Method and system for analyzing filling for karst reservoir based on spectrum decomposition and machine learning

A system for analyzing filling for a karst reservoir based on spectrum decomposition and machine learning, and describing a structure of the karst reservoir by adopting a lithology shielding technology and a 3D engraving technology for reservoir identification, comprising:

• at least one processor and a memory that is in communication connection with the at least one processor, wherein the memory is configured to store instructions capable of being executed by the processor;
• an original geophysical logging data acquisition module, a seismic data acquisition module, an original geophysical logging data preprocessing module, a seismic data preprocessing module, a well-to-seismic calibration and characteristic parameter selection module, an isochronous framework model constructing module, an inter-well characteristic parameter simulation module, a module for characterizing the boundary of the inter-well karst cave system, a module for characterizing the lithological and physical boundaries of internal filling of the karst cave and a 3D module for describing the structure and the filling characteristics of the paleokarst cave;
• the original geophysical logging data acquisition module is configured to acquire the original logging data of each sample well by the logging equipment, comprising:
• measuring a self-potential SP of each sample well by a measuring electrode, measuring a natural gamma ray GR of each sample well by a natural gamma ray downhole device and a natural gamma ray ground instrument, and acquiring a caliper CAL of each sample well by a caliper arm;
• and acquiring data of a resistivity curve:
• deep lateral logging RLLD, shallow lateral logging RLLS and micro-lateral logging RLLM, and data of a physical property characterization curve:
• compensated neutrons CNL, a compensated acoustic curve AC and a density curve DEN by traditional logging equipment;
• and
• acquire the definite lithological information and physical information of the individual depth segment based on the imaging logging information, the drilling information, the mud logging information and the rock-core information, so as to determine the depth data of the marker bed at the target horizon;
• the seismic data acquisition module is configured to acquire the data of the original seismic wave reflected signal by the seismic wave excitation device and the receiving device and acquire isochronous 3D distribution of the marker bed at the target horizon according to a waveform of the data of the original seismic wave reflected signal;
• the original geophysical logging data preprocessing module is configured to draw the data of the logging curves based on the original logging data of the sample wells and carry out abnormal value processing and standardization processing, so as to obtain the data of the standardized logging curves;
• the seismic data preprocessing module is configured to obtain the high-precision 3D seismic amplitude data body by mixed-phase wavelet estimation and maximum posteriori deconvolution and enhancing diffusion filtering based on the data of the seismic wave reflected signal;
• the well-to-seismic calibration and selecting characteristic parameters:
• acquiring a wave impedance curve of each sample well based on the compensated acoustic curve AC and the density curve DEN in the data of the standardized logging curve, so as to calculate a reflection coefficient curve;
• acquiring a preferred frequency of a ricker wavelet, so that the preferred frequency is consistent with a dominant frequency of the high-precision 3D seismic amplitude data body;
• carrying out convolution operation on the ricker wavelet and the reflection coefficient curve, so as to obtain a synthetic seismic record;
• comparing the depth data of the marker bed at the target horizon and a isochronous 3D distribution of the marker bed at the target horizon for well-to-seismic calibration;
• calculating a correlation between the synthetic seismic record and a waveform of a near-well seismic channel;
• and when the correlation is more than or equal to a preset first threshold, judging that a result of well-to-seismic calibration is qualified, so as to obtain the time-depth transforming relationship between the data of the logging curve and the seismic record and the characteristic parameters sensitive to the reservoir,
• wherein a method for obtaining the characteristic parameters sensitive to the reservoir comprises:
• drawing a columnar statistical graph of responses of the logging parameters, which are generated to different geological bodies near the wells;
• and when a numerical value of data of a certain standardized logging curve used for distinguishing data points that are more than a second threshold preset by different logging interpretation conclusions, selecting the data of the standardized logging curve as the characteristic parameters sensitive to the reservoir, wherein the characteristic parameters sensitive to the reservoir at least comprise the wave impedance IMP and also comprise one or more of the caliper CAL, the natural gamma ray GR, the self-potential SP, the data of the resistivity curve:
• the deep lateral logging RLLD, the shallow lateral logging RLLS and the micro-lateral logging RLLM, and the data of the physical property characterization curve:
• the compensated neutrons CNL, the compensated acoustic curve AC and the density curve DEN;
• the isochronous framework model constructing module is configured to construct the isochronous framework model based on law of sedimentary strata reflected by the high-precision 3D seismic amplitude data body and the time-depth transforming relationship;
• the inter-well characteristic parameter simulation module is configured to determine the parameter of optimal sample number, which reflect the overall geological condition, based on the data of the standardized logging curve of each sample well, determining the parameter of an optimal sample number, which reflect an overall geological condition;
• selecting data of the characteristic parameters sensitive to the reservoir of the sample wells to construct an initial model, wherein the data has the highest correlation of the seismic waveforms, and a number of pieces of data is equal to the parameter of the optimal sample number;
• constantly correcting the parameters of the initial model;
• and outputting high-precision characteristic value simulation result data bodies, wherein the high-precision characteristic value simulation result data bodies are data bodies corresponding to the characteristic parameters sensitive to the reservoir one to one;
• the module for characterizing a boundary of the inter-well karst cave system is configured to draw the histogram of the wave impedance curves based on the wave impedance curves IMP in the characteristic parameters sensitive to the reservoir of the sample wells, and determine the wave impedance threshold for distinguishing the paleokarst cave and the surrounding rock;
• and
• based on the wave impedance threshold of the paleokarst cave and the surrounding rock, define the region that is greater than the impedance threshold in the high-precision characteristic value simulation result data bodies as the bed rock of the carbonate rock, and define the region that is lower than the impedance threshold as the development position of the paleokarst cave type reservoir, so as to obtain the structure characteristics and a spatial distribution law of the paleokarst cave type reservoir;
• the module for characterizing the lithological and physical boundaries of internal filling of the karst cave is configured to characterize the lithological and physical boundaries of internal filling of the cave:
• constructing the cross plot based on the characteristic parameters sensitive to the reservoir, and carrying out box selection on the data points in the cross plot as the logging interpretation sample points according to the lithological information and the physical information of the individual depth segment, so as to obtain the interpretation conclusion threshold of the filling degree and the types of the fillings;
• and
• carrying out cross analysis on the high-precision characteristic value simulation result data bodies by the interpretation conclusion threshold, so as to obtain the internal filling degree of the cave and the characteristics of a 3D spatial morphologies of the combination of the fillings;
• the 3D module for describing the structure and the filling characteristics of the paleokarst cave is configured to describe the structure and the filling of the paleokarst cave:
• engraving the spatial distribution of the paleokarst cave and the development characteristics of the internal filling, by adopting the lithology shielding technology and the 3D engraving technology, based on structure characteristics and the spatial distribution law of the paleokarst cave type reservoir as well as the internal filling degree of the cave and the characteristics of the 3D spatial morphologies of the combination of the fillings, to generate a diagram characterizing the paleokarst cave;
• wherein the system is configured to determine a position of the paleokarst cave of an interval of interest, divide the paleokarst cave into a fully filled reservoir, a semi-filled reservoir and a bed rock reservoir according to the filling degree, and divide a type of the reservoir into a sedimentation filled reservoir, a collapse filled reservoir, a chemically filled reservoir and a mixed filled reservoir by utilizing lithological and physical characteristics of geological bodies, thereby enabling reservoir identification.