CN-121721722-B - Waveform-curve dynamic matching-based inversion method and system for thin coal seam and gangue

CN121721722BCN 121721722 BCN121721722 BCN 121721722BCN-121721722-B

Abstract

The invention relates to the technical field of oil and gas geophysical exploration, and discloses a method and a system for inverting a thin coal seam and gangue inclusion based on waveform-curve dynamic matching, which adopt sliding time window waveform matching based on a Dynamic Time Warping (DTW) algorithm, excavate the thin lithology change of the thin layer contained in the seismic waveform while overcoming waveform distortion caused by local stratum speed difference, introduce a dual constraint weighting formula for fusing waveform similarity and space distance, take the DTW similarity as a matched centering weight, the space smoothness of the inversion result is limited by using space distance weight constraint, so that the inversion result not only can be better fit to all known well points, but also can be used for predicting unknown stratum mutation zones among wells more reliably, finally, the inversion result optimizes the horizontal well track design, and the drilling track is successfully guided to effectively avoid a gangue-bearing development interval in the bisection coal seam, so that the earthquake inversion method capable of improving the prediction precision of the thin coal seam and the gangue is provided.

Inventors

LI ZHIXIANG
CHEN YANHU
ZHAO HAISHAN
ZHANG LINGLING
JIN GUOYU
ZHANG XUSHENG
CAO LIANYU
JI XIAOCHAO
LI NA
NIE XINJIA

Assignees

北京中恒利华石油技术研究所

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (10)

1. The method for inverting the thin coal seam and the gangue based on the waveform-curve dynamic matching is characterized by comprising the following steps of: Acquiring logging data of a plurality of drilled wells in a work area, correcting and standardizing the logging data, calculating to obtain wave impedance curves of the drilled wells in a time domain, and generating target curve data for distinguishing a thin coal seam from gangue; Acquiring well-passing seismic channel data and estimated wavelet data of a well, performing well-shock calibration operation on the target curve data to determine the top boundary time and the bottom boundary time of a target interval, and performing time domain expansion operation based on the top boundary time and the bottom boundary time to generate a working interval; Extracting seismic waveform data corresponding to each drilled well in the working layer section, performing time alignment on the seismic waveform data and target curve data, and then performing pairing storage to construct a waveform-wave impedance curve database; For any seismic channel to be inverted in a three-dimensional seismic data volume, segmenting the seismic waveform data to be inverted extracted from the working layer section based on a sliding time window strategy, performing similarity calculation on the segmented seismic waveform data to be inverted and the drilled seismic waveform data in the database by adopting a dynamic time warping algorithm, and selecting the front N drilled wells with the highest similarity measurement value as a sample well group; Calculating fusion weights based on the wave impedance curves corresponding to the sample well groups, the similarity measurement values and space distance data between the sample wells and the seismic channels to be inverted, and performing weighted fusion operation on the wave impedance curves of the sample well groups to generate wave impedance inversion curves of the seismic channels to be inverted; and establishing a three-dimensional wave resistance body based on wave impedance inversion curves of all the seismic channels to be inverted, and generating a coal seam structure plan by utilizing the three-dimensional wave resistance body.
2. The inversion method of thin coal seam and gangue comprising the steps of obtaining logging data of a plurality of drilled wells in a work area, correcting and standardizing the logging data, calculating wave impedance curves of the drilled wells in a time domain, and generating target curve data for distinguishing the thin coal seam from the gangue, wherein the inversion method comprises the following steps: Acquiring acoustic time difference logging data and density logging data of a plurality of drilled wells in a work area; performing environment correction processing and standardization processing on the acoustic time difference logging data and the density logging data, calculating a wave impedance sequence based on the standardization result, and performing uniform sampling interval resampling processing on the wave impedance sequence to generate a wave impedance curve; And taking the wave impedance curve as target curve data for distinguishing the thin coal seam from the gangue.
3. The method for inversion of thin coal and gangue systems based on waveform-curve dynamic matching as claimed in claim 1, wherein obtaining the well-drilled seismic trace data and estimated wavelet data, performing well-shock calibration operation on the target curve data to determine the top boundary time and bottom boundary time of the target interval, comprises: Acquiring the well-drilled seismic trace data and estimated wavelet data, converting a wave impedance curve into a reflection coefficient sequence, and performing convolution operation in combination with the estimated wavelet to generate a synthetic seismic record; Performing time alignment operation on the synthetic seismic records and the through-well seismic trace data and maximizing output of calibration alignment results by correlation coefficients; And determining the top boundary time and the bottom boundary time of the target interval based on the calibration alignment result.
4. The method for inversion of thin coal and gangue based on waveform-curve dynamic matching as claimed in claim 1, wherein the method for generating the working interval comprises the steps of: Calculating the time thickness of the target interval based on the top boundary time and the bottom boundary time; based on a preset edging factor and a target interval time thickness, performing expansion operation on the top boundary time and the bottom boundary time to generate a working interval top boundary time and a working interval bottom boundary time; and determining the working layer section based on the working layer section top boundary time and the working layer section bottom boundary time.
5. A thin coal seam and gangue inversion method based on waveform-curve dynamic matching as claimed in claim 3, wherein in said working interval, extracting the corresponding seismic waveform data of each well, and performing time alignment and then pairing storage of the seismic waveform data and the target curve data, constructing a waveform-wave impedance curve database, comprising: Performing mapping conversion from a depth domain to a time domain on each well-drilled wave impedance curve based on a well shock calibration alignment result to obtain a time domain wave impedance curve; Intercepting seismic waveform data in the working interval from the three-dimensional seismic data volume along each well trajectory; Respectively intercepting and executing time sampling point alignment on the seismic waveform data and the time domain wave impedance curve in the working layer section to generate well point waveform-curve data pairs; and storing the waveform-curve data pairs of each well point according to the well identification to form a waveform-wave impedance curve database.
6. The method for inversion of thin coal seams and gangue inclusion based on dynamic matching of waveforms-curves as claimed in claim 1, wherein for any seismic trace to be inverted in a three-dimensional seismic data volume, the seismic waveform data to be inverted extracted in the working interval is segmented based on a sliding time window strategy, and the method specifically comprises the following steps: calculating the average time thickness of the gangue inclusion based on a time thickness set of the gangue inclusion in the drilled well, and setting the sliding time window length and the sliding step length based on the average time thickness; And in the working layer section top boundary time and the working layer section bottom boundary time of the working layer section, sliding window interception is carried out on the seismic waveform data to be inverted according to the sliding time window length and the sliding step length, and a short waveform sequence to be inverted is generated.
7. The method for inversion of thin coal seam and gangue inclusion based on dynamic matching of waveform-curve as claimed in claim 6, wherein the method for inversion of the segmented seismic waveform data to be inverted and the seismic waveform data in the database by adopting dynamic time warping algorithm performs similarity calculation, and selects the front N drilled wells with the highest similarity measurement value as a sample well group, specifically comprising: Respectively acquiring a short waveform to be inverted and well point short waveforms corresponding to all drilled wells in each sliding time window, and executing dynamic time warping operation on the short waveform to be inverted and the well point short waveforms to generate distance values; Performing aggregation operation on the distance values of the same drilled well in a plurality of time windows to obtain an overall distance value, and taking the overall distance value as a similarity measurement value; And sorting from small to large according to the similarity measurement values, selecting the first N drilling holes with the smallest total distance value to form a sample well group, and outputting a total distance value set of the sample well group.
8. The method for inversion of thin coal seam and gangue based on dynamic matching of waveform-curve as claimed in claim 7, wherein the calculating of the comprehensive weight based on the waveform impedance curve corresponding to the sample well group, the similarity measure and the spatial distance data between the sample well and the seismic trace to be inverted specifically comprises: calculating waveform similarity weight based on the similarity measurement value, wherein the expression is as follows: ; In the formula, For the weight of the similarity of the waveforms, Is the first The overall distance value of the individual sample wells from the waveform of the channel to be inverted, Is the standard deviation of candidate well spacing; Calculating a spatial distance weight based on the planar distance between the sample well and the seismic trace to be inverted, wherein the expression is as follows: ; In the formula, As the weight of the spatial distance, Is the first The planar distance of the well of the port sample from the channel to be inverted, Is a geology related radius; And calculating comprehensive weights based on the waveform similarity weights and the spatial distance weights, wherein the expression is as follows: ; In the formula, In order to integrate the weights of the weights, The coefficients are adjusted for the weights.
9. A method of inversion of thin coal and gangue based on dynamic matching of waveforms-curves as claimed in claim 8, wherein performing a weighted fusion operation on the sample well group wave impedance curves to generate the wave impedance inversion curve of the seismic trace to be inverted, comprises: At each time sampling point in the working layer section, acquiring a wave impedance value of a corresponding time sampling point in a sample well group, and carrying out weighted average calculation on the wave impedance value according to the comprehensive weight corresponding to each sample well, wherein the expression is as follows: ; In the formula, Is the pseudo well curve generated at time Is used to determine the wave impedance value of the (c) wave, Is the first The weight of the well of the mouth sample, Is the first Time of the well profile of the mouth sample Is a value of (2); and taking the calculation result as an inversion wave impedance value of the seismic channel to be inverted at the time sampling point to form a complete wave impedance inversion curve.
10. A thin coal seam and gangue inversion system based on waveform-curve dynamic matching, the system comprising: the acquisition module is used for acquiring logging data of a plurality of drilled wells in a work area, correcting and standardizing the logging data, calculating to obtain wave impedance curves of the drilled wells in a time domain, and generating target curve data for distinguishing a thin coal seam from gangue; the operation module is used for acquiring well-passing seismic channel data and estimated wavelet data of a well, performing well-shock calibration operation on the target curve data to determine the top boundary time and the bottom boundary time of a target interval, and performing time domain expansion operation based on the top boundary time and the bottom boundary time to generate a working interval; the extraction module is used for extracting the seismic waveform data corresponding to each drilled well in the working layer section, performing time alignment on the seismic waveform data and the target curve data, and then performing pairing storage to construct a waveform-wave impedance curve database; The selection module is used for segmenting the seismic waveform data to be inverted extracted from the working layer section based on a sliding time window strategy aiming at any seismic channel to be inverted in the three-dimensional seismic data volume, performing similarity calculation on the segmented seismic waveform data to be inverted and the drilled seismic waveform data in the database by adopting a dynamic time warping algorithm, and selecting the front N drilled wells with the highest similarity measurement value as a sample well group; The generation module is used for calculating fusion weights based on the wave impedance curves corresponding to the sample well groups, the similarity measurement values and the space distance data between the sample wells and the seismic channels to be inverted, and performing weighted fusion operation on the wave impedance curves of the sample well groups to generate the wave impedance inversion curves of the seismic channels to be inverted; The building module is used for building a three-dimensional wave resistance body based on wave resistance inversion curves of all the seismic channels to be inverted, and generating a coal seam structure plan by using the three-dimensional wave resistance body.

Description

Waveform-curve dynamic matching-based inversion method and system for thin coal seam and gangue Technical Field The invention relates to the technical field of oil and gas geophysical exploration, in particular to a waveform-curve dynamic matching-based inversion method and system for thin coal seams and gangue inclusion. Background Earthquake prediction methods for thin coal seams and gangue are mainly classified into several types. The method is based on model constrained sparse pulse inversion, well logging wave impedance is utilized to constrain seismic inversion, and a wave impedance data volume is obtained for interpretation. And secondly, geostatistical inversion, combining geostatistical simulation and seismic inversion, can generate a plurality of high-resolution lithology models with equal probability. Thirdly, conventional waveform indication inversion utilizes seismic waveform similarity to extrapolate well log information in space. In addition, identification techniques based on qualitative seismic attributes such as frequency division interpretation, attribute analysis, etc. are also often used. However, the current technology has a great limitation in use. The model-based inversion method has low resolution due to the seismic frequency band, has insufficient resolution capability on thin interbings (particularly sub-meter thin coal seams and gangue), is highly dependent on an initial model and wavelets, and has strong multi-solution. Geostatistical inversion is computationally expensive, while the variogram is difficult to characterize well complex thin interbings. In the conventional waveform indication method, the problem is that the accurate matching degree cannot be ensured because one or more fixed waveforms are compared with the matching how to realize the local stretching/compression of the stratum, and in addition, the attribute analysis method is more qualitative, so that the accurate depiction like lithology boundaries is difficult to realize. Aiming at the defects, the current requirements of coal bed gas/coal rock gas fine geology cannot be met, and the application is difficult. Therefore, there is a need to develop a seismic inversion method that can improve the accuracy of thin coal and gangue inclusion predictions. Disclosure of Invention The invention provides a waveform-curve dynamic matching-based inversion method and system for thin coal seams and gangue inclusion, and aims to solve at least one technical problem. In order to achieve the above purpose, the invention provides a waveform-curve dynamic matching-based inversion method for thin coal seams and gangue, which comprises the following steps: Acquiring logging data of a plurality of drilled wells in a work area, correcting and standardizing the logging data, calculating to obtain wave impedance curves of the drilled wells in a time domain, and generating target curve data for distinguishing a thin coal seam from gangue; Acquiring well-passing seismic channel data and estimated wavelet data of a well, performing well-shock calibration operation on the target curve data to determine the top boundary time and the bottom boundary time of a target interval, and performing time domain expansion operation based on the top boundary time and the bottom boundary time to generate a working interval; Extracting seismic waveform data corresponding to each drilled well in the working layer section, performing time alignment on the seismic waveform data and target curve data, and then performing pairing storage to construct a waveform-wave impedance curve database; For any seismic channel to be inverted in a three-dimensional seismic data volume, segmenting the seismic waveform data to be inverted extracted from the working layer section based on a sliding time window strategy, performing similarity calculation on the segmented seismic waveform data to be inverted and the drilled seismic waveform data in the database by adopting a dynamic time warping algorithm, and selecting the front N drilled wells with the highest similarity measurement value as a sample well group; Calculating fusion weights based on the wave impedance curves corresponding to the sample well groups, the similarity measurement values and space distance data between the sample wells and the seismic channels to be inverted, and performing weighted fusion operation on the wave impedance curves of the sample well groups to generate wave impedance inversion curves of the seismic channels to be inverted; and establishing a three-dimensional wave resistance body based on wave impedance inversion curves of all the seismic channels to be inverted, and generating a coal seam structure plan by utilizing the three-dimensional wave resistance body. Optionally, acquiring logging data of a plurality of drilled wells in a work area, correcting and standardizing the logging data, calculating to obtain wave impedance curves of the drilled wells in a time domain, and generating ta