Rapid assessment method and system for dynamic reserves of fractured-vuggy reservoir

Abstract

The invention provides a rapid assessment method and system for dynamic reserves of a fracture-vuggy reservoir. The method comprises the following steps: selecting all lower-layer characteristic factors possibly related to the dynamic reserves by considering different characteristic types on the basis of selected well groups which are put into production and have been produced for set time; performing clustering analysis on each well of the selected well group by taking each feature type as a unit on the basis of a lower-layer feature factor, and selecting a target feature type related to the dynamic reserves; performing clustering processing on each well according to the lower-layer feature factors of the target feature type to obtain different classified well groups, further selecting related feature factors, selecting master control feature parameters according to partial correlation coefficients, and training dynamic reserve assessment models corresponding to the reserve classified well groups based on the master control feature parameters to quickly assess dynamic reserves; by adopting the scheme, the defects of oil reservoir dynamic energy storage assessment lag and insufficient reliability in the prior art are overcome, the dynamic reserves are rapidly assessed according to oil reservoir geologic features in the early stage of production, and important support is provided for production strategy decision.

Description

Technical Field

[0001] The invention relates to the technical field of reliability testing and evaluation, and in particular to a method and system for quickly evaluating the dynamic reserves of a fracture-cavity oil reservoir. Background Art

[0002] Fracture-cave oil reservoirs are significantly different from conventional sandstone clastic rock oil reservoirs. Fracture-cave oil reservoirs are mainly developed along fault zones and nearby fracture zones. After multiple periods of karst water infiltration along faults or local hydrothermal upwelling, the main reservoir storage space is formed by dissolution. The unique geological characteristics lead to the obvious characteristic of small reserves controlled by single wells. The production dynamics of the reservoir in the later stage are not suitable for the production strategy adopted in the early stage; and the decision on the production strategy mainly depends on the reserves controlled by a single well in the reservoir.

[0003] The reserves controlled by oil wells can be calculated by two methods: geological static parameters and production dynamic data. The reserves calculated by geological static parameters are called static reserves, and the reserves calculated by production dynamic data are called dynamic reserves. Dynamic reserve calculation is usually not possible in the early stage of production and can only be realized in the middle and late stages of production. For conventional sandstone reservoirs, the ratio between static reserves and dynamic reserves is roughly about 1 / 3 to 2 / 3, and it presents a regional distribution, which is relatively easy to estimate. Therefore, the production strategy of conventional sandstone reservoirs can be roughly estimated and formulated based on the results of static reserve calculation. However, the reservoir space of fracture-cavity reservoirs is complex, the relationship between dynamic reserves and static reserves varies greatly, and there are obvious differences between single wells, which cannot be directly estimated. It is even more difficult to clarify the scale of dynamic reserves in the early stage of production. The production strategy formulated in this way is difficult to achieve efficient development of the reservoir and provide support for the later production strategy decision. If the dynamic reserve scale of a new well (not yet put into production) can be quickly evaluated based on the geological dynamic data of an existing well (which has been put into production for a certain period of time), it will be of great significance to the development strategy of the block.

[0004] The information disclosed in the background technology section of the present invention is only intended to deepen the understanding of the general background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs, which is to cluster the selected production well groups into multiple categories of production well groups based on dynamic reserves, respectively determine whether the lower layer characteristic factors under the reservoir characteristic type are meaningful to each type of production well group, calculate the partial correlation coefficient of the dynamic reserves of the corresponding category of production well groups, select the main control factors with significant correlation and the specific data values ​​of the dynamic reserves as samples for neural network training, and perform neural network training to calculate the dynamic reserves of the remaining well groups in the oil reservoir; using this method, new wells in fracture-cavity oil reservoirs can be classified according to geological characteristics and engineering characteristics, and the dynamic reserve values ​​can be obtained, and the assessment is efficient and accurate. Preferably, in one embodiment, the method includes:

[0006] Analyze the characteristic decision step, select the well group that has been put into production at a set scale and has been in production for a set time from the overall oil reservoir to be evaluated as the analysis production well group, and select all the lower layer characteristic factors that may be related to dynamic reserves based on the analysis production well group and considering different characteristic types;

[0007] Characteristic data acquisition step, obtaining data of each underlying characteristic factor by analyzing geological analysis files and well test report data of each well in the production well group;

[0008] The relevant feature type decision step is to perform cluster analysis on each well in the analysis production well group based on its underlying feature factors, and select the feature type whose correlation with the dynamic reserves meets the set requirements as the target feature type according to the clustering result;

[0009] Well group classification and factor selection steps, clustering the wells of the analyzed production well group according to the lower layer characteristic factors of the target characteristic type and the dynamic reserve data to obtain different reserve classification well groups, and then analyzing the distribution characteristics of all the lower layer characteristic factors under the target characteristic type around each reserve classification well group, and selecting the lower layer characteristic factors that meet the set requirements as relevant characteristic factors;

[0010] The main control parameter determination step is to further calculate the partial correlation coefficient of each relevant characteristic factor relative to the dynamic reserves of each well in the energy storage classification well group to which it belongs, and select the main control characteristic parameter according to the partial correlation coefficient;

[0011] The evaluation model establishment step is to use the main control characteristic parameters and the corresponding dynamic reserve data as samples for well groups with different reserve classifications, and to implement the training of the neural network model based on the set neural network configuration to obtain the dynamic reserve evaluation model corresponding to each well group with different reserve classifications;

[0012] The reserve assessment application steps are as follows: for the unproduced new wells to be assessed, the data of each underlying characteristic factor is obtained based on the geological analysis files and well test report materials of each well in the production well group, and then the data is substituted into the dynamic reserve assessment model to determine the corresponding dynamic energy storage data.

[0013] Preferably, in one embodiment, in the feature analysis decision step, the feature types include geological feature factors, engineering feature factors and production feature factors; lower layer feature factors related to dynamic reserves are set respectively for the geological feature factors, engineering feature factors and production feature factors.

[0014] In an optional embodiment, each well in the analyzed production well group is a production well in the middle of production and later, and in the relevant feature type decision step, the dynamic reserve data of each well is first obtained from the production data.

[0015] Furthermore, in one embodiment, in the relevant feature type decision step, a hierarchical clustering method is used to perform cluster analysis on the lower-layer feature factors and dynamic reserves under different feature types, to determine whether the clustering results of the lower-layer feature factors and dynamic reserves of each well have a set distribution law. If the clustering results of all lower-layer feature factors and dynamic reservoirs do not meet the set distribution law, it indicates that the current feature type has no correlation with the dynamic reserves and is not used as a target feature type.

[0016] In a preferred embodiment, in the well group classification and factor selection step, the process of obtaining different reserve classification well groups through clustering processing includes:

[0017] After integrating the underlying characteristic factors under the target characteristic type, all underlying characteristic factors are used to cluster the analyzed well groups until each well is classified into multiple reserve classification well groups with different levels of dynamic reserves.

[0018] Furthermore, in one embodiment, in the well group classification and factor selection step, the process of selecting relevant characteristic factors includes:

[0019] For each level of reserve classification well group, the phase difference amplitude of each lower layer characteristic factor under the target characteristic type is calculated respectively to determine whether the phase difference amplitude meets the set requirements. If the set requirements are met, it is confirmed that the lower layer characteristic factor is correlated with the reserve classification well group.

[0020] Optionally, in one embodiment, before calculating the phase difference amplitude of each lower layer characteristic factor under the target characteristic type, it also includes: using the wire box method to eliminate abnormal data points of various lower layer characteristic factors of well groups with different reserve classifications.

[0021] Furthermore, in an optional embodiment, the phase difference amplitude R of each lower-layer feature factor under the target feature type is calculated according to the following formula:

[0022]

[0023] Among them, y 最大值 is the specific value of the larger value of the two sets of reservoir lower layer characteristic factors being compared, y 最小值 It is the specific value of the smaller number among the two groups of reservoir lower layer characteristic factors being compared.

[0024] Based on other aspects of the method described in any one or more of the above embodiments, the present invention further provides a storage medium storing program codes that can implement the method described in any one or more of the above embodiments.

[0025] Based on the application aspects of the method described in any one or more of the above embodiments, the present invention also provides a rapid assessment system for dynamic reserves of fracture-vuggy oil reservoirs, which executes the method described in any one or more of the above embodiments.

[0026] Compared with the closest prior art, the present invention also has the following beneficial effects:

[0027] The present invention provides a method and system for quickly evaluating the dynamic reserves of fracture-cavity oil reservoirs. The method selects all lower-layer characteristic factors that may be related to the dynamic reserves based on the selected production analysis well group that meets the set conditions and takes different characteristic types into consideration; cluster analysis is performed on each well in the production analysis well group based on its lower-layer characteristic factors, and a target characteristic type whose correlation with the dynamic reserves meets the set requirements is selected; cluster processing is performed on each well according to the lower-layer characteristic factors of the target characteristic type to obtain different reserve classification well groups, and then relevant characteristic factors are selected around each reserve classification well group; further, main control characteristic parameters are selected according to partial correlation coefficients, and dynamic reserve evaluation models corresponding to each reserve classification well group are trained based on the main control characteristic parameters, so that the dynamic reserves of other well groups of new wells that have not been put into production can be quickly predicted; by adopting the scheme, the dynamic reserves can be quickly evaluated according to the geological characteristics of the oil reservoir in the early stage of production, the production fluctuation in the production characteristics will basically not affect the stability of the dynamic reserve calculation, the operation flexibility is large, the calculation accuracy of the dynamic reserves is high, and it is suitable for oil reservoirs of various types and various regions, which is of great significance to the production strategy decision.

[0028] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0030] Figure 1 It is a schematic flow chart of a method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs provided in one embodiment of the present invention;

[0031] Figure 2 A pedigree diagram for clustering production well groups using geological characteristic factors in a rapid assessment method for dynamic reserves of fracture-cavity oil reservoirs provided in an embodiment of the present invention;

[0032] Figure 3 A pedigree diagram for clustering production well groups using engineering characteristic factors in a rapid assessment method for dynamic reserves of fracture-cavity oil reservoirs provided in an embodiment of the present invention;

[0033] Figure 4 A pedigree diagram for clustering production well groups using production characteristic factors in a rapid assessment method for dynamic reserves of fracture-cavity oil reservoirs provided in an embodiment of the present invention;

[0034] Figure 5 A cluster geological zoning characteristic map of the Shunbei Oilfield Area 1 in the rapid assessment method for dynamic reserves of fracture-cavity oil reservoirs provided in an embodiment of the present invention;

[0035] Figure 6 A neural network structure of the first type of wells in a method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs provided in an embodiment of the present invention;

[0036] Figure 7 A neural network structure of the second type of wells in a method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs provided by another embodiment of the present invention;

[0037] Figure 8 The neural network structure of the third type of wells in the rapid assessment method of dynamic reserves of fracture-cavity oil reservoirs provided in the embodiment of the present invention;

[0038] Fig. 9 It is a schematic diagram of the structure of a rapid assessment system for dynamic reserves of fracture-cavity oil reservoirs provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will describe the implementation methods of the present invention in detail in conjunction with the accompanying drawings and embodiments, so that the implementers of the present invention can fully understand how the present invention applies technical means to solve technical problems and achieve the implementation process of technical effects and implement the present invention specifically according to the above implementation process. It should be noted that as long as there is no conflict, the various embodiments and various features of the embodiments in the present invention can be combined with each other, and the technical solutions formed are all within the protection scope of the present invention.

[0040] Although the flowcharts describe the operations as sequential processes, many of the operations may be performed in parallel, concurrently, or simultaneously. The order of the operations may be rearranged. A process may be terminated when its operations are completed, but may also have additional steps not included in the accompanying drawings. A process may correspond to a method, function, procedure, subroutine, subprogram, etc.

[0041] Computer devices include user devices and network devices. Among them, user devices or clients include but are not limited to computers, smart phones, PDAs, etc.; network devices include but are not limited to a single network server, a server group consisting of multiple network servers, or a cloud composed of a large number of computers or network servers based on cloud computing. Computer devices can be operated alone to implement the present invention, or they can be connected to the network and implement the present invention through interactive operations with other computer devices in the network. The network where the computer device is located includes but is not limited to the Internet, wide area network, metropolitan area network, local area network, VPN network, etc.

[0042] The terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms "one", "one" and "item" used herein are also intended to include plural numbers. It should also be understood that the terms "include" and / or "comprise" used herein specify the existence of stated features, integers, steps, operations, units and / or components, without excluding the existence or addition of one or more other features, integers, steps, operations, units, components and / or combinations thereof.

[0043] Fracture-cave oil reservoirs are significantly different from conventional sandstone clastic rock oil reservoirs. Fracture-cave oil reservoirs are mainly developed along fault zones and nearby fracture zones. After multiple periods of karst water infiltration along faults or local hydrothermal upwelling, the main reservoir storage space is formed by dissolution. The unique geological characteristics lead to the obvious characteristic of small reserves controlled by single wells. The production dynamics of the reservoir in the later stage are not suitable for the production strategy adopted in the early stage; and the decision on the production strategy mainly depends on the reserves controlled by a single well in the reservoir.

[0044] The reserves controlled by oil wells can be calculated by two methods: geological static parameters and production dynamic data. The reserves calculated by geological static parameters are called static reserves, and the reserves calculated by production dynamic data are called dynamic reserves. Dynamic reserve calculation usually cannot be performed in the early stages of production and needs to be achieved in the middle and late stages of production.

[0045] For conventional sandstone reservoirs, the ratio between static reserves and dynamic reserves is roughly 1 / 3 to 2 / 3, and presents a regional distribution, which is relatively easy to estimate. Therefore, the production strategy of conventional sandstone reservoirs can be roughly estimated and formulated based on the static reserve calculation results. However, the reservoir space of fracture-cavity reservoirs is complex, the relationship between dynamic reserves and static reserves varies greatly, and there are obvious differences between single wells, which cannot be directly estimated, and it is even more difficult to determine the scale of dynamic reserves in the early stage of production. Therefore, the production strategy formulated is difficult to achieve efficient development of the reservoir.

[0046] If the dynamic reserves controlled by the oil wells in the reservoir can be estimated in the early stage of production based on the geological characteristics of the reservoir and the results of static reserve calculation, it will be of great significance for the subsequent production strategy decision. Therefore, a solution is urgently needed to scientifically evaluate the dynamic reserves of fracture-cavity reservoirs in the early stage of production.

[0047] To solve the above problems, the present invention provides a method and system for quickly evaluating the dynamic reserves of fracture-cavity oil reservoirs. By selecting some production well groups in the entire oil reservoir, clustering analysis is performed on the production well groups separately until the production well groups are clustered into multiple production well groups based on dynamic reserves, and whether the lower layer characteristic factors under the oil reservoir characteristic type are meaningful to each type of production well group is determined respectively, and the partial correlation coefficient of the dynamic reserves of the corresponding type of production well group is calculated, and it is determined whether there is a significant correlation between the lower layer characteristic factors of the oil reservoir and the dynamic reserves of the production well group. The significantly correlated main controlling factors and the specific data values ​​of the dynamic reserves are used as samples for neural network training to perform neural network training, which is used to calculate the dynamic reserves of the remaining well groups in the oil reservoir. The dynamic reserve estimation method of the present invention can classify new wells in fracture-cavity oil reservoirs according to geological characteristics and engineering characteristics, and obtain the numerical value of dynamic reserves, so that the evaluation is efficient and accurate.

[0048] Next, the detailed process of the method of the embodiment of the present invention is described in detail based on the accompanying drawings. The steps shown in the flowchart of the accompanying drawings can be executed in a computer system including a set of computer executable instructions. Although the logical order of each step is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0049] Embodiment 1

[0050] Figure 1The schematic diagram of the process of the rapid evaluation method of the dynamic reserves of the fracture-cavity oil reservoir provided in the first embodiment of the present invention is shown. Figure 1 It can be seen that the method includes the following steps.

[0051] Analyze the characteristic decision step, select the well group that has been put into production at a set scale and has been in production for a set time from the overall oil reservoir to be evaluated as the analysis production well group, and select all the lower layer characteristic factors that may be related to dynamic reserves based on the analysis production well group and considering different characteristic types;

[0052] Characteristic data acquisition step, obtaining data of each underlying characteristic factor by analyzing geological analysis files and well test report data of each well in the production well group;

[0053] The relevant feature type decision step is to perform cluster analysis on each well in the analysis production well group based on its underlying feature factors, and select the feature type whose correlation with the dynamic reserves meets the set requirements as the target feature type according to the clustering result;

[0054] Well group classification and factor selection steps, clustering the wells of the analyzed production well group according to the lower layer characteristic factors of the target characteristic type and the dynamic reserve data to obtain different reserve classification well groups, and then analyzing the distribution characteristics of all the lower layer characteristic factors under the target characteristic type around each reserve classification well group, and selecting the lower layer characteristic factors that meet the set requirements as relevant characteristic factors;

[0055] The main control parameter determination step is to further calculate the partial correlation coefficient of each relevant characteristic factor relative to the dynamic reserves of each well in the energy storage classification well group to which it belongs, and select the main control characteristic parameter according to the partial correlation coefficient;

[0056] The evaluation model establishment step is to use the main control characteristic parameters and the corresponding dynamic reserve data as samples for well groups with different reserve classifications, and to implement the training of the neural network model based on the set neural network configuration to obtain the dynamic reserve evaluation model corresponding to each well group with different reserve classifications;

[0057] The reserve assessment application steps are as follows: for the unproduced new wells to be assessed, the data of each underlying characteristic factor is obtained based on the geological analysis files and well test report materials of each well in the production well group, and then the data is substituted into the dynamic reserve assessment model to determine the corresponding dynamic energy storage data.

[0058] In actual application, in a preferred embodiment, in the analysis feature decision step, a part of the production well group in the entire reservoir is selected as the analysis production well group, and all major reservoir characteristic factors related to dynamic reserves of this part of the production well group are counted, that is, the characteristic type, wherein the reservoir characteristic type includes geological characteristic factors, engineering characteristic factors and production characteristic factors, and according to the actual situation of the reservoir, the reservoir sub-category characteristic factors related to dynamic reserves are further determined in the reservoir characteristic type, that is, the lower layer characteristic factors; the selected reservoir lower layer characteristic factors are characteristic factors that may be related to dynamic reserves. For well groups under different conditions, targeted selection can be made according to the actual situation. In actual application, when selecting the analysis production well group, all existing wells can be selected in the target reservoir, and the data of the existing wells can be used to analyze and predict the new wells based on this.

[0059] Among them, in an optional embodiment, in the feature analysis decision step, the feature types include geological feature factors, engineering feature factors and production feature factors; the lower layer feature factors related to dynamic reserves are determined respectively for the geological feature factors, engineering feature factors and production feature factors.

[0060] Furthermore, the data of each lower layer characteristic factor is obtained through the characteristic data acquisition step, specifically, the data of each lower layer characteristic factor is obtained based on analyzing the geological analysis files and well test report data of each well in the production well group.

[0061] In addition, each well in the analyzed production well group is a production well in the middle production stage or later, and the dynamic reserve data of each well is first obtained from the production data.

[0062] Next, the relevant feature type decision step is executed, and each well in the analyzed production well group is clustered based on its underlying characteristic factors, and the feature type whose correlation with dynamic reserves meets the set requirements is selected as the target feature type according to the clustering results.

[0063] In a preferred embodiment, in the relevant feature type decision step, a hierarchical clustering method is used to perform cluster analysis on the lower layer feature factors and dynamic reserves under different feature types, and determine whether the clustering results of the lower layer feature factors and dynamic reserves of each well have a set distribution law. If the clustering results of all lower layer feature factors and dynamic reservoirs do not meet the set distribution law, it indicates that the current feature type has no correlation with the dynamic reserves and is not used as a target feature type.

[0064] Furthermore, the well group classification and factor selection steps are executed. First, the wells of the analyzed production well group are clustered according to the lower-layer characteristic factors of the target characteristic type and the dynamic reserve data to obtain different reserve classification well groups. Then, the distribution characteristics of all the lower-layer characteristic factors under the target characteristic type are analyzed around each reserve classification well group, and the lower-layer characteristic factors whose distribution characteristics meet the set requirements are selected as relevant characteristic factors.

[0065] For well groups with more complex geological conditions, the dynamic reserve distribution of well locations is often not uniform, and it is difficult to obtain a more regular distribution of results by directly calculating the relationship between the dynamic reserves and reservoir characteristic factors. Therefore, it is necessary to classify the dynamic reserves and calculate them separately. The basis for well group classification is to examine the correlation between the reservoir characteristic type and the well group, and to classify the dynamic reserves using the reservoir characteristic type with strong correlation.

[0066] Therefore, in one embodiment, in the well group classification and factor selection step, clustering processing is first used to obtain different reserve classification well groups, and the process includes:

[0067] After integrating the underlying characteristic factors under the target characteristic type, all underlying characteristic factors are used to cluster the analyzed well groups until each well is classified into multiple reserve classification well groups with different levels of dynamic reserves.

[0068] In actual application, the reservoir characteristic types are clustered and analyzed separately for the production well groups to determine whether there is a correlation between the clustering results and the dynamic reserves of the production well groups. After integrating the correlated reservoir characteristic types, the production well groups are clustered again until the production well groups are clustered into multiple types based on dynamic reserves.

[0069] In the well group classification and factor selection steps, the process of selecting relevant characteristic factors includes:

[0070] For each level of reserve classification well group, the phase difference amplitude of the lower layer characteristic factors of different wells under the reserve classification is calculated respectively to determine whether the phase difference amplitude meets the set requirements, and the distribution of wells in the reserve classification well group is analyzed. If the phase difference amplitude meets the set requirements and the number of wells whose distribution meets the requirements reaches the set level, it is confirmed that the lower layer characteristic factor is correlated with the reserve classification well group.

[0071] When analyzing the well group data of the reservoir lower layer characteristic factors, some well groups may have irregular abnormal points with a data difference of 5% to 10% compared with the majority of well groups, which need to be manually removed.

[0072] Therefore, before calculating the phase difference amplitude of each lower layer characteristic factor under the target characteristic type, it also includes: using the wire box method to eliminate abnormal data points of various lower layer characteristic factors of well groups with different reserve classifications.

[0073] In this embodiment, it is determined whether the reservoir lower layer characteristic factors in the geological characteristic factors and the production characteristic factors are meaningful to each type of production well group obtained by clustering. The judgment method is: for the reservoir lower layer characteristic factors in the production characteristic factors, a type of well group obtained by clustering is selected. When the difference in the data values ​​of the reservoir lower layer characteristic factors of any two groups of wells in this type of well group after removing abnormal values ​​is within 30%, it is determined that the reservoir lower layer characteristic factors are meaningful to this type of production well group. In a preferred embodiment, the difference amplitude R of each lower layer characteristic factor under the target characteristic type is calculated according to the following formula:

[0074]

[0075] The maximum value of y is the specific value of the larger value of the two sets of reservoir lower layer characteristic factors being compared, and the minimum value of y is the specific value of the smaller value of the two sets of reservoir lower layer characteristic factors being compared.

[0076] Optionally, for the reservoir sub-category characteristics in the geological characteristic factors, a type of well group is randomly clustered, and when 80% of the wells in the well group are in a mutually adjacent state, the reservoir sub-category characteristics in the geological characteristic factors are considered to be meaningful for this type of production well group and are relevant characteristic factors.

[0077] By adopting the means of this embodiment, the production well group obtained after clustering based on the selected relevant characteristic factors is not only in a relatively concentrated specific range in geological space, but also in a state of approximate range between the production dynamics of the characteristic factors of the lower layer of the reservoir in each group of wells, and needs to be classified and studied to more accurately reflect the connection between the characteristic factors of the lower layer of the reservoir and the production well group. Therefore, the well groups with a difference amplitude of less than 30% under any factor are classified into one category, and a connection is established with the factor and further judged by the distribution of the number of adjacent wells in the clustered well group. When the difference amplitude of the clustered data corresponding to the current lower layer characteristic factor meets the requirement of less than or equal to 30% and the number of adjacent wells distributed in the corresponding clustered well group meets the requirement of more than 80%, the current lower layer characteristic factor is determined to be a relevant characteristic factor with significance relative to the dynamic reserves of the reservoir.

[0078] It is worth noting that the engineering characteristic factors in the reservoir characteristic type are only reflected in the production dynamics. Therefore, for any oil and gas reservoir, it should be considered that its engineering characteristic factors are meaningful to the well group.

[0079] Further, through the main control parameter determination step, for each relevant characteristic factor, the partial correlation coefficient relative to the dynamic reserves of each well in the energy storage classification well group to which it belongs is further calculated, and the main control characteristic factor is selected, among which the relevant characteristic factor whose partial correlation coefficient meets the set conditions is selected as the main control characteristic parameter.

[0080] The identified reservoir lower layer characteristic factors that are significant to the corresponding category of production well groups are used to calculate the Pearson partial correlation coefficients of their dynamic reserves relative to the corresponding category of production well groups.

[0081] Existing studies have shown that the reservoir lower layer characteristic factors that affect the dynamic reserves of fracture-vuggy reservoirs are almost not independent of each other. For example, there is a certain relationship between the common drilling loss and the mechanism of fracture formation. Therefore, in the case of two or more factors being correlated, it is difficult to accurately obtain the relationship between each influencing factor and the dynamic reserves. Therefore, it is necessary to further use the partial correlation analysis method to control other variables that may affect the correlation between the two variables, that is, to study the correlation between the two variables after eliminating the interference of other variables.

[0082] Similar to simple correlation analysis, partial correlation analysis also uses statistical indicators to study the correlation between variables; the partial correlation coefficient is used in partial correlation analysis.

[0083] In an optional embodiment, in the main control parameter determination step, the process of calculating the partial correlation coefficient of the relevant characteristic factors relative to the dynamic reserves of each well in the energy storage classification well group to which they belong can adopt the following ideas:

[0084] Assuming there are three variables x1, x2, and x3, after eliminating the influence of variable x3, the partial correlation coefficient between x1 and x2 is:

[0085]

[0086] Where r ij Represents variable x i With x j The simple correlation coefficient between them. As can be seen from the formula, the partial correlation coefficient is determined by the simple correlation coefficient. The partial correlation coefficient and the simple correlation coefficient are often different. When calculating the simple correlation coefficient, all other variables are not considered, while the partial correlation coefficient treats other variables as constants.

[0087] Suppose a variable x4 is added, then the second-order partial correlation coefficient between x1 and x2 is:

[0088]

[0089] In general, assuming there are p variables, the p-2 order partial correlation coefficient between x1 and x2 is:

[0090]

[0091] The meaning and significance test of the partial correlation coefficient are similar to those of the simple correlation coefficient and will not be elaborated here.

[0092] In actual application, in an optional embodiment, when the relevant characteristic factors whose partial correlation coefficients meet the set conditions are selected as the main controlling characteristic parameters, the absolute value of the obtained Pearson partial correlation coefficient can be calculated, and the lower layer characteristic factors of the reservoir corresponding to the absolute value greater than the correlation judgment value are identified as significantly correlated with the dynamic reserves of the corresponding category of production well groups, and are used as the main controlling characteristic factors.

[0093] The Pearson correlation coefficient takes values ​​between -1 and 1, and it describes the direction and degree of linear correlation between two variables: r>0 means that the two variables are positively correlated; r<0 means that the two variables are negatively correlated; r=1 means that the two variables are completely correlated; r=0 means that there is no linear correlation between the two variables. Moreover, the closer |r| is to 1, the higher the degree of linear correlation between the two variables, and the closer it is to 0, the weaker the degree of linear correlation. Therefore, when the absolute value of the Pearson partial correlation coefficient between a certain reservoir lower layer characteristic factor and dynamic reserves is greater than the correlation judgment value, it means that |r| is closer to 1, proving that there is a higher degree of linear correlation between the two, that is, the two are significantly correlated.

[0094] The evaluation model establishment step is to respectively use the main control characteristic parameters and the corresponding dynamic reserve data as samples for well groups with different reserve classifications, implement the training of the neural network model based on the set neural network configuration, and obtain the dynamic reserve evaluation model;

[0095] Specifically, the lower layer characteristic factors that are identified as significantly correlated are taken as the main controlling factors, and the specific data values ​​of the main controlling factors and dynamic reserves of all production wells in the corresponding category of production well groups are taken as samples for neural network training to establish the relationship between the main controlling factors and dynamic reserves and form a dynamic reserve assessment model. In actual application, based on the relationship between the main controlling factors and dynamic reserves, the dynamic reserves of other well groups in the reservoir can be calculated, and then their classification in terms of dynamic reserves can be determined.

[0096] The reservoir structure of fracture-cavity oil fields is complex, the fluid components of the reservoirs are complex, the flow conditions are multi-scale coupled, the relationship between the factors affecting production capacity and their impact on dynamic reserves are unclear and extremely complex. It is difficult to obtain the impact of various factors on dynamic reserves using traditional reservoir engineering. Therefore, the present invention adopts a neural network learning method to establish the relationship between various main controlling factors affecting the well group and the dynamic reserves of the well group.

[0097] When performing dynamic reserve assessment, by executing the reserve assessment application steps, for other well groups to be assessed, after obtaining data on each underlying characteristic factor based on the geological analysis files and well test report materials of each well in the production well group, the data are substituted into the dynamic reserve assessment model to determine the corresponding dynamic energy storage data.

[0098] By using the rapid assessment method for dynamic reserves of fracture-cavity oil reservoirs according to the embodiment of the present invention, new wells in fracture-cavity oil reservoirs can be classified according to geological characteristics and engineering characteristics, and the value of dynamic reserves can be calculated using the neural network calculation method of each category. The calculation speed is high, the processing efficiency is fast, the output fluctuation in the production characteristics will basically not affect the stability of the dynamic reserve calculation, the operation flexibility is large, the calculation accuracy of the dynamic reserves is high, and it is basically consistent with the results calculated based on long-term dynamic data, and the calculation accuracy is close to 100%.

[0099] The present invention is further described below in conjunction with examples of implementation. The scope of the present invention is not limited by the examples, but is set forth in the claims.

[0100] Taking the evaluation process of a well group in Shunbei Oilfield as an example, due to the complex geological conditions of Shunbei Oilfield, the types and dynamic reserves obtained simply through geological and well location classification are not uniform;

[0101] Firstly, some production well groups in the whole reservoir are selected, and all reservoir characteristic types related to dynamic reserves of these production well groups are counted. Then, the reservoir lower layer characteristic factors (sub-category characteristic factors) related to dynamic reserves are further determined in the reservoir characteristic types.

[0102] Therefore, in this embodiment, the fault distance, reflection type, stress mechanism, fracture-cavity combination, and reservoir development change of the well group are selected as the reservoir lower layer characteristic factors among the geological characteristic factors, as shown in Table 1;

[0103] The depth of the well group into the mountain, drilling loss, production increase method, and production increase scale are selected as the reservoir lower layer characteristic factors in the engineering characteristic factors, as shown in Table 2.

[0104] The average oil production, average gas-oil ratio, production decline rate, energy utilization rate, and energy decline rate are selected as the reservoir lower layer characteristic factors in the production characteristic factors, as shown in Table 3.

[0105] The data of the above reservoir lower layer characteristic factors are all derived from the actual results of geological analysis + well test report. All factors that may affect the dynamic reserves are selected, and for well groups under different conditions, targeted selection can be made according to actual conditions.

[0106] Table 1 Statistics of geological characteristics of oil wells in fault-karst reservoirs in Shunbei Oilfield

[0107]

[0108]

[0109] In Table 1 , the numbers in reflection type, stress mechanism, fracture-vuggy combination, and reservoir development change represent: weak compression = 1, translation = 2, strong compression = 3, and pull-apart = 4, respectively.

[0110] Table 2 Statistics of oil well engineering characteristics of fault-karst reservoirs in Shunbei Oilfield Area 1

[0111]

[0112]

[0113] In Table 2 , the numbers in reflection type, stress mechanism, fracture-vuggy combination, and reservoir development change represent: chaotic = 1, beaded = 2, chaotic + beaded = 3, respectively.

[0114] Table 3 Statistics of oil well production characteristics of fault-karst reservoirs in Shunbei Oilfield Area 1

[0115]

[0116]

[0117] Furthermore, cluster analysis is performed on the production well groups according to the reservoir characteristic types (major characteristic factors) to determine whether there is a correlation between the clustering results and the dynamic reserves of the production well groups; then, the relevant reservoir characteristic types are integrated and the production well groups are clustered again until the production well groups are clustered into multiple categories based on dynamic reserves.

[0118] For well groups with complex geological conditions, the dynamic reserve distribution of well locations is often not uniform. It is difficult to obtain a more regular distribution of results by directly calculating the relationship between the dynamic reserves and reservoir characteristic factors. Therefore, it is necessary to classify the dynamic reserves and calculate them separately. The basis for well group classification is to examine the correlation between reservoir characteristic types and well groups, and use reservoir characteristic types with strong correlation to classify dynamic reserves. Therefore, this implementation conducted cluster analysis on geological characteristic factors, engineering characteristic factors, and production characteristic factors respectively. The clustering method adopted the Ward clustering method commonly used in hierarchical clustering. The clustering results are as follows: Figure 1 , Figure 2 , Figure 3 shown.

[0119] Combining Table 1 and Figure 2 The analysis found that for the clustering results of geological characteristic factors, the dynamic reserves are basically distributed according to the pulling and squeezing conditions of the fault zone, that is, there is a correlation between them and the geological characteristic factors.

[0120] According to Figure 3As shown in Table 2, the cluster analysis of engineering characteristic factors yielded rather messy results. The clustering was basically carried out according to the production increase method and scale of the production wells. The clustering results based on engineering characteristic factors showed no correlation with dynamic reserves.

[0121] In the right Figure 4 From the analysis of Table 3, it can be found that the clustering results based on production characteristic factors are basically distributed according to the size of dynamic reserves and production decline, indicating that there is also a correlation between production characteristic factors and dynamic reserves.

[0122] Therefore, the two reservoir characteristic types of geological characteristic factors and production characteristic factors were re-selected and integrated for clustering, that is, all the reservoir lower layer characteristic factors under the geological characteristic factors and production characteristic factors were used to cluster the well groups until they were divided into three types of production well groups with high dynamic reserves, medium dynamic reserves and low dynamic reserves. The first type of well group is the production well group with low dynamic reserves, the second type of well group is the production well group with high dynamic reserves, and the third type of well group is the production well group with medium dynamic reserves. The classification reference Figure 5 The same as Tables 1, 2, and 3 above.

[0123] Among them, the first type of well group includes: SHB5-11H, SHB5-3, SHB5-2CH, SHB5-13H, Shunbei 71X, SHB5-5H, SHB5-15H, Shunbei 53X. The second type of well group includes: Shunbei 5, SHB5-1X, SHB1-9, SHB1-17H, SHB1-3, SHB1-19H, SHB1-10H, SHB1-6CH, SHB1-7H, SHB1-20H, SHB1-1H, SHB1-4HCH, SHB1-5H, SHB1-11. The third type of well group includes: SHB5-4H, SHB5-12H, Shunbei 51X, SHB5-7, SHB5-10, SHB5-6H, SHBP3H, SHB1-8H, SHB1-18H, SHB1-22H, SHB1-2H, SHB1-13H, SHB1-24X, SHB1-14, SHB1-12, and SHB1-15.

[0124] Furthermore, it is determined whether the geological characteristic factors and the reservoir lower layer characteristic factors in the production characteristic factors are meaningful to each type of production well group obtained by clustering.

[0125] The clustered production well groups are not only in a relatively concentrated specific range in geological space (80% of the wells are clustered in the form of well groups, each well group has 3-6 wells, and the wells in the well group are all adjacent wells), that is, they have similar geological characteristics; but this type of wells also has similar production dynamics, such as production decline rates that differ by less than 30%, stable production times that are close (differences that are less than 30%), and pressure retention levels that are close (differences that are less than 30%). Therefore, it is necessary to determine whether the geological characteristic factors and the reservoir lower layer characteristic factors in the production characteristic factors are meaningful to each type of production well group.

[0126] In the production process of Shunbei Oil and Gas Field, the adjacent wells are quite different; therefore, the scale of closeness is selected to be less than 30%. When the difference in the data value of the reservoir lower layer characteristic factor under the production characteristic factor of any well is within 30%, it means that the reservoir lower layer characteristic factor is in line with the production dynamic law, and thus it is considered to be meaningful to the corresponding production well group;

[0127] The phase difference amplitude calculation formula is as follows:

[0128]

[0129] Among them, y 最大值 is the specific value of the larger value of the two sets of reservoir lower layer characteristic factors being compared, y 最小值 It is the specific value of the smaller number among the two groups of reservoir lower layer characteristic factors being compared.

[0130] For the three types of production well groups with high dynamic reserves, medium dynamic reserves and low dynamic reserves, it is also necessary to determine whether the reservoir lower layer characteristic factors under the geological characteristic factors are meaningful to each type of well group. In this embodiment, the proximity of the production characteristic factor data of the three types of production well groups with high dynamic reserves, medium dynamic reserves and low dynamic reserves is within 30%, that is, the production characteristic factors are meaningful to the three types of well groups; and 80% of the wells in the well groups are in a state of near wells, indicating that the geological characteristic factors are also meaningful to the three types of well groups. For the engineering characteristic factors, it can be considered that they are also meaningful to the three types of well groups, and finally it is proved that all the reservoir lower layer characteristic factors are meaningful to the corresponding types of production well groups.

[0131] It is worth noting that when analyzing the well group data of the characteristic factors of the lower layer of the reservoir, some well groups will have irregular abnormal points with a data difference of 5% to 10% that is larger than that of most well groups, which need to be manually removed. The removal method can adopt the wire box method commonly used in this field.

[0132] Then, the identified reservoir lower layer characteristic factors that are significant to the corresponding category of production well groups are used to calculate the Pearson partial correlation coefficients of their dynamic reserves relative to the corresponding category of production well groups.

[0133] When calculating the partial correlation coefficient between the dynamic reserves of each type of well group and the meaningful reservoir lower layer characteristic factors, the interrelated reservoir lower layer characteristic factors are determined through the geological analysis and well test reports of the actual well group. The calculation results of the partial correlation coefficients of the dynamic reserves of different types of well groups to the lower layer characteristic factors of different characteristic types are shown in Tables 4 to 12.

[0134] Table 4 Calculation results of partial correlation coefficients of the first type of well group dynamic reserves and geological characteristic factors

[0135]

[0136] Table 5 Calculation results of partial correlation coefficients of the first type of well group dynamic reserves and engineering characteristic factors

[0137]

[0138] Table 6 Calculation results of partial correlation coefficients of the first type of well group dynamic reserves and production characteristic factors

[0139]

[0140] Table 7 Calculation results of partial correlation coefficients of the second type of well group dynamic reserves and geological characteristic factors

[0141]

[0142] Table 8 Calculation results of partial correlation coefficients of the second type of well group dynamic reserves and engineering characteristic factors

[0143]

[0144]

[0145] Table 9 Calculation results of partial correlation coefficients of the second type of well group dynamic reserves and production characteristic factors

[0146]

[0147] Table 10 Calculation results of partial correlation coefficients of the third type of well group dynamic reserves and geological characteristic factors

[0148]

[0149]

[0150] Table 11 Calculation results of partial correlation coefficients of the third type of well group dynamic reserves and engineering characteristic factors

[0151]

[0152] Table 12 Calculation results of partial correlation coefficients of the third type of well group dynamic reserves and production characteristic factors

[0153]

[0154] Tables 4 to 12 show the partial correlation coefficients between the significant reservoir lower layer characteristic factors and the dynamic reserves of each type of well group. Table 8 shows that in the second type of well group, the size of the dynamic reserves has no correlation with the engineering factors.

[0155] In the calculation of partial correlation coefficient of geological factor characteristics, the numerical code is directly substituted as the numerical value. The partial correlation coefficient value finally calculated will be used for subsequent determination of significant correlation.

[0156] Furthermore, the absolute value of the obtained Pearson partial correlation coefficient is calculated, and the reservoir lower layer characteristic factors corresponding to the absolute value greater than the correlation judgment value are identified as being significantly correlated with the dynamic reserves of the corresponding category of production well groups;

[0157] According to the Pearson correlation analysis principle, it is considered that the absolute value of the main controlling factor is greater than 0.6, which is considered to be significantly correlated. That is, the absolute value of the Pearson partial correlation coefficient is between 0.6 and 1, which indicates that the main controlling factor is significantly correlated with the dynamic reserves. Therefore, the main controlling factors significantly correlated with the dynamic reserves of the first type of well group are: reflection type, stress mechanism, fracture-cavity combination, depth into the mountain, production increase method, average oil production, production decline rate, energy utilization rate, and energy decline rate.

[0158] The main controlling factors significantly correlated with the dynamic reserves of the second type of well group are: stress mechanism, fault distance, reservoir development changes, average oil production, average gasoline ratio, and production decline rate.

[0159] The main controlling factors significantly correlated with the dynamic reserves of the third type of well group are: geology, reflection type, stress mechanism, fault distance, fracture-cavity combination, reservoir development changes, depth into the mountain, drilling leakage, production increase method, production increase scale, production decline rate, and energy utilization rate.

[0160] Furthermore, the reservoir lower layer characteristic factors identified as significantly correlated are taken as main controlling factors, and the main controlling factors and specific data values ​​of dynamic reserves of all production wells in the corresponding category of production well groups are taken as samples for neural network training to obtain the neural network structure and threshold set corresponding to different wells in each typical well group; the relationship between the main controlling factors and dynamic reserves is established, and based on the relationship between the main controlling factors and dynamic reserves, the dynamic reserves of the remaining well groups in the reservoir are calculated to determine their classification in terms of dynamic reserves.

[0161] The neural network used to establish the relationship between the main controlling factors and the dynamic reserves in the present invention is a BP neural network. The neural network structures of the three types of well groups are as follows: Figure 5 to Figure 7 As shown in the figure, the learning rate is set to 0.01, and the target error of the model is set to 1×10 -3 The maximum number of iterations is 1000. Finally, the relationship between the dynamic reserves of the well group and the main controlling factors is obtained, which is used to estimate the dynamic reserves of the remaining well groups during the oil field development process. The neural network training prediction results of the three types of production well groups are shown in Tables 13 to 15.

[0162] Table 13 Neural network prediction results of the first type of wells

[0163]

[0164]

[0165] Table 14 Neural network prediction results of the second type of wells

[0166]

[0167] Table 15 Neural network prediction results of the third type of wells

[0168]

[0169] According to the prediction results in Tables 13 to 15, the prediction results of the second and third types of production wells are better, which proves that the evaluation method can be effectively used to evaluate well groups with higher production.

[0170] For the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0171] It should be pointed out that in other embodiments of the present invention, the method can also obtain a new rapid assessment method for the dynamic reserves of fracture-cavity oil reservoirs by combining one or several of the above embodiments, so as to achieve early and accurate prediction of the dynamic reserves of fracture-cavity oil reservoirs.

[0172] It should be noted that, based on the method in any one or more of the above embodiments of the present invention, the present invention also provides a storage medium, on which is stored a program code that can implement the method as described in any one or more of the above embodiments, and when the code is executed by the operating system, it can implement the rapid assessment method of dynamic reserves of fracture-cavity oil reservoirs as described above.

[0173] Embodiment 2

[0174] The method is described in detail in the embodiments disclosed in the present invention. The method of the present invention can be implemented by various devices or systems. Therefore, based on other aspects of the method described in any one or more embodiments, the present invention also provides a system for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs. The system is used to execute the method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs described in any one or more embodiments. Specific embodiments are given below for detailed description.

[0175] Specifically, Fig. 9 FIG. 4 is a schematic diagram showing a structure of a rapid evaluation system for dynamic reserves of a fracture-cavity oil reservoir provided in an embodiment of the present invention. Fig. 9 As shown, the system includes:

[0176] An analysis feature decision module is configured to select a well group that has been put into production at a set scale and has been in production for a set time from the overall reservoir to be evaluated as an analysis production well group, and based on the analysis production well group, select all lower layer feature factors that may be related to dynamic reserves by considering different feature types;

[0177] A characteristic data acquisition module, which is configured to acquire data of each underlying characteristic factor based on analyzing geological analysis files and well test report data of each well in the production well group;

[0178] A relevant feature type decision module is configured to perform cluster analysis on each well in the analysis production well group based on its underlying feature factors, taking each feature type as a unit, and selecting a feature type whose correlation with dynamic reserves meets the set requirements as a target feature type according to the clustering result;

[0179] The well group classification and factor selection module is configured to cluster the wells of the analyzed production well group according to the lower layer characteristic factors of the target characteristic type and the dynamic reserve data to obtain different reserve classification well groups, and then analyze the distribution characteristics of all the lower layer characteristic factors under the target characteristic type around each reserve classification well group, and select the lower layer characteristic factors that meet the set requirements as relevant characteristic factors;

[0180] A main control parameter determination module is configured to further calculate the partial correlation coefficient of each relevant characteristic factor relative to the dynamic reserves of each well in the energy storage classification well group to which it belongs, and select the main control characteristic parameter according to the partial correlation coefficient;

[0181] An evaluation model building module is configured to respectively use the main control characteristic parameters and the corresponding dynamic reserve data as samples for well groups with different reserve classifications, implement training of the neural network model based on the set neural network configuration, and obtain a dynamic reserve evaluation model corresponding to each well group with different reserve classifications;

[0182] The reserve assessment application module is configured to obtain data of various underlying characteristic factors for unproduced new wells to be assessed based on the geological analysis files and well test report data of each well in the production well group, and then substitute them into the dynamic reserve assessment model to determine the corresponding dynamic energy storage data.

[0183] In an optional embodiment, the analysis feature decision module sets the feature types to include geological feature factors, engineering feature factors and production feature factors; and sets lower layer feature factors related to dynamic reserves for the geological feature factors, engineering feature factors and production feature factors respectively.

[0184] Preferably, in one embodiment, each well in the analyzed production well group is a production well in the middle of production and later, and the relevant feature type decision module is configured to first obtain the dynamic reserve data of each well from the production data.

[0185] Furthermore, in one embodiment, the relevant feature type decision module is configured to use a hierarchical clustering method to perform cluster analysis on the lower-layer feature factors and dynamic reserves under different feature types, respectively, to determine whether the clustering results of the lower-layer feature factors and dynamic reserves of each well have a set distribution law. If the clustering results of all lower-layer feature factors and dynamic reservoirs do not meet the set distribution law, it indicates that the current feature type has no correlation with the dynamic reserves and is not used as a target feature type.

[0186] Preferably, in one embodiment, the well group classification and factor selection module performs clustering processing according to the following operations to obtain well groups with different reserve classifications:

[0187] After integrating the underlying characteristic factors under the target characteristic type, all underlying characteristic factors are used to cluster the analyzed well groups until each well is classified into multiple reserve classification well groups with different levels of dynamic reserves.

[0188] Furthermore, in one embodiment, the well group classification and factor selection module selects relevant characteristic factors by the following operations:

[0189] For each level of reserve classification well group, the phase difference amplitude of each lower layer characteristic factor under the target characteristic type is calculated respectively to determine whether the phase difference amplitude meets the set requirements. If the set requirements are met, it is confirmed that the lower layer characteristic factor is correlated with the reserve classification well group.

[0190] Optionally, in one embodiment, before calculating the phase difference amplitude of each lower layer characteristic factor under the target characteristic type, the well group classification and factor selection module also includes: using the wire box method to eliminate abnormal data points of various lower layer characteristic factors of well groups with different reserve classifications.

[0191] Specifically, in an optional embodiment, the phase difference amplitude R of each lower-layer feature factor under the target feature type is calculated according to the following formula:

[0192]

[0193] Among them, y 最大值 is the specific value of the larger value of the two sets of reservoir lower layer characteristic factors being compared, y 最小值 It is the specific value of the smaller number among the two groups of reservoir lower layer characteristic factors being compared.

[0194] In the rapid assessment system for dynamic reserves of fracture-vuggy oil reservoirs provided by the embodiment of the present invention, each module or unit structure can be operated independently or in combination according to actual factor analysis requirements and data calculation requirements to achieve corresponding technical effects.

[0195] It should be understood that the embodiments disclosed in the present invention are not limited to the specific structures, processing steps or materials disclosed herein, but should be extended to equivalent substitutions of these features understood by ordinary technicians in the relevant field. It should also be understood that the terms used herein are only used for the purpose of describing specific embodiments and are not meant to be limiting.

[0196] The "one embodiment" mentioned in the specification means that a particular feature, structure or characteristic described in conjunction with the embodiment is included in at least one embodiment of the present invention. Therefore, the phrase "one embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0197] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for rapid assessment of dynamic reserves of fracture-cavity oil reservoirs, characterized in that: The method comprises: Analyze the characteristic decision step, select the well group that has been put into production at a set scale and has been in production for a set time from the overall oil reservoir to be evaluated as the analysis production well group, and select all the lower layer characteristic factors that may be related to dynamic reserves based on the analysis production well group and considering different characteristic types; Characteristic data acquisition step, obtaining data of each underlying characteristic factor by analyzing geological analysis files and well test report data of each well in the production well group; The relevant feature type decision step is to perform cluster analysis on each well in the analysis production well group based on its underlying feature factors, and select the feature type whose correlation with the dynamic reserves meets the set requirements as the target feature type according to the clustering result; Well group classification and factor selection steps, clustering the wells of the analyzed production well group according to the lower layer characteristic factors of the target characteristic type and the dynamic reserve data to obtain different reserve classification well groups, and then analyzing the distribution characteristics of all the lower layer characteristic factors under the target characteristic type around each reserve classification well group, and selecting the lower layer characteristic factors that meet the set requirements as relevant characteristic factors; The main control parameter determination step is to further calculate the partial correlation coefficient of each relevant characteristic factor relative to the dynamic reserves of each well in the energy storage classification well group to which it belongs, and select the main control characteristic parameter according to the partial correlation coefficient; The evaluation model establishment step is to use the main control characteristic parameters and the corresponding dynamic reserve data as samples for well groups with different reserve classifications, and to implement the training of the neural network model based on the set neural network configuration to obtain the dynamic reserve evaluation model corresponding to each well group with different reserve classifications; The reserve assessment application steps are as follows: for the unproduced new wells to be assessed, the data of each underlying characteristic factor is obtained based on the geological analysis files and well test report materials of each well in the production well group, and then the data is substituted into the dynamic reserve assessment model to determine the corresponding dynamic energy storage data.

2. The method according to claim 1, characterized in that: In the feature analysis decision step, the feature types include geological feature factors, engineering feature factors and production feature factors; lower layer feature factors related to dynamic reserves are set for the geological feature factors, engineering feature factors and production feature factors respectively.

3. The method according to claim 1, characterized in that Each well in the analyzed production well group is a production well in the middle stage of production or later. In the relevant feature type decision step, the dynamic reserve data of each well is first obtained from the production data.

4. The method according to claim 1, characterized in that: In the relevant feature type decision step, the hierarchical clustering method is used to perform cluster analysis on the lower layer characteristic factors and dynamic reserves under different feature types to determine whether the clustering results of the lower layer characteristic factors and dynamic reserves of each well have the set distribution law. If the clustering results of all lower layer characteristic factors and dynamic reservoirs do not meet the set distribution law, it indicates that the current feature type has no correlation with the dynamic reserves and is not used as the target feature type.

5. The method according to claim 1, characterized in that In the well group classification and factor selection steps, the process of clustering to obtain different reserve classification well groups includes: After integrating the underlying characteristic factors under the target characteristic type, all underlying characteristic factors are used to cluster the analyzed well groups until each well is classified into multiple reserve classification well groups with different levels of dynamic reserves.

6. The method according to claim 1 or 5, characterized in that: In the well group classification and factor selection steps, the process of selecting relevant characteristic factors includes: For each level of reserve classification well group, the phase difference amplitude of each lower layer characteristic factor under the target characteristic type is calculated respectively to determine whether the phase difference amplitude meets the set requirements. If the set requirements are met, it is confirmed that the lower layer characteristic factor is correlated with the reserve classification well group.

7. The method according to claim 1, characterized in that Before calculating the difference amplitude of each lower layer characteristic factor under the target characteristic type, it also includes: using the wire box method to eliminate abnormal data points of various lower layer characteristic factors of well groups with different reserve classifications.

8. The method according to claim 1, characterized in that The phase difference amplitude R of each lower-level feature factor under the target feature type is calculated according to the following formula: Among them, y 最大值 is the specific value of the larger value of the two sets of reservoir lower layer characteristic factors being compared, y 最小值 It is the specific value of the smaller number among the two groups of reservoir lower layer characteristic factors being compared.

9. A storage medium, characterized in that: The storage medium stores program codes that can implement the method as claimed in any one of claims 1 to 8.

10. A rapid assessment system for dynamic reserves of fracture-cavity oil reservoirs, characterized in that: The system executes the method according to any one of claims 1 to 8.

Images