Low-permeability oil reservoir comprehensive evaluation method and device

Abstract

This specification provides a comprehensive evaluation method and apparatus for low-permeability reservoirs. The method involves acquiring multi-source heterogeneous parameters of the target low-permeability reservoir; determining a multi-dimensional evaluation index system and corresponding target evaluation results based on these parameters; identifying the contribution relationship type between each evaluation index and the target evaluation result; mapping each evaluation index to a unified evaluation direction based on its contribution relationship type to obtain unified evaluation direction characteristics; using a pre-defined monotonic constraint comprehensive evaluation model, obtaining a comprehensive evaluation index based on the unified evaluation direction characteristics; and determining the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index. This improves the accuracy and stability of the comprehensive evaluation results for the target low-permeability reservoir.

Description

Technical Field

[0001] This manual belongs to the field of oil and gas field development technology, and in particular relates to a comprehensive evaluation method and apparatus for low-permeability reservoirs. Background Technology

[0002] With the continued development of unconventional low-permeability reservoirs such as shale oil, tight sandstone reservoirs, and tight carbonate reservoirs, the evaluation of low-permeability reservoirs requires the integration of multiple dimensions. However, existing evaluation methods struggle to unify the contribution directions of different indicators to the overall evaluation results when dealing with multi-source heterogeneous evaluation indices, easily leading to inconsistencies in evaluation logic and thus affecting the accuracy and stability of the overall evaluation results.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This specification provides a comprehensive evaluation method and apparatus for low-permeability reservoirs, which solves the problem that existing comprehensive evaluation methods for low-permeability reservoirs are unable to unify the contribution direction between multi-source heterogeneous evaluation indicators and target evaluation results, resulting in poor accuracy and insufficient stability of the comprehensive evaluation results.

[0005] This specification provides a comprehensive evaluation method and apparatus for low-permeability reservoirs, including: Obtain multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; Based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation results corresponding to the multi-dimensional evaluation index system are determined; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; Based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type, and non-monotonic correlation type; Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. Using a pre-defined monotonic constraint comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction characteristics; wherein, the pre-defined monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, and the contribution consistency constraint condition is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. Based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined.

[0006] In one embodiment, determining the contribution relationship type corresponding to each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result includes: Based on the correspondence between the changes in the values ​​of each evaluation indicator and the changes in the target evaluation results, the contribution change relationship of each evaluation indicator is determined. When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result improves, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonic positive correlation type. When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result decreases, the contribution relationship type corresponding to the evaluation indicator is determined to be monotonically negatively correlated. When the contribution change relationship indicates that there is an optimal contribution interval between the indicator value and the target evaluation result, the contribution relationship type corresponding to the evaluation indicator is determined to be a non-monotonic correlation type.

[0007] In one embodiment, the step of performing unified evaluation direction mapping processing on each evaluation indicator based on the contribution relationship type corresponding to each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator includes: Based on the contribution relationship type corresponding to each evaluation indicator, the evaluation direction mapping rule corresponding to each evaluation indicator is determined; wherein, the evaluation direction mapping rule includes the positive retention rule, the monotonically decreasing mapping rule, and the optimal contribution interval mapping rule; Based on the evaluation direction mapping rules corresponding to each evaluation indicator, each evaluation indicator is mapped to obtain a unified direction evaluation value corresponding to each evaluation indicator. Based on the unified direction evaluation value corresponding to each evaluation indicator, determine the unified evaluation direction feature corresponding to each evaluation indicator. The positive retention rule is used to determine the evaluation index corresponding to the monotonically positive correlation type as the corresponding unified direction evaluation value; the monotonically decreasing mapping rule is used to perform monotonically decreasing mapping processing on the evaluation index corresponding to the monotonically negative correlation type to obtain the corresponding unified direction evaluation value; and the optimal contribution interval mapping rule is used to perform optimal contribution interval mapping processing on the evaluation index corresponding to the non-monotonically correlated type to obtain the corresponding unified direction evaluation value.

[0008] In one embodiment, determining the unified evaluation direction feature corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator includes: Based on the unified directional evaluation value corresponding to each evaluation indicator, determine the distribution parameters of the evaluation value corresponding to each evaluation indicator; Based on the evaluation value distribution parameters corresponding to each evaluation indicator, the unified directional evaluation values ​​corresponding to each evaluation indicator are normalized to obtain the unified dimensional evaluation values ​​corresponding to each evaluation indicator. Based on the unified evaluation value of each evaluation indicator, the unified evaluation direction characteristics corresponding to each evaluation indicator are determined.

[0009] In one embodiment, the step of obtaining the comprehensive evaluation index using a preset monotonic constraint comprehensive evaluation model based on the unified evaluation direction characteristics includes: Based on the unified evaluation direction characteristics corresponding to each evaluation indicator, determine the set of unified evaluation direction characteristics; Based on the unified evaluation direction feature set, determine the comprehensive evaluation feature data; The comprehensive evaluation feature data is input into the preset monotonic constraint comprehensive evaluation model to obtain the initial evaluation index output by the preset monotonic constraint comprehensive evaluation model. Based on the aforementioned contribution consistency constraint, the initial evaluation index is constrained to obtain the comprehensive evaluation index.

[0010] In one embodiment, determining the multidimensional evaluation index system for the target low-permeability reservoir and the target evaluation result corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters includes: Based on the multi-source heterogeneous parameters, determine the parameter category data of the target low-permeability reservoir; Based on the parameter category data, candidate evaluation index data are determined; Based on the candidate evaluation index data, a multidimensional evaluation index system for the target low-permeability reservoir is determined; Based on the multidimensional evaluation index system, determine the evaluation result type corresponding to the multidimensional evaluation index system; Based on the evaluation result type, determine the target evaluation result corresponding to the multidimensional evaluation index system.

[0011] In one embodiment, after obtaining the multi-source heterogeneous parameters of the target low-permeability reservoir, the method further includes: Based on the multi-source heterogeneous parameters, determine the parameter preprocessing rules corresponding to the multi-source heterogeneous parameters; According to the parameter preprocessing rules, the multi-source heterogeneous parameters are preprocessed to obtain the preprocessed multi-source heterogeneous parameters; The parameter preprocessing rules include at least one of the following: outlier removal rules, missing value completion rules, noise filtering rules, unit unification rules, and data consistency verification rules.

[0012] This specification provides a comprehensive evaluation device for low-permeability reservoirs, including: The parameter acquisition module is used to acquire multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters and engineering parameters; The system determination module is used to determine the multidimensional evaluation index system of the target low-permeability reservoir and the target evaluation result corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters; wherein, the multidimensional evaluation index system includes multiple evaluation indicators; The type determination module is used to determine the contribution relationship type corresponding to each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type and non-monotonic correlation type; The feature determination module is used to perform unified evaluation direction mapping processing on each evaluation indicator according to the contribution relationship type corresponding to each evaluation indicator, so as to obtain the unified evaluation direction feature corresponding to each evaluation indicator. The index determination module is used to obtain a comprehensive evaluation index based on the unified evaluation direction characteristics using a preset monotonic constraint comprehensive evaluation model; wherein, the preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, and the contribution consistency constraint condition is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. The result determination module is used to determine the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index.

[0013] This specification also provides an electronic device, including a processor and a memory for storing processor-executable instructions, wherein the processor, when executing the instructions, implements a comprehensive evaluation method for low-permeability reservoirs.

[0014] This specification also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed, implement a comprehensive evaluation method for low-permeability reservoirs.

[0015] Based on the comprehensive evaluation method for low-permeability reservoirs provided in this specification, multi-source heterogeneous parameters of the target low-permeability reservoir are obtained. These multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters. Based on these multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the corresponding target evaluation results are determined. The multi-dimensional evaluation index system includes multiple evaluation indicators. Based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined. The contribution relationship type includes monotonically positive correlation and monotonically negative correlation. The evaluation indicators are classified into two types: one is the contribution relationship type and the other is the non-monotonic correlation type. Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. Using a preset monotonic constraint comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. The preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, which is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction feature corresponding to each evaluation indicator. Based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined. In this way, by acquiring multi-source heterogeneous parameters of the target low-permeability reservoir and determining the multi-dimensional evaluation index system and its corresponding target evaluation results, the contribution relationship between each evaluation index and the target evaluation results can be clarified. By determining the contribution relationship type corresponding to each evaluation index and performing unified evaluation direction mapping processing on each evaluation index based on the contribution relationship type, a unified evaluation direction feature can be obtained. This can unify the contribution direction between different evaluation indicators and the target evaluation results, reducing the impact of inconsistent evaluation logic on the comprehensive evaluation results. At the same time, by using a pre-set monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint condition, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. This ensures that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction feature corresponding to each evaluation index, thereby improving the accuracy and stability of the comprehensive evaluation results of the target low-permeability reservoir. Attached Figure Description

[0016] To more clearly illustrate the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. The drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a comprehensive evaluation method for low-permeability reservoirs provided in one embodiment of this specification. Figure 2This is a schematic diagram of the electronic device structure provided in one embodiment of this specification; Figure 3 This is a schematic diagram of the structural composition of a comprehensive evaluation device for low-permeability oil reservoirs provided in one embodiment of this specification. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0019] With the continuous development of unconventional oil and gas resources such as shale oil, tight sandstone reservoirs, and tight carbonate reservoirs, low-permeability reservoirs have become an important target for increasing oil and gas reserves and production. These reservoirs are typically characterized by strong reservoir heterogeneity, complex pore structure, weak permeability, and significant differences in development response. Their evaluation process requires comprehensive consideration of multiple factors, including geological conditions, reservoir properties, fluid properties, and engineering modifiability.

[0020] Existing reservoir evaluation methods mainly include empirical scoring methods, statistical regression methods, grey relational analysis methods, analytic hierarchy process (AHP) methods, and machine learning evaluation methods, but they still have the following shortcomings: Evaluation index systems typically contain both positive and negative contribution indicators. Different evaluation indicators have different effects on the evaluation results. If they are directly used in comprehensive evaluation modeling, it can easily lead to logical conflicts in the evaluation. The relationship between some evaluation indicators and evaluation results is not a simple linear or monotonic relationship, but rather exhibits interval contribution characteristics or non-monotonic variation patterns. Traditional normalization methods are insufficient to accurately characterize the contribution characteristics of such evaluation indicators. Existing evaluation models often focus on statistical fitting ability and lack constraints on the consistency of contributions between evaluation indicators and evaluation results, resulting in insufficient stability and interpretability of evaluation results. For the numerous heterogeneous parameters that exist in low-permeability reservoirs, current technology lacks a comprehensive evaluation method that can uniformly express the contribution direction of different evaluation indicators and maintain the physical rationality of the evaluation results.

[0021] To address the root causes of the aforementioned problems, by acquiring multi-source heterogeneous parameters of the target low-permeability reservoir and determining a multi-dimensional evaluation index system and its corresponding target evaluation results, the contribution relationship between each evaluation index and the target evaluation results can be clarified. By determining the contribution relationship type corresponding to each evaluation index and performing unified evaluation direction mapping processing on each evaluation index based on the contribution relationship type, a unified evaluation direction feature can be obtained. This can unify the contribution direction between different evaluation indicators and the target evaluation results, reducing the impact of inconsistent evaluation logic on the comprehensive evaluation results. At the same time, using a pre-set monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint condition, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. This ensures that the comprehensive evaluation index satisfies a monotonically non-decreasing relationship with the unified evaluation direction feature corresponding to each evaluation index, thereby improving the accuracy and stability of the comprehensive evaluation results of the target low-permeability reservoir.

[0022] See Figure 1 As shown in the embodiments of this specification, a comprehensive evaluation method for low-permeability reservoirs is provided, wherein the method is specifically applied to the server side. In specific implementation, the method may include the following: S101: Obtain multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; S102: Based on the multi-source heterogeneous parameters, determine the multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation results corresponding to the multi-dimensional evaluation index system; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; S103: Based on the contribution relationship between each evaluation indicator and the target evaluation result, determine the contribution relationship type corresponding to each evaluation indicator; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type, and non-monotonic correlation type; S104: Based on the contribution relationship type corresponding to each evaluation indicator, perform unified evaluation direction mapping processing on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. S105: Using a preset monotonic constraint comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction characteristics; wherein, the preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, and the contribution consistency constraint condition is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. S106: Determine the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index.

[0023] Among them, the aforementioned target low-permeability reservoirs refer to low-permeability reservoirs that require comprehensive evaluation, which can be low-permeability reservoirs such as shale oil reservoirs, tight sandstone oil reservoirs, and tight carbonate rock oil reservoirs.

[0024] The aforementioned multi-source heterogeneous parameters refer to parameter data obtained from the target low-permeability reservoir, which have different sources and different physical meanings, including at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters. Geological parameters can reflect reservoir sedimentation, hydrocarbon source, and oil-bearing conditions; reservoir parameters can reflect reservoir space and seepage capacity; fluid parameters can reflect fluid flow characteristics; and engineering parameters can reflect fracturing and development conditions. The geological parameters include at least one or more of total organic carbon content, vitrinite reflectance, effective thickness, and oil saturation; the reservoir parameters include at least one or more of porosity, permeability, brittleness index, fracture development degree, pore throat radius, nanopore tortuosity, and nanopore connectivity; the fluid parameters include at least one or more of crude oil viscosity, fluid flow resistance, starting pressure gradient, and gas content; and the engineering parameters include at least one or more of fracturing difficulty, construction displacement, proppant addition scale, and single-well production decline rate.

[0025] The aforementioned multidimensional evaluation index system refers to a set of indicators determined based on multi-source heterogeneous parameters, used to evaluate the comprehensive characteristics of a target low-permeability reservoir. This multidimensional evaluation index system includes multiple evaluation indicators, which characterize the comprehensive development conditions of the target low-permeability reservoir from geological, reservoir, fluid, and engineering dimensions.

[0026] The aforementioned target evaluation results can refer to the evaluation output objects corresponding to the multidimensional evaluation index system, used as reference results for judging the contribution relationship of each evaluation index. The target evaluation results can correspond to one or more of the following: reservoir quality evaluation results, sweet spot evaluation results, target optimization results, development potential prediction results, fracturing effect prediction results, or single-well productivity prediction results.

[0027] The aforementioned contribution relationship refers to the influence relationship between various evaluation indicators and the target evaluation result, used to characterize the direction or pattern of the effect of changes in evaluation indicators on the target evaluation result. For example, if the target evaluation result improves when a certain evaluation indicator increases, then there is a positive contribution relationship between the evaluation indicator and the target evaluation result; if the target evaluation result decreases when a certain evaluation indicator increases, then there is a negative contribution relationship; if a certain evaluation indicator contributes significantly to the target evaluation result within a specific range, then there is a non-monotonic contribution relationship.

[0028] The aforementioned contribution relationship types refer to the classification results determined based on the contribution relationship between each evaluation indicator and the target evaluation result. These contribution relationship types include monotonically positive correlation, monotonically negative correlation, and non-monotonic correlation, which serve as a basis for subsequent unified evaluation direction mapping processing.

[0029] The aforementioned monotonic positive correlation type refers to a contribution relationship where, as the value of the evaluation indicator increases, the target evaluation result improves or tends to be better. For example, indicators such as porosity, permeability, effective thickness, and oil saturation can generally be used as evaluation indicators of the monotonic positive correlation type.

[0030] The aforementioned monotonically negative correlation type refers to a contribution relationship where the target evaluation result decreases or tends to worsen as the value of the evaluation indicator increases. For example, indicators such as crude oil viscosity, starting pressure gradient, fluid flow resistance, fracturing stimulation difficulty, and single-well production decline rate can generally be used as evaluation indicators of the monotonically negative correlation type. It should include at least one or more of the following: water cut, starting pressure gradient, fluid flow resistance, fracturing stimulation difficulty, single-well production decline rate, nanopore tortuosity, and clay mineral content.

[0031] The aforementioned non-monotonic correlation type refers to a relationship where the evaluation index and the target evaluation result are not simply monotonically positively or monotonically negatively correlated, but rather exhibit a contribution range. For example, indices such as vitrinite reflectance, formation pressure coefficient, brittleness index, and fracture density may exhibit range-based contribution characteristics, meaning that the evaluation result is better when the index value is within a certain range, and the evaluation result decreases when it deviates from that range. This includes at least one or more of vitrinite reflectance, formation pressure coefficient, brittleness index, and fracture density.

[0032] The aforementioned unified evaluation direction mapping process refers to a data processing procedure that unifies the direction of evaluation indicators based on the contribution relationship type corresponding to each indicator. This process allows evaluation indicators with different contribution directions or patterns to be converted into a unified expression, thereby reducing conflicts in evaluation directions among multi-source heterogeneous evaluation indicators.

[0033] The aforementioned unified evaluation direction feature refers to the feature data obtained after mapping the evaluation indicators to a unified evaluation direction. This unified evaluation direction feature is used to characterize the degree of unified positive contribution of the corresponding evaluation indicator to the target evaluation result. That is, after mapping, the larger the unified evaluation direction feature, the better the corresponding target evaluation result tends to be.

[0034] The aforementioned pre-defined monotonic constraint comprehensive evaluation model refers to an evaluation model used to calculate a comprehensive evaluation index based on the characteristics of a unified evaluation direction. This model satisfies the contribution consistency constraint, ensuring that the comprehensive evaluation index output by the model has a monotonically non-decreasing relationship relative to each unified evaluation direction characteristic, thereby avoiding a situation where the comprehensive evaluation index decreases as the unified evaluation direction characteristics improve. Specifically, the pre-defined monotonic constraint comprehensive evaluation model can be one or more of the following: a weighted summation model, a statistical model, a machine learning model, a random forest model, a gradient boosting tree model, a tree model with monotonic constraints, or a neural network model.

[0035] The aforementioned comprehensive evaluation index refers to the numerical evaluation result obtained using a pre-defined monotonic constraint comprehensive evaluation model based on the characteristics of a unified evaluation direction. This comprehensive evaluation index is used to comprehensively characterize the evaluation level of the target low-permeability reservoir and serves as the basis for determining the comprehensive evaluation result.

[0036] The aforementioned contribution consistency constraint can refer to the condition that constrains the output of a pre-defined monotonic constraint comprehensive evaluation model, ensuring that the comprehensive evaluation index satisfies a monotonically non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. In other words, under otherwise unchanged conditions, when a certain unified evaluation direction characteristic increases, the comprehensive evaluation index does not decrease.

[0037] The aforementioned comprehensive evaluation results can refer to the evaluation output results of the target low-permeability reservoir determined based on the comprehensive evaluation index. These comprehensive evaluation results can be used to characterize the reservoir quality, sweet spot selection, development potential, fracturing effect, or single-well productivity of the target low-permeability reservoir, among other comprehensive evaluation conclusions. This includes one or more of the following: reservoir quality evaluation; sweet spot evaluation and target selection; development potential prediction; fracturing effect prediction; and single-well productivity prediction.

[0038] In some embodiments, determining the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index may specifically include: Based on the comprehensive evaluation index, determine the evaluation result type corresponding to the target low-permeability reservoir; Based on the evaluation result type, determine the evaluation level classification rules corresponding to the target low-permeability reservoir; The evaluation level corresponding to the target low-permeability reservoir is determined based on the evaluation level classification rules and the comprehensive evaluation index. Based on the evaluation result type and the evaluation level, the comprehensive evaluation result of the target low-permeability reservoir is determined; The evaluation result types include at least one of the following: reservoir quality evaluation type, sweet spot evaluation type, target selection type, development potential prediction type, fracturing effect prediction type, and single well productivity prediction type.

[0039] Specifically, after obtaining the comprehensive evaluation index of the target low-permeability reservoir, the comprehensive evaluation index can be used as a unified evaluation quantity for the target low-permeability reservoir under the current evaluation task. The current evaluation task can be determined based on the aforementioned target evaluation results. For example, when the target evaluation result corresponds to reservoir quality evaluation, the comprehensive evaluation index is used as the reservoir quality evaluation quantity; when the target evaluation result corresponds to sweet spot evaluation, the comprehensive evaluation index is used as the sweet spot evaluation quantity; when the target evaluation result corresponds to development potential prediction, fracturing effect prediction, or single-well production capacity prediction, the comprehensive evaluation index is used as the development potential evaluation quantity, fracturing effect evaluation quantity, or production capacity prediction evaluation quantity, respectively.

[0040] For reservoir quality evaluation types, the classification rules for reservoir quality evaluation grades can be determined based on the distribution of comprehensive evaluation indices of known wells or known stratigraphic intervals within the study area. For example, the comprehensive evaluation index can be divided into high-quality reservoir grades, medium-quality reservoir grades, and low-quality reservoir grades. Based on the comprehensive evaluation index corresponding to the stratigraphic interval to be evaluated in the target low-permeability reservoir, and the aforementioned reservoir quality evaluation grade classification rules, the reservoir quality grade corresponding to the stratigraphic interval to be evaluated is determined, and this reservoir quality grade is used as the reservoir quality evaluation result for the target low-permeability reservoir. This approach allows evaluation results from different stratigraphic intervals to be compared based on the same comprehensive evaluation index, avoiding classification bias caused by inconsistent contribution directions of evaluation indicators.

[0041] For sweet spot evaluation types, the sweet spot evaluation level of different evaluation units in the target low-permeability reservoir can be determined based on the comprehensive evaluation index. Specifically, the target low-permeability reservoir can be divided into multiple evaluation units, and the comprehensive evaluation index corresponding to each evaluation unit can be obtained; a sweet spot level classification rule can be determined according to the sweet spot evaluation type; based on the sweet spot level classification rule and the comprehensive evaluation index corresponding to each evaluation unit, the sweet spot level corresponding to each evaluation unit can be determined; the multiple evaluation units can be sorted according to the sweet spot level to obtain the sweet spot evaluation result or target optimization result. Among them, the evaluation units with higher comprehensive evaluation indices are identified as priority target areas, and the evaluation units with lower comprehensive evaluation indices are identified as non-priority target areas or secondary target areas.

[0042] For development potential prediction, the development potential level of a target low-permeability reservoir can be determined based on a comprehensive evaluation index. Specifically, a development potential level classification rule can be determined based on the comprehensive evaluation index ranges corresponding to high, medium, and low development potential areas in historical development data. The development potential level corresponding to the target low-permeability reservoir is then determined based on this classification rule and the comprehensive evaluation index. The development potential level can include high, medium, and low development potential levels, providing an evaluation basis for optimal development schemes and well location deployment.

[0043] For fracturing effect prediction, the fracturing effect level of the target low-permeability reservoir can be determined based on a comprehensive evaluation index. Specifically, at least one of the following can be used as the basis for classifying the fracturing effect level: post-fracturing production increase, post-fracturing production stabilization time, fracturing stimulation volume response degree, or post-fracturing overall production capacity improvement degree. A fracturing effect level classification rule is then established. Based on the fracturing effect level classification rule and the comprehensive evaluation index, the corresponding fracturing effect level for the target low-permeability reservoir is determined. The fracturing effect level can include an excellent fracturing effect level, a moderate fracturing effect level, and a poor fracturing effect level.

[0044] For single-well production capacity prediction, the production capacity prediction result of the target well can be determined based on a comprehensive evaluation index. Specifically, a correspondence between the comprehensive evaluation index and production capacity indicators can be established based on historical production data. The production capacity indicators include at least one of the following: initial daily oil production, cumulative oil production, predicted final recoverable reserves, or stable production period output. Based on the comprehensive evaluation index corresponding to the target well and the correspondence, the production capacity prediction value or production capacity level corresponding to the target well is determined, and the production capacity prediction value or production capacity level is used as the single-well production capacity prediction result.

[0045] Based on the above embodiments, the comprehensive evaluation index is no longer just a single numerical output, but can be transformed into reservoir quality evaluation results, sweet spot evaluation results, target selection results, development potential prediction results, fracturing effect prediction results, or single-well productivity prediction results according to different evaluation result types. Since the comprehensive evaluation index is obtained from a unified evaluation direction feature through a monotonically constrained comprehensive evaluation model, and satisfies a monotonically non-decreasing relationship relative to the unified evaluation direction feature, it can reduce the impact of inconsistent contribution directions of different evaluation indicators on evaluation grading and target selection, and improve the accuracy, stability, and interpretability of the comprehensive evaluation results for low-permeability reservoirs.

[0046] In some embodiments, the method for determining the contribution relationship type of each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result may further include the following: S1: Determine the contribution change relationship of each evaluation indicator based on the correspondence between the changes in the indicator values ​​of each evaluation indicator and the changes in the target evaluation results; S2: When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result improves, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonic positive correlation type; S3: When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result decreases, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonically negative correlation type; S4: When the contribution change relationship indicates that there is an optimal contribution interval between the indicator value of the evaluation indicator and the target evaluation result, the contribution relationship type corresponding to the evaluation indicator is determined to be a non-monotonic correlation type.

[0047] The optimal contribution range mentioned above is determined based on one or more of the following methods: statistical analysis of historical production capacity data, dynamic analysis of reservoir development, domain knowledge constraints, machine learning fitting, or numerical simulation inversion.

[0048] Specifically, based on historical development data, historical production data, reservoir development dynamics data, and domain knowledge of the target low-permeability reservoir, the contribution relationship between each evaluation indicator and the target evaluation result can be identified. Specifically, for each evaluation indicator in the multi-dimensional evaluation indicator system, the indicator value in different wells, different layers, or different evaluation units is obtained, along with the target evaluation result change data for the corresponding evaluation unit; where the target evaluation result change data can be one of the following: reservoir quality grade change, sweet spot grade change, development potential grade change, fracturing effect grade change, or single-well production capacity change.

[0049] For each evaluation indicator, the indicator values ​​can be sorted according to their numerical values, and the changing trend of the target evaluation results within different value ranges can be statistically analyzed. If the target evaluation result of the corresponding evaluation unit shows an upward trend as the indicator value increases, then the contribution relationship of the evaluation indicator is determined to be a positive monotonic contribution relationship. For example, increases in total organic carbon content, effective thickness, porosity, permeability, oil saturation, and nanopore connectivity are generally beneficial to oil and gas enrichment, improved storage capacity, or increased production capacity. Therefore, based on the correspondence between the changes in the indicator values ​​of this type of indicator and the changes in the target evaluation results, this type of indicator can be identified as a monotonic positive correlation type.

[0050] For certain evaluation indicators, if the target evaluation result of the corresponding evaluation unit decreases as the indicator value increases, then the contribution relationship of the corresponding evaluation indicator is determined to be a negative monotonic contribution relationship. For example, when crude oil viscosity, starting pressure gradient, fluid flow resistance, fracturing stimulation difficulty, single-well production decline rate, and nanopore tortuosity increase, fluid migration resistance usually increases, development effectiveness decreases, or production decline accelerates. Therefore, based on the correspondence between the changes in the indicator values ​​and the changes in the target evaluation results, these indicators can be identified as monotonically negatively correlated.

[0051] For evaluation indicators that do not exhibit monotonic variation, the existence of a contribution range for the indicator can be determined based on the correspondence between the indicator's value range and the target evaluation result. Specifically, the indicator's value is divided into multiple continuous intervals, and the target evaluation result level corresponding to each continuous interval is calculated. If the target evaluation result is high within one or more intervals, but decreases when it is below or above that interval, then the contribution relationship corresponding to the evaluation indicator is determined to be a non-monotonic contribution relationship. For example, vitrinite reflectivity, formation pressure coefficient, brittleness index, and fracture density are not simply a matter of the higher the better; rather, there are suitable contribution ranges. When such indicators deviate from their suitable contribution ranges, their contribution to the target evaluation result decreases, thus these indicators can be identified as non-monotonic correlation types.

[0052] When determining the type of contribution relationship, one or more of the following can be used: historical production data statistical analysis, reservoir development dynamic analysis, domain knowledge constraints, machine learning fitting, or numerical simulation inversion results. For example, for the brittleness index, the fracturing effect level corresponding to different brittleness index intervals can be determined based on the increase in production after fracturing, the time to stabilize production after fracturing, or the degree of volumetric response of fracturing stimulation. If the fracturing effect corresponding to a certain brittleness index interval is better, and the fracturing effect decreases when it is lower or higher than that interval, then the brittleness index is determined to be a non-monotonic correlation type. For the starting pressure gradient, the starting pressure gradient can be determined to be a monotonically negative correlation type based on the corresponding relationship between the decrease in single-well production capacity or development potential when the starting pressure gradient increases.

[0053] The above implementation method clarifies the contribution relationship of each evaluation indicator to the target evaluation result before the unified evaluation direction mapping process. Each evaluation indicator is then categorized into monotonically positively correlated, monotonically negatively correlated, and non-monotonicly correlated types. This allows for the application of different unified evaluation direction mapping methods for different contribution relationship types. Consequently, it avoids directly inputting evaluation indicators with different contribution directions or interval contribution characteristics into the evaluation model, which could lead to evaluation logic conflicts and improves the accuracy, stability, and interpretability of subsequent comprehensive evaluation index calculations.

[0054] In some embodiments, the method of performing unified evaluation direction mapping on each evaluation indicator based on the contribution relationship type corresponding to each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator may further include the following: S1: Based on the contribution relationship type corresponding to each evaluation indicator, determine the evaluation direction mapping rule corresponding to each evaluation indicator; wherein, the evaluation direction mapping rule includes the positive retention rule, the monotonically decreasing mapping rule, and the optimal contribution interval mapping rule; S2: Based on the evaluation direction mapping rules corresponding to each evaluation indicator, perform mapping processing on each evaluation indicator to obtain the unified direction evaluation value corresponding to each evaluation indicator. S3: Determine the unified evaluation direction characteristics corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator. The positive retention rule is used to determine the evaluation index corresponding to the monotonically positive correlation type as the corresponding unified direction evaluation value; the monotonically decreasing mapping rule is used to perform monotonically decreasing mapping processing on the evaluation index corresponding to the monotonically negative correlation type to obtain the corresponding unified direction evaluation value; and the optimal contribution interval mapping rule is used to perform optimal contribution interval mapping processing on the evaluation index corresponding to the non-monotonically correlated type to obtain the corresponding unified direction evaluation value.

[0055] The above monotonically decreasing mapping rules include one or more of the following: reciprocal mapping function; linear difference mapping function; exponential decay mapping function; logarithmic decay mapping function.

[0056] The above-mentioned optimal contribution interval mapping rules include one or more of the following: interval mapping function; peak mapping function; Gaussian mapping function; piecewise mapping function.

[0057] In practical implementation, after determining the contribution relationship type corresponding to each evaluation indicator, evaluation direction mapping rules can be configured for different contribution relationship types to transform each evaluation indicator in the multi-dimensional evaluation indicator system into a feature expression with a unified evaluation direction. Specifically, for evaluation indicators with a monotonically positive correlation contribution relationship type, a positive retention rule is configured; for evaluation indicators with a monotonically negative correlation contribution relationship type, a monotonically decreasing mapping rule is configured; and for evaluation indicators with a non-monotonic correlation contribution relationship type, a contribution interval mapping rule is configured. The evaluation direction mapping rule corresponds one-to-one with each evaluation indicator and is used to determine the unified directional evaluation value of each evaluation indicator after mapping processing.

[0058] For evaluation indicators of the monotonic positive correlation type, the indicator value can be used as the corresponding unified direction evaluation value according to the positive retention rule. For example, when indicators such as total organic carbon content, effective thickness, porosity, permeability, oil saturation, and nanopore connectivity increase, it usually indicates that the oil and gas enrichment conditions, storage capacity, or seepage conditions tend to be better. Therefore, such indicators do not need to be reversed in terms of contribution direction and can retain their original evaluation direction. For such evaluation indicators, their original indicator value or the indicator value after dimension unification can be determined as the corresponding unified direction evaluation value, so that when the unified direction evaluation value increases, it still indicates that the target evaluation result tends to be better.

[0059] For evaluation indicators of the monotonically negative correlation type, the indicator value can be reversed according to the monotonically decreasing mapping rule to obtain the corresponding unified directional evaluation value. For example, when indicators such as crude oil viscosity, starting pressure gradient, fluid flow resistance, fracturing stimulation difficulty, single-well production decline rate, and nanopore tortuosity increase, it usually indicates increased fluid migration difficulty, reduced development effect, or intensified production decline. Therefore, such indicators cannot be directly used as positive evaluation inputs. In specific implementation, one of the following methods can be used: reciprocal mapping, linear difference mapping, exponential decay mapping, or logarithmic decay mapping, to perform a monotonically decreasing transformation on the indicator value of this type of evaluation indicator. This ensures that the transformed unified directional evaluation value decreases as the original negative indicator worsens and increases as the original negative indicator improves.

[0060] For evaluation indicators that are not monotonically correlated, their values ​​can be mapped to intervals according to contribution interval mapping rules to obtain corresponding unified direction evaluation values. For example, indicators such as vitrinite reflectance, formation pressure coefficient, brittleness index, and fracture density are not simply better the higher they are; rather, their contribution to the target evaluation result is higher within a suitable contribution interval, and the contribution decreases when deviating from this interval. In practice, the contribution interval corresponding to this type of evaluation indicator can be determined based on historical production data statistical analysis, reservoir development dynamic analysis, domain knowledge constraints, machine learning fitting, or numerical simulation inversion results. Then, one of interval mapping, peak mapping, Gaussian mapping, or piecewise mapping can be used to map the indicator values ​​within the contribution interval to higher unified direction evaluation values, and the indicator values ​​below or above the contribution interval to lower unified direction evaluation values.

[0061] After mapping each evaluation indicator, a unified evaluation direction feature can be determined based on the unified evaluation value corresponding to each indicator. This unified evaluation direction feature serves as the input feature for the subsequent monotonic constraint comprehensive evaluation model, ensuring that evaluation indicators from different sources, with different physical meanings, and with different contribution relationships are expressed according to a unified evaluation direction. In other words, after the unified evaluation direction mapping process, each unified evaluation direction feature can represent the degree of positive contribution of the corresponding evaluation indicator to the target evaluation result.

[0062] Through the above implementation methods, before comprehensive evaluation modeling, evaluation indicators of monotonically positive correlation, monotonically negative correlation, and non-monotonic correlation types can be adapted by mapping their evaluation directions. This transforms evaluation indicators with inconsistent contribution directions or interval contribution patterns into indicators with unified evaluation direction characteristics. This reduces conflicts in evaluation directions among multi-source heterogeneous evaluation indicators, enabling the subsequent monotonically constrained comprehensive evaluation model to perform comprehensive evaluation based on a unified contribution direction, thereby improving the accuracy, stability, and interpretability of the comprehensive evaluation index and results.

[0063] In some embodiments, the method for determining the unified evaluation direction feature corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator may further include the following: S1: Determine the distribution parameters of the evaluation values ​​corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator; S2: Based on the evaluation value distribution parameters corresponding to each evaluation indicator, the unified direction evaluation value corresponding to each evaluation indicator is normalized to obtain the unified dimension evaluation value corresponding to each evaluation indicator. S3: Determine the unified evaluation direction characteristics of each evaluation indicator based on the unified evaluation value of each indicator.

[0064] In practice, after completing the unified evaluation direction mapping process for each evaluation indicator, the obtained unified direction evaluation values ​​can be subjected to dimensional unification processing to avoid the impact of different evaluation indicators on subsequent comprehensive evaluation due to differences in value range, unit, or order of magnitude. For multiple evaluation units in the target low-permeability reservoir, the unified direction evaluation values ​​corresponding to each evaluation indicator can be obtained separately, and the unified direction evaluation values ​​can be statistically analyzed according to the evaluation indicator dimensions to obtain the evaluation value distribution parameters corresponding to each evaluation indicator.

[0065] For any evaluation index, the evaluation value distribution parameter may include at least one of the following: the maximum unified direction evaluation value, the minimum unified direction evaluation value, the mean, or the standard deviation of the evaluation index across multiple evaluation units. Taking normalization as an example, the normalization parameter corresponding to the evaluation index can be determined based on the maximum and minimum unified direction evaluation values. Based on the normalization parameter, the unified direction evaluation values ​​of the evaluation index in each evaluation unit are normalized to obtain the dimensional unified evaluation value corresponding to the evaluation index. Through this process, the dimensional unified evaluation values ​​corresponding to different evaluation indices can fall within a consistent data range, facilitating subsequent unified input into the monotonic constraint comprehensive evaluation model.

[0066] For monotonically positively correlated evaluation indicators, such as porosity, permeability, effective thickness, or oil saturation, a unified directional evaluation value can be obtained after retaining the positive correlation. This value can then be normalized according to the distribution parameters of the corresponding evaluation values ​​to obtain a unified dimensional evaluation value. For monotonically negatively correlated evaluation indicators, such as crude oil viscosity, starting pressure gradient, fluid flow resistance, or fracturing difficulty, a unified directional evaluation value can be obtained first through a monotonically decreasing mapping. This unified directional evaluation value can then be normalized. Since the direction of this type of indicator has already been transformed, the normalized unified dimensional evaluation value still represents the degree of positive contribution of the evaluation indicator to the target evaluation result.

[0067] For evaluation indicators that are not monotonically correlated, such as vitrinite reflectance, formation pressure coefficient, brittleness index, or fracture density, their values ​​can first be converted into uniform-direction evaluation values ​​according to contribution interval mapping rules. Values ​​within the contribution interval correspond to higher uniform-direction evaluation values, while values ​​deviating from the contribution interval correspond to lower uniform-direction evaluation values. Based on the distribution parameters of the evaluation values ​​corresponding to this non-monotonic correlation type of evaluation indicator, the uniform-direction evaluation values ​​are normalized to obtain dimensionally unified evaluation values, enabling these dimensionally unified evaluation values ​​to characterize the contribution of the evaluation indicator on a unified data scale.

[0068] After obtaining the unified evaluation values ​​for each evaluation indicator, these unified evaluation values ​​can be determined as the unified evaluation direction features for each evaluation indicator. These unified evaluation direction features retain the positive contribution meaning of each evaluation indicator after mapping to the unified evaluation direction, while eliminating differences in dimensions and value ranges between different evaluation indicators. They can be used as input data for a pre-defined monotonic constraint comprehensive evaluation model to calculate the comprehensive evaluation index.

[0069] Through the above implementation methods, after the unified evaluation direction mapping process, the unified evaluation value can be further normalized, so that evaluation indicators from different sources, with different physical meanings and different orders of magnitude can participate in the comprehensive evaluation under a unified data scale, reducing the impact of dimensional differences on the comprehensive evaluation index, thereby improving the accuracy and stability of the comprehensive evaluation results of the target low-permeability reservoir.

[0070] In some embodiments, the method of obtaining a comprehensive evaluation index by utilizing a preset monotonic constraint comprehensive evaluation model based on the unified evaluation direction characteristics may further include the following: S1: Determine the set of unified evaluation direction features based on the unified evaluation direction features corresponding to each evaluation indicator; S2: Determine the comprehensive evaluation feature data based on the unified evaluation direction feature set; S3: Input the comprehensive evaluation feature data into the preset monotonic constraint comprehensive evaluation model to obtain the initial evaluation index output by the preset monotonic constraint comprehensive evaluation model; S4: Based on the aforementioned contribution consistency constraint, the initial evaluation index is constrained to obtain the comprehensive evaluation index.

[0071] In practical implementation, the unified evaluation direction features corresponding to each evaluation indicator can be organized according to the evaluation units of the target low-permeability reservoir to form a unified evaluation direction feature set. The evaluation unit can be a well, a section, a grid unit, or a reservoir evaluation area. For any evaluation unit, the unified evaluation direction features obtained after the corresponding evaluation indicators such as porosity, permeability, oil saturation, starting pressure gradient, crude oil viscosity, fracturing difficulty, brittleness index, or fracture density have been mapped by the unified evaluation direction and processed by dimension unification are combined according to a preset indicator order to obtain the comprehensive evaluation feature data corresponding to that evaluation unit.

[0072] After obtaining the comprehensive evaluation feature data, it can be input into a preset monotonic constraint comprehensive evaluation model. The preset monotonic constraint comprehensive evaluation model can be one of the following: a tree model with monotonic constraints, a gradient boosting tree model, a random forest model, a neural network model, a statistical model, or a weighted summation model. The model performs comprehensive evaluation processing on the comprehensive evaluation feature data to obtain the initial evaluation index for the corresponding evaluation unit. The initial evaluation index is used to characterize the initial comprehensive evaluation level of the target low-permeability reservoir under the current evaluation task.

[0073] Since all unified evaluation direction features have been processed into feature expressions aligned with the target evaluation result, the initial evaluation index can be constrained during model calculation based on the contribution consistency constraint. This contribution consistency constraint stipulates that, with other unified evaluation direction features remaining unchanged, the evaluation index output by the model does not decrease when any unified evaluation direction feature increases. This constraint allows for the correction of initial evaluation indices that do not satisfy a monotonically non-decreasing relationship, resulting in a comprehensive evaluation index that satisfies a monotonically non-decreasing relationship relative to the unified evaluation direction features corresponding to each evaluation indicator.

[0074] For example, in reservoir quality evaluation, when the unified evaluation direction characteristics formed after mapping evaluation indicators such as porosity, permeability, and oil saturation through a unified evaluation direction increase, the comprehensive evaluation index should not decrease. Similarly, in fracturing effect prediction, when the unified evaluation direction characteristics obtained after mapping fracturing stimulation difficulty through a monotonically decreasing process increase, it indicates that fracturing stimulation conditions are becoming more favorable, and the comprehensive evaluation index should not decrease. For non-monotonic correlated indicators such as brittleness index or fracture density, their values ​​are mapped through contribution intervals to obtain unified evaluation direction characteristics. If these unified evaluation direction characteristics increase, it indicates that the indicator's contribution to the target evaluation result is increasing, and the comprehensive evaluation index should also remain unchanged.

[0075] Through the above implementation methods, unified evaluation directional features can be organized into comprehensive evaluation feature data, and a comprehensive evaluation index can be generated using a monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint condition. This ensures that the comprehensive evaluation index maintains a monotonic non-decreasing relationship with each unified evaluation directional feature, thereby reducing the impact of inconsistent evaluation index directions or reverse fluctuations in model output on the comprehensive evaluation results and improving the accuracy, stability, and interpretability of the comprehensive evaluation of low-permeability reservoirs.

[0076] In some embodiments, the method for determining the multidimensional evaluation index system of the target low-permeability reservoir and the target evaluation result corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters may further include the following: S1: Determine the parameter category data of the target low-permeability reservoir based on the multi-source heterogeneous parameters; S2: Determine candidate evaluation index data based on the parameter category data; S3: Based on the candidate evaluation index data, determine the multidimensional evaluation index system for the target low-permeability reservoir; S4: Determine the evaluation result type corresponding to the multidimensional evaluation index system based on the multidimensional evaluation index system; S5: Determine the target evaluation result corresponding to the multidimensional evaluation index system based on the evaluation result type.

[0077] In practice, the multi-source heterogeneous parameters of the target low-permeability reservoir can be categorized first, into at least one of the following: geological data, reservoir data, fluid data, and engineering data. Geological data may include total organic carbon content, vitrinite reflectance, effective thickness, and oil saturation; reservoir data may include porosity, permeability, brittleness index, fracture development degree, pore throat radius, nanopore tortuosity, and nanopore connectivity; fluid data may include crude oil viscosity, starting pressure gradient, fluid flow resistance, and gas content; and engineering data may include fracturing stimulation difficulty, operational displacement, proppant addition scale, and single-well production decline rate. This categorization allows parameters from different sources and with different physical meanings to be placed into their corresponding data categories, facilitating the subsequent construction of evaluation indicators.

[0078] After determining the parameter categories, candidate evaluation index data can be identified based on the characterization effect of each parameter on the target low-permeability reservoir. For example, candidate evaluation indexes such as total organic carbon content, vitrinite reflectance, effective thickness, and oil saturation can be determined based on geological category data; porosity, permeability, brittleness index, fracture development degree, pore throat radius, nanopore tortuosity, and nanopore connectivity can be determined based on reservoir category data; crude oil viscosity, starting pressure gradient, fluid flow resistance, and gas content can be determined based on fluid category data; and fracturing stimulation difficulty, operational displacement, proppant addition scale, and single-well production decline rate can be determined based on engineering category data. These candidate evaluation index data can cover the geological conditions, reservoir properties, fluid flow characteristics, and engineering stimulation conditions of the target low-permeability reservoir.

[0079] After obtaining the candidate evaluation index data, a multidimensional evaluation index system can be determined based on the correlation between the candidate evaluation index data and the evaluation requirements of the target low-permeability reservoir. For reservoir quality evaluation, candidate evaluation indices that can characterize reservoir enrichment conditions, storage capacity, and permeability can be selected; for sweet spot evaluation or target selection, candidate evaluation indices that can reflect oil and gas enrichment, reservoir availability, and engineering modification potential can be selected; for development potential prediction, fracturing effect prediction, or single-well production capacity prediction, candidate evaluation indices that can reflect development response, fracturing stimulation conditions, and production capacity change trends can be selected. Through this process, the multidimensional evaluation index system includes multiple evaluation indices, and each evaluation index is derived from the aforementioned candidate evaluation index data.

[0080] After determining the multidimensional evaluation index system, the corresponding evaluation result type can be determined based on the evaluation content represented by the multidimensional evaluation index system. If the multidimensional evaluation index system is mainly used to characterize reservoir geology and reservoir physical properties, the evaluation result type can be determined as reservoir quality evaluation type; if the multidimensional evaluation index system is used to characterize favorable reservoir areas or preferred target areas, the evaluation result type can be determined as sweet spot evaluation type or target preference type; if the multidimensional evaluation index system is used to characterize development response capability, fracturing effect, or production capacity, the evaluation result type can be determined as development potential prediction type, fracturing effect prediction type, or single well production capacity prediction type.

[0081] After determining the evaluation result type, the target evaluation result corresponding to the multi-dimensional evaluation index system can be determined based on the evaluation result type. For example, when the evaluation result type is reservoir quality evaluation, the target evaluation result can be reservoir quality grade or reservoir quality evaluation value; when the evaluation result type is sweet spot evaluation or target optimization, the target evaluation result can be sweet spot grade, favorable area identification result, or target optimization result; when the evaluation result type is development potential prediction, the target evaluation result can be development potential grade; when the evaluation result type is fracturing effect prediction, the target evaluation result can be fracturing effect grade; when the evaluation result type is single-well production capacity prediction, the target evaluation result can be single-well production capacity prediction value or single-well production capacity grade. Therefore, the target evaluation result can serve as a reference for subsequently determining the contribution relationship type of each evaluation index.

[0082] Through the above implementation methods, the multi-source heterogeneous parameters of the target low-permeability reservoir can be classified into parameter category data. Candidate evaluation index data, a multi-dimensional evaluation index system, evaluation result types, and target evaluation results are then gradually determined from the parameter category data, establishing a clear correspondence between the multi-dimensional evaluation index system and the target evaluation results. This provides a foundation for subsequently identifying the contribution relationship between each evaluation index and the target evaluation result, avoiding problems such as unclear sources of evaluation indicators and ambiguous evaluation result objects, thereby improving the logical consistency of the comprehensive evaluation process for low-permeability reservoirs and the accuracy of the evaluation results.

[0083] In some embodiments, after obtaining the multi-source heterogeneous parameters of the target low-permeability reservoir, the method may further include the following: S1: Determine the parameter preprocessing rules corresponding to the multi-source heterogeneous parameters based on the multi-source heterogeneous parameters; S2: According to the parameter preprocessing rules, the multi-source heterogeneous parameters are preprocessed to obtain the preprocessed multi-source heterogeneous parameters; The parameter preprocessing rules include at least one of the following: outlier removal rules, missing value completion rules, noise filtering rules, unit unification rules, and data consistency verification rules.

[0084] In practice, after acquiring the multi-source heterogeneous parameters of the target low-permeability reservoir, the data source, data format, unit type, and value status of the multi-source heterogeneous parameters can be identified first to determine the parameter preprocessing rules corresponding to the multi-source heterogeneous parameters. The multi-source heterogeneous parameters can include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters. Different types of parameters have different data sources and physical meanings; therefore, corresponding preprocessing methods need to be configured according to the data characteristics of the parameters. For example, for parameters obtained from well logging interpretation, core experiments, production dynamics, or fracturing operation data, it is possible to identify whether there are outliers, missing values, noisy data, inconsistent units, or inconsistent data calibers.

[0085] For multi-source heterogeneous parameters with outlier values, outlier removal rules can be applied to process the parameters. Specifically, the normal value range of a parameter can be determined based on its value distribution in different wells, different sections, or different evaluation units within the target low-permeability reservoir. When a parameter value exceeds the normal value range and does not match the geological understanding or development dynamics of the target low-permeability reservoir, the parameter value is identified as an outlier and removed or replaced. For example, for parameters such as porosity, permeability, oil saturation, crude oil viscosity, or starting pressure gradient, significantly deviating outliers can be identified by combining historical measured data and regional statistical ranges.

[0086] For multi-source heterogeneous parameters with missing values, the parameters can be processed according to missing value completion rules. Specifically, the completion value corresponding to the missing parameter can be determined based on other parameters already obtained in the same evaluation unit, parameters of adjacent layers, parameters of adjacent wells, or historical parameters of the same type of reservoir; the missing parameter is then completed based on the completion value to obtain complete parameter data. For parameters such as total organic carbon content, vitrinite reflectance, pore throat radius, nanopore connectivity, construction displacement, or sand addition scale, if corresponding data are missing in some evaluation units, they can be completed based on statistical values ​​of adjacent evaluation units or the correlation between similar parameters.

[0087] For multi-source heterogeneous parameters that are subject to acquisition noise or fluctuation interference, the parameters can be processed according to noise filtering rules. Specifically, noise data can be determined based on the time series changes, spatial distribution changes, or repeated measurement results of the parameters; the noise data can then be smoothed, filtered, or statistically corrected to obtain the parameter values ​​after noise filtering. For example, for single-well production decline rate, fluid flow resistance in production dynamics data, or construction displacement in fracturing construction data, short-term abnormal fluctuations can be removed based on their changing trends to make them more consistent with the actual development response of the target low-permeability reservoir.

[0088] For heterogeneous parameters from different sources with inconsistent units, the parameters can be processed according to a unified unit rule. Specifically, a standard unit can be determined based on the physical quantity type of each parameter; then, unit conversion can be performed on parameter values ​​from different sources using the standard unit to obtain parameter data with consistent units. For example, unit conversion can be performed on parameters such as permeability, porosity, crude oil viscosity, starting pressure gradient, effective thickness, construction discharge rate, and sand addition scale to avoid parameters with the same physical meaning affecting the determination of subsequent multidimensional evaluation index systems due to different units.

[0089] For heterogeneous parameters from multiple sources with inconsistent data calibers or parameter correspondences, data consistency verification rules can be applied. Specifically, based on the correspondence between wells, intervals, and evaluation units in the target low-permeability reservoir, the spatial location, stratigraphic range, and evaluation unit consistency of geological parameters, reservoir parameters, fluid parameters, and engineering parameters can be verified. When a parameter does not match the target evaluation unit, it is re-matched, removed, or corrected. This process ensures that heterogeneous parameters from multiple sources within the same evaluation unit have consistent data calibers.

[0090] After the above preprocessing, preprocessed multi-source heterogeneous parameters can be obtained. These preprocessed multi-source heterogeneous parameters are used to determine the multidimensional evaluation index system for the target low-permeability reservoir, ensuring that the data used to determine subsequent evaluation indicators, contribution relationship types, unified evaluation direction characteristics, and comprehensive evaluation indices has good completeness, consistency, and usability.

[0091] Through the above implementation methods, outlier removal, missing value completion, noise filtering, unit unification, and data consistency verification can be performed on multi-source heterogeneous parameters before constructing a multi-dimensional evaluation index system. This reduces the impact of differences in the quality of original parameters on the determination of evaluation indicators and the calculation of comprehensive evaluation index, thereby improving the accuracy and stability of the comprehensive evaluation results of the target low-permeability reservoir.

[0092] As can be seen from the above, the embodiments of this specification provide a comprehensive evaluation method for low-permeability reservoirs, which obtains multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation result corresponding to the multi-dimensional evaluation index system are determined; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined; wherein, the contribution relationship type includes monotonic positive correlation type, single... The evaluation criteria are: negative correlation type and non-monotonic correlation type; based on the contribution relationship type corresponding to each evaluation index, a unified evaluation direction mapping process is performed on each evaluation index to obtain the unified evaluation direction feature corresponding to each evaluation index; using a preset monotonic constraint comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction feature; wherein, the preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, which is used to ensure that the comprehensive evaluation index satisfies a monotonically non-decreasing relationship with respect to the unified evaluation direction feature corresponding to each evaluation index; based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined. In this way, by acquiring multi-source heterogeneous parameters of the target low-permeability reservoir and determining the multi-dimensional evaluation index system and its corresponding target evaluation results, the contribution relationship between each evaluation index and the target evaluation results can be clarified. By determining the contribution relationship type corresponding to each evaluation index and performing unified evaluation direction mapping processing on each evaluation index based on the contribution relationship type, a unified evaluation direction feature can be obtained. This can unify the contribution direction between different evaluation indicators and the target evaluation results, reducing the impact of inconsistent evaluation logic on the comprehensive evaluation results. At the same time, by using a pre-set monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint condition, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. This ensures that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction feature corresponding to each evaluation index, thereby improving the accuracy and stability of the comprehensive evaluation results of the target low-permeability reservoir.

[0093] See Figure 2 As shown in the embodiments of this specification, a specific electronic device is also provided, wherein the electronic device includes a network communication port 201, a processor 202 and a memory 203, and the above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0094] Specifically, the network communication port 201 can be used to acquire multi-source heterogeneous parameters of the target low-permeability reservoir; wherein the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters.

[0095] The processor 202 is specifically configured to determine, based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation result corresponding to the multi-dimensional evaluation index system; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; determine the contribution relationship type corresponding to each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result; wherein, the contribution relationship type includes monotonically positive correlation, monotonically negative correlation, and non-monotonic correlation; perform unified evaluation direction mapping processing on each evaluation indicator based on the contribution relationship type corresponding to each evaluation indicator to obtain unified evaluation direction features corresponding to each evaluation indicator; use a preset monotonically constrained comprehensive evaluation model to obtain a comprehensive evaluation index based on the unified evaluation direction features; wherein, the preset monotonically constrained comprehensive evaluation model satisfies the contribution consistency constraint condition, the contribution consistency constraint condition being used to ensure that the comprehensive evaluation index satisfies a monotonically non-decreasing relationship with respect to the unified evaluation direction features corresponding to each evaluation indicator; and determine the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index.

[0096] The memory 203 can be used to store the corresponding instruction program.

[0097] Based on the above method, the relevant structural performance of electronic equipment can be effectively utilized to improve the data processing speed of electronic equipment and efficiently realize a comprehensive evaluation method for low-permeability reservoirs.

[0098] In this embodiment, the network communication port 201 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0099] In this embodiment, the processor 202 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0100] In this embodiment, the memory 203 may include a hierarchy. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0101] This specification also provides a computer-readable storage medium based on the above-described comprehensive evaluation method for low-permeability reservoirs, which acquires multi-source heterogeneous parameters of a target low-permeability reservoir; wherein the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and a target evaluation result corresponding to the multi-dimensional evaluation index system are determined; wherein the multi-dimensional evaluation index system includes multiple evaluation indicators; based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined; wherein the contribution relationship type includes monotonically positive phase. The evaluation indicators are categorized into three types: correlation type, monotonically negative correlation type, and non-monotonic correlation type. Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. Using a preset monotonically constrained comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. The preset monotonically constrained comprehensive evaluation model satisfies the contribution consistency constraint condition, which is used to ensure that the comprehensive evaluation index satisfies a monotonically non-decreasing relationship with respect to the unified evaluation direction feature corresponding to each evaluation indicator. Based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined.

[0102] In this embodiment, the storage medium includes, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured according to standards specified in the communication protocol for network connection communication.

[0103] In this embodiment, the specific functions and effects implemented by the program instructions stored in the computer-readable storage medium can be explained in comparison with other embodiments, and will not be repeated here.

[0104] See Figure 3 At the software level, this specification also provides a comprehensive evaluation device for low-permeability reservoirs, which may specifically include the following structural modules: The parameter acquisition module 301 is used to acquire multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters and engineering parameters; The system determination module 302 is used to determine the multidimensional evaluation index system of the target low-permeability reservoir and the target evaluation result corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters; wherein, the multidimensional evaluation index system includes multiple evaluation indicators; The type determination module 303 is used to determine the contribution relationship type corresponding to each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type and non-monotonic correlation type; The feature determination module 304 is used to perform unified evaluation direction mapping processing on each evaluation indicator according to the contribution relationship type corresponding to each evaluation indicator, so as to obtain the unified evaluation direction feature corresponding to each evaluation indicator. The index determination module 305 is used to obtain a comprehensive evaluation index based on the unified evaluation direction characteristics using a preset monotonic constraint comprehensive evaluation model; wherein, the preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, and the contribution consistency constraint condition is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. The result determination module 306 is used to determine the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index.

[0105] In some embodiments, the type determination module 303, in specific implementation, determines the contribution change relationship corresponding to each evaluation indicator based on the correspondence between the changes in the indicator values ​​of each evaluation indicator and the changes in the target evaluation result; when the contribution change relationship indicates that the indicator value of the evaluation indicator increases and the target evaluation result improves, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonically positive correlation type; when the contribution change relationship indicates that the indicator value of the evaluation indicator increases and the target evaluation result decreases, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonically negative correlation type; when the contribution change relationship indicates that there is an optimal contribution interval between the indicator value of the evaluation indicator and the target evaluation result, the contribution relationship type corresponding to the evaluation indicator is determined to be a non-monotonic correlation type.

[0106] In some embodiments, the feature determination module 304, in specific implementation, determines the evaluation direction mapping rule corresponding to each evaluation indicator based on the contribution relationship type corresponding to each evaluation indicator; wherein, the evaluation direction mapping rule includes a positive retention rule, a monotonically decreasing mapping rule, and an optimal contribution interval mapping rule; according to the evaluation direction mapping rule corresponding to each evaluation indicator, the evaluation indicators are mapped to obtain a unified direction evaluation value corresponding to each evaluation indicator; the direction feature determination module is used to determine the unified evaluation direction feature corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator; wherein, the positive retention rule is used to determine the evaluation indicators corresponding to the monotonically positive correlation type as the corresponding unified direction evaluation value, the monotonically decreasing mapping rule is used to perform monotonically decreasing mapping processing on the evaluation indicators corresponding to the monotonically negative correlation type to obtain the corresponding unified direction evaluation value; the optimal contribution interval mapping rule is used to perform optimal contribution interval mapping processing on the evaluation indicators corresponding to the non-monotonically correlated type to obtain the corresponding unified direction evaluation value.

[0107] In some embodiments, the directional feature determination module, in specific implementation, determines the evaluation value distribution parameters corresponding to each evaluation indicator based on the unified directional evaluation value corresponding to each evaluation indicator; performs normalization processing on the unified directional evaluation value corresponding to each evaluation indicator based on the evaluation value distribution parameters corresponding to each evaluation indicator to obtain the dimensional unified evaluation value corresponding to each evaluation indicator; and determines the unified evaluation directional feature corresponding to each evaluation indicator based on the dimensional unified evaluation value corresponding to each evaluation indicator.

[0108] In some embodiments, the index determination module 305, in specific implementation, determines a unified evaluation direction feature set based on the unified evaluation direction features corresponding to each evaluation indicator; determines comprehensive evaluation feature data based on the unified evaluation direction feature set; inputs the comprehensive evaluation feature data into the preset monotonic constraint comprehensive evaluation model to obtain the initial evaluation index output by the preset monotonic constraint comprehensive evaluation model; and performs constraint processing on the initial evaluation index according to the contribution consistency constraint condition to obtain the comprehensive evaluation index.

[0109] In some embodiments, the result determination module 306, in specific implementation, determines the parameter category data of the target low-permeability reservoir based on the multi-source heterogeneous parameters; determines candidate evaluation index data based on the parameter category data; determines a multi-dimensional evaluation index system for the target low-permeability reservoir based on the candidate evaluation index data; determines the evaluation result type corresponding to the multi-dimensional evaluation index system based on the multi-dimensional evaluation index system; and determines the target evaluation result corresponding to the multi-dimensional evaluation index system based on the evaluation result type.

[0110] In some embodiments, after the parameter acquisition module 301 described above, in specific implementation, the parameter preprocessing rules corresponding to the multi-source heterogeneous parameters are determined according to the multi-source heterogeneous parameters; the multi-source heterogeneous parameters are preprocessed according to the parameter preprocessing rules to obtain preprocessed multi-source heterogeneous parameters; wherein, the parameter preprocessing rules include at least one of outlier removal rules, missing value completion rules, noise filtering rules, unit unification rules, and data consistency verification rules.

[0111] It should be noted that the units, devices, or modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above devices are described by dividing them into various modules according to their functions. Of course, in implementing this specification, the functions of each module can be implemented in the same software and / or hardware, or modules that implement the same function can be implemented by a combination of sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. For example, units or components can be combined or integrated into another system, or some features can be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection between the devices or units shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0112] As can be seen from the above, based on the low-permeability reservoir comprehensive evaluation device provided in the embodiments of this specification, multi-source heterogeneous parameters of the target low-permeability reservoir are obtained; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation result corresponding to the multi-dimensional evaluation index system are determined; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined; wherein, the contribution relationship type includes monotonic positive correlation type, The evaluation criteria are: monotonically negatively correlated and non-monotonically correlated. Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. Using a preset monotonically constrained comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction feature. The preset monotonically constrained comprehensive evaluation model satisfies the contribution consistency constraint condition, which ensures that the comprehensive evaluation index, relative to the unified evaluation direction feature corresponding to each evaluation indicator, satisfies a monotonically non-decreasing relationship. Based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined.

[0113] In a specific scenario example, the comprehensive evaluation method and apparatus for low-permeability reservoirs provided in this specification can be applied. This solves the problem that existing comprehensive evaluation methods for low-permeability reservoirs struggle to unify the contribution directions between multi-source heterogeneous evaluation indicators and target evaluation results, leading to poor accuracy and insufficient stability of the comprehensive evaluation results. The specific implementation process may include the following:

[0114] In some embodiments, a shale oil block is selected as the target low-permeability reservoir, and the comprehensive evaluation method for low-permeability reservoirs described in this specification is used to evaluate the reservoir quality, sweet spot, select the target, and identify favorable areas for the target low-permeability reservoir.

[0115] S1: Data Acquisition.

[0116] Collect multi-source heterogeneous parameters of the target low-permeability reservoir, including geological parameters, reservoir parameters, fluid parameters and engineering parameters, and establish an evaluation index database based on the multi-source heterogeneous parameters.

[0117] The geological parameters include total organic carbon content, vitrinite reflectance, effective thickness, and oil saturation; the reservoir parameters include porosity, permeability, brittleness index, fracture development degree, pore throat radius, nanopore tortuosity, and nanopore connectivity; the fluid parameters include crude oil viscosity, starting pressure gradient, fluid flow resistance, and gas content; and the engineering parameters include fracturing stimulation difficulty, construction displacement, sand addition scale, and single-well production decline rate.

[0118] S2: Contribution Relationship Identification.

[0119] Based on historical development data, domain knowledge, and statistical analysis results, the contribution relationship between each evaluation indicator in the multidimensional evaluation indicator system and the target evaluation result is identified, and the contribution relationship type corresponding to each evaluation indicator is determined.

[0120] Among them, the evaluation indicators with monotonic positive correlation include total organic carbon content, effective thickness, porosity, permeability, oil saturation, and nanopore connectivity. Increased values ​​of these evaluation indicators generally contribute positively to oil and gas enrichment, improved storage capacity, or increased production capacity, thus showing a positive contribution to the target evaluation results.

[0121] Evaluation indicators exhibiting monotonically negative correlations include crude oil viscosity, initiation pressure gradient, fluid flow resistance, fracturing stimulation difficulty, single-well production decline rate, and nanopore tortuosity. Increased values ​​of these indicators typically increase fluid migration difficulty, reduce development effectiveness, or accelerate production decline, thus contributing negatively to the target evaluation results. Specifically, increased nanopore tortuosity implies an increased fluid migration path and seepage resistance, generally showing a negative correlation with development effectiveness.

[0122] Evaluation indicators that are not monotonically correlated include vitrinite reflectance, formation pressure coefficient, brittleness index, and fracture density. The relationship between these indicators and the target evaluation results is not a simple monotonically positive or negative correlation, but rather typically exhibits a suitable contribution range. This suitable contribution range can be determined using one or more methods, such as historical production data statistical analysis, reservoir development dynamic analysis, domain knowledge constraints, machine learning fitting, or numerical simulation inversion.

[0123] S3: Construction of unified evaluation direction features.

[0124] Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator.

[0125] For evaluation indicators of the monotonic positive correlation type, the positive retention rule is adopted to determine the indicator value or the indicator value after dimension unification as the corresponding unified direction evaluation value.

[0126] For evaluation indicators of the monotonically negative correlation type, a monotonically decreasing mapping rule is used to transform the indicator value to obtain a corresponding unified direction evaluation value. The monotonically decreasing mapping rule may include one or more of the following: reciprocal mapping rule, linear difference mapping rule, exponential decay mapping rule, or logarithmic decay mapping rule.

[0127] For evaluation indicators that are not monotonically correlated, a contribution interval mapping rule is used to transform the indicator value to obtain a corresponding unified direction evaluation value. The contribution interval mapping rule may include one or more of the following: interval mapping rule, peak mapping rule, Gaussian mapping rule, or piecewise mapping rule.

[0128] After the above mapping process, the evaluation indicators of different contribution relationship types are all converted into unified evaluation direction features, so that each unified evaluation direction feature can characterize the positive contribution of the corresponding evaluation indicator to the target evaluation result.

[0129] S4: Unify the dimensions of the evaluation direction characteristics.

[0130] The evaluation values ​​corresponding to each evaluation indicator are processed to unify their dimensions, resulting in the unified evaluation direction characteristics for each indicator. This unification process can employ range standardization, Z-score standardization, or normalization.

[0131] By unifying the dimensions, the impact of differences in units, value ranges, or orders of magnitude between different evaluation indicators can be reduced, enabling evaluation indicators from different sources such as geology, reservoirs, fluids, and engineering to participate in comprehensive evaluation under a unified data scale.

[0132] S5: Determination of the comprehensive evaluation index.

[0133] Based on the unified evaluation direction characteristics corresponding to each evaluation indicator, a unified evaluation direction characteristic set is determined, and comprehensive evaluation characteristic data is determined based on the unified evaluation direction characteristic set.

[0134] The comprehensive evaluation feature data is input into a preset monotonic constraint comprehensive evaluation model to obtain the comprehensive evaluation index. The preset monotonic constraint comprehensive evaluation model can be one or more of the following: weighted summation model, statistical model, machine learning model, random forest model, gradient boosting tree model, tree model with monotonic constraints, or neural network model.

[0135] The preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, which ensures that the comprehensive evaluation index, relative to the unified evaluation direction feature corresponding to each evaluation indicator, satisfies a monotonic, non-decreasing relationship. In other words, while other unified evaluation direction features remain unchanged, the comprehensive evaluation index does not decrease when any unified evaluation direction feature increases.

[0136] S6: Output of comprehensive evaluation results.

[0137] Based on the comprehensive evaluation index, the comprehensive evaluation results of the target low-permeability reservoir are determined. Specifically, the comprehensive evaluation index can be used to classify reservoir quality, evaluate sweet spots, select the best reservoirs, and identify favorable areas for the target low-permeability reservoir.

[0138] For example, evaluation units can be divided into high-quality reservoirs, medium-quality reservoirs, and low-quality reservoirs based on the magnitude of the comprehensive evaluation index; different evaluation units can also be ranked according to the comprehensive evaluation index, and evaluation units with higher comprehensive evaluation indices can be identified as sweet spots or preferred target areas.

[0139] This embodiment can convert multiple evaluation indicators corresponding to geological parameters, reservoir parameters, fluid parameters, and engineering parameters into unified evaluation direction features, and use a monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint to determine the comprehensive evaluation index. This reduces the impact of inconsistent contribution directions of evaluation indicators on reservoir quality evaluation and sweet spot evaluation, and improves the accuracy, stability, and physical interpretability of sweet spot identification results.

[0140] In some embodiments, the same data acquisition, contribution relationship identification, unified evaluation direction feature construction, and comprehensive evaluation index determination process as described in the above embodiments are used to predict the development potential and fracturing effect of the target low-permeability reservoir.

[0141] Unlike the above embodiments, in this embodiment, the target evaluation result corresponds to the development potential prediction result or the fracturing effect prediction result. The preset monotonic constraint comprehensive evaluation model outputs a comprehensive evaluation index to characterize the development potential or fracturing effect based on the unified evaluation direction characteristics corresponding to each evaluation index.

[0142] In development potential prediction, the development potential level of each evaluation unit in a target low-permeability reservoir can be determined based on a comprehensive evaluation index. For example, based on the magnitude of the comprehensive evaluation index, each evaluation unit can be divided into high development potential areas, medium development potential areas, and low development potential areas. Among them, evaluation units with higher comprehensive evaluation indices can be designated as priority development areas, while evaluation units with lower comprehensive evaluation indices can be designated as secondary development areas or areas where development is temporarily suspended.

[0143] In predicting fracturing effectiveness, the fracturing effectiveness level of each evaluation unit in a target low-permeability reservoir can be determined based on a comprehensive evaluation index. For example, based on the magnitude of the comprehensive evaluation index, each evaluation unit can be divided into a region with excellent fracturing effectiveness, a region with moderate fracturing effectiveness, and a region with poor fracturing effectiveness. The fracturing effectiveness level can be characterized by one or more of the following: post-fracturing production increase, post-fracturing production stabilization time, fracturing stimulation volume response degree, or post-fracturing overall production capacity improvement degree.

[0144] The comprehensive evaluation index can also be used for fracturing parameter optimization, fracturing construction scheme selection, and modification effect prediction. Specifically, evaluation units with higher comprehensive evaluation indices can be identified as preferred fracturing targets, and fracturing construction schemes can be optimized by combining engineering parameters such as construction displacement, sand addition scale, and fracturing modification difficulty.

[0145] This embodiment enables a graded evaluation of the development potential and fracturing effect of target low-permeability reservoirs based on the unified contribution direction of different evaluation indicators. It reduces the impact of inconsistent contribution directions of multi-source heterogeneous evaluation indicators on the prediction results and improves the reliability, consistency and interpretability of the prediction results of development potential and fracturing effect.

[0146] In some embodiments, this embodiment uses the same data acquisition, contribution relationship identification, unified evaluation direction feature construction, and comprehensive evaluation index determination process as the above embodiments to predict the single-well productivity of the well to be predicted in the target low-permeability reservoir.

[0147] In this embodiment, historical production data is used as training labels, and a mapping relationship between the unified evaluation direction features and single-well production capacity is established based on the unified evaluation direction features corresponding to each evaluation indicator. The historical production data may include at least one of the following: initial daily oil production, cumulative oil production, predicted final recoverable reserves, and stable production period output.

[0148] Specifically, multi-source heterogeneous parameters of known wells can be obtained, and the multi-dimensional evaluation index system, contribution relationship type, and unified evaluation direction characteristics corresponding to the known wells can be determined according to the aforementioned method. Based on the unified evaluation direction characteristics and historical production data corresponding to the known wells, a preset monotonic constraint comprehensive evaluation model is trained so that the model can output a comprehensive evaluation index to characterize the productivity level of a single well based on the unified evaluation direction characteristics.

[0149] After model training is completed, evaluation index data for the well to be predicted is acquired, and the unified evaluation direction feature corresponding to the well is determined based on the evaluation index data. Using the trained monotonic constraint comprehensive evaluation model, a comprehensive evaluation index corresponding to the well is obtained based on the unified evaluation direction feature. The production capacity prediction result for the well is then determined based on the comprehensive evaluation index. The production capacity prediction result can be at least one of the following: initial daily oil production prediction result, cumulative oil production prediction result, predicted final recoverable reserves result, or stable production period production prediction result; it can also be a high-production-capacity level, a medium-production-capacity level, or a low-production-capacity level.

[0150] This embodiment enables the unification of the contribution directions of different evaluation indicators during single-well production capacity prediction. By outputting a comprehensive evaluation index through a monotonic constraint comprehensive evaluation model that satisfies the contribution consistency constraint, the impact of inconsistent directions of multi-source heterogeneous evaluation indicators on production capacity prediction results can be reduced, thereby improving the stability and interpretability of single-well production capacity prediction results and providing a basis for decision-making on well location deployment and development scheme optimization.

[0151] While this specification provides the steps of operation for the methods described in the embodiments or flowcharts, more or fewer steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible order of execution among many steps and does not represent the only possible order. In actual device or client product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded. The terms "first," "second," etc., are used to denote names and do not indicate any particular order.

[0152] Those skilled in the art will also know that, besides implementing the controller using purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller function as logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices within it used to implement various functions can also be considered structures within that hardware component. Alternatively, the devices used to implement various functions can be considered as both software modules implementing the method and structures within a hardware component.

[0153] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments of this specification.

[0154] Although this specification has been described by way of examples, those skilled in the art will recognize that many variations and modifications are possible without departing from the spirit of this specification, and it is intended that the appended claims cover such variations and modifications without departing from the spirit of this specification.

Claims

1. A comprehensive evaluation method for low-permeability oil reservoirs, characterized in that, include: Obtain multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters, and engineering parameters; Based on the multi-source heterogeneous parameters, a multi-dimensional evaluation index system for the target low-permeability reservoir and the target evaluation results corresponding to the multi-dimensional evaluation index system are determined; wherein, the multi-dimensional evaluation index system includes multiple evaluation indicators; Based on the contribution relationship between each evaluation indicator and the target evaluation result, the contribution relationship type corresponding to each evaluation indicator is determined; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type, and non-monotonic correlation type; Based on the contribution relationship type corresponding to each evaluation indicator, a unified evaluation direction mapping process is performed on each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator. Using a pre-defined monotonic constraint comprehensive evaluation model, a comprehensive evaluation index is obtained based on the unified evaluation direction characteristics. The pre-defined monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, which is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. Based on the comprehensive evaluation index, the comprehensive evaluation result of the target low-permeability reservoir is determined.

2. The method of claim 1, wherein, The step of determining the contribution relationship type of each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result includes: Based on the correspondence between the changes in the values ​​of each evaluation indicator and the changes in the target evaluation results, the contribution change relationship of each evaluation indicator is determined. When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result improves, the contribution relationship type corresponding to the evaluation indicator is determined to be a monotonic positive correlation type. When the contribution change relationship indicates that the value of the evaluation indicator increases and the target evaluation result decreases, the contribution relationship type corresponding to the evaluation indicator is determined to be monotonically negatively correlated. When the contribution change relationship indicates that there is an optimal contribution interval between the indicator value and the target evaluation result, the contribution relationship type corresponding to the evaluation indicator is determined to be a non-monotonic correlation type.

3. The method of claim 1, wherein, The step of performing a unified evaluation direction mapping process on each evaluation indicator based on the contribution relationship type corresponding to each evaluation indicator to obtain the unified evaluation direction feature corresponding to each evaluation indicator includes: Based on the contribution relationship type corresponding to each evaluation indicator, the evaluation direction mapping rule corresponding to each evaluation indicator is determined; wherein, the evaluation direction mapping rule includes the positive retention rule, the monotonically decreasing mapping rule, and the optimal contribution interval mapping rule; Based on the evaluation direction mapping rules corresponding to each evaluation indicator, each evaluation indicator is mapped to obtain a unified direction evaluation value corresponding to each evaluation indicator. Based on the unified direction evaluation value corresponding to each evaluation indicator, determine the unified evaluation direction feature corresponding to each evaluation indicator. The positive retention rule is used to determine the evaluation index corresponding to the monotonically positive correlation type as the corresponding unified direction evaluation value; the monotonically decreasing mapping rule is used to perform monotonically decreasing mapping processing on the evaluation index corresponding to the monotonically negative correlation type to obtain the corresponding unified direction evaluation value; and the optimal contribution interval mapping rule is used to perform optimal contribution interval mapping processing on the evaluation index corresponding to the non-monotonically correlated type to obtain the corresponding unified direction evaluation value.

4. The method of claim 3, wherein, The step of determining the unified evaluation direction characteristics corresponding to each evaluation indicator based on the unified direction evaluation value corresponding to each evaluation indicator includes: Based on the unified directional evaluation value corresponding to each evaluation indicator, determine the distribution parameters of the evaluation value corresponding to each evaluation indicator; Based on the evaluation value distribution parameters corresponding to each evaluation indicator, the unified directional evaluation values ​​corresponding to each evaluation indicator are normalized to obtain the unified dimensional evaluation values ​​corresponding to each evaluation indicator. Based on the unified evaluation value of each evaluation indicator, the unified evaluation direction characteristics corresponding to each evaluation indicator are determined.

5. The method of claim 1, wherein, The comprehensive evaluation index obtained by utilizing the preset monotonic constraint comprehensive evaluation model and based on the unified evaluation direction characteristics includes: Based on the unified evaluation direction characteristics corresponding to each evaluation indicator, determine the set of unified evaluation direction characteristics; Based on the unified evaluation direction feature set, determine the comprehensive evaluation feature data; The comprehensive evaluation feature data is input into the preset monotonic constraint comprehensive evaluation model to obtain the initial evaluation index output by the preset monotonic constraint comprehensive evaluation model. Based on the aforementioned contribution consistency constraint, the initial evaluation index is constrained to obtain the comprehensive evaluation index.

6. The method of claim 1, wherein, The step of determining the multidimensional evaluation index system for the target low-permeability reservoir and the target evaluation results corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters includes: Based on the multi-source heterogeneous parameters, determine the parameter category data of the target low-permeability reservoir; Based on the parameter category data, candidate evaluation index data are determined; Based on the candidate evaluation index data, a multidimensional evaluation index system for the target low-permeability reservoir is determined; Based on the multidimensional evaluation index system, determine the evaluation result type corresponding to the multidimensional evaluation index system; Based on the evaluation result type, determine the target evaluation result corresponding to the multidimensional evaluation index system.

7. The method of claim 1, wherein, After obtaining the multi-source heterogeneous parameters of the target low-permeability reservoir, the method further includes: Based on the multi-source heterogeneous parameters, determine the parameter preprocessing rules corresponding to the multi-source heterogeneous parameters; According to the parameter preprocessing rules, the multi-source heterogeneous parameters are preprocessed to obtain the preprocessed multi-source heterogeneous parameters; The parameter preprocessing rules include at least one of the following: outlier removal rules, missing value completion rules, noise filtering rules, unit unification rules, and data consistency verification rules.

8. A low-permeability oil reservoir comprehensive evaluation device, characterized in that, include: The parameter acquisition module is used to acquire multi-source heterogeneous parameters of the target low-permeability reservoir; wherein, the multi-source heterogeneous parameters include at least one of geological parameters, reservoir parameters, fluid parameters and engineering parameters; The system determination module is used to determine the multidimensional evaluation index system of the target low-permeability reservoir and the target evaluation result corresponding to the multidimensional evaluation index system based on the multi-source heterogeneous parameters; wherein, the multidimensional evaluation index system includes multiple evaluation indicators; The type determination module is used to determine the contribution relationship type corresponding to each evaluation indicator based on the contribution relationship between each evaluation indicator and the target evaluation result; wherein, the contribution relationship type includes monotonic positive correlation type, monotonic negative correlation type and non-monotonic correlation type; The feature determination module is used to perform unified evaluation direction mapping processing on each evaluation indicator according to the contribution relationship type corresponding to each evaluation indicator, so as to obtain the unified evaluation direction feature corresponding to each evaluation indicator. The index determination module is used to obtain a comprehensive evaluation index based on the unified evaluation direction characteristics using a preset monotonic constraint comprehensive evaluation model; wherein, the preset monotonic constraint comprehensive evaluation model satisfies the contribution consistency constraint condition, and the contribution consistency constraint condition is used to ensure that the comprehensive evaluation index satisfies a monotonic non-decreasing relationship with respect to the unified evaluation direction characteristics corresponding to each evaluation indicator. The result determination module is used to determine the comprehensive evaluation result of the target low-permeability reservoir based on the comprehensive evaluation index.

9. An electronic device, comprising: It includes a processor and a memory for storing processor-executable instructions, wherein the processor, when executing the instructions, implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images