Residual gas distribution research based on dynamic fine-grained K-means recursive algorithm

Abstract

The invention discloses a method for applying a dynamic fine-grained K-means recursive algorithm to residual gas type distribution of a tight sandstone reservoir. The method comprises the following steps: acquiring main influence factors of each grid influencing residual gas at a tight gas t moment; taking the main influence factors on all the grids as input, adopting K-means to carry out unsupervised learning on all the grids through the main control factors, and pasting labels; and then a new algorithm, namely a dynamic fine-grained K-means recursive algorithm, is utilized, space coordinate parameters of the grids are fused, a threshold value is set, dynamic recursive search is carried out on each grid to form a residual gas area, an area with high residual gas, an area with medium residual gas and an area with low residual gas are depicted, and finally residual gas type distribution of each layer is obtained. According to the new method for determining the distribution type of the residual gas, multiple types of dynamic and static data can be processed, the residual gas can be dynamically analyzed, unsupervised learning and recursion are carried out through main control factors, the distribution type of the residual gas is solved, and a basis is provided for subsequent potential tapping measures.

Description

Technical Field

[0001] The present invention belongs to the field of unconventional natural gas development and tight sandstone residual gas type distribution, and specifically provides a method for tight sandstone reservoir residual gas type distribution based on a dynamic fine-grained K-means recursive algorithm. Background Art

[0002] With the growth of global energy demand and the increasing depletion of conventional oil and gas resources, tight gas has become an important area for the development of unconventional natural gas resources. It is widely distributed in major oil and gas basins in the world, and in recent years it has become the main source of global natural gas reserves and production growth. Due to the poor physical properties and complex storage space of tight sandstone gas reservoirs, among which the porosity and permeability are low and the degree of fracture development is poor, the development is difficult and the final recovery rate is low. There is a problem of local enrichment of residual gas that is difficult to accurately characterize. When faced with a large amount of drilling, mining and development data, traditional numerical simulation and qualitative analysis methods can usually only use part of the data to evaluate the residual gas unilaterally, and cannot achieve comprehensive evaluation of high-dimensional data. With the development of artificial intelligence, machine learning methods have good nonlinearity, self-learning, adaptability and fault tolerance, can handle multiple types of dynamic and static data, quantitative models to determine the distribution type of residual gas, and dynamically analyze the residual gas, providing a new method for determining the distribution of residual gas types.

[0003] The K-means algorithm is an iterative unsupervised analysis algorithm. Its core idea is to divide the n objects in the data set into K clusters so that the sum of the distances from each object to the center of the cluster to which it belongs is minimized. It continuously optimizes the clustering results in an iterative manner so that the objects in each cluster are as close as possible, while the objects in different clusters are as separated as possible. The advantages of this algorithm are that it is intuitive, easy to understand, fast in calculation, and easy to implement. However, it also has some limitations, such as being sensitive to the selection of the initial cluster centroid, being likely to fall into a local optimal solution, and requiring the number of clusters K to be set in advance. In addition, the K-means algorithm cannot meet the requirements for forming various types of regional blocks of residual gas.

[0004] Therefore, the K-means algorithm is improved and upgraded. After unsupervised learning is performed on all grids using K-means, labels are attached. Then, the main control factors and spatial parameters of each grid are integrated, the threshold is set, and each grid is dynamically recursively searched to form various types of residual gas regional blocks, characterizing areas with high residual gas, areas with medium residual gas, and areas with low residual gas, and finally obtaining the residual gas distribution type of tight sandstone gas reservoirs. This provides a theoretical basis for adding wells and opening layers in the reservoir in the later stage. Summary of the invention

[0005] The existing technologies either evaluate the remaining gas from a unilateral perspective, fail to achieve comprehensive evaluation of high-dimensional data, or fail to meet the requirements for forming various types of regional blocks of remaining gas. The purpose of the present invention is to provide a method for the distribution of remaining gas types in tight sandstone reservoirs based on an improved K-means algorithm—a dynamic fine-grained K-means recursive algorithm, which aims to solve the problem of being able to use factors affecting the remaining gas to form various types of regional blocks of remaining gas when facing high-dimensional data of tight sandstone gas reservoirs.

[0006] In order to solve the above problems, the present invention specifically provides the following technical solutions:

[0007] Based on the improved K-means algorithm - dynamic fine-grained K-means recursive algorithm applied to the distribution of remaining gas types in tight sandstone reservoirs, the implementation steps are as follows:

[0008] Step S1: Obtain the main controlling factors affecting the residual gas in each grid at time t of dense gas, and construct a grid feature data set; that is, given a grid data set A at time t i ={x i ;y i ;z i ;h i ;p fi ;Ω gi ; G i}The spatial parameter coordinate set is F i ={x i ;y i ;z i}, the set of main controlling factors is M i ={h i ;p fi ;Ω gi ; G i}, then A i ={F i ;M i};

[0009] Where i is the i-th grid (i=1,2,...,n); x i is the x-axis coordinate of the ith grid; y i is the y-axis coordinate of the i-th grid; z i is the z-axis coordinate of the ith grid; h i is the effective thickness of the ith grid (m); p fi is the formation pressure of the ith grid (MPa); Ω gi is the abundance of the ith grid (m 3 / m 2 );G i is the residual gas of the ith grid (10 8 m 3 );

[0010] Step S2: take the main controlling factors on all grids as input, and use K-means to cluster all grids by the main controlling factors; for a given data set M of the i-th grid i ={h i ;p fi ;Ω gi ; G i} is normalized to improve the quality and stability of the clustering results. The specific formula is as follows:

[0011]

[0012] In the formula, l i is the original data of a certain type of data in grid i; l i * is the normalized data of a certain type of data in grid i; l min is the minimum value of this type of data; l max It is the maximum value of this type of data;

[0013] Step S3: Divide all grids into k sets Calculate the center of each set Solve the following formula for the goal:

[0014]

[0015] In the formula, It is the minimum value of the sum of the distances from all data to their respective cluster centers;

[0016] Step S4: Calculate the distance between each grid and each cluster center, assign each grid to the cluster center closest to it, and assign the grid M to the cluster center closest to it. i With cluster center Distance, the specific formula is as follows:

[0017]

[0018] Step S5: Recalculate the cluster center of each cluster The specific formula is as follows:

[0019]

[0020] In the formula, is a cluster with p samples;

[0021] Step S6: Repeat steps S4 to S5 until the partitioning stops changing or the maximum number of iterations is reached to obtain clusters. Different colors are assigned to grids in different clusters, denoted as Where i→j means that the i-th cluster corresponds to the j-th color;

[0022] Step S7: Add coordinate set parameter F to the clusters obtained by clustering i ={x i ;y i ;z i Get a new data set Continue recursively on this basis;

[0023] Step S8: i The categories within a certain distance of a grid in are counted, and the dataset W i Now there are n samples, each grid corresponds to a category The training data set W i It can be expressed as:

[0024]

[0025] p i =(x i ,y i ,z i )

[0026] The distance between any two points is:

[0027]

[0028] For any point p i Now, its ε neighborhood is defined as:

[0029] N ε ={p i ∈W i |dist<ε}

[0030] In the formula, dist<ε means the distance is less than ε;

[0031] Step S9: Data set W i The other grid categories in a grid area are counted, and the grid categories are reassigned according to whether the number of other categories in the grid area is ≥ 80%, until all grids are recursively completed;

[0032] Step S10, dividing the residual gas into three categories: reservoirs with high residual gas, reservoirs with medium residual gas, and reservoirs with low residual gas, and finally obtaining the residual gas type distribution of each layer in the tight gas reservoir.

[0033] As a further description of the above technical solution:

[0034] The main controlling factors include: permeability, porosity, depth, formation pressure, gas saturation, gas mass density, remaining gas volume, reserve abundance and effective thickness.

[0035] As a further description of the above technical solution:

[0036] Two machine learning methods (XGBoost and Random Forest RF) and four statistical criteria were used for evaluation, and finally the importance ranking of factors affecting the remaining gas was obtained. Through comprehensive analysis, reserve abundance, effective thickness and formation pressure were selected as the main controlling factors.

[0037] As a further description of the above technical solution:

[0038] The four statistical criteria include mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE) and coefficient of determination (R 2 ).

[0039] As a further description of the above technical solution:

[0040] The grid categories in step S9 also include grids without residual gas, so as to ensure that the number of grids within a certain distance of each grid is the same.

[0041] As a further description of the above technical solution:

[0042] The dataset W i The dataset is randomly divided into two parts: 80% for training and 20% for testing.

[0043] As a further description of the above technical solution:

[0044] 80% of the training set is used to train XGBoost and Random Forest RF and determine their optimal hyperparameter values, while 20% of the test set is used to evaluate the stability and generalization ability of the constructed model in predicting data.

[0045] In summary, due to the adoption of the above technical solution, the present invention can achieve the following technical effects:

[0046] The present invention first obtains the main influencing factors affecting the residual gas of each grid of tight gas at time t, and constructs a grid feature data set; then takes the main influencing factors on all grids as input, and uses K-means to cluster all grids through the main control factors; finally, a new algorithm, the dynamic fine-grained K-means recursive algorithm, is used, combined with the spatial coordinate parameters added to the grid, to divide the residual gas into three categories: reservoirs with high residual gas, reservoirs with medium residual gas, and reservoirs with low residual gas, and finally obtains the residual gas type distribution of each layer in the tight gas reservoir. This method is an improvement on the K-means algorithm, adds spatial coordinate parameters, and recursively performs each grid in its neighborhood, which increases the rationality and interpretability of the results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1It is a flow chart of the present invention;

[0048] Figure 2 To invent and improve the K-means algorithm flowchart;

[0049] Figure 3 Fine-grained k-means clustering result diagram (red, green, and blue correspond to high, medium, and low residual gas reservoirs);

[0050] Figure 4 Dynamic fine-grained K-means recursive result diagram (red, green, and blue correspond to high, medium, and low residual gas reservoirs);

[0051] Figure 5 The third layer has local magnified and enhanced contrast images of various places;

[0052] Figure 6 Local magnified and enhanced contrast images of various parts of the fourth layer. Specific implementation methods

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further illustrated below in conjunction with the accompanying drawings and embodiments. Obviously, the embodiments described are only partial embodiments of the present invention, and the protection scope of the present invention is not limited to the specific embodiments disclosed below. For those skilled in the art, all other embodiments obtained without creative work are within the protection scope of the present invention.

[0055] A method based on improved K-means algorithm - dynamic fine-grained K-means recursive algorithm applied to the distribution of remaining gas types in tight sandstone reservoirs, such as Figure 2 (Invention improved K-means algorithm flow chart), which includes the following steps:

[0056] Step S1: Obtain the main controlling factors affecting the residual gas in each grid at time t of dense gas, and construct a grid feature data set; that is, given a grid data set A at time t i ={x i ;y i ;z i ;h i ;p fi ;Ω gi ; G i}The spatial parameter coordinate set is F i ={x i ;y i ;z i}, the set of main controlling factors is M i ={h i ;p fi ;Ω gi ; G i}, then Ai ={F i ;M i};

[0057] Where i is the i-th grid (i=1,2,...,n); x i is the x-axis coordinate of the ith grid; y i is the y-axis coordinate of the i-th grid; z i is the z-axis coordinate of the ith grid; h i is the effective thickness of the ith grid (m) ;p fi is the formation pressure of the ith grid (MPa); Ω gi is the abundance of the ith grid (m 3 / m 2 );G i is the residual gas of the ith grid (10 8 m 3 );

[0058] The main controlling factors include permeability, porosity, depth, formation pressure, gas saturation, gas mass density, remaining gas volume, reserve abundance and effective thickness. Two machine learning methods (XGBoost and random forest RF) and four statistical criteria (mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE) and determination coefficient (R 2 )) was evaluated and finally the importance ranking of factors affecting the remaining gas was obtained. Through comprehensive analysis, reserve abundance, effective thickness and formation pressure were selected as the main controlling factors.

[0059] Step S2: take the main controlling factors on all grids as input, and use K-means to cluster all grids by the main controlling factors; for a given data set M of the i-th grid i ={h i ;p fi ;Ω gi ; G i} is normalized to improve the quality and stability of the clustering results. The specific formula is as follows:

[0060]

[0061] In the formula, l i is the original data of a certain type of data in grid i; l i * is the normalized data of a certain type of data in grid i; l min is the minimum value of this type of data; l max It is the maximum value of this type of data;

[0062] Step S3: Divide all grids into k sets Calculate the center of each set Solve the following formula for the goal:

[0063]

[0064] In the formula, It is the minimum value of the sum of the distances from all data to their respective cluster centers;

[0065] Step S4: Calculate the distance between each grid and each cluster center, assign each grid to the cluster center closest to it, and assign the grid M to the cluster center closest to it. i With cluster center Distance, the specific formula is as follows:

[0066]

[0067] Step S5: Recalculate the cluster center of each cluster The specific formula is as follows:

[0068]

[0069] In the formula, is a cluster with p samples;

[0070] Step S6: Repeat steps S4 to S5 until the partitioning stops changing or the maximum number of iterations is reached to obtain clusters. Different colors are assigned to grids in different clusters, denoted as Where i→j means that the i-th cluster corresponds to the j-th color;

[0071] Step S7: Add coordinate set parameter F to the clusters obtained by clustering i ={x i ;y i ;z i Get a new data set Continue recursively on this basis;

[0072] Step S8: i The categories within a certain distance of a grid in are counted, and the dataset W i Now there are n samples, each grid corresponds to a category The training data set W i It can be expressed as:

[0073]

[0074] p i =(x i ,y i ,z i )

[0075] The distance between any two points is:

[0076]

[0077] For any point p i Now, its ε neighborhood is defined as:

[0078] N ε ={p i ∈W i |dist<ε}

[0079] In the formula, dist<ε means the distance is less than ε;

[0080] Step S9: Data set W i The other grid categories in a grid area are counted, and the grid category is reassigned according to whether the number of other categories in the grid area is ≥ 80% (the grid category also includes grids with no remaining gas to ensure that the number of grids within a certain distance of each grid is the same) until all grids are recursively completed; Dataset W i It is randomly divided into two parts: 80% training set and 20% test set, where 80% of the training set is used to train XGBoost and Random Forest RF and determine their optimal hyperparameter values, and 20% of the test set is used to evaluate the stability and generalization ability of the constructed model in predicting data.

[0081] Step S10, dividing the residual gas into three categories: reservoirs with high residual gas, reservoirs with medium residual gas, and reservoirs with low residual gas, and finally obtaining the residual gas type distribution of each layer in the tight gas reservoir. Example 1

[0082] This example is implemented in a tight sandstone gas reservoir in a gas field. The reservoir is divided into 6 small layers, with a total of 1.72 million effective grids. The main influencing factors of each grid: reserve abundance, effective thickness and formation pressure are used as inputs, combined with spatial coordinate parameters, and the distribution of remaining gas types is recursively obtained. These three features are used as input parameters, and the output parameters are the proportion of each type of reservoir in each layer and the range of each main controlling factor (see Table 1):

[0083] Table 1 Grid coordinates and main control factors

[0084]

[0085] The dynamic fine-grained K-means recursive method is used to obtain the residual gas distribution type of each layer. First, all grids are divided into three sets Calculate the center of each set

[0086] Then use the formula:

[0087]

[0088] Calculate the distance from each grid to the center of each set and assign it to each set;

[0089] Then use the formula:

[0090]

[0091] Recalculate the center of the set Then, the grids are repeatedly allocated until the division does not change, that is, the minimum value of the following formula is obtained:

[0092]

[0093] And assign red, green and blue colors to the grids in the three sets.

[0094] Finally, each layer of the tight sandstone gas reservoir is divided into three categories: red, green, and blue, which are respectively for the reservoir with high remaining gas: higher reserve abundance, reservoir thickness, and slightly higher formation pressure; the reservoir with medium remaining gas: medium reserve abundance, reservoir thickness, and medium formation pressure; and the reservoir with low remaining gas: less reserve abundance, reservoir thickness, and slightly lower formation pressure. The results of clustering all grids by the main controlling factors are visualized as follows Figure 3 ( Figure 3 The distribution of residual gas types in each layer is obtained by using only the fine-grained K-means method. Although the residual gas type can be judged in detail, the fine-grained K-means method only attaches different color labels to each grid, and it is impossible to form area labels in some areas).

[0095] The results show that after K-means clustering of the main control factors of the grid, the remaining reserves of each layer are marked. The approximate distribution of the remaining gas in each layer can be seen. Among them, the first and second layers are mainly high-residual gas reservoirs, and the rest are a small number of low-residual gas reservoirs. The third layer has more high-residual gas reservoirs, and the remaining residual gas reservoirs are distributed to a certain extent. There are basically no high-residual gas reservoirs in the fourth and fifth layers, which are mainly medium-residual gas reservoirs, with a small number of low-residual gas reservoirs, and the sixth layer has very little residual gas.

[0096] On this basis, add the spatial coordinate parameters and use the formula:

[0097] N ε ={p i ∈W i |dist<ε}

[0098] Set the domain ε = 0.7m, count the number of grids of various colors in the grid domain, and if it is greater than or equal to 80% of the total number of grids in the domain, then assign it this color until all grids have completed the recursion. The result is as follows Figure 4 ( Figure 4 The dynamic fine-grained K-means recursive method is used, combined with the spatial coordinate parameters of each grid, to find that the residual gas type results of each layer are clearer, and the location of the residual gas type can be better determined in order to formulate relevant potential tapping plans).

[0099] After dynamic fine-grained K-means recursion, the results are obviously clearer and more effective, and the location of residual gas can be accurately determined. The first and second layers are still dominated by reservoirs with high residual gas, and the rest are a small number of reservoirs with low residual gas, which have great development potential and belong to areas that need to be developed. For the upper left area of ​​the first layer and the local areas of the second layer, there are more residual gas reserves, and more wells can be added, or for areas with large wells, this layer can be opened to improve the degree of exploitation. The third layer has more reservoirs with high residual gas and medium residual gas, and has great development potential. It belongs to the layer that can be developed. Wells can be added to the left middle and right middle positions of the third layer, and layers can be opened for other positions with more residual gas. The fourth and fifth layers are mostly medium-yield reservoirs, containing some low-yield reservoirs, with average development potential. Layers are mainly opened in places with medium residual gas, and there is no need to add wells. The sixth layer has very little residual gas, and there is basically no need to develop it. The effect after recursion can be seen more clearly through the local enlarged comparison charts in the third and fourth layers (the results of comparing the three positions separately are shown in the figure below). Figure 5 , compare the four positions separately and the results are as follows Figure 6 , Figure 5 and Figure 6 These are the local magnified enhanced comparison diagrams of the third and fourth layers, in which the more dispersed remaining gas is classified as a grid without reserves during recursion, and the grid is recursively classified according to the discrimination conditions). Finally, the average values ​​of each main controlling factor of the three types of remaining gas are obtained (see Table 2):

[0100] Table 2 Average values ​​of three types of residual gas

[0101]

[0102] The conclusion is drawn through the fine-grained K-means recursive method: the first and second layers have large remaining gas reserves and are the key reservoirs for development; the third layer has large remaining gas reserves and needs to be developed in some areas; the fourth and fifth layers have medium remaining gas reserves and have certain development value, and the sixth layer has no development value.

[0103] 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 method for the distribution of remaining gas types in tight sandstone reservoirs based on a dynamic fine-grained K-means recursive algorithm, the implementation steps of which are as follows: Step S1: Obtain the main controlling factors affecting the residual gas in each grid at time t of dense gas, and construct a grid feature data set; that is, given a grid data set A at time t i ={x i ;y i ; z i ;h i ;p fi ;Ω gi ; G i}The spatial parameter coordinate set is F i ={x i ;y i ; z i}, the set of main controlling factors is M i ={h i ;p fi ;Ω gi ; G i}, then A i ={F i ;M i}; Where i is the i-th grid (i=1,2,...,n); x i is the x-axis coordinate of the ith grid; y i is the y-axis coordinate of the i-th grid; z i is the z-axis coordinate of the ith grid; h i is the effective thickness of the ith grid (m) ;p fi is the formation pressure of the ith grid (MPa); Ω gi is the abundance of the ith grid (m 3 / m 2 );G i is the residual gas of the ith grid (10 8 m 3 ); Step S2: take the main controlling factors on all grids as input, and use K-means to cluster all grids by the main controlling factors; for a given data set M of the i-th grid i ={h i ;p fi ;Ω gi ; G i} is normalized to improve the quality and stability of the clustering results. The specific formula is as follows: In the formula, l i is the original data of a certain type of data in grid i; l i *is the normalized data of a certain type of data in grid i; l min is the minimum value of this type of data; l max It is the maximum value of this type of data; Step S3: Divide all grids into k sets Calculate the center of each set Solve the following formula for the goal: In the formula, It is the minimum value of the sum of the distances from all data to their respective cluster centers; Step S4: Calculate the distance between each grid and each cluster center, assign each grid to the cluster center closest to it, and assign the grid M to the cluster center closest to it. i With cluster center Distance, the specific formula is as follows: Step S5: Recalculate the cluster center of each cluster The specific formula is as follows: In the formula, is a cluster with p samples; Step S6: Repeat steps S4 to S5 until the partitioning stops changing or the maximum number of iterations is reached to obtain clusters. Different colors are assigned to grids in different clusters, denoted as Where i→j means that the i-th cluster corresponds to the j-th color; Step S7: Add coordinate set parameter F to the clusters obtained by clustering i ={x i ;y i ; z i Get a new data set Continue recursively on this basis; Step S8: i The categories within a certain distance of a grid in are counted, and the dataset W i Now there are n samples, each grid corresponds to a category The training data set W i It can be expressed as: p i =(x i ,y i ,z i ) The distance between any two points is: For any point p i Now, its ε neighborhood is defined as: N ε ={p i ∈W i |dist<e} In the formula, dist<ε means the distance is less than ε; Step S9: Data set W i The other grid categories in a grid area are counted, and the grid categories are reassigned according to whether the number of other categories in the grid area is ≥ 80%, until all grids are recursively completed; Step S10, dividing the residual gas into three categories: reservoirs with high residual gas, reservoirs with medium residual gas, and reservoirs with low residual gas, and finally obtaining the residual gas type distribution of each layer in the tight gas reservoir.

2. The method for the distribution of remaining gas types in tight sandstone reservoirs based on the dynamic fine-grained K-means recursive algorithm according to claim 1 is characterized in that: The main controlling factors include: permeability, porosity, depth, formation pressure, gas saturation, gas mass density, remaining gas volume, reserve abundance and effective thickness.

3. The method for the distribution of remaining gas types in tight sandstone reservoirs based on the dynamic fine-grained K-means recursive algorithm according to claim 2 is characterized in that: Two machine learning methods (XGBoost and Random Forest RF) and four statistical criteria were used for evaluation, and finally the importance ranking of factors affecting the remaining gas was obtained. Through comprehensive analysis, reserve abundance, effective thickness and formation pressure were selected as the main controlling factors.

4. The method for the distribution of remaining gas types in tight sandstone reservoirs based on the dynamic fine-grained K-means recursive algorithm according to claim 3 is characterized in that: The four statistical criteria include mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE) and coefficient of determination (R 2 ).

5. The method for distributing remaining gas types in tight sandstone reservoirs based on dynamic fine-grained K-means recursive algorithm according to claim 1, characterized in that: The grid categories in step S9 also include grids without residual gas, so as to ensure that the number of grids within a certain distance of each grid is the same.

6. The method for the distribution of remaining gas types in tight sandstone reservoirs based on the dynamic fine-grained K-means recursive algorithm according to claim 5, characterized in that: The dataset W i The dataset is randomly divided into two parts: 80% for training and 20% for testing.

7. The method for the distribution of remaining gas types in tight sandstone reservoirs based on the dynamic fine-grained K-means recursive algorithm according to claim 6 is characterized in that: 80% of the training set is used to train XGBoost and Random Forest RF and determine their optimal hyperparameter values, while 20% of the test set is used to evaluate the stability and generalization ability of the constructed model in predicting data.

Images