Method for improving one-time interpretation coincidence rate of oil reservoir porosity and permeability parameters

Abstract

The invention relates to the technical field of oil reservoir development, and discloses a method for improving the one-time interpretation coincidence rate of oil reservoir porosity and permeability parameters, which comprises the following steps: step 1, determining data distribution: collecting coring well data of a target block, and counting and determining numerical value intervals of porosity and permeability parameters of oil layers under different classifications; 2, determining a calculation object: combing factors with high correlation with pore permeability parameters in combination with a business thought, selecting an optimal division attribute from all features, and constructing a classification node; and step 3, determining a calculation method: after the classification is completed in the step 2, the grouping functions are not in a parallel relationship, but a plurality of grouping functions jointly determine the porosity and permeability parameters of a certain reservoir. According to a software program written through the method, the one-time interpretation coincidence rate of the porosity and permeability parameters reaches 85% or above, and the coincidence rate is increased by 65% compared with the coincidence rate of 20% before the method is applied. The workload of manual secondary interpretation is remarkably reduced, and the working efficiency is remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of oil reservoir development, in particular to a method for improving the one-time interpretation coincidence rate of oil reservoir porosity and permeability parameters. Background Art

[0002] The interpretation of porosity and permeability parameters is an important part of well logging interpretation. The main purpose is to give the porosity and permeability of the interpretation layer based on the well logging curve and some well logging interpretation results. The difficulty of interpretation lies in:

[0003] 1. Due to the differences and complexity of reservoir development in planes, between layers and within layers, any single set of porosity and permeability parameter calculation methods cannot be applied to all blocks and well networks. Therefore, the consistency rate of primary interpretation based on conventional software automation is low (generally around 20%), and a large amount of manual secondary interpretation is still required.

[0004] 2. The main basis of manual secondary interpretation is expert experience and core well data of the block where the interpretation object is located. Secondary interpretation can give more reasonable porosity and permeability parameters according to local conditions, but there are two disadvantages: First, the efficiency is low. The number of parameter interpretations for each well is 170-300, and each interpretation parameter needs to be given after comprehensive analysis of multiple logging curves, which generally takes about 1.5 to 3 hours (including the time for computer input after secondary interpretation); second, the difference is large. It is mainly reflected in the cognitive differences of expert experience among logging interpreters. Different people will give different interpretation results for the same well and the same interpretation layer. Although these numerical results can meet the standard of the falling zone compliance rate, the value of the porosity and permeability parameter of a certain section of rock must be unique objectively. The falling zone compliance rate is essentially an indicator to measure whether the interpretation result reaches the allowable error range. In actual work, we hope that this difference is as small as possible, and we also hope to obtain higher work efficiency. The most effective way is to improve the first interpretation compliance rate and reduce the proportion of manual secondary interpretation.

[0005] In summary, improving the consistency rate of the first interpretation can effectively reduce the disadvantages of manual secondary interpretation, but it also faces challenges and conventional methods do not work. Summary of the invention

[0006] In view of the deficiencies of the prior art, the present invention provides a method for improving the first-time interpretation compliance rate of reservoir porosity and permeability parameters, thereby solving the problems of poor versatility and low first-time interpretation compliance rate of conventional methods.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for improving the first interpretation compliance rate of reservoir porosity and permeability parameters, comprising the following steps:

[0008] Step 1: Determine data distribution: collect core well data from the target block and statistically determine the numerical range of porosity and permeability parameters of oil layers under different classifications;

[0009] Step 2: Determine the calculation object: Combine business ideas to sort out the factors with high correlation with porosity and permeability parameters, select the best partitioning attributes from all features, and construct classification nodes;

[0010] Step 3: Determine the calculation method: After completing the classification in step 2, each grouping function is not in a parallel relationship, but multiple grouping functions jointly determine the porosity and permeability parameters of a reservoir. When calculating, multiple grouping functions need to be calculated in stages through sequential iteration;

[0011] Step 4: Implementation of the calculation method: Write software based on a programming language, take each classification node as a grouping function, and define each of its sub-functions as a calculation module. Each calculation module consists of four parts: data screening, data aggregation, equation calculation, and data assignment.

[0012] Step 5. Improve the versatility of the software: Each block is configured with an independent configuration file, which contains several configuration parameters. The configuration parameters are determined by the numerical distribution of the target block determined in the first step. When the software calculates, it automatically reads the configuration parameters in the file, and the configuration parameters participate in the calculation in the form of variables.

[0013] Preferably, the construction of classification nodes in the step 2 is to regard each classification node as a grouping function. Then, from the perspective of reservoir category, it can be divided into three sub-functions: inside the table, one type outside the table, and two types outside the table; from the perspective of oil layer group, a sub-function is constructed with each oil layer group or oil layer group with similar physical properties as the object; from the perspective of layer thickness, it can be divided into four sub-functions of 0.2 meters, 0.3 meters, 0.4 meters and above 0.4 meters; from the perspective of flooding level, it can be divided into two sub-functions of high flooding and non-high flooding.

[0014] Preferably, in the step three, the permeability is calculated first, and then the porosity is derived based on the permeability result combined with the density and the acoustic wave curve.

[0015] Preferably, the data screening in step 4 is to locate the classified data governed by the current sub-function, and when calculating the permeability of the reservoir in the table, it is necessary to first screen out all the data rows containing the thickness in the table.

[0016] Preferably, the data aggregation in step 4 is to aggregate the qualified data rows according to the interpreted thickness. When aggregating, due to the different processing purposes of each calculation module, the statistical methods adopted for different logging curves are also different, and obtaining the maximum value, minimum value and average value is the most commonly used.

[0017] Preferably, the equation calculation in step 4 is to use the aggregated well logging curve data as calculation parameters to obtain the permeability or porosity value. In order to obtain a high-quality equation, the derivation of the equation is achieved through machine learning, and the learning samples are checked by business experts well by well.

[0018] Preferably, the data assignment in step 4 is to assign the calculation results to each data row that meets the conditions, aggregate them according to the interpretation thickness, and use the aggregated data as the independent variable during calculation to ensure that each data row under the interpretation thickness can obtain a unified porosity and permeability knot.

[0019] Preferably, in step 5, each block will form a personalized algorithm based on the core wells in the block, and the equation derivation of the sub-function will be subject to sample review by business experts.

[0020] The present invention provides a method for improving the first-time interpretation coincidence rate of reservoir porosity and permeability parameters.

[0021] Beneficial effects:

[0022] The software program written by the present invention using the method has a primary interpretation conformity rate of more than 85% for porosity and permeability parameters, which is 65 percentage points higher than the 20% conformity rate before the method is applied. The workload of manual secondary interpretation is significantly reduced, and the work efficiency is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram showing an example of the method for ladder calculation of the present invention;

[0024] Figure 2 A schematic diagram of the main technical route developed for the program of the present invention;

[0025] Figure 3 A schematic diagram of an application program written based on the method of the present invention;

[0026] Figure 4 A schematic diagram of the best embodiment of the explanation result of the present invention;

[0027] Figure 5 This is a schematic diagram of an embodiment of the present invention with relatively different interpretation results;

[0028] Figure 6 A schematic diagram of a code example for equation derivation of the present invention;

[0029] Figure 7 A schematic diagram of an example of a machine learning code for equation derivation of the present invention;

[0030] Figure 8 Schematic diagram of an example of the fitting results of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. 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.

[0032] Embodiment 1:

[0033] Please see attached Figure 1 -Attached Figure 8 The embodiment of the present invention provides a method for improving the first-time interpretation conformity rate of reservoir porosity and permeability parameters, comprising the following steps:

[0034] Step 1: Determine data distribution: collect core well data from the target block and statistically determine the numerical range of porosity and permeability parameters of oil layers under different classifications;

[0035] Step 2: Determine the calculation object: Combine business ideas to sort out the factors with high correlation with porosity and permeability parameters, select the best partitioning attributes from all features, and construct classification nodes;

[0036] Combined with business ideas, we sorted out the factors with high correlation with porosity and permeability parameters, selected the best partitioning attributes from all features, and constructed classification nodes. Considering each classification node as a grouping function, it can be divided into three sub-functions: inside the table, one type outside the table, and two types outside the table from the perspective of reservoir category; from the perspective of oil layer group, sub-functions are constructed for each oil layer group or oil layer group with similar physical properties; from the perspective of small layer thickness, it can be divided into four sub-functions: 0.2 meters, 0.3 meters, 0.4 meters, and above 0.4 meters; from the perspective of flooding level, it can be divided into two sub-functions: high flooding and non-high flooding. In addition, for old oil fields, after long-term development, the bottom of the thick oil layer is severely washed, and the porosity and permeability parameters are higher than those in the early stage of oil field development. Therefore, it is necessary to calculate the bottom of the thick oil layer and other parts separately, and at the same time, the influence of the rhythmicity of the thick oil layer on the porosity and permeability parameters must be considered. These all require the establishment of sub-functions.

[0037] Step 3: Determine the calculation method: After completing the classification in step 2, each grouping function is not in a parallel relationship, but multiple grouping functions jointly determine the porosity and permeability parameters of a reservoir. When calculating, multiple grouping functions need to be calculated in stages through sequential iteration;

[0038] Step 4: Implementation of the calculation method: Write software based on a programming language, take each classification node as a grouping function, and define each of its sub-functions as a calculation module. Each calculation module consists of four parts: data screening, data aggregation, equation calculation, and data assignment.

[0039] Step 5: Improve the versatility of the software: Each block is configured with an independent configuration file, which contains several configuration parameters. The configuration parameters are determined by the numerical distribution of the target block determined in the first step. When the software calculates, it automatically reads the configuration parameters in the file, and the configuration parameters participate in the calculation in the form of variables;

[0040] After the program development based on the data sample of a certain block is completed according to this method, the block is defined as the benchmark block. The parameters of the benchmark block will be calculated in the form of constants, and the corresponding parameters of the interpretation block will be calculated in the form of variables. The variables are obtained from the configuration file. Before interpreting each new well block, the core well data can be statistically analyzed in the form of Table 1, and then split into several parameters and written into the configuration file to form the configuration parameters of the interpretation block. After the parameters are ready, the implementation of the calculation method can be considered. The following is an example of calculating the permeability of the oil layer in the table:

[0041]

[0042] K——Permeability

[0043] E——Calculation equation for the permeability of the oil layer in the benchmark block based on machine learning fitting

[0044] mean a ——The average permeability of the oil layer in the benchmark block (variable, from the configuration file)

[0045] mean b ——Interpret the average permeability of the oil layer in the table of the block (constant, written directly by the program)

[0046] W——Oil layer adjustment coefficient in the table (variable, derived from the configuration file, default value 1).

[0047] The software is written based on programming language, with each classification node as a grouping function, and each of its sub-functions is defined as a calculation module. Each calculation module consists of four parts: data screening, data aggregation, equation calculation and data assignment. The main purpose of data screening is to locate the classification data under the jurisdiction of the current sub-function. For example, when calculating the permeability of the reservoir in the table, it is necessary to first screen out all the data rows containing the thickness in the table. The main purpose of data aggregation is to aggregate the qualified data rows according to the interpreted thickness. During aggregation, due to the different processing purposes of each calculation module, the statistical methods adopted for different well logging curves are also different. The maximum, minimum and average values ​​are the most commonly used. The main purpose of equation calculation is to use the aggregated well logging curve data as the calculation parameter to obtain the permeability or porosity value. In order to obtain high-quality equations, the derivation of the equation is realized through machine learning, and the learning samples are checked by business experts well by well. The main purpose of data assignment is to assign the calculation results to each qualified data row. Because the aggregation is performed according to the interpreted thickness, and the aggregated data is used as the independent variable during the calculation, it can ensure that each data row under the jurisdiction of the interpreted thickness can obtain a unified porosity and permeability result.

[0048] Embodiment 2:

[0049] refer to Figure 4 This is the result map of the N1-111-S2P22 well after a one-time interpretation based on this method. There are 175 interpreted thicknesses and 350 porosity and permeability parameters in the whole well, 16 of which were manually corrected. After calculation, it was found that the one-time interpretation conformity rate of the well reached 95%, which greatly reduced the workload of map modification and computer modification.

[0050] Embodiment three:

[0051] refer to Figure 5 , is the result map of the N1-121-S2P025 well after one-time interpretation based on this method. The whole well has 217 interpreted thicknesses and 434 porosity and permeability parameters, of which 63 were manually corrected. After calculation, it is concluded that the one-time interpretation conformity rate of this well is 85.5%. This well has the lowest one-time interpretation conformity rate among all the wells currently counted, but it can still reach more than 85%.

[0052] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for improving the first interpretation compliance rate of reservoir porosity and permeability parameters, characterized in that: The following steps are involved: Step 1: Determine data distribution: collect core well data from the target block and statistically determine the numerical range of porosity and permeability parameters of oil layers under different classifications; Step 2: Determine the calculation object: Combine business ideas to sort out the factors with high correlation with porosity and permeability parameters, select the best partitioning attributes from all features, and construct classification nodes; Step 3: Determine the calculation method: After completing the classification in step 2, each grouping function is not in a parallel relationship, but multiple grouping functions jointly determine the porosity and permeability parameters of a reservoir. When calculating, multiple grouping functions need to be calculated in stages through sequential iteration; Step 4: Implementation of the calculation method: Write software based on a programming language, take each classification node as a grouping function, and define each of its sub-functions as a calculation module. Each calculation module consists of four parts: data screening, data aggregation, equation calculation, and data assignment. Step 5. Improve the versatility of the software: Each block is configured with an independent configuration file, which contains several configuration parameters. The configuration parameters are determined by the numerical distribution of the target block determined in the first step. When the software calculates, it automatically reads the configuration parameters in the file, and the configuration parameters participate in the calculation in the form of variables.

2. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: The construction of classification nodes in the step 2 is to regard each classification node as a grouping function. From the perspective of reservoir category, it can be divided into three sub-functions: inside the table, one type outside the table, and two types outside the table; from the perspective of oil layer group, a sub-function is constructed with each oil layer group or oil layer group with similar physical properties as the object; from the perspective of small layer thickness, it can be divided into four sub-functions of 0.2 meters, 0.3 meters, 0.4 meters and above 0.4 meters; from the perspective of flooding level, it can be divided into two sub-functions of high flooding and non-high flooding.

3. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: In the calculation of step three, the permeability is calculated first, and then the porosity is derived based on the permeability result combined with the density and the acoustic wave curve.

4. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: The data screening in step 4 is to locate the classified data under the jurisdiction of the current sub-function. When calculating the permeability of the reservoir in the table, it is necessary to first screen out all the data rows containing the thickness in the table.

5. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: The data aggregation in step 4 is to aggregate the qualified data rows according to the interpreted thickness. When aggregating, due to the different processing purposes of each calculation module, the statistical methods adopted for different logging curves are also different. The maximum value, minimum value and average value are most commonly used.

6. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: The equation calculation in step 4 is to use the aggregated well logging curve data as calculation parameters to obtain the permeability or porosity value. In order to obtain high-quality equations, the derivation of the equations is achieved through machine learning, and the learning samples are checked by business experts well by well.

7. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: The data assignment in step 4 is to assign the calculation results to each data row that meets the conditions, aggregate them according to the interpreted thickness, and use the aggregated data as the independent variable during calculation to ensure that each data row under the jurisdiction of the interpreted thickness can obtain a unified porosity and permeability structure.

8. A method for improving the first interpretation conformity rate of reservoir porosity and permeability parameters according to claim 1, characterized in that: In the step 5, each block will form a personalized algorithm based on the core wells in the block, and the equation derivation of the sub-function will be checked by business experts for samples.

Images