CN-122021481-A - Reservoir post-lamination yield prediction method based on microcosmic gas-water competitive adsorption

CN122021481ACN 122021481 ACN122021481 ACN 122021481ACN-122021481-A

Abstract

The application discloses a reservoir pressure post-production prediction method based on microcosmic gas-water competitive adsorption, which relates to the technical field of coalbed methane development, and comprises the steps of constructing a three-dimensional model of a coal-rock pore structure and a molecular model of three molecules, and establishing a molecular simulation system; the method comprises the steps of carrying out molecular dynamics simulation on a molecular simulation system based on a giant regular Monte Carlo principle, solving a molecular motion equation, determining micro characteristic parameters of methane gas, determining a simulation predicted value of methane gas yield based on a coal-rock pore structure three-dimensional model and the micro characteristic parameters, optimizing the molecular simulation system based on the simulation predicted value, simultaneously constructing a multi-factor response curved surface model, constructing a yield prediction coupling model based on the optimized molecular simulation system and the multi-factor response curved surface model, and carrying out yield prediction of the methane gas. The method realizes the efficient and accurate prediction of the methane gas output rule in the high-temperature high-pressure in-situ environment after hydraulic fracturing.

Inventors

TANG JIZHOU
ZHAO ZHENGGUANG
HUANG LEI
LI KUNJIE
TAO JIAPING
ZHANG FENGYUAN
MENG SIWEI
HE CHUEN
LI YUWEI
CHEN MIN
LI NING
ZHAO LUANXIAO
ZHANG FENGSHOU

Assignees

同济大学

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (10)

1. The method for predicting the post-reservoir-pressure production based on microscopic gas-water competitive adsorption is characterized by comprising the following steps of: acquiring focused ion beam-scanning electron microscope pore structure images and CT scanning data of coal and rock samples from reservoirs with preset depths of different sample blocks after hydraulic fracturing, and acquiring geological parameters of the sample blocks where the coal and rock samples are located; adopting a voxel reconstruction method to construct a coal-rock pore structure three-dimensional model based on the focused ion beam-scanning electron microscope pore structure image, the CT scanning data and the geological parameters; Respectively constructing molecular models of three molecules, defining a molecular simulation space based on a coal-rock pore structure three-dimensional model, and quantifying interaction relation among the molecules to establish a molecular simulation system, wherein the three molecules comprise methane molecules, water molecules and coal-rock surface molecules; based on the giant regular Monte Carlo principle, carrying out molecular dynamics simulation on a molecular simulation system and solving a molecular motion equation so as to determine micro-characteristic parameters of methane gas, wherein the micro-characteristic parameters comprise adsorption quantity and desorption probability; Determining a simulation predicted value of methane gas yield based on the coal-rock pore structure three-dimensional model and the micro-characteristic parameters of methane gas, and optimizing force field parameters and boundary conditions of the molecular simulation system based on the simulation predicted value of methane gas yield to obtain an optimized molecular simulation system; Constructing a multi-factor response curved surface model, and constructing a yield prediction coupling model based on the optimized molecular simulation system and the multi-factor response curved surface model, wherein the multi-factor response curved surface model is a model related to environmental parameters; and inputting the environmental parameters of the target reservoir of the target block into a yield prediction coupling model, and outputting a yield prediction result of methane gas.
2. The method for predicting post-reservoir production based on microscopic gas-water competitive adsorption of claim 1, wherein obtaining focused ion beam-scanning electron microscope pore structure images and CT scan data of coal and rock samples from reservoirs of predetermined depths of different sample blocks after hydraulic fracturing and obtaining geological parameters of the sample blocks where each coal and rock sample is located specifically comprises: collecting coal and rock samples of reservoirs with preset depths from different sample blocks after hydraulic fracturing, and scanning each coal and rock sample by adopting a focused ion beam-scanning electron microscope to obtain pore structure images of the sections of each coal and rock sample; acquiring three-dimensional pore structure data in a core of each coal rock sample by adopting a CT scanning method, and taking the three-dimensional pore structure data in the core as CT scanning data; And collecting geological parameters of sample blocks where the coal rock samples are located, wherein the geological parameters at least comprise stratum temperature, stratum pressure, vitrinite content in the coal rock components and ash content in the coal rock components.
3. The method for predicting post reservoir pressure production based on microscopic gas-water competitive adsorption of claim 1, wherein a voxel reconstruction method is adopted to construct a coal-rock pore structure three-dimensional model based on the focused ion beam-scanning electron microscope pore structure image, the CT scan data and the geological parameters, and the method specifically comprises: carrying out standardization processing on the focused ion beam-scanning electron microscope pore structure image, the CT scanning data and the geological parameters to obtain a standardized focused ion beam-scanning electron microscope pore structure image, standardized CT scanning data and standardized geological parameters; and adopting a voxel reconstruction method, and constructing and obtaining a coal rock pore structure three-dimensional model based on the standardized focused ion beam-scanning electron microscope pore structure image, the standardized CT scanning data and the standardized geological parameters.
4. The method for predicting postreservoir pressure production based on microscopic gas-water competitive adsorption of claim 1, wherein the intermolecular interactions include intermolecular van der waals forces and intermolecular electrostatic interactions; Respectively constructing molecular models of three molecules, defining a molecular simulation space based on a coal-rock pore structure three-dimensional model, quantifying an interaction relation among the molecules, and constructing a molecular simulation system, wherein the molecular simulation system specifically comprises: Respectively constructing a molecular model of methane molecules, a molecular model of water molecules and a molecular model of coal rock surface molecules, defining a molecular simulation space by using a coal rock pore structure three-dimensional model, and initializing the positions of the methane molecules and the positions of the water molecules in the molecular simulation space according to the fluid density of a reservoir of a sample block; After the position initialization of the methane molecules and the position initialization of the water molecules are completed, the Van der Waals force between the methane molecules and the methane molecules, between the methane molecules and the water molecules, between the water molecules and the water molecules is calculated by adopting a Lennard-Jones potential energy model, and the electrostatic interaction between the water molecules and the methane molecules and the electrostatic interaction between the polar groups on the surface of the coal rock and the methane molecules are calculated by adopting a coulomb force interaction energy formula, so that a molecular simulation system comprising the molecular simulation space, all molecular models and the intermolecular interaction relation is established.
5. The method for predicting post-reservoir pressure production based on microscopic gas-water competitive adsorption of claim 4, wherein the method comprises the steps of respectively constructing a molecular model of methane molecules, a molecular model of water molecules and a molecular model of coal-rock surface molecules, defining a molecular simulation space by using a three-dimensional model of coal-rock pore structure, initializing the positions of the methane molecules and the positions of the water molecules in the molecular simulation space according to the fluid density of the reservoir of the sample block, and specifically comprises the following steps: defining molecular model parameters of methane molecules by adopting a transferable molecular force field, and constructing to obtain a molecular model of the methane molecules; defining molecular model parameters of water molecules by adopting four-point transferable intermolecular potential, and constructing a molecular model of the water molecules; Based on the organic matter components of the coal rock and the X-ray photoelectron spectroscopy data, constructing a molecular model of molecules on the surface of the coal rock by adopting a COMPASS force field; mapping the coal rock pore structure three-dimensional model into a molecular simulation space for molecular dynamics simulation, and setting a periodic boundary condition; and generating initial spatial coordinates of methane molecules and initial spatial coordinates of water molecules in the molecular simulation space by adopting a uniformly distributed random function based on the methane gas density and the formation water density of the reservoir of the sample block, thereby completing the initialization of the positions of the methane molecules and the initialization of the positions of the water molecules.
6. The method for predicting the post-reservoir pressure production based on microscopic gas-water competitive adsorption according to claim 1, wherein the molecular dynamics simulation is performed on a molecular modeling system based on the megacanonical monte carlo principle and a molecular motion equation is solved, so as to determine microscopic characteristic parameters of methane gas, and the method specifically comprises the following steps: Based on the megarule Monte Carlo principle, carrying out molecular dynamics simulation on a molecular simulation system; Solving a molecular motion equation by adopting Verlet algorithm, and updating the position and the speed of each molecule in each unit time step; and counting the adsorption quantity of methane molecules in pores in a molecular simulation space in real time, recording the desorption probability of the methane molecules, and determining the adsorption quantity and the desorption probability as micro-characteristic parameters of methane gas.
7. The method for predicting post-reservoir pressure production based on microscopic gas-water competitive adsorption of claim 6, wherein determining the simulated predicted value of methane gas production based on the three-dimensional model of the coal-rock pore structure and the microscopic characteristic parameters of methane gas specifically comprises: determining the total volume of pores of the coal rock based on the three-dimensional model of the pore structure of the coal rock; and calculating to obtain the effective desorption amount of methane based on the adsorption amount, the desorption probability and the total volume of pores of the coal and rock in the micro-characteristic parameters of the methane, and taking the effective desorption amount of methane as a simulation predicted value of the methane yield.
8. The method for predicting post-reservoir pressure production based on microscopic gas-water competitive adsorption according to claim 1, wherein the method for optimizing the force field parameters and boundary conditions of the molecular simulation system based on the simulated predicted value of methane gas production specifically comprises the steps of: calculating a relative error between the simulated predicted value of methane gas yield and the actual value of methane gas yield of the corresponding sample block; when the relative error exceeds a preset error threshold, optimizing and adjusting force field parameters and boundary conditions of the molecular simulation system; And updating the molecular simulation system based on the optimized and adjusted force field parameters and boundary conditions to obtain an optimized molecular simulation system.
9. The method of claim 8, wherein when the relative error exceeds a preset error threshold, an optimization adjustment strategy is used to optimally adjust force field parameters and boundary conditions of the molecular modeling system, wherein the optimization adjustment strategy comprises one or more of the following optimization adjustment sub-strategies: Adjusting action energy parameters and/or molecular collision diameters in a Lennard-Jones potential energy model; Correcting charge distribution of polar groups in the molecular model of the coal rock surface; And optimizing the temperature boundary condition and the pressure boundary condition set by the molecular dynamics simulation.
10. The method of reservoir pressure post production prediction based on microscopic gas-water competitive adsorption of claim 1, wherein the environmental parameters include temperature, pressure, and porosity; constructing a multi-factor response surface model, and constructing a yield prediction coupling model based on the optimized molecular simulation system and the multi-factor response surface model, wherein the method specifically comprises the following steps of: taking temperature, pressure and porosity as independent variables, and taking methane gas yield as dependent variables, constructing an initial expression of a multi-factor response curved surface model; Acquiring microscopic characteristic parameters of methane gas under different environmental parameters by using an optimized molecular simulation system; based on the mapping relation between the micro characteristic parameters and the environmental parameters of methane gas, determining each coefficient in the initial expression of the multi-factor response curved surface model through multiple regression fitting to obtain a target expression of the multi-factor response curved surface model; and performing associated coupling on the target expression of the multi-factor response curved surface model and the optimized molecular simulation system to form a yield prediction coupling model.

Description

Reservoir post-lamination yield prediction method based on microcosmic gas-water competitive adsorption Technical Field The application relates to the technical field of coalbed methane exploitation, in particular to a method for predicting the post-reservoir-pressure production based on microcosmic gas-water competitive adsorption, and more particularly relates to a method for predicting the methane gas production of a hydraulically-fractured coalbed methane reservoir by combining microcosmic characterization, molecular dynamics simulation and intelligent calculation. Background Along with the transformation of global energy structures to clean, methane gas is used as an efficient low-carbon unconventional methane gas resource, and the development and the utilization of the methane gas have important significance for guaranteeing energy safety and reducing greenhouse gas emission. However, deep methane gas reservoirs have the characteristics of high temperature and high pressure (the temperature is 70-120 ℃ and the pressure is 15-30 MPa), complex pore structures and uneven gas-water distribution, and in order to mine methane gas in coal bed gas reservoirs (deeper formations of 2000m and above), hydraulic fracturing treatment is required along with the bottom layer to promote methane gas desorption and output. Methane gas output law is influenced by intermolecular action, pore restriction and coupling of environmental parameters, and the traditional prediction method faces three major core bottlenecks: 1) The cognition of the molecular mechanism is insufficient, namely, the competitive adsorption, diffusion and desorption processes of methane gas and water molecules in coal-rock pores lack of an accurate quantitative model, and the association mechanism of molecular behavior-macroscopic output cannot be revealed. 2) The experimental characterization is limited in that the conventional means such as a Focused Ion Beam-scanning electron microscope (FIB-SEM) and low-temperature nitrogen adsorption can only obtain a static pore structure, and the dynamic movement track of molecules under the in-situ conditions of high temperature and high pressure after hydraulic fracturing can not be observed. 3) The prediction accuracy is low, the related model is mostly based on ideal condition assumptions (such as normal temperature and normal pressure and single pore type), microcosmic factors such as force field action, molecular collision and the like in a real reservoir are not considered, so that the prediction error of methane gas yield exceeds 30%, and the optimization of a field development scheme is difficult to guide. In recent years, molecular dynamics simulation technology provides a new path for breaking deep coal bed methane development problems, but prediction accuracy is still low. Therefore, a method for predicting the post-reservoir-pressure production based on microcosmic gas-water competitive adsorption is needed, so that efficient and accurate prediction of methane gas production rules in a high-temperature and high-pressure in-situ environment after hydraulic fracturing is realized, and technical support is provided for deep coal bed gas development. Disclosure of Invention The application aims to provide a reservoir pressure post-production prediction method based on microcosmic gas-water competitive adsorption, which realizes efficient and accurate prediction of methane gas production rules in a high-temperature high-pressure in-situ environment after hydraulic fracturing. In order to achieve the above object, the present application provides the following solutions: the application provides a method for predicting post-reservoir-pressure production based on microscopic gas-water competitive adsorption, which comprises the following steps: acquiring focused ion beam-scanning electron microscope pore structure images and CT scanning data of coal and rock samples from reservoirs with preset depths of different sample blocks after hydraulic fracturing, and acquiring geological parameters of the sample blocks where the coal and rock samples are located; adopting a voxel reconstruction method to construct a coal-rock pore structure three-dimensional model based on the focused ion beam-scanning electron microscope pore structure image, the CT scanning data and the geological parameters; Respectively constructing molecular models of three molecules, defining a molecular simulation space based on a coal-rock pore structure three-dimensional model, and quantifying interaction relation among the molecules to establish a molecular simulation system, wherein the three molecules comprise methane molecules, water molecules and coal-rock surface molecules; based on the giant regular Monte Carlo principle, carrying out molecular dynamics simulation on a molecular simulation system and solving a molecular motion equation so as to determine micro-characteristic parameters of methane gas, wherein the micro-characterist