CN-121634204-B - DAS data self-adaptive scale distance reconstruction and multi-resolution imaging method

CN121634204BCN 121634204 BCN121634204 BCN 121634204BCN-121634204-B

Abstract

The invention provides a DAS data self-adaptive gauge length reconstruction and multi-resolution imaging method, and belongs to the technical field of optical fiber sensing and signal processing. The method comprises the following steps of S1, exciting a seismic source vehicle in an active source mode to generate seismic waves, and S2, enabling a DAS system to perform scale distance assignment The method comprises the steps of completing VSP data acquisition to obtain original strain rate data with high spatial sampling rate, preprocessing the original strain rate data to obtain preprocessed strain rate data, normalizing and denoising, reconstructing a continuous wave field to obtain a continuous wave field function, adaptively generating a gauge length, and fusing and imaging multi-resolution data, wherein the preprocessing comprises the step of S4. The method only relies on the minimum gauge length to collect data, and optical fibers or demodulation hardware is not required to be changed, so that the adaptability of the existing DAS-VSP system can be directly improved, the method can be used for other distributed optical fiber sensing scenes in a migration mode, and higher observation quality and better engineering usability are achieved.

Inventors

AN SHUJIE
RAN ZENGLING
CHEN HAIFENG
CHEN YUNFEI
WANG YUQI
YAO WEIYI
HU YILEI
YUAN XIANGKAI

Assignees

电子科技大学

Dates

Publication Date: 20260508
Application Date: 20260204

Claims (6)

1. A method for adaptive gauge length reconstruction and multi-resolution imaging of DAS data, comprising the steps of: step S1, exciting a seismic source vehicle in an active source mode to generate seismic waves; step S2 DAS System to specify the gauge length The method comprises the steps of completing VSP data acquisition to obtain original strain rate data with high spatial sampling rate, wherein DAS refers to distributed optical fiber acoustic wave sensing, and VSP refers to vertical seismic profile; step S3, preprocessing the original strain rate data to obtain preprocessed strain rate data, wherein the preprocessing comprises normalization and denoising; s4, performing continuous wave field reconstruction on the preprocessed strain rate data to obtain a continuous wave field function; s5, adaptively generating a scale distance, and carrying out multi-resolution data fusion and imaging; Step S5 includes the steps of: s51, generating virtual channel data by adopting high signal-to-noise ratio data with large gauge length for a deep region to form a stratum structure image so as to realize high signal-to-noise ratio imaging; S52, generating high-resolution virtual channel data by adopting high-resolution data with small gauge length for a shallow layer region to form a stratum structure image so as to realize high-resolution imaging; step S53, imaging is carried out in a transition area by adopting smooth transition of data with different scale distance resolutions, virtual track data with different scale distance resolutions are generated, multi-resolution imaging is realized, and final self-adaptive fusion data are formed; step S51 includes the steps of: s511, determining the target position of a virtual track for a deep region according to imaging target requirements, and generating a scale array by adopting a large scale; Step S512, generating virtual track data for all virtual gauges in the gauge length array by simulating a DAS physical measurement process; The virtual track data is calculated as follows: Wherein, the Representing virtual track data; Is the first The positions of the sensing channels represent the target positions of the virtual channels; Is a proportionality coefficient; representing the first of the array of gauges Virtual gauge lengths; Indicating movement to the right Is a continuous wave field function of (2); Indicating movement to the left Is a continuous wave field function of (2); step S513, calculating the signal-to-noise ratio of all the virtual channel data, adaptively selecting an optimal scale distance through the signal-to-noise ratio, taking the virtual scale distance corresponding to the maximum signal-to-noise ratio as the optimal scale distance, and taking the virtual channel data corresponding to the maximum signal-to-noise ratio as the stratum structure image; the signal-to-noise ratio calculation formula of the virtual channel data is as follows: Wherein, the Representing a window of noise and, Representing a signal window; Representing virtual track data A noise segment in (a); Representing virtual track data Is a signal segment of the (b); Is the window length.
2. The DAS data adaptive gauge length reconstruction and multi-resolution imaging method of claim 1, wherein the raw strain rate data in step S2 is represented as , wherein, Indicating the position of the sensing channel, Is time series, the first The original strain rate data of each sensing channel is as follows Represents the first The positions of the individual sensing channels Original strain rate data at which, wherein, Represent the first The position of the individual sensing channels.
3. The DAS data adaptive gauge length reconstruction and multi-resolution imaging method of claim 2, wherein step S3 comprises the steps of: step S31, normalizing the original strain rate data, and compressing the original strain rate data in a section Obtaining normalized strain rate data; First, the The positions of the individual sensing channels The original strain rate data normalization processing formula is as follows: Wherein, the To normalized strain rate data, represent the first The positions of the individual sensing channels Processing normalized strain rate data; the representation takes the minimum value of the value, Indicating that the maximum value is taken; step S32, carrying out frequency domain bandpass filtering on the normalized strain rate data to suppress low-frequency noise and high-frequency noise which can be seriously amplified in the integration process, so as to obtain preprocessed strain rate data; First, the The positions of the individual sensing channels Normalized strain rate data The frequency domain bandpass filtering formula is performed as follows: Wherein, the Frequency spectrum of strain rate data, representing the first The positions of the individual sensing channels Frequency spectrum of strain rate data at; representing a fourier transform function; Representing the frequency; representing imaginary units; For the data band-pass filtered in the frequency domain, represent the first The positions of the individual sensing channels Data after band-pass filtering in a frequency domain; representing a bandpass filter response function; Representing the filtered data, i.e. the pre-processed strain rate data, representing the first The positions of the individual sensing channels The pre-processed strain rate data at the location; Preprocessing the original strain rate data of all the sensing channels to obtain preprocessed strain rate data 。
4. The DAS data adaptive gauge length reconstruction and multi-resolution imaging method of claim 3, wherein step S4 comprises the steps of: Step S41, performing integral transformation on the preprocessed strain rate data to obtain a particle fluctuation function: Wherein, the Representing the first particle fluctuation function The positions of the individual sensing channels A particle fluctuation function at; Representing a scaling factor related to the formation velocity, The integrated starting depth; step S42, performing continuous wave field recovery on the quality point fluctuation function by using a spatial interpolation algorithm to obtain a continuous wave field function: Wherein, the Representing a continuous wave field function; Representing a spatial interpolation operator.
5. The DAS data adaptive gauge length reconstruction and multi-resolution imaging method of claim 4, wherein step S52 comprises the steps of: step S521, determining the target position of the virtual track for the shallow region according to the imaging target requirement, and generating a scale array by adopting a small scale; Step S522, generating virtual track data for all virtual gauges in the gauge length array by simulating a DAS physical measurement process; step S523, calculating the signal-to-noise ratio of all the virtual channel data, adaptively selecting an optimal scale distance through the signal-to-noise ratio, taking the virtual scale distance corresponding to the maximum signal-to-noise ratio as the optimal scale distance, and taking the virtual channel data corresponding to the maximum signal-to-noise ratio as the stratum structure image.
6. The DAS data adaptive gauge length reconstruction and multi-resolution imaging method of claim 5, wherein step S53 comprises the steps of: step S531, determining the target position of a virtual track in a transition area according to imaging target requirements, dividing the stratum of the transition area into 4 layers averagely, and generating a scale array by adopting virtual scale distances in the middle of the large scale distance and the small scale distance of each layer; S532, generating virtual track data for all virtual scale distances in each layer of scale distance array by simulating a DAS physical measurement process; In step S533, the signal-to-noise ratio of all the virtual channel data is calculated, the optimal scale distance is selected in a self-adaptive mode through the signal-to-noise ratio, the virtual scale distance corresponding to the maximum signal-to-noise ratio of each layer is taken as the optimal scale distance of the layer, and the virtual channel data corresponding to the maximum signal-to-noise ratio of each layer is taken as the stratum structure image of the layer.

Description

DAS data self-adaptive scale distance reconstruction and multi-resolution imaging method Technical Field The invention belongs to the technical field of optical fiber sensing and signal processing, in particular to a distributed optical fiber acoustic wave sensing (DAS) and a data processing and reconstruction method thereof in Vertical Seismic Profile (VSP) observation, and particularly relates to a DAS data self-adaptive gauge length reconstruction and multi-resolution imaging method. Background Distributed fiber acoustic wave sensing (DAS) technology is widely used for Vertical Seismic Profile (VSP) observation, oil and gas wellbore imaging, subsurface structure monitoring, and other geophysical exploration tasks. DAS systems obtain axial strain along the fiber direction by demodulating the phase difference of adjacent scattering segments on the fiber, so gauge length (gauge length) is an important parameter for spatial sampling characteristics and waveform response. The gauge length determines the integrated length of the system to the along-path strain, has a direct effect on spatial resolution, signal-to-noise ratio (SNR), and sensitivity to seismic waves of different wavelengths. Existing researches and practical engineering show that when the DAS gauge length is too large, strain in the range along the gauge length is averaged, which may cause waveform flattening, reflected energy attenuation and high-frequency component loss, especially in the vicinity of a wellhead or in a region with rapid formation change, and the spatial resolution is reduced. And when the gauge length is too small, the number of scattering points is reduced, the noise is relatively enhanced, so that the SNR is reduced, and weak signals are difficult to reliably identify. In addition, by comparison with conventional geophone recordings and cross-product cross-correlation analysis, the optimal gauge length is affected by factors such as formation velocity, wavelength, source type, and wellbore location. At the same time, the tap length of the DAS should be close to 1/3 of the seismic dominant wavelength to achieve a higher SNR, but if the tap length exceeds a significant proportion of the wavelength, the waveform will be severely smoothed. The existing method can not continuously adjust the gauge length according to the well section change in the same underground record, and can not flexibly reconstruct the gauge length in the data processing stage. For VSP observations where formation velocity changes are significant, wellbore structures are complex, or both high and low frequency events need to be identified, a fixed gauge length often has difficulty in achieving both spatial resolution and signal-to-noise ratio. The existing DAS-VSP technology generally adopts fixed gauge length for acquisition, however, the optimal gauge length depends on stratum wave velocity, dominant frequency, shaft position and noise condition, and different gauge lengths are often needed for different well sections to obtain optimal reflection imaging quality and waveform reliability. Therefore, how to break through the limitation of fixed gauge length acquisition, realize the flexible adjustment of gauge length, can be used for customizing the change of the position along a shaft, and obtain multi-resolution DAS data corresponding to different gauge lengths in a data processing stage, and is a technical problem to be solved in the field. Disclosure of Invention The invention aims to provide a DAS data self-adaptive scale distance reconstruction and multi-resolution imaging method, which can generate multi-resolution DAS data reconstruction methods with arbitrary scale distances at arbitrary positions, dynamically provide optimal resolution, enhance engineering adaptability and observation efficiency of a DAS system, and solve the technical problem that spatial resolution and signal to noise ratio cannot be compatible due to fixed scale distances in DAS-VSP in the prior art. In order to solve the technical problems, the specific technical scheme of the invention is as follows: A DAS data adaptive scale distance reconstruction and multi-resolution imaging method, the method comprising the steps of: step S1, exciting a seismic source vehicle in an active source mode to generate seismic waves; step S2 DAS System to specify the gauge length Completing VSP data acquisition to obtain original strain rate data with high spatial sampling rate; step S3, preprocessing the original strain rate data to obtain preprocessed strain rate data, wherein the preprocessing comprises normalization and denoising; s4, performing continuous wave field reconstruction on the preprocessed strain rate data to obtain a continuous wave field function; and S5, adaptively generating a scale distance, and carrying out multi-resolution data fusion and imaging. Further, the raw strain rate data in step S2 is expressed as, wherein,Indicating the position of the sensing channel,Is time se