Priori knowledge-based reservoir fracturing comprehensive evaluation method

Abstract

The invention relates to the technical field of oil and gas yield increase, in particular to a reservoir fracturing performance comprehensive evaluation method based on prior knowledge, which comprises the following steps: 1) optimizing fracturing performance evaluation indexes by using the prior knowledge, determining the nature of each fracturing performance evaluation index, and constructing virtual positive and negative ideal point labels; 2) constructing an original evaluation matrix based on the optimized fracturing performance evaluation indexes; (3) carrying out standardization treatment on the optimized fracturing performance evaluation indexes; 4) determining the weight of each fracturing performance evaluation index by adopting an entropy weight method; 5) weighting the weight calculated by the entropy weight method into the original evaluation matrix, and constructing a weighted standardized matrix; and 6) calculating the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point and the Euclidean sticking rate, and evaluating the fracturing property of the reservoir. According to the method, the concept of priori knowledge is introduced into reservoir comprehensive evaluation, so that a large amount of field data is effectively utilized, and the fracturing performance of the reservoir can be scored and quantified.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas production increase, and in particular relates to a comprehensive evaluation method for reservoir fracturing based on prior knowledge. Background Art

[0002] Unconventional oil and gas resources account for 80% of the total oil and gas reserves in the world, and more efficient production of shale gas plays an important role in my country's energy strategic security. Due to the extremely poor physical properties of shale reservoirs, there is basically no natural production capacity, and it is often necessary to use horizontal wells and large-scale hydraulic fracturing to have a certain industrial production capacity. The scale-effective exploitation of shale gas reservoirs depends on the comprehensive evaluation of geological-engineering fracturability. However, in the multi-scale and multi-factor comprehensive evaluation process, due to the high heterogeneity of shale reservoirs, it is difficult for conventional evaluation methods to integrate the key factors required for information from different data types and different scales. There is also some controversy about the relative importance of geological sweet spots and engineering sweet spots in different regions. The main controlling factors are selected based on experience, and the comprehensive evaluation is only carried out through simple parameter interval division, which is not very interpretable.

[0003] Sweet spots refer to areas where existing exploration and development technologies can produce economic oil and gas production and have actual development benefits. According to the role of sweet spots in the development of unconventional oil and gas reservoirs, they can be further subdivided into geological sweet spots and engineering sweet spots. Geological sweet spots refer to areas with rich oil and gas resources, good physical properties, and high fluidity, usually characterized by indicators such as organic matter content, total porosity, gas saturation, and permeability. Engineering sweet spots refer to areas with good fracturing properties, which are conducive to low-cost and high-efficiency fracturing construction and easy to form fracture networks. They are usually characterized by indicators such as rock mechanics parameters, brittleness, degree of natural fracture development, and anisotropic characteristics.

[0004] Accurate evaluation of geological sweet spots and engineering sweet spots is the key to economically and efficiently developing unconventional reservoirs. In terms of geological sweet spot evaluation, Zhang et al. (Zhang S, Liu H, Liu Y, et al. Main controls and geological sweetspot types in Paleogene shale oil rich areas of the Jiyang Depression, Bohai Bay basin, China [J]. Marine and Petroleum Geology, 2020, 111 576-587) combined X-ray diffraction and geochemical indicators, and analyzed the characteristics of shale reservoirs in detail based on well core sample thin sections, field emission scanning electron microscopy observations and mercury injection experiments, determined the sweet spot value range of each main controlling factor, and selected the sweet spot by directly superimposing the indicator value range, but different indicators have different degrees of influence on production. Jia et al. (JIA Y, HUANG W, WANG P, et al. Prediction of Longitudinal Superimposed "Sweet Spot" of Tight Gas Reservoir: A Case Study of Block G, Canada [J]. Processes, 2023, 11 (3): 666) calculated the index weights based on the hierarchical analysis method, optimized the key parameters of production capacity using the Pearson-MIC comprehensive evaluation method, and classified the geological "sweet spot" evaluation indicators. However, the result is still to divide the sweet spot distribution by superimposing indicators, and it is impossible to achieve multi-indicator and multi-parameter fusion. In the evaluation of engineering sweet spots, that is, the evaluation of fracturing properties, many scholars have mainly considered brittleness, rock mechanics parameters, fracture toughness, and the degree of natural fracture development to establish an engineering sweet spot evaluation method. However, the fracturability of the reservoir is affected by intrinsic properties such as reservoir brittleness, fracture toughness, natural fractures, and internal factors such as confining pressure; the external stress induced by hydraulic fracturing will cause the rock to have different fracturability characteristics (Zeng F, Zhang Y, Guo J, et al. Optimized completion design fortriggering a fracture network to enhance horizontal shale well production[J]. Journal of Petroleum Science and Engineering, 2020, 190 107043-107043).Zeng et al. (Fanhui Z, Gaobin G, Yu Z, et al. Fracability evaluation of shale reservoirs considering rock brittleness, fracture toughness, and hydraulic fracturing-induced effects [J]. Geoenergy Science and Engineering, 2023, 229) first established a rock mechanical parameter brittleness evaluation method based on in-situ triaxial mechanical experiments and logging data, introduced a displacement discontinuity method considering the percolation effect, and established a net pressure calculation model considering fracture extension; by introducing the fractal calculation of fracture toughness of bifurcated irregular fracture rocks and applying the product method, a comprehensive reservoir compressibility evaluation method considering the comprehensive effects of reservoir brittleness, fracture toughness, induced fracture index, etc. was established. The above sweet spot evaluation method studies the geological sweet spot and engineering sweet spot separately based on the physical meaning, and does not consider the correlation between the sweet spot index and production.

[0005] It can be seen that unconventional reservoirs are highly heterogeneous, involve many evaluation factors, and the resolution and accuracy of physical models are difficult to meet actual needs. During the development of oil and gas fields, a large amount of rich oil and gas field data was collected on site, which laid a good foundation for establishing a sweet spot evaluation method based on data fusion drive. Some researchers have attempted to establish a comprehensive evaluation method for geological and engineering sweet spots by establishing the relationship between production and geological sweet spots and engineering sweet spots. However, these methods are all based on data-driven with labeled data to help the model identify data features. However, commonly used labeled data relies heavily on the user's subjective and consistent judgment, and requires large-scale, high-quality sample training data sets for constrained training, which is not suitable for the application scenarios of unlabeled samples in practical applications.

[0006] Some scholars use unsupervised machine learning to solve the problem of unlabeled data. Unsupervised unlabeled samples are processed by calculating the similarity, distance or probability distribution between the indicators of geological engineering sweet spots, and performing anomaly detection, self-organizing mapping algorithm, clustering, dimensionality reduction, association rule mining, etc. on the data. There is no need to label the data or give specific goals, and the same type of reservoir compressibility well sections are preferred. Unsupervised learning does not require any prior model training, and high-precision classification effects are obtained through clustering analysis of the data itself. However, clustering is an unsupervised learning process that can cluster reservoir sweet spots into different categories, but the performance of each category cannot be directly judged based on the clustering results, and the classification effect cannot be achieved; and clustering analysis mainly focuses on how to group data objects, rather than how to group them according to the importance of indicators or variables, without considering the fusion of multiple factors. At the same time, in the clustering process, the weights of various indicators or variables are regarded as equal, regardless of their importance differences. How to effectively classify a large amount of unlabeled data in horizontal wells is a technical problem that needs to be solved urgently. Summary of the invention

[0007] The purpose of the present invention is to provide a comprehensive evaluation method for reservoir fracturability based on prior knowledge.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is:

[0009] A comprehensive evaluation method for reservoir fracturability based on prior knowledge, characterized by comprising the following steps:

[0010] 1) Utilize prior knowledge to select the optimal fracturability evaluation indexes, characterize the various fracturability evaluation indexes, and construct virtual positive and negative ideal point labels;

[0011] 2) constructing an original evaluation matrix based on the preferred fracturability evaluation index;

[0012] 3) Standardizing the preferred fracturability evaluation indexes;

[0013] 4) The entropy weight method is used to determine the weight of each fracturing performance evaluation index;

[0014] 5) Add the weights calculated by the entropy weight method to the original evaluation matrix to construct a weighted standardized matrix;

[0015] 6) Calculate the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point, as well as the Euclidean distance, to evaluate the fracturability of the reservoir.

[0016] Preferably, the prior knowledge in step 1) includes physical laws, engineering control factors, and expert experience.

[0017] Preferably, constructing virtual positive and negative ideal point labels in step 1) includes:

[0018] Positive and negative indicators are used to generate labeled evaluation points with certain labels. The expression is as follows:

[0019] Ideal point:

[0020]

[0021] Negative ideal point:

[0022]

[0023] Where: is a positive and negative ideal point, y kj is the evaluation index value.

[0024] Preferably, the fracturability evaluation indexes in step 1) include TOC, total gas content, total hydrocarbon content, porosity, Young's modulus, Poisson's ratio, horizontal principal stress difference, and brittle mineral content.

[0025] Preferably, in step 1), the positive indicators include total gas content, total hydrocarbon content, porosity, TOC, brittle mineral content, and Young's modulus, and the negative indicators include Poisson's ratio and horizontal principal stress difference.

[0026] Preferably, in step 3), a normalization method is used for standardization.

[0027] Preferably, the step 6) of calculating the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point and the Euclidean distance of the evaluation point to the positive ideal point and the negative ideal point includes:

[0028] Positive Euclidean distance:

[0029]

[0030] Negative Euclidean distance:

[0031]

[0032] Euclidean closeness:

[0033]

[0034] Where:

[0035] d i ——Euclidean closeness;

[0036] ——The distance between a data point and the positive ideal point;

[0037] ——The distance between a data point and the negative ideal point.

[0038] In summary, the present invention proposes a comprehensive evaluation method for reservoir fracturability based on prior knowledge. The improved technical solution first uses prior physical knowledge to optimize field indicators and data, and qualitatively analyzes various evaluation indicators through geophysical chemistry and reservoir engineering data to construct virtual positive and negative ideal point labels; according to the preferred fracturability evaluation indicators, the original evaluation matrix is ​​constructed and standardized; the weights of each fracturability evaluation indicator are determined by the entropy weight method; based on the constructed virtual positive and negative ideal point labels, the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point and the Euclidean distance are further calculated to effectively evaluate the reservoir fracturability, and a new comprehensive evaluation method for reservoir fracturability is provided. The method introduces the concept of prior knowledge into the comprehensive evaluation of reservoirs, so that a large amount of data generated on the oil and gas field site can be effectively utilized, and the reservoir fracturability can be scored, the reservoir fracturability evaluation results can be quantified, and the reservoir fracturability can be judged more intuitively. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0040] Figure 2 Graph showing the distribution of TOC with segment numbers in an embodiment of the present invention;

[0041] Figure 3 This is a diagram showing the distribution of total gas content with section number in an embodiment of the present invention;

[0042] Figure 4 This is a diagram showing the distribution of porosity with segment number in an embodiment of the present invention;

[0043] Figure 5 This is a diagram showing the distribution of total hydrocarbon content with respect to segment number in an embodiment of the present invention;

[0044] Figure 6 This is a diagram showing the distribution of brittle mineral content with respect to segment number in an embodiment of the present invention;

[0045] Figure 7 Graph showing the distribution of Young's modulus with segment number in an embodiment of the present invention;

[0046] Figure 8 Graph showing the distribution of Poisson's ratio with segment number in an embodiment of the present invention;

[0047] Fig. 9 Graph showing the distribution of horizontal principal stress difference with segment number in an embodiment of the present invention;

[0048] Fig.10 Graph showing the weights of various evaluation indicators in the embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] The present invention provides a comprehensive evaluation method for reservoir fracturing based on prior knowledge, comprising the following steps:

[0052] 1) Utilize prior knowledge to select the optimal fracturability evaluation indexes, characterize the various fracturability evaluation indexes, and construct virtual positive and negative ideal point labels;

[0053] 2) constructing an original evaluation matrix based on the preferred fracturability evaluation index;

[0054] 3) Standardizing the preferred fracturability evaluation indexes;

[0055] 4) The entropy weight method is used to determine the weight of each fracturing performance evaluation index;

[0056] 5) Add the weights calculated by the entropy weight method to the original evaluation matrix to construct a weighted standardized matrix;

[0057] 6) Calculate the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point, as well as the Euclidean distance, to evaluate the fracturability of the reservoir.

[0058] Furthermore, in step 1), prior physical knowledge such as physical laws, engineering control, and expert experience are used to optimize on-site fracturability evaluation indicators and data, and each evaluation indicator is qualitatively characterized through geophysical chemistry and reservoir engineering data to construct virtual positive and negative ideal point labels.

[0059] Positive and negative indicators are used for artificial supervision and constraint to generate labeled evaluation points with certain labels. The expression is as follows:

[0060] Ideal point:

[0061]

[0062] Negative ideal point:

[0063]

[0064] Where: It is a positive and negative ideal point, which is essentially a 1-row n-column matrix, and the data in each column consists of the maximum or minimum value of each positive and negative evaluation index.

[0065] Furthermore, in step 2), an original evaluation matrix is ​​constructed based on the preferred fracturability evaluation index, which specifically includes:

[0066] Based on the preferred evaluation indicators, there are n evaluation indicators, each of which has m data points. The initial evaluation matrix X constructed by n indicators and m data points is:

[0067]

[0068] Where: i = 1, 2, ... m, j = 1, 2, ... n; x ij Represents the specific value of the j-th evaluation indicator at the i-th data point.

[0069] Furthermore, in step 3), a normalization method is used to standardize the preferred fracturability evaluation index, specifically:

[0070]

[0071] In the formula, is the standardized value; x ij is the value of the factor affecting production; minx ij is the minimum value of the factor in the sample; maxx ij is the maximum value of this factor in the sample.

[0072] Furthermore, in step 4), the entropy weight method is used to determine the weights of each fracturability evaluation index, specifically:

[0073] The original evaluation matrix constructed in step 2) is processed according to the standardized processing method described in step 3) to construct a standardized evaluation matrix R = (r ij ) m×n :

[0074]

[0075] Where: i = 1, 2, ..., m; j = 1, 2, ..., n; r ij is the evaluation value of the i-th item under the j-th indicator.

[0076] For positive indicators affecting fracturability:

[0077]

[0078] For negative indicators that affect fracturability:

[0079]

[0080] Where: X max , X min They are the maximum and minimum values ​​under the same indicator respectively.

[0081] According to the standardized data, calculate the proportion p of the index value of each project under each index ij :

[0082]

[0083] Calculate the information entropy e of each indicator j :

[0084]

[0085] Where: e j is the entropy value of the evaluation index; when p ij =0, p ij lnp ij =0; m is the number of data.

[0086] Calculate the weight c of each evaluation index j :

[0087]

[0088] Where: c j is the weight of the jth evaluation index, dimensionless; n is the number of evaluation indicators.

[0089] Furthermore, in step 5), the weights calculated by the entropy weight method are weighted into the original evaluation matrix to construct a weighted standardized matrix, specifically:

[0090]

[0091] Y is the weighted normalization matrix.

[0092] Furthermore, in 6), the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point and the Euclidean distance are calculated to evaluate the fracturability of the reservoir, specifically:

[0093] Positive Euclidean distance:

[0094]

[0095] Negative Euclidean distance:

[0096]

[0097] Euclidean closeness:

[0098]

[0099] d i ——Euclidean closeness;

[0100] ——The distance between a data point and the positive ideal point;

[0101] ——The distance between a data point and the negative ideal point.

[0102] The greater the Euclidean closeness, the better the fracturability.

[0103] Example verification:

[0104] Taking 16 horizontal sections of a well as the research object, the characteristic parameters related to the previous fracturing transformation were extracted and statistically analyzed based on prior knowledge, and TOC, total gas content, total hydrocarbon content, porosity, Young's modulus, Poisson's ratio, horizontal principal stress difference, and brittle mineral content were selected as fracturing evaluation indicators. The positive indicators were total gas content, total hydrocarbon content, porosity, TOC, brittle mineral content, and Young's modulus; the negative indicators were Poisson's ratio and horizontal principal stress difference. A database was established (Table 1).

[0105] Table 1 Geological parameter database

[0106]

[0107] The distribution of each evaluation index with segment number is as follows Figure 2-9 After each evaluation index is standardized according to step 3), the weight is determined according to the entropy weight method in step 4. The results are shown in Table 2. Fig.10 shown.

[0108] Table 2 Weights of evaluation indicators

[0109]

[0110]

[0111] The weight calculation results of the entropy weight method show that the weight of TOC is 12.17%, the weight of Young's modulus is 17.958%, the weight of total gas content is 7.09%, the weight of brittle mineral content is 16.871%, the weight of porosity is 6.974%, the weight of total hydrocarbon content is 15.128%, the weight of horizontal principal stress difference is 6.963%, and the weight of Poisson's ratio is 16.846%. The maximum value of the indicator weight is Young's modulus (17.958%), and the minimum value is the horizontal principal stress difference (6.963%).

[0112] After the above steps, the selected fracturability evaluation index is processed according to the method described in step 5) and step 6) to calculate the comprehensive score (Euclidean paste progress) of each section, and the calculation results are shown in Table 3. The calculated comprehensive scores of each section are distributed between 0.3872 and 0.5364. The distribution interval is divided according to the comprehensive score, and then the fracturability of each section is divided into excellent and general, corresponding to Class I (0.4518-0.5364) and Class II (0.3872-0.4375), respectively.

[0113] Table 3 Comprehensive score results of each level

[0114]

[0115] Comparing the comprehensive scores of each section with their respective production index (gas extraction index), the comprehensive score change trend of 10 horizontal sections, namely, horizontal sections 3, 4, 6, 7, 9, 10, 11, 12, 14, and 16, is consistent with the change trend of their production index (gas extraction index), accounting for 62.5%. Due to parameter reasons, the comprehensive score change trend of 6 horizontal sections, namely, horizontal sections 1, 2, 5, 8, 13, and 15, is inconsistent with the change trend of their production index (gas extraction index), accounting for 37.5%. It can be seen that the horizontal sections with large gas extraction indexes also have high scores, which have good correlation, indicating that this method has a high degree of fit with actual application and has strong field practicability.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A comprehensive evaluation method for reservoir fracturability based on prior knowledge, characterized in that: The following steps are involved: 1) Utilize prior knowledge to select the optimal fracturability evaluation indexes, characterize the various fracturability evaluation indexes, and construct virtual positive and negative ideal point labels; 2) constructing an original evaluation matrix based on the preferred fracturability evaluation index; 3) Standardizing the preferred fracturability evaluation indexes; 4) The entropy weight method is used to determine the weight of each fracturing performance evaluation index; 5) Add the weights calculated by the entropy weight method to the original evaluation matrix to construct a weighted standardized matrix; 6) Calculate the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point, as well as the Euclidean distance, to evaluate the fracturability of the reservoir.

2. The comprehensive evaluation method for reservoir fracturability as claimed in claim 1, wherein the prior knowledge in step 1) includes physical laws, engineering control factors, and expert experience.

3. The comprehensive evaluation method for reservoir fracturability according to claim 1, wherein the step 1) constructs virtual positive and negative ideal point labels, including: Positive and negative indicators are used to generate labeled evaluation points with certain labels. The expression is as follows: Ideal point: Negative ideal point: Where: is a positive and negative ideal point, y kj is the evaluation index value.

4. The comprehensive evaluation method for reservoir fracturability according to claim 1, wherein the fracturability evaluation indexes in step 1) include TOC, total gas content, total hydrocarbon content, porosity, Young's modulus, Poisson's ratio, horizontal principal stress difference, and brittle mineral content.

5. The comprehensive evaluation method for reservoir fracturability according to claim 4, wherein the positive indicators in step 1) include total gas content, total hydrocarbon content, porosity, TOC, brittle mineral content, and Young's modulus, and the negative indicators include Poisson's ratio and horizontal principal stress difference.

6. The comprehensive evaluation method for reservoir fracturability according to claim 1, wherein the normalization method is used for standardization in step 3).

7. The comprehensive evaluation method for reservoir fracturability according to claim 1, wherein the step 6) of calculating the Euclidean distance from the evaluation point to the positive ideal point and the negative ideal point and the Euclidean distance of the evaluation point comprises: Positive Euclidean distance: Negative Euclidean distance: Euclidean closeness: Where: d i ——Euclidean closeness; ——The distance between a data point and the positive ideal point; ——The distance between a data point and the negative ideal point.

Images