse transformation on the separated effective signal in the FK domain, and finally outputting the purified noise cross-correlation function.

The invention relates to the technical field of geological disaster monitoring, and discloses a slope crack depth monitoring method, device and equipment and a readable storage medium. The method comprises the following steps: acquiring environmental seismic noise through a noise sensor preset at a monitored slope; screening the environmental seismic noise based on a preset spectrum peak detection algorithm to obtain a target noise spectrum signal; and determining a target slope crack depth based on the target noise spectrum signal and a preset crack depth inversion model. Environmental seismic noise of a monitored slope is collected, a target noise spectrum signal is obtained in combination with a preset spectrum peak detection algorithm, and then the target slope crack depth is determined based on the target noise spectrum signal and a preset crack depth inversion model. The problems of low efficiency and high safety risk caused by traditional manual contact type measurement adopted in existing slope crack depth monitoring are solved, the efficiency of slope crack depth monitoring is improved, and the monitoring danger coefficient is reduced.

The utility model provides an anti-seismic noise reduction structure of a transformer in a booster station, which comprises a transformer main body and a support shell, and further comprises a damping structure arranged in the support shell. By rotating a third rotating plate, two clamping plates are driven to be clamped into two first clamping grooves, damping cannot be affected while the transformer is conveniently and well installed and fixed, and by clamping a clamping frame into a second clamping groove, the situation that the third rotating plate rotates, the clamping plates are disengaged from the first clamping grooves, and consequently a transformer body is not firmly fixed is conveniently avoided. Vibration of the transformer body is reduced and buffered through the damping springs and the damping dampers, working noise of the transformer is well reduced due to reduction of the vibration, meanwhile, collision noise can be reduced through the second buffer silencing pad, the third buffer silencing pad and the second buffer silencing pad, the vibration of the transformer can be well reduced, and the service life of the transformer is prolonged. And the noise of the transformer can be well reduced while the transformer can be well damped.

The invention discloses a multi-dimensional seismic background noise imaging evaluation method and system based on big data, and the method comprises the following steps: obtaining seismic background noise imaging data; constructing an imaging evaluation model; and evaluating real-time seismic background noise imaging data. According to the method, the seismic background noise data are obtained by synchronously performing data imaging preprocessing on the obtained seismic background noise original data, then the processing effect of the data imaging preprocessing is evaluated, whether the reference processing effect is achieved or not is judged, and then the corresponding seismic background noise imaging data are obtained according to the seismic background noise data. An imaging evaluation model is constructed according to the seismic noise imaging data, and the real-time seismic background noise imaging data is evaluated in combination with the imaging evaluation model, so that the effect of reducing the influence of the background noise data on the accuracy of the imaging result is achieved, and the problem that the background noise data has great influence on the accuracy of the imaging result in the prior art is solved.

A background noise differential accompanying tomography system (1) is provided. The system (1) comprises a seismic wave recording device (101, 102, 103) and a processor (100). The processor (100) designates three of the seismic wave recording devices (101, 102, 103) as a seismic source (101), a first receiver (102), and a second receiver (103), respectively, and the seismic source (101) is arranged in a linear array with the first receiver (102) and the second receiver (103), the first receiver (102) and the second receiver (103) being arranged adjacent to each other. A processor (100) receives seismic noise and obtains noise interferometry, calculates a synthetic dispersion surface wave signal, and generates differential adjoint tomography based on observed dispersion surface waves obtained from the noise interferometry. A processor (100) estimates a shear wave velocity based on differential accompanying tomography, estimates a feature of the region using the estimated tomography results, and provides a graph with the feature of the region through a display or printer.

The application discloses a seismic data denoising model establishing method, a seismic data denoising method and related devices. The seismic data denoising model establishing method comprises the following steps: based on multiple sets of original seismic data, denoised seismic data corresponding to the original seismic data is obtained by combining a denoising method and a synthetic seismic method according to a set method ratio, as label data, and a sample set is obtained; the original seismic data of each sample in the sample set is subjected to wavelet frequency division processing to obtain multi-band data, and a converted sample set is obtained; a selected deep neural network model is trained by using the converted sample set, and a seismic data denoising model is obtained, which is used for denoising processing of input seismic data. Through the established seismic data denoising model, multiple seismic noises can be suppressed, and on the basis of obviously improving the noise suppression efficiency, the noise recognition precision is effectively improved.

The embodiment of the invention relates to a seismic imaging profile resolution improving method and device, and the method comprises the steps: collecting and analyzing a plurality of seismic imaging profiles, and constructing a plurality of seismic velocity models; obtaining a first target seismic imaging profile based on the seismic convolution model and the plurality of seismic velocity models; by randomly setting different migration imaging parameters, carrying out migration imaging on the plurality of seismic velocity models by using a reverse time migration imaging method to obtain a second target seismic imaging profile; expanding a seismic imaging profile with diversity characteristics by using adaptive synthesis sampling, and adding seismic noise to the seismic imaging profile; designing an initial generative adversarial network in combination with the generative adversarial network, and training to obtain an initial resolution improvement model; and carrying out transfer learning on the initial resolution improvement model to obtain a target resolution improvement model. Therefore, the signal-to-noise ratio and the resolution ratio of the offset profile can be improved, the calculation efficiency is high, and no parameter dependence exists.