CN-121765981-B - Complex coal seam group gas occurrence dynamic reconstruction method based on space-time evolution field theory

CN121765981BCN 121765981 BCN121765981 BCN 121765981BCN-121765981-B

Abstract

The invention relates to the technical field of coal mine safety engineering and geological disaster prevention and control, and discloses a complex coal seam group gas occurrence dynamic reconstruction method based on a space-time evolution field theory, which comprises the following steps of S1, constructing a multi-scale geological structure evolution basic model; the method comprises the steps of S2, establishing a space-time evolution equation of a dual-medium gas occurrence parameter, S3, carrying out multi-source data dynamic correction and multi-scale Bayesian inversion reconstruction, wherein in the step S1, a permeability distribution model and geological constraints are provided for the step S2, a forward core framework is provided for the step S3, and the step S3 carries out reverse correction on the evolution equation parameter of the step S2 through real-time data to form closed loop linkage. The invention realizes high-precision and real-time dynamic reconstruction of the gas occurrence state of the coal seam with complex structure, and effectively improves the accurate decision capability of mine gas disaster prevention and control.

Inventors

Su Caiquan
LI QINGSONG
ZHOU DONGPING
HENG XIANWEI
ZHANG PENG
SHEN ZHENHUA

Assignees

贵州省矿山安全科学研究院有限公司
贵州省煤矿设计研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260302

Claims (7)

1. The dynamic reconstruction method for the occurrence of the gas in the complex coal seam group based on the space-time evolution field theory is characterized by comprising the following steps: S1, establishing a three-dimensional geological model of a coal bed based on geological structure data, and generating a coal bed initial permeability distribution field matched with a structure space by quantifying the curvature and interface complexity of the structure; in step S1, generating an initial permeability distribution field includes: Quantifying curvature and interface complexity of a structure, namely calculating a structure curvature sigma reflecting the bending degree of the structure based on seismic interpretation data, and calculating a fractal dimension D reflecting the irregularity degree of the structure based on structural interface data revealed by drilling; Based on a statistical regression method, establishing a permeability control model taking the construction curvature sigma and the fractal dimension D as variables: , wherein, The permeability control function is used for calculating the gridding permeability distribution of the coal bed, and the output value is the coal bed permeability of the corresponding grid node; the permeability basic value of the original coal bed; Substituting the values of the construction curvature sigma and the fractal dimension D of each position in space into the permeability control model K (sigma, D), and calculating to generate the initial permeability distribution field; S2, taking the initial permeability distribution field as input, and establishing a space-time evolution model for describing occurrence and migration of gas in a matrix and a fracture based on a dual medium theory, wherein the space-time evolution model is coupled with a dynamic damage mechanism driven by mining stress to correct the fracture permeability in real time; In the step S2, the space-time evolution model is a dynamic model based on a dual medium theory, and the construction comprises the steps of dividing a coal bed into a coal matrix system and a fracture system, and respectively establishing a control equation for describing the free gas pressure diffusion in the fracture system; s3, collecting real-time monitoring data of mine gas emission, extracting different frequency characteristic components of the mine gas emission through a multi-scale signal decomposition technology, and constructing a Bayesian inversion framework by taking the space-time evolution model as physical forward modeling constraint, wherein the Bayesian inversion framework carries out differentiated space constraint on inversion parameters of an excavation influence region and a non-influence region by integrating space transformation priori information; The method comprises the steps of implementing differentiated space constraint on inversion parameters of a mining influence area and a non-influence area, and specifically realizing differentiated setting according to a space relation between grid nodes and a mining working face by constructing space variation covariance prior distribution, wherein the prior distribution is formed by setting prior variance of a first numerical range for grid nodes in a dynamic mining influence range, setting prior variance of a second numerical range for grid nodes outside the influence range, and applying space smoothness constraint to keep regional continuity, the first numerical range is larger than the second numerical range, and the dynamic mining influence range is dynamically determined according to real-time propelling speed of the working face.
2. The method according to claim 1, wherein in step S1, when constructing the three-dimensional geologic model, a tetrahedral mesh is used for subdivision, and a different mesh size is set for the construction dense region than for the conventional region, wherein the mesh size of the construction dense region is smaller than for the conventional region.
3. The method according to claim 1, wherein the space-time evolution model is coupled with a mining damage mechanism and is used for correcting the permeability of the fracture system in real time according to the mining process, and specifically comprises the following steps: calculating the damage variable and the volume strain of the coal body according to the mining stress; Dynamically modifying the porosity of the fracture system based on the damage variable and the volume strain; and updating the permeability of the fracture system in real time by utilizing the corrected porosity according to the fracture cube law.
4. A method according to claim 1 or 3, characterized in that the space-time evolution model implements numerical simulation by solving the following control equations simultaneously: a fracture system gas pressure diffusion equation based on gas quasi-pressure linearization; A gas exchange rate equation between the matrix system and the fracture system based on a linear driving force hypothesis; A dynamic equation of the gas content of adsorption in the matrix system; and coupling a fracture system permeability evolution equation of the mining damage.
5. The method according to claim 1, wherein in step S3, different frequency characteristic components are extracted by a multi-scale signal decomposition technique, in particular: And carrying out multi-scale decomposition on the gas emission amount time sequence data based on a matching tracking principle, separating the gas emission amount time sequence data into different frequency components representing large-scale background change, medium-scale transition change and small-scale abnormal fluctuation, and carrying out weighted fusion on each decomposed component based on the weight determined by historical data verification so as to form a fused data vector for inversion.
6. The method according to claim 1, characterized in that the bayesian inversion framework constructed in step S3, whose posterior probability distribution maximization problem to be solved is equivalent to the following optimization problem: Wherein, the For the optimal estimation of the gas content parameters to be inverted, For a forward operator constructed based on the space-time evolution model, In order to observe the data vector, Fitting a residual term to the weighted data; Based on a spatially varying covariance matrix Is used to determine the prior constraint terms of (c), The method is characterized in that the method is a sparse regular term for promoting local anomaly characterization, wherein m is a parameter vector to be inverted; parameter vector to be inverted Is a transposed matrix of (a).
7. The method of claim 6, wherein the optimization problem is numerically solved using an alternate direction multiplier method to obtain an optimal estimate of the gas content parameter.

Description

Complex coal seam group gas occurrence dynamic reconstruction method based on space-time evolution field theory Technical Field The invention relates to the technical field of coal mine safety engineering and geological disaster prevention and control, in particular to a dynamic reconstruction method for gas occurrence of a complex coal seam group based on a space-time evolution field theory. Background Accurate perception of coalbed methane occurrence state is a core precondition for preventing and controlling coal mine methane disasters. The complex coal seam group has the characteristics of geological structure development, strong coal seam permeability space-time heterogeneity, obvious coupling effect of a mining stress field and a gas seepage field and the like, so that the occurrence of gas presents a strong nonlinear and multi-scale dynamic change rule. The existing gas occurrence reconstruction method is mostly dependent on static geologic modeling and single-scale monitoring data, and has the defects that firstly, a data processing mode is single, multi-scale characteristics of time sequence data such as gas emission quantity are not considered, large-scale background trend and small-scale abnormal fluctuation are difficult to effectively separate and are easy to be interfered by observation noise, secondly, an inversion frame is lack of pertinence constraint, the traditional inversion method mostly adopts single smooth constraint or sparse constraint, spatial continuity and local abnormal characteristics of gas occurrence cannot be considered, reconstruction accuracy is limited, thirdly, suitability for complex structure and dynamic influence of mining is insufficient, real-time dynamic update of gas occurrence state is difficult to achieve, deviation exists between a reconstruction result and on-site reality, and accurate control requirements cannot be met. Although some methods attempt to introduce a Bayesian inversion theory to improve stability, the stability is improved without combining a multi-scale data decomposition technology and designing a differential constraint mechanism aiming at the spatial distribution characteristics of gas occurrence, and the problems of accuracy and dynamics of gas occurrence reconstruction in complex coal seam groups are still difficult to solve. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a dynamic reconstruction method for the occurrence of gas in a complex coal seam group based on a space-time evolution field theory, which can realize high-precision and real-time dynamic monitoring and reconstruction of the occurrence state of gas, provides technical support for gas extraction optimization, mine ventilation adjustment and safe production decision, and is suitable for coal mine wells with complex geological structures (faults, folds, collapse columns and the like). The method comprises the following steps of S1, establishing a coal seam three-dimensional geological model based on geological structure data, generating a coal seam initial permeability distribution field matched with a structure space through quantifying curvature and interface complexity of the structure, S2, taking the initial permeability distribution field as input, establishing a space-time evolution model describing occurrence and migration of gas in matrixes and cracks based on a dual-medium theory, wherein the space-time evolution model is coupled with a dynamic damage mechanism driven by mining stress to correct crack permeability in real time, S3, collecting real-time monitoring data of mine gas emission, extracting different frequency characteristic components of the real-time monitoring data through a multi-scale signal decomposition technology, and constructing a Bayesian inversion frame by taking the space-time evolution model as physical forward modeling constraint, wherein the Bayesian inversion frame carries out differentiated space constraint on inversion parameters of a mining influence area and a non-influence area through fusing space-variant priori information, and dynamically reconstructing the space-time evolution parameter field through solving the frame and updating the evolution model to realize closed loop; The method comprises the steps of implementing differentiated space constraint on inversion parameters of a mining influence area and a non-influence area, and specifically realizing differentiated setting according to a space relation between grid nodes and a mining working face by constructing space variation covariance prior distribution, wherein the prior distribution is formed by setting prior variance of a first numerical range for grid nodes in a dynamic mining influence range, setting prior variance of a second numerical range for grid nodes outside the influence range, and applying space smoothness constraint to keep regional continuity, the first numerical range is larger than the second numerical range, and the