Method for improving hydraulic fracturing numerical simulation fitting precision based on experimental analysis

Abstract

The invention relates to a method for improving hydraulic fracturing numerical simulation fitting precision based on experimental analysis, and belongs to the technical field of oil and gas exploitation. The method comprises the following specific steps: step 1, establishing an ultra-low permeability reservoir equal probability model; 2, determining an equal probability geologic model; and step 3, performing numerical simulation on the conventional hydraulic fracturing fracture of the ultra-low permeability reservoir. Compared with the conventional hydraulic fracturing numerical simulation of the ultra-low permeability reservoir, the method can improve the single well historical fitting precision of the conventional hydraulic fracturing of the ultra-low permeability reservoir, thereby outputting accurate pore pressure and fluid saturation field for fracture network fracturing simulation. The method is more scientific, the precision of historical fitting is higher, the prediction of the pore pressure and the fluid saturation field is more accurate and reliable, and the method has important significance for guiding the production and development of an oil field.

Description

Technical Field

[0001] The invention relates to a method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis, and belongs to the technical field of oil and gas exploitation. Background Art

[0002] The conventional exploitation method of ultra-low permeability reservoirs is usually based on the initial application of conventional hydraulic fracturing, and secondary large-scale fracture network fracturing is often used in the later stage to achieve the purpose of increasing production. Conventional hydraulic fracturing and production processes will change the original stress field. Since fracture network fracturing is a new technology, the impact of stress field changes on secondary large-scale fracture network fracturing is unclear, the crack morphology after fracture network fracturing is unclear, and the optimization ideas for post-fracture production water injection adjustment are unclear. Therefore, it is necessary to carry out numerical simulation of conventional hydraulic pressure before fracture network fracturing, carry out four-dimensional fluid-solid coupling through simulated pore pressure and fluid saturation field, and clarify the changing laws of stress fields in different time periods. The accuracy of numerical simulation before fracture network fracturing is particularly important.

[0003] At present, for such a situation, the method usually adopted is to first establish a certain phase-controlled geological property model of conventional single-porous media, and carry out numerical simulation based on this. In the fitting process, the permeability of the fracture is artificially increased by local grid encryption, so as to fit the single fracture of equivalent conventional hydraulic fracturing without considering the uncertainty of the static model and the conductivity of conventional fractures. Therefore, the existing technology is not applicable to the numerical simulation of conventional hydraulic fractures in ultra-low permeability reservoirs, and the overall and single well history fitting accuracy is low. Summary of the invention

[0004] In order to solve the problem of low overall fitting accuracy of numerical simulation of conventional hydraulic fracturing in the early stage of ultra-low permeability reservoirs, the present invention discloses a method for improving the fitting accuracy of numerical simulation of hydraulic fracturing based on experimental analysis, and clarifies the method for improving the fitting accuracy of numerical simulation of conventional hydraulic fracturing in ultra-low permeability reservoirs.

[0005] The technical solution adopted by the present invention is a method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis, and the specific steps are as follows:

[0006] Step 1: Establish a probability model of ultra-low permeability reservoir;

[0007] Step 2, determine the equiprobability geological model;

[0008] Step 3: Numerical simulation of conventional hydraulic fractures in ultra-low permeability reservoirs.

[0009] Furthermore, the step 1 specifically includes:

[0010] ① Through the verification of the spatial fault model of seismic interpretation and the spatial breakpoint data of a single well, the occurrence of the spatial faults in the block is described. According to the geological stratification data of a single well, the seismic interpretation layer data is used as a constraint condition. Through the convergence method interpolation calculation, a stratified layer model is established to describe the structural fluctuation characteristics of the stratification. The fault and layer models are gridded to establish a gridded structural model containing structural and fault information;

[0011] ② Using well data, establish an intersection diagram of the relationship between acoustic time difference and core porosity, regress the relationship curve between acoustic time difference and porosity, establish a relationship chart between porosity and permeability through core data, analyze the correlation between the two, regress the relationship between permeability and porosity, and use the multivariate statistical regression method of core analysis saturation and logging data to establish an equation for calculating the original oil saturation of the logging data layered section. Using the above regression relationship, the porosity and permeability of the sand body are corrected layer by layer and well by well through the electrical logging curve of a single well;

[0012] ③ Each small layer of the structural model is subdivided into multiple sections with a length of 1m.

[0013] Furthermore, the specific requirement for comprehensively considering the crack direction is that the transverse grid is parallel to the crack direction, and the longitudinal grid is perpendicular to the crack direction.

[0014] Furthermore, the step 2 specifically includes adding the gridded sedimentary facies data body to the work area, establishing a sedimentary facies model by directly assigning values ​​to the grid, applying sedimentary facies model constraints, establishing phase-controlled equal probability multi-realization attribute modeling through a sequential Gaussian algorithm, calculating geological reserves using different attribute models, selecting the attribute model that is closest to the calibrated geological reserves as the final attribute model, and outputting the final geological model.

[0015] Furthermore, in the process of outputting the geological model, in order to meet the requirements of numerical simulation on the number of geological model grids, the subdivided sublayers are re-merged for non-target sublayers.

[0016] Further, different property models refer to porosity, effective thickness and permeability models.

[0017] Furthermore, the selection of the attribute model closest to the calibrated geological reserves as the final attribute model means calculating different geological reserves using different geological models, and the geological model whose calculated geological reserves are closest to the geological reserves calculated by the static volumetric method is the final model.

[0018] Furthermore, the step three specifically includes:

[0019] Nano-CT scanning experiments and nuclear magnetic resonance experiments are used to systematically study the microscopic pore structure characteristics and movable fluid distribution characteristics of ultra-low permeability reservoirs, providing basic reservoir parameters for refined numerical simulations;

[0020] The output structural model and attribute model are used to establish the numerical model, including the attribute volume data such as skeleton grid, active grid distribution, grid depth and height, grid step size, grid porosity, permeability and oil saturation;

[0021] Reserve fitting is carried out based on the numerical model, and the perforation and production information of the single well are added to the numerical model established in the early stage to carry out historical matching.

[0022] Furthermore, in the historical matching process, the conventional cracks in the first fracturing are treated equivalently, and the productivity index (WPIMULT) is appropriately changed in the perforation section to simulate the negative skin; in the ultra-low permeability reservoir, there is often a phenomenon of mudstone absorbing water. The percentage of some invalid water injection volume in the total water injection volume is determined by test data, and the invalid water injection volume in this part of the water injection well needs to be deducted in the historical matching. The specific requirement of appropriately changing the productivity index is that when the skin coefficient of the well is greater than 0, the well is imperfect, and when the skin coefficient of the well is less than or equal to 0, the well is perfect. Therefore, the appropriate change of the productivity index is to make the skin coefficient of the perforation section of a single well as small as possible. Less than 0, and near 0.

[0023] Furthermore, in the history matching process, when the fitting effect cannot be achieved by changing the perforation section productivity index, the change in permeability in the maximum principal stress direction caused by conventional fracturing is simulated by increasing the grid conductivity, so as to achieve the purpose of improving the history matching accuracy of the entire block and single well.

[0024] Furthermore, the structural model refers to the grid structural model containing structural and fault information established in the early stage, and the attribute model refers to the porosity, permeability and saturation model.

[0025] The present invention discloses a method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis, and the beneficial effect is that compared with the conventional numerical simulation of hydraulic fracturing in ultra-low permeability reservoirs, the method can improve the historical fitting accuracy of conventional hydraulic fracturing single wells in ultra-low permeability reservoirs, thereby outputting accurate pore pressure and fluid saturation fields for fracture network fracturing simulation. The method is more scientific, the accuracy of historical fitting is higher, and the prediction of pore pressure and fluid saturation fields is more accurate and reliable, which is of great significance for guiding oilfield production and development. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0027] Figure 1 Shown is a schematic diagram of the permeability geological model obtained in Example 1;

[0028] Figure 2 Shown is a schematic diagram of the porosity model obtained in Example 1;

[0029] Figure 3 Shown is a schematic diagram of the effective thickness model obtained in Example 1;

[0030] Figure 4 Shown is a schematic diagram of a multi-stage well model of a horizontal well in Example 1;

[0031] Figure 5 It is a schematic diagram of the negative skin of a single well simulated by changing the productivity index of the perforation section in Example 1;

[0032] Figure 6 The figure shows the statistical diagram of the invalid water injection ratio of the ultra-low permeability reservoir in XX Oilfield;

[0033] Figure 7 Shown is a schematic diagram of local grid conductivity correction in Example 2;

[0034] Figure 8 Shown is a schematic diagram of the history matching results of Example 2.

[0035] Specific implementation methods

[0036] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] In order to further understand the content of the present invention, the present invention is further described below in combination with specific embodiments. The present invention is based on the ultra-low permeability reservoir numerical simulation technology based on the black oil model, and takes the conventional hydraulic fracturing numerical simulation of the ultra-low permeability reservoir as the research object, and clarifies the method of improving the accuracy of the conventional hydraulic fracturing numerical simulation of the ultra-low permeability reservoir from three aspects: static fine model establishment and attribute model optimization, and conventional hydraulic fracturing equivalent simulation.

[0038] Example 1: Figures 1 to 6 As shown, a method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis specifically includes:

[0039] Step 1: Establish a probability model of ultra-low permeability reservoir;

[0040] The spatial fault model of seismic interpretation is combined with the verification of single well spatial breakpoint data to describe the occurrence of block spatial faults. Based on the geological stratification data of single wells, the seismic interpretation layer data is used as constraints. Through the convergence method interpolation calculation, a stratified layer model is established to describe the stratified structural fluctuation characteristics. The fault and layer models are gridded to establish a gridded structural model containing structural and fault information.

[0041] Using the coring well data, we established an intersection diagram of the relationship between the acoustic time difference and the core porosity, regressed the relationship curve between the acoustic time difference and the porosity, established a relationship chart between porosity and permeability through the core data, analyzed the correlation between the two, regressed the relationship between permeability and porosity, and established an equation for calculating the original oil saturation of the logging data layered section using the multivariate statistical regression method of the core analysis saturation and the logging data. Using the above regression relationship, we corrected the porosity and permeability of the sand body layer by layer and well by well through the electrical logging curve of a single well;

[0042] Each small layer of the structural model is subdivided into multiple sections with a length of 1 m in the longitudinal direction, so as to describe the heterogeneity within the small layer.

[0043] Step 2, determine the equiprobability geological model;

[0044] The gridded sedimentary facies data volume is added to the work area, and a sedimentary facies model is established by directly assigning values ​​to each grid. The corrected single well porosity, permeability and effective thickness data are assigned to the established grid. Based on the random function theory, the random simulation technology controlled by the sedimentary facies model is adopted. Through the variogram analysis and the application of the sequential Gaussian interpolation method, multiple geological models of equal probability of porosity, permeability and effective thickness are generated respectively.

[0045] The main range direction, range length, and random seed points are uncertain parameters generated by the formation of an equal probability geological model. Therefore, through the reserve fitting of the geological model, the main range direction, range length, and random seed points of the geological model close to the reported reserves are obtained as the parameters of the optimal model. By inputting the determined main range direction, range length, and random seed point simulation parameters, the determined geological model such as porosity, permeability, and effective thickness is obtained. Figure 1 , Figure 2 and Figure 3 The optimal geological model is shown; the output of the optimized geological model is performed. During the output of the geological model, in order to meet the requirements of numerical simulation on the number of geological model grids, the subdivided sublayers are re-merged for non-target sublayers, and the target sublayers are still kept in a subdivided state to reflect the heterogeneity within the layer.

[0046] Step 3: numerical simulation of conventional hydraulic fractures in ultra-low permeability reservoirs;

[0047] Using the output geological model, a numerical model is constructed, including data such as skeleton grid, active grid distribution, grid depth and height, grid step size, grid porosity, permeability and oil saturation attributes;

[0048] The fracture development characteristics of the block were observed intuitively from core observation and imaging logging, and it was determined that the block was a closed fracture before fracture water injection, so the numerical model selected a single-pore medium model;

[0049] Add well information and static attribute parameters to the numerical model. Vertical wells use uniform static attribute parameters, and horizontal wells generally use multi-segment well description technology, such as Figure 4 As shown, the wellbore is discretized into four segments, each segment consists of a node and a flow path, and each segment has an independent node length, pressure, and roughness. The node length, pressure, and roughness parameters of the wellbore are defined by segmentation to accurately describe the fluid flow characteristics inside the horizontal wellbore. The fluid flow characteristics are affected by the pressure and the roughness of the wellbore. Different pressures and roughness can be given for each segment through segmentation settings, so that the flow state of the fluid in each segment is closer to the actual situation.

[0050] Carry out reserve fitting based on the numerical model, compare it with the submitted geological reserves, and calculate the geological reserve fitting accuracy of the layers. Because the geological model is selected based on the geological reserves in the early stage, the geological reserve fitting accuracy of the numerical model is generally required to be greater than 95%;

[0051] The numerical model established in the early stage was applied to add the perforation and production information of a single well to carry out historical matching. In the process of historical matching, due to the lack of actual construction parameters of conventional hydraulic fracturing, the change of productivity index (WPIMULT) was used to equate conventional cracks during hydraulic fracturing. Conventional hydraulic fracturing improves the permeability of the oil layer near the bottom of the well, causing the additional flow resistance near the bottom of the well to be negative (negative skin coefficient). By changing Figure 5 The productivity index (WPIMULT) in the wellbore can improve the production capacity of a single well. The improvement of the production capacity of a single well can be considered as the enhancement of the oil layer's seepage capacity, that is, the permeability of the oil layer near the bottom of the well is improved, thereby achieving a better single well fitting effect and being consistent with the actual dynamic understanding. In ultra-low permeability reservoirs, there is often a phenomenon of mudstone absorbing water. The percentage of some invalid water injection volume to the total water injection volume is determined through test data, such as Figure 6 As shown in the figure, the invalid water injection volume in this part of the injection wells needs to be deducted in the historical matching.

[0052] Embodiment 2:

[0053] like Figures 1 to 4 and Figures 6 to 8 As shown, the difference from Example 1 is that in step 3, when the productivity index of the single well perforation section is changed, the historical fitting accuracy of the single well liquid production, oil production and water content is still less than 80%, as shown in Figure 7 As shown, the change in permeability in the direction of the maximum principal stress caused by conventional fracturing wells is simulated by increasing the grid conductivity, so as to improve the accuracy of history fitting for the entire block and a single well.

[0054] The extent of the conductivity increase is determined based on the empirical value of the region, and a range of conductivity values ​​is determined by floating up and down. Based on this, historical fitting is carried out. Each conductivity value in the conductivity range corresponds to a historical fitting result, such as Figure 8 The corresponding set of conductivity values ​​with the best historical matching effect is shown as the final result. The conductivity value range is determined based on the permeability, effective grid thickness and grid thickness in the region. First, the average conductivity in the region is calculated, and then the average conductivity is adjusted up and down by 20% to obtain the conductivity range.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis, characterized in that: The specific steps are as follows: Step 1: Establish a probability model of ultra-low permeability reservoir; Step 2, determine the equiprobability geological model; Step 3: Numerical simulation of conventional hydraulic fractures in ultra-low permeability reservoirs.

2. A method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 1, characterized in that: Step 1 specifically includes: ① Through the verification of the spatial fault model of seismic interpretation and the spatial breakpoint data of a single well, the occurrence of the spatial faults in the block is described. According to the geological stratification data of a single well, the seismic interpretation layer data is used as a constraint condition. Through the convergence method interpolation calculation, a stratified layer model is established to describe the structural fluctuation characteristics of the stratification. The fault and layer models are gridded to establish a gridded structural model containing structural and fault information; ② Using well data, establish an intersection diagram of the relationship between acoustic time difference and core porosity, regress the relationship curve between acoustic time difference and porosity, establish a relationship chart between porosity and permeability through core data, analyze the correlation between the two, regress the relationship between permeability and porosity, and use the multivariate statistical regression method of core analysis saturation and logging data to establish an equation for calculating the original oil saturation of the logging data layered section. Using the above regression relationship, the porosity and permeability of the sand body are corrected layer by layer and well by well through the electrical logging curve of a single well; ③ Each small layer of the structural model is subdivided into multiple sections with a length of 1m.

3. A method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 2, characterized in that: Taking the crack direction into comprehensive consideration, the specific requirements are that the horizontal grid is parallel to the crack direction, and the vertical grid is perpendicular to the crack direction.

4. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 1, characterized in that: Step 2 specifically includes adding the gridded sedimentary facies data volume to the work area, establishing a sedimentary facies model through direct grid assignment, applying sedimentary facies model constraints, establishing phase-controlled equiprobability multi-realization attribute modeling through a sequential Gaussian algorithm, calculating geological reserves using different attribute models, selecting the attribute model that is closest to the calibrated geological reserves as the final attribute model, and outputting the final geological model.

5. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 4, characterized in that: In the process of geological model output, in order to meet the requirements of numerical simulation on the number of geological model grids, the subdivided sublayers are re-merged for non-target sublayers.

6. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 4, characterized in that: Different attribute models refer to porosity model, effective thickness model and permeability model; selecting the attribute model closest to the calibrated geological reserves as the final attribute model means calculating different geological reserves using different geological models, and the geological model whose calculated geological reserves are closest to the geological reserves calculated by the static volume method is the final model.

7. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 1, characterized in that: Step three specifically includes: Nano-CT scanning experiments and nuclear magnetic resonance experiments are used to systematically study the microscopic pore structure characteristics and movable fluid distribution characteristics of ultra-low permeability reservoirs, providing basic reservoir parameters for refined numerical simulations; The output structural model and attribute model are used to establish the numerical model, including the attribute volume data such as skeleton grid, active grid distribution, grid depth and height, grid step size, grid porosity, permeability and oil saturation; Reserve fitting is carried out based on the numerical model, and the perforation and production information of the single well are added to the numerical model established in the early stage to carry out historical matching.

8. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 7, characterized in that: In the process of historical matching, the conventional cracks in the first fracturing are treated equivalently, and the production capacity index is appropriately changed in the perforation section to simulate the negative skin. In ultra-low permeability reservoirs, there is often a phenomenon of mudstone absorbing water. The percentage of some invalid water injection volume in the total water injection volume is determined through test data. The invalid water injection volume in this part of the injection well needs to be deducted in historical matching.

9. The method for improving the fitting accuracy of hydraulic fracturing numerical simulation based on experimental analysis according to claim 8, characterized in that: During the history matching process, when the fitting effect cannot be achieved by changing the productivity index of the perforation section, the change of permeability in the direction of the maximum principal stress caused by conventional fracturing is simulated by increasing the grid conductivity, so as to improve the accuracy of history matching for the entire block and a single well.

Images