CN-120762138-B - Deep stratum rock physical modeling method under complex lithology background

CN120762138BCN 120762138 BCN120762138 BCN 120762138BCN-120762138-B

Abstract

The invention relates to the technical field of geological modeling, in particular to a deep stratum petrophysical modeling method under a complex lithology background, which comprises the following steps: preprocessing logging data, optimizing a mineral model, carrying out petrophysical modeling and transverse wave prediction, and verifying the accuracy of the transverse wave prediction of the petrophysical model. According to the invention, four mineral models of quartz-clay model, calcite-clay model, coal and bauxite are respectively adopted, and optimized multi-mineral model processing is carried out in a layered manner, so that rock physical modeling for realizing complex lithology combination of deep coal-based stratum and transverse wave prediction under a pore structure is obtained, and the accuracy of transverse wave prediction is improved.

Inventors

XU XINGKE
LIU XIAOHONG
YU YINGSHUN
LAN ZHICHAO
WAN XIN

Assignees

北京地大博创科技有限公司

Dates

Publication Date: 20260508
Application Date: 20250717

Claims (7)

1. The method for the physical modeling of the deep stratum rock under the complex lithology background is characterized by comprising the following steps of: preprocessing logging data, namely selecting a high-quality standard well and a mark layer to preprocess original logging data to obtain logging data for building a mineral model; The mineral model optimization processing comprises the steps of determining a multi-mineral model by combining logging data with core data, analyzing lithology analysis data and all-rock X-ray diffraction data, optimizing the multi-mineral model by a layering section to obtain an optimal multi-mineral model, identifying lithology of special lithology in the optimal multi-mineral model, marking out a well section of the special lithology, and independently performing primary data backfilling and secondary processing to obtain petrophysical parameters for establishing a petrophysical model; According to Gassman equation, according to the rock physical parameters and rock elastic response characteristics under the conditions of simulating different lithology, different physical properties and different fluid properties, constructing a rock physical model based on multi-mineral model improvement, and carrying out transverse wave prediction through repeated iteration; the accuracy verification of the rock physical model prediction comprises the steps of comparing the longitudinal wave speed, the transverse wave speed and the longitudinal and transverse wave speed ratio obtained by calculating the rock physical model based on the improvement of the multi-mineral model with logging measured data, and verifying the accuracy of the transverse wave prediction based on the average error of the longitudinal wave speed and the transverse wave speed ratio and the longitudinal and transverse wave speed ratio; The multi-mineral model optimization processing method comprises the following steps: carrying out layering section processing on the logging data, and dividing the stratum into different layers according to lithology change and geological characteristics of the stratum; in each interval, through deep analysis of lithology analysis data and all-rock X-ray diffraction data, four mineral models of quartz-clay model, calcite-clay model, coal and bauxite are adopted to carry out optimized multi-mineral model treatment; identifying special lithology, namely accurately marking out a well section of the special lithology, wherein the special lithology comprises coal, limestone and bauxite; And (3) for the well section with special lithology, the primary data backfill and secondary treatment are independently carried out to obtain the porosity, the gas saturation and the multi-mineral component percentage content.
2. The method for modeling deep stratum petrophysical under a complex lithology background of claim 1, wherein the preprocessing comprises outlier rejection, depth correction, curve stitching, environmental correction, and consistency correction, wherein: abnormal value eliminating, namely identifying and eliminating samples which obviously deviate from other observed values in the data set through a statistical or physical method; Depth correction, namely comparing logging curves of a reference layer or a mark layer, combining core analysis data, and adjusting the depths of different logging curves to be aligned with each other by adopting sliding window matching and combining manual verification; curve splicing, namely eliminating jump at joints of curve data acquired in a segmented way by adopting an overlapping area matching method, and splicing by adopting smooth transition of relevant analysis matching characteristic points in an overlapping area; The method comprises the steps of environmental correction, namely adopting a curve with good sand and shale recognition effect as a lithology logging interpretation sensitive parameter curve, intersecting acoustic time difference curves of adjacent well sections of a collapsed well section with the lithology logging interpretation sensitive parameter curve to obtain a fitted acoustic time difference curve, and replacing the well logging curve of the collapsed section with the fitted acoustic time difference curve and a fitted density curve to obtain a corrected well logging curve; And (3) consistency correction, namely selecting a preset number of key wells, determining a standard layer, comparing the well logging data distribution of the treatment wells with the corresponding data distribution of the key wells to determine the correlation and the difference degree of the two, further solving a group of conversion values required by correction, and obtaining unified inter-well data based on the standard wells.
3. The method for modeling deep formation petrophysical in a complex lithology context of claim 1, wherein the determining the multi-mineral model comprises: carrying out depth alignment and standardization on the pretreated logging data and logging lithology description and core experimental data; primarily determining formation lithology by using logging data and core description, dividing mineral major categories by combining logging curve characteristics, and establishing a mineral database by using core experimental data as a multi-mineral model calibration basis; Selecting logging parameters sensitive to minerals, establishing a model corresponding to logging response and mineral content by using a statistical method, and verifying the precision of the multi-mineral model by core data; and constructing a multi-mineral model by combining physical property constraints of the mineral combination, verifying model prediction effects by core samples or oil test data which do not participate in modeling, and adjusting model parameters according to the prediction effects.
4. The method for rock physical modeling of deep stratum in complex lithology context according to claim 1, wherein the rock physical modeling method comprises the following steps: Step 1, inputting the porosity, the gas saturation and the percentage content of the multi-mineral components obtained by optimizing the multi-mineral model, and obtaining the elastic modulus of clay, quartz, calcite and coal through a laboratory; Step 2, calculating the modulus of the rock matrix according to the component parts of the rock minerals through a Gassman equation or a theoretical model of a Biot theory, calculating the bulk modulus of the mixed fluid by using a Wood formula, giving the water saturation, and building the modulus of the dry rock skeleton; step 3, taking the rock matrix modulus, the mixed fluid bulk modulus and the dry rock skeleton modulus obtained in the step 2 as inputs, and establishing a relation among input quantities based on Gassman equation; and 4, taking the density, the clay content, the porosity and the water saturation curve as basic data, and obtaining an improved petrophysical model based on a multi-mineral model through Gassman equation on the basis of obtaining the rock matrix modulus, the mixed fluid bulk modulus and the dry rock skeleton modulus.
5. The method for deep stratum petrophysical modeling in complex lithology background of claim 1, wherein the method for shear wave prediction comprises the following steps: And inputting the petrophysical parameters into a petrophysical model improved based on a multi-mineral model, and predicting longitudinal and transverse wave speeds and longitudinal and transverse wave speed ratios through repeated iterative simulation calculation.
6. The method of claim 1, wherein the raw log data comprises 11 log curves, which are respectively a borehole diameter curve, a drill bit diameter curve, a natural gamma curve, a deep and shallow resistivity curve, a compensated neutron curve, a longitudinal wave time difference curve, a transverse wave time difference curve, a volume density curve, a lithology density curve and a well deviation curve.
7. The method for modeling deep stratum petrophysical under the complex lithology background of claim 1, wherein the standard well selection rules comprise: the standard well has complete and accurate various data, including logging data, geological data, logging data and rock core experimental data; The quality control of the logging data is mainly based on the evaluation of the measured data, wherein the evaluation content comprises whether the well diameter is expanded or not, and whether a better corresponding relation exists between each two curves in a lithology stable section or not.

Description

Deep stratum rock physical modeling method under complex lithology background Technical Field The invention relates to the technical field of geological modeling, in particular to a deep stratum petrophysical modeling method under a complex lithology background. Background Because the Hudous basin has large burial depth, the shallow layer has more development benefit under the condition of similar natural gas yield according to the past exploration thinking mode, so the focus of exploration and development is placed on the coal bed gas in the middle and shallow layer for a long time, and the natural gas in deep coal rocks in the basin is not explored on a large scale. In recent years, the natural gas yield in deep coal rock is far higher than that of middle and shallow layers, and good exploration and development prospects are shown, so that the method has very important significance for petrophysical modeling of deep coal-based stratum. The deep coal-based stratum has complex lithology combination, in the front edge deposition environment of the Taiyuan group delta, the general sea phase and land phase deposition transition area is thicker in mudstone, and the coal-mudstone gas gathering combination is mainly formed, so that the sealing performance is good, and the gas gathering condition is good. In the shallow sea deposition environment of the Taiyuan group, a limestone compact area is formed, a coal rock-limestone gas gathering combination is formed, and the compact limestone is covered to play a good role in sealing. Deep coal rock is generally a dual pore medium with matrix pores and cracks. In the coal rock matrix of the Erdos basin #8, pores, residual plant tissue pores, intercrystalline pores, inter-grain pores, intra-grain pores and other inorganic mineral pores mainly develop, and in addition, cutting lines and micro-cracks also develop in a large amount. In summary, a single petrophysical model cannot realize complex lithology combinations of deep coal formations and transverse wave prediction under pore structures, and if modeling is performed by using only a single model, obvious error values of petrophysical modeling results occur in complex lithology combinations of limestone, coal, bauxite and the like. Disclosure of Invention The invention aims to provide a deep stratum petrophysical modeling method under a complex lithology background, which aims to solve the technical problems of complex lithology combination of a deep coal-based stratum and low transverse wave prediction accuracy under a pore structure in the prior art. In order to solve the technical problems, the invention specifically provides the following technical scheme: a method for rock physical modeling of deep stratum in complex lithology background comprises the following steps: preprocessing logging data, namely selecting a high-quality standard well and a mark layer to preprocess original logging data to obtain logging data for building a mineral model; The mineral model optimization processing comprises the steps of determining a multi-mineral model by combining logging data with core data, analyzing lithology analysis data and all-rock X-ray diffraction data, optimizing the multi-mineral model by a layering section to obtain an optimal multi-mineral model, identifying lithology of special lithology in the optimal multi-mineral model, marking out a well section of the special lithology, and independently performing primary data backfilling and secondary processing to obtain petrophysical parameters for establishing a petrophysical model; According to Gassman equation, according to the rock physical parameters and rock elastic response characteristics under the conditions of simulating different lithology, different physical properties and different fluid properties, constructing a rock physical model based on multi-mineral model improvement, and carrying out transverse wave prediction through repeated iteration; and verifying the accuracy of the predicted transverse wave of the petrophysical model, namely comparing the longitudinal and transverse wave speeds and the longitudinal and transverse wave speed ratio obtained by calculating the petrophysical model based on the improvement of the multi-mineral model with logging measured data, and verifying the accuracy of the predicted transverse wave based on the average error of the longitudinal and transverse wave speeds and the longitudinal and transverse wave speed ratio. As a preferred embodiment of the present invention, the preprocessing includes outlier rejection, depth correction, curve stitching, environmental correction, and consistency correction, wherein: abnormal value eliminating, namely identifying and eliminating samples which obviously deviate from other observed values in the data set through a statistical or physical method; Depth correction, namely comparing logging curves of a reference layer or a mark layer, combining core analysis data, and adjusting the