CN-121980654-A - Unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling

CN121980654ACN 121980654 ACN121980654 ACN 121980654ACN-121980654-A

Abstract

The invention relates to the technical field of oil and gas exploitation, and particularly discloses an unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling, which is used for calculating the half length of a fracture by combining reservoir parameters at a perforation position obtained by well logging and other methods with an analysis type fracture expansion simulation method, carrying out complex fracture network simulation by using the method as constraint and combining a fracture growth algorithm with fractal probability random distribution characteristics to obtain the shape distribution of the complex fracture network of the reservoir, and finally carrying out fracturing construction monitoring pump pressure fitting based on an intelligent algorithm, inverting and correcting the volume and the shape of the complex fracture network, so as to realize the real-time expansion simulation of the complex fracture network in the fracturing process. The method can more finely describe the complex branch crack morphology, has higher calculation speed, can realize the real-time simulation of the complex fracture network morphology in the fracturing process, realizes the efficient and accurate simulation of the crack morphology under the condition of unclear reservoir stratum ground stress knowledge, and provides an effective technical means for unconventional reservoir stratum fracturing construction design.

Inventors

SHENG GUANGLONG
GUO SHUHANG
ZHAO HUI
CHEN KAI
HUANG PENGFEI
LIAN XINYI

Assignees

长江大学

Dates

Publication Date: 20260505
Application Date: 20260120

Claims (9)

1. The unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling is characterized by comprising the following steps of: s1, data preprocessing, namely collecting logging interpretation data, fracturing construction data and fracturing monitoring data of a perforation position, establishing a reservoir node discretization grid, and determining perforation position coordinates; S2, calculating the crack length of the perforation position, namely calculating the dynamic evolution of the crack in time steps through a variable superposition principle by adopting an improved KGD analysis model to calculate the fluid loss based on the rock mechanics, the ground stress parameters and the fracturing construction data of the perforation position interpreted by logging data, and calculating the dynamic crack length of the perforation position; S3, reservoir complex fracture network expansion simulation, namely dispersing a reservoir into grid nodes, determining the expansion direction of the crack tips of the fracture network branches by adopting a maximum energy release rate theory MERRT based on ground stress, rock mechanics and fracturing construction parameters, constructing a fractal probability model of the expansion direction, determining an expansion path by accumulating probability distribution and random sampling, and circularly iterating until the total length of the crack approaches to the dynamic crack length of a perforation position to generate a complex fracture network form; S4, performing inversion correction on the stitch network shape, namely calculating a theoretical pumping pressure based on a volume balance principle, adopting an SPSA algorithm, taking an actual pumping pressure curve as a fitting target, taking the total volume of the stitch network as a main control variable, and performing inversion correction on the stitch network volume and shape by minimizing Root Mean Square Error (RMSE) of the theoretical pumping pressure and the actual pumping pressure, so as to realize real-time expansion simulation of the stitch network.
2. The unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling of claim 1, wherein in step S1, the logging interpretation data comprises rock mechanical parameters, ground stress parameters, elastic modulus, poisson' S ratio, the fracturing construction data comprises injection displacement, construction time, fracturing fluid properties, and the fracturing monitoring data comprises wellhead pressure, bottom hole pressure.
3. The method for modeling unconventional fracturing complex fracture network expansion simulation of uncertainty-based according to claim 1, wherein in step S2, the improved KGD analytical model is: Deducing a KGD model accurate analysis solution under the action of fluid-solid coupling, wherein the KGD model accurate analysis solution is as follows: ; wherein L is the crack length, Q 0 is the injection displacement, h is the crack height, E is the elastic modulus, v is the Poisson's ratio, and t is the construction time; is the dynamic viscosity coefficient of the fracturing fluid.
4. The method for simulating unconventional fracturing complex fracture network expansion based on uncertainty modeling according to claim 3, wherein in step S2, a Carter fluid loss model is introduced to calculate fluid loss, so as to obtain a dynamic fracture length of a perforation position considering fluid loss, specifically: Decomposing the whole fracturing process into a series of small time periods, calculating by adopting the equivalent displacement generated by subtracting the fluid loss from the actual displacement as a formula parameter, and superposing the results of all the time periods to obtain the fracture length considering the fluid loss characteristics, wherein the fracture length is as follows: ; In the formula, The fracturing displacement at the ith moment; And (5) the fluid loss displacement of the fracturing fluid at the i time.
5. The method for simulating the expansion of an unconventional fracturing complex fracture network based on uncertainty modeling according to claim 1, wherein in the step S3, the expansion direction of the fracture tip of the fracture network branch is determined by using a maximum energy release rate theory MERRT, specifically: At each node, the energy release rates in each direction are compared Whether the critical energy release rate Gc of the material is satisfied Gc, if satisfied, there is a probability of crack propagation in that direction, and if not satisfied, there is no probability of crack propagation in that direction.
6. The unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling of claim 5, wherein the energy release rate The calculation process of the critical energy release rate Gc of the material and the fractal probability model is as follows: S31, setting a main crack tip to deviate from the main crack direction Is the micro-branch of the energy release rate Expressed as a function of the type I and type II stress intensity factors in that direction: ; In the formula, Is the direction of the branch Upper energy release rate; E is elastic modulus, v is Poisson's ratio; 、 respectively the branch directions Type I and type II stress intensity factors above; the tip of the main crack is in the branching direction The above type I and type II stress intensity factors are expressed as: ; ; In the middle of , Type I and type II stress intensity factors for the fracture tip; S32, the relation between the critical energy release rate Gc of the material and the type I fracture toughness K Ic is as follows: ; S33, defining a normalized fractal probability model as follows: ; wherein P (theta i ) is the expansion probability in the direction of theta i , and eta is the fractal index.
7. The method for modeling irregular fracturing complex fracture network expansion simulation of uncertainty based on claim 6, wherein in step S3, the expansion path is determined by cumulative probability distribution and random sampling, specifically: s34, let the scalable direction set be { θ 1 ,θ 2 ,...,θ n }, its corresponding expansion probability be P (θ 1 ),P(θ 2 ),. P (θn), and define the cumulative probability as: ; wherein C 0 = 0; S35, generating a uniform random number r of a [0,1 ] interval, wherein the finally selected expansion direction theta k meets the following conditions: 。
8. The method for modeling unconventional fracturing complex fracture network expansion simulation according to claim 1, wherein the calculating theoretical pump pressure based on the volume balance principle in step S4 is specifically as follows: in the fracturing injection process, the volume balance relationship between the injected liquid for filling the fracture and the fluid loss to the reservoir is expressed as follows: ; Wherein Q (t) is injection displacement, V f (t) is the volume of liquid in a fracture at the moment t, and Q leakoff (t) is the fluid loss rate; Q leakoff (t) is described as using the Carter model: ; Wherein, C L,i is the fluid loss coefficient of the ith crack, t 0,i is the opening time of the ith crack, N f is the number of communication nodes, namely the number of cracks, and h i (t)、l i (t) is the height and the length of the ith branch crack respectively; the effective flow rates actually used for crack propagation are: ; Crack opening pressure analysis solution based on classical KGD isopiestic crack model is: ; Wherein P frac (t) represents the pressure in the fracture or the fracture opening pressure, alpha is a dimensionless model constant related to the properties and geometric characteristics of the fluid, n is the flow index of the fluid, pi is the circumference ratio, K represents the related parameter of the elastic modulus of the rock, E is the elastic modulus, q (t) is the injection flow rate, and H represents the fracture height; On the basis, perforation friction, vertical/horizontal shaft friction and net liquid column pressure are comprehensively considered to obtain wellhead pressure: ; Wherein P inj (t) is the surface injection pressure, namely the pumping pressure, deltaP fric (t) is the friction pressure drop, and DeltaP hyd (t) is the hydrostatic column pressure drop.
9. The method for modeling irregular fracturing complex fracture network expansion simulation according to claim 1, wherein in step S4, the total volume of the fracture network is used as a main control variable, and the correction of the fracture network volume and morphology is performed by minimizing the root mean square error RMSE of the theoretical pump pressure and the actual pump pressure, specifically: Taking the total volume of a crack network, namely the number of nodes, as a main control variable, adopting an SPSA intelligent optimization algorithm, and realizing the rough fitting of a pumping pressure curve by continuously adjusting the number of nodes and minimizing the pumping pressure RMSE; based on the determination of the number of nodes, generating a plurality of groups of different crack forms by utilizing the probability fractal crack expansion algorithm in the step S2, calculating a corresponding theoretical pumping pressure curve for each group of forms, selecting the form with the minimum RMSE as a final simulation result, realizing the fine fitting of the pumping pressure, and finally obtaining the complex seam net form conforming to the pumping pressure curve.

Description

Unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling Technical Field The invention relates to the technical field of oil and gas exploitation, in particular to an unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling. Background The large-scale hydraulic fracturing technology is a core means for realizing efficient reservoir reconstruction by using unconventional oil and gas resources (shale gas, compact oil, coal bed gas and the like). However, unconventional reservoirs generally have the characteristics of low permeability, strong heterogeneity, complex natural fracture network and the like, so that a hydraulic fracture expansion path presents high uncertainty, and a fracture expansion calculation method with high development precision, weak geological parameter dependence and capability of realizing complex fracture network shape characterization is needed. The current mainstream fracture expansion simulation calculation method can be divided into three types of analytical model, numerical simulation and data driving, and the applicability and limitation of the method are obvious due to the characteristic difference of the reservoir stratum (1) the analytical method is suitable for the rapid evaluation of a homogeneous reservoir stratum, but the core defect is that the prediction error is obvious in unconventional reservoir stratum such as shale, coal seam and the like due to neglecting the heterogeneous reservoir stratum, the natural fracture interference and the multi-fracture competition expansion effect. (2) Numerical simulation methods mainly include an extended finite element method (XFEM), a Discrete Element Method (DEM), and a Boundary Element Method (BEM). The expansion of finite element calculation efficiency is limited by the complexity of multi-field coupling, the discrete element method has the problem of high calculation cost in large-scale simulation, and the application of the boundary element method is limited by the high dependence on basic solutions. Meanwhile, the numerical simulation method relies on a high-precision geological model and strict mechanical parameters (such as a ground stress field and fracture toughness), when the uncertainty of reservoir parameters is strong, an input error can be transmitted through the model to obviously amplify an output deviation, and in the later stage, the geological modeling is required to be carried out again when the monitoring data is adjusted, so that time and effort are consumed, and a good prediction effect can not be necessarily obtained. (3) Data-driven methods that reduce reliance on priori geologic parameters, but because of insufficient physical mechanism embedment, it is difficult to explain the mechanical cause of crack propagation and are subject to noise interference from the monitored data. Therefore, development of a crack propagation calculation method with mechanical mechanism and data adaptability is needed to be developed, wherein dependence on reservoir parameter completeness is weakened, geological uncertainty is represented through probability modeling or fractal theory, real-time data (such as pump pressure, displacement and microseism event) of fracturing construction are needed to be fused deeply, a dynamic feedback mechanism is constructed to restrict a crack propagation path, and technical support with efficiency and accuracy is provided for unconventional reservoir fracturing design. Disclosure of Invention The invention aims to solve the problems that the fracture morphology simulation result is poor in matching with the actual reservoir rock, ground stress and other parameter distribution due to the fact that the prior art is imperfect in consideration of geological parameters. In order to achieve the above object, the present invention provides an unconventional fracturing complex fracture network expansion simulation method based on uncertainty modeling, comprising the following steps: s1, data preprocessing, namely collecting logging interpretation data, fracturing construction data and fracturing monitoring data of a perforation position, establishing a reservoir node discretization grid, and determining perforation position coordinates; S2, calculating the crack length of the perforation position, namely calculating the dynamic evolution of the crack in time steps through a variable superposition principle by adopting an improved KGD analysis model to calculate the fluid loss based on the rock mechanics, the ground stress parameters and the fracturing construction data of the perforation position interpreted by logging data, and calculating the dynamic crack length of the perforation position; S3, reservoir complex fracture network expansion simulation, namely dispersing a reservoir into grid nodes, determining the expansion direction of the crack tips of the fracture network branches by adopting a