A CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS

Abstract

The application discloses a CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS, and relates to the technical field of CO2 capture, utilization, storage and oil and gas field development risk evaluation. The method can solve the problems of strong direct monitoring data dependency, insufficient multi-source information fusion, single evaluation level and large subjectivity of index weighting in the existing CCUS injection development process. The method fuses geological, logging, fracturing reconstruction, injection-production system, production dynamic and interwell relationship and other multi-source information, and establishes a single-well reservoir physical property evaluation and well group gas channeling risk evaluation double-layer system. The single-well layer adopts CRITIC objective weighting combined with the TOPSIS method to obtain the reservoir physical property comprehensive score. The well group layer superimposes injection-production dynamic indexes, and uses CRITIC-TOPSIS again to carry out evaluation, so that the advantage channel identification, five-level division and spatial partitioning of the gas channeling risk are realized.

Description

Technical Field

[0001] This invention belongs to the technical field of CO2 capture, utilization and storage and oil and gas field development risk assessment, specifically involving a CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS. Background Technology

[0002] Carbon dioxide capture, utilization, and storage (CO2 capture, utilization, and storage) technology is one of the important technological pathways to achieve low-carbon development and carbon emission reduction goals for fossil energy. CO2 injection not only improves reservoir recovery but also serves as a geological storage function for carbon dioxide, showing broad application prospects in low-permeability reservoirs, tight reservoirs, and the stable production and efficiency enhancement of old reservoirs. Injecting CO2 into the formation can effectively reduce crude oil viscosity, decrease interfacial tension, enhance crude oil expansion capacity, and improve fluid flow, thereby significantly improving displacement efficiency and swept volume. However, during CO2 injection development, due to strong reservoir heterogeneity, significant differences in fracture development, complex inter-well connectivity, and insufficient matching of injection and production parameters, injected CO2 tends to preferentially migrate along high-permeability zones, fracture systems, or existing advantageous seepage paths, forming dominant channels. This can induce gas channeling, water channeling, displacement front surge, and injection-production imbalance, which not only reduces the CO2 oil displacement effect and swept efficiency but may also affect the effectiveness of storage and the overall development benefits of the block. Therefore, gas channeling risk identification has become a key link in the dynamic control and risk prevention of CO2 injection development.

[0003] Existing gas channeling identification technologies mainly rely on direct dynamic monitoring data such as wellhead gas production, gas-oil ratio, and CO2 concentration in produced gas, or employ methods such as production dynamic analysis, tracer monitoring, and empirical judgment. While these methods have some reference value for local well points or typical well groups, they still have significant shortcomings in practical applications. These shortcomings are mainly manifested in the following ways: direct field monitoring data is often discontinuous and incomplete, with some wells even lacking effective monitoring data, severely limiting the applicability of the methods; more critically, existing methods severely underutilize multi-source data, including geological, logging, engineering, injection-production dynamics, and inter-well relationships, making it difficult to consider both single-well reservoir conditions and well group connectivity characteristics, resulting in a single evaluation level and an inability to form a systematic risk identification capability; furthermore, some evaluation methods rely on manual experience-based weighting, which is highly subjective, making it difficult to guarantee the stability and repeatability of the results. Under conditions of limited or even missing direct monitoring data, how to integrate multi-source static and dynamic information to objectively and quantitatively identify dominant channels and gas channeling risks has become a core technical bottleneck restricting the risk management of CO2 injection development.

[0004] Based on this, the present invention proposes a CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS to solve the problems existing in the prior art. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a CCUS injection gas channeling risk identification method based on CRITIC-dual-layer TOPSIS, which solves the problems of existing CCUS injection development gas channeling risk identification methods, such as strong dependence on direct monitoring data, insufficient fusion of multi-source information, single evaluation level, and high subjectivity in indicator weighting.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS includes the following steps: Step S1: Perform multi-source data acquisition and preprocessing; Acquire multi-source data related to CCUS injected gas channeling risk identification within the target block, and after preprocessing the data, form raw datasets for single injection wells, raw datasets for single production wells, and evaluation datasets for injection-production well pairs. Step S2: Based on the original datasets of injection wells and production wells formed in Step S1, construct the comprehensive evaluation index system for injection wells and the comprehensive evaluation index system for production wells, respectively, and establish the original data matrix with each well as the evaluation object. Step S3: Determine the attributes of the single-well evaluation indicators obtained in Step S2, classify them into positive indicators and negative indicators, and convert all indicators into a uniform form, perform dimensionless processing, and establish a standardized matrix for injection wells and a standardized matrix for oil production wells. Step S4: Based on the standardized matrix of injection wells and oil production wells obtained in step S3, the weights of each evaluation index are determined using the CRITIC method, and the injection wells and oil production wells are comprehensively evaluated using the TOPSIS method. Step S5: Based on the TOPSIS comprehensive evaluation scores of the injection well and the production well obtained in Step S4, output the comprehensive evaluation result table of the injection well and the comprehensive evaluation result table of the production well respectively, so as to provide basic input data for subsequent gas channeling risk identification of the injection and production wells. Step S6: Based on the multi-source data obtained in Step S5, establish a well group layer evaluation index system to characterize inter-well connectivity, the degree of development of dominant channels, and the risk of CO2 gas channeling; Step S7: Perform attribute determination and standardization on the well group layer evaluation indicators to obtain the well group layer standardized evaluation matrix; Step S8: Based on the well group standardized evaluation matrix obtained in step S7, the CRITIC method is used to assign overall weight to all injection-production well pairs, and the proximity coefficient of each injection-production well pair is calculated as the comprehensive evaluation score of the well group. Step S9: Sort all injection and production well pairs according to the comprehensive evaluation score of the well group obtained in step S8, and classify the CO2 gas channeling risk level using a fixed threshold. Step S10: Compare and analyze the CO2 advantageous channels and high-risk injection-production wells identified in Step S9 with the development dynamic data, field monitoring data and historical understanding of the target block, and output differentiated control suggestions based on the verification results.

[0007] In a preferred embodiment, in step S1, The static data for a single well includes geological data, well logging data, and engineering data; The geological data include at least reservoir thickness, perforation thickness, porosity, permeability, permeability gradient, hydrocarbon saturation, and clay content. The well logging data shall include at least resistivity, sonic transit time, and rock density; the engineering data shall include at least sand addition volume, sand ratio, displacement, and fluid injection volume. The dynamic parameters of well group injection and production include at least the injection-production well spacing, average daily injection volume, average daily fluid production, water cut, and corresponding parameters of well group.

[0008] In a preferred embodiment, in step S1, The preprocessing process includes, in sequence, well name unification, coordinate unification, layer alignment, unit conversion, outlier identification, missing value completion, and duplicate data removal. The outlier identification process includes one or more of the statistical threshold method, box plot method, or 3σ method; the missing value completion process includes one or more of the mean substitution method, interpolation method, or regression estimation method.

[0009] In a preferred embodiment, in step S2 The comprehensive evaluation index system for injection wells includes perforation thickness, porosity, permeability, permeability gradient, hydrocarbon saturation, resistivity, sonic transit time, rock density, and clay content. The comprehensive evaluation index system for a single oil well includes static evaluation indexes and parameters for well completion and fracturing stimulation engineering. The static evaluation indicators include perforation thickness, porosity, permeability, permeability gradient, oil and gas saturation, resistivity, sonic transit time, rock density, and clay content; the well completion and fracturing engineering parameters include sand addition volume, sand ratio, displacement, and fluid injection volume.

[0010] In a preferred embodiment, step S2, the process of establishing the original data matrix with a single well as the evaluation object, includes: Assume there are a total of n IInjection wells, each injection well selected m I Based on these indicators, the original data matrix of the injection wells is as follows: ; Where, x i,j (I) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n I ; j = 1,2,···, m I ; Assume there are a total of n P Each oil well is selected as a production well. m P Based on these indicators, the original data matrix of the oil wells is as follows: ; Where, x i,j (P) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n P ; j = 1,2,···, m P .

[0011] In a preferred embodiment, step S3, which unifies the positive and negative indicators of a single-well evaluation index into a unidirectional form, includes: Let the original data matrix of any single well evaluation object be: ; in, N Indicates the number of evaluation objects; M Indicates the number of evaluation indicators; x i,j Indicates the first i The evaluation object is in the first j The original values ​​of the indicators; The matrix after the unification process is denoted as: ; in, Determine as follows: ; in, εThis represents a small constant used to prevent the denominator from being zero.

[0012] In a preferred embodiment, step S3 involves performing dimensionless processing on the homogenized index data using the Z-score standardization method, specifically including: Let the evaluation matrix after homogenization be... X , obtained the j Sample mean of each indicator and standard deviation They are respectively: ; ; The standardized index based on the Z-score standardization method is: ; Let the standardized matrix be: ; in, Z This represents the basic input matrix for subsequent calculation of CRITIC index information, determination of objective weights, and evaluation of TOPSIS closeness.

[0013] In a preferred embodiment, step S4, the process of determining the objective weights of the single-well evaluation index based on the CRITIC method, includes: For any type of single-well normalized matrix obtained in step S3 Z ; Step S4-1-1: Calculate the degree of variation of the index: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] j The sample standard deviation of the evaluation index is denoted as . The calculation formula is as follows: ; in, Indicates the first j The sample mean of each evaluation indicator across all evaluation objects; z i,j Indicates the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; N Indicates the number of evaluation objects; ; Step S4-1-2: Calculate the conflict of indices: Calculate the Pearson correlation coefficient matrix R between indices. r j,k ]: ; Using the CRITIC method to measure indicatorsj Other indicators k Conflictality is defined as: ; in, r j,k Indicates the first j Evaluation indicators and the first k Pearson correlation coefficient among the evaluation indicators; z i,j For the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first j The sample mean of each evaluation indicator across all evaluation objects; z i,k Indicates the first i The evaluation object is in the first k The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first k The sample mean of each evaluation indicator across all evaluation objects; M Indicates the number of evaluation indicators; H j Indicates the conflict coefficient; Step S4-1-3: Calculate the information content of the indicators: [This step involves] calculating the information content of the indicators. j Comprehensive information content C j Defined as the product of its standard deviation and its conflict rate: ; Step S4-1-4: Determine objective weights: Normalize the information content of each indicator to obtain the first... j CRITIC objective weight of the evaluation indicators : ; Therefore, the weight vector is obtained as follows: .

[0014] In a preferred embodiment, step S4, which involves comprehensively evaluating the injection well and the production well using the TOPSIS method, includes: Step S4-2-1: Construct the weighted evaluation matrix V: ; ; in: v i,j Indicates the first i The evaluation object is in the first jWeighted values ​​on the indicators; N Indicates the number of evaluation objects; M Indicates the number of evaluation indicators; w j Indicates the first j CRITIC objective weights of the evaluation indicators; z i,j Indicates the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; Step S4-2-2: Determine the ideal solution V + and negative ideal solution V - : ; ; Step S4-2-3: Calculate the distance measure: Calculate the first... i Euclidean distance from each evaluation object to the positive and negative ideal solutions and They are respectively: ; ; Step S4-2-4: Calculate the proximity coefficient and overall score: Use the TOPSIS method to map the degree of excellence to the proximity coefficient. C i : .

[0015] In a preferred embodiment, in step S6, the comprehensive evaluation index system for well groups includes basic indicators and derived indicators, specifically: For the i For each injection-production well pair, the original evaluation index vector of its well group is constructed as follows: ; If a total of P injection-production well pairs are constructed within the target block, a raw evaluation data matrix of the well group layer can be further formed: ; in, Indicates the first i The comprehensive physical property score of the corresponding injection well in each injection-production well pair; Indicates the first i The comprehensive physical property score of each injection-production well relative to the corresponding production well; Indicates the first i The well spacing between injection and production wells in a single injection-production well pair; Indicates the first iThe average daily injection volume of the corresponding injection well in each injection-production well during the evaluation period; Indicates the first i The average daily fluid production of each injection-production well and its corresponding production well during the evaluation period; Indicates the first i Water cut parameters of each injection-production well and its corresponding production well; Indicates the first i Average daily water production of each injection-production well pair; Indicates the first i Average daily oil production of each injection-production well pair; Indicates the first i The production / injection ratio of each injection-production well to its corresponding production well during the evaluation period; Indicates the first i The water production / injection ratio of each injection-production well to the corresponding production well during the evaluation period.

[0016] Compared with existing technologies, this invention provides a CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS, which has the following advantages: 1. This invention addresses the lack of direct monitoring data on wellhead gas production, gas-oil ratio, and CO2 concentration in produced gas during CCUS injection development. It utilizes multi-source data, including geological, logging, engineering, injection-production dynamics, and inter-well relationships, for comprehensive analysis. Under conditions of limited direct gas channeling characterization data, a two-layer progressive evaluation framework is employed—first, a comprehensive evaluation of reservoir properties in a single well layer is conducted to obtain a comprehensive score for each well's properties. Then, this score is integrated with injection-production well spacing, injection dynamic parameters, production dynamic parameters, and derived response parameters to construct a well group layer dominant channel and gas channeling risk evaluation index system. This effectively identifies CO2 dominant channels and gas channeling risks, significantly expanding the applicability and data adaptability of this method.

[0017] 2. This invention constructs a two-layer identification framework for single-well comprehensive evaluation and injection-production well comprehensive evaluation, and employs the CRITIC method for objective weighting. The CRITIC method, by comprehensively considering the sample variability and inter-indicator conflicts of each evaluation indicator, uses the amount of information in each indicator as the basis for weight determination, avoiding the subjectivity problem of traditional methods relying on manual experience for weighting, and improving the objectivity, stability, and repeatability of gas channeling risk identification results. Combined with the TOPSIS method, it realizes comprehensive score calculation, ranking analysis, and risk level classification, simultaneously taking into account both the static reservoir conditions of a single well and the dynamic connectivity characteristics of the well group, forming a complete technical chain from reservoir property characterization to inter-injection well connectivity evaluation.

[0018] 3. This invention can output the results of identifying the advantageous channels, ranking the gas channeling risks, and classifying the risks of injection and production wells, and can generate corresponding control suggestions based on these results. It can further generate multi-dimensional visualization outputs such as gas channeling risk score bar charts, risk score distribution maps, and spatial distribution maps of gas channeling risks in blocks, providing a quantitative basis for on-site implementation of measures such as injection and production structure optimization, layered and zoned channeling control, water and gas blocking, and injection parameter optimization. It has good engineering application value and promotion prospects. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall process of the CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS provided by the present invention.

[0020] Figure 2 This invention provides a bar chart of gas channeling risk scores for injection and production wells.

[0021] Figure 3 This invention provides a distribution map of gas channeling risk scores for target blocks.

[0022] Figure 4 This invention provides a spatial distribution map of block gas channeling risk. Detailed Implementation

[0023] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0024] Please see Figures 1-4 As shown, this invention proposes a CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS, comprising the following steps: Step S1: Multi-source data acquisition and preprocessing; specifically including: Step S1-1: First, acquire multi-source data related to CCUS injection gas channeling risk identification within the target block, and preprocess the multi-source data to form the original dataset of injection wells, the original dataset of production wells, and the evaluation dataset of injection-production well pairs, providing basic data support for the subsequent construction of a comprehensive evaluation index system for single wells and the identification of gas channeling risk in injection-production well pairs.

[0025] Among them, (1) the multi-source data includes single-well static data and well group layer injection and production dynamic parameters. Among them: The single-well static data includes geological data, logging data, and engineering data. The geological data includes reservoir thickness, perforation thickness, porosity, permeability, permeability gradient, hydrocarbon saturation, and clay content; the logging data includes resistivity, sonic transit time, and rock density; and the engineering data includes sand addition volume, sand ratio, displacement, and fluid injection volume.

[0026] (2) The dynamic parameters of well group injection and production include injection and production well spacing, average daily injection volume, average daily liquid production, water cut and corresponding parameters of well group.

[0027] Step S1-2: Then, the data collected in step S1-1 above is preprocessed to form the original dataset of injection wells, the original dataset of production wells, and the evaluation dataset of injection-production well pairs.

[0028] The preprocessing process includes, in sequence, unifying well names, unifying coordinates, aligning layers, converting units, identifying outliers, filling in missing values, and removing duplicate data.

[0029] Wherein: the well name unification is used to establish the correspondence between well numbers in different databases; the coordinate unification is used to ensure the consistency of the spatial relationship of well locations; the layer alignment is used to ensure the matching of data from different sources on the same target layer; the unit conversion is used to unify the measurement caliber of each evaluation parameter; the outlier identification is used to remove data that does not conform to the actual engineering situation; the missing value completion is used to improve the integrity of the sample data; and the duplicate data removal is used to eliminate redundant records.

[0030] Specifically, the outlier identification includes one or more of the following methods: statistical thresholding, box plot discrimination, and 3σ discrimination; the missing value completion includes one or more of the following methods: mean substitution, interpolation, and regression estimation.

[0031] Step S1-3: After the preprocessing in step S1-2, the original datasets of injection wells, production wells, and injection-production well pairs are generated, providing a data foundation for the subsequent construction of a comprehensive evaluation index system for single wells and the identification of gas channeling risks in injection-production well pairs.

[0032] Step S2: Based on the original datasets of injection wells and production wells formed in Step S1, construct the comprehensive evaluation index system for injection wells and production wells respectively, and establish the original data matrix with each well as the evaluation object.

[0033] Step S2-1: First, considering the availability of injection well data and its sensitivity to reservoir permeability and CO2 migration behavior, select the following parameters: perforation thickness (m), porosity (%), permeability (mD) and its gradient, hydrocarbon saturation (%), resistivity (Ω·m), sonic transit time (μs / m), and rock density (g / cm³). 3 Nine key geological and logging parameters, including argillaceous content (%), were used to construct a comprehensive characterization and evaluation index system for reservoir properties in injection wells. Among them: Perforation thickness characterizes the effective connectivity between the injection well and the target reservoir; porosity characterizes the reservoir's pore space size and fluid storage capacity; permeability characterizes the reservoir's conductivity; permeability gradient characterizes the vertical heterogeneity of the target interval, and the permeability gradient can be characterized by the ratio of the maximum to minimum permeability values, the range, or other equivalent differences, depending on the available data; hydrocarbon saturation reflects the fluid occurrence state of the target reservoir; resistivity and acoustic transit time reflect the rock-fluid integrated response characteristics and differences in reservoir pore structure; rock density characterizes the properties and compactness of the rock skeleton; and clay content reflects the proportion of ineffective reservoir components and potential blockage tendency.

[0034] The above nine indicators comprehensively characterize the physical properties of the reservoir controlled by the injection well and the CO2 injection and migration potential from multiple dimensions, including reservoir connectivity scale, pore space, conductivity, heterogeneity, fluid occurrence state, rock-fluid response characteristics, rock skeleton properties, and non-reservoir content. This forms a relatively complete and complementary comprehensive evaluation index system for single injection wells.

[0035] Step S2-2: For oil production wells, to characterize the reservoir's ability to be affected by CO2 displacement, the formation conditions of potential dominant channels, and the impact of wellbore stimulation, the following geological, logging, and completion-related properties are selected: perforation thickness (m), porosity (%), permeability (mD), permeability gradient, hydrocarbon saturation (%), resistivity (Ω·m), sonic transit time (μs / m), and rock density (g / cm³). 3 Nine static evaluation indicators were used, including sand content (%) and clay content (%). Based on these, the volume of sand added (m³) was further introduced. 3 ), sand ratio (%), displacement (m³) 3 / min) and the amount of liquid entering the ground (m 3 Four parameters related to well completion and fracturing stimulation were used to construct a comprehensive evaluation index system for single oil wells. Among them: The first nine basic static indicators mentioned above are used to characterize the effective reservoir thickness, reservoir space size, conductivity, vertical heterogeneity, fluid occurrence state, and rock skeleton and pore structure characteristics within the control range of the oil well.

[0036] Among the last four well completion and fracturing engineering parameters mentioned above, the sand addition volume, sand ratio, discharge rate, and fluid injection rate are static engineering parameters related to fracturing. They are used to reflect the fracture support scale, proppant configuration intensity, fracturing construction intensity, and fluid injection scale, respectively, and have an important impact on the fracture extension range, fracture conductivity, and effective seepage channel formation capacity.

[0037] The above 13 indicators jointly characterize the effective basis of the reservoir on the oil well side, fracture conduction conditions and potential gas channeling sensitivity from three dimensions: geology, logging and completion and fracturing engineering. They have a comprehensive characterization effect on the migration path, sweep efficiency, effectiveness and gas channeling risk of CO2 on the oil well side.

[0038] Specifically, in some specific implementation scenarios, if there are differences in the type or completeness of data in the target block, the above indicators can be replaced equivalently, appropriately added or subtracted, or refined and layered without changing the evaluation objectives and the physical meaning of the indicators. However, at least key indicators that are representative of reservoir connectivity, pore structure, conductivity, heterogeneity, and engineering stimulation intensity should be retained.

[0039] Step S2-3: Using a single well as the evaluation object, establish the original data matrices for injection wells and production wells respectively. Specifically: In the original data matrix, each row corresponds to a well, and each column corresponds to an evaluation index.

[0040] Assume there are a total of n I Injection wells, each injection well selected m I If there are several indicators, then the original data matrix of the injection well is denoted as: ; Where, x i,j (I) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n I ; j = 1,2,···, m I .

[0041] Assume there are a total of n P Each oil well is selected as a production well. m P If there are 10 indicators, then the original data matrix of the oil wells is denoted as: ; Where, x i,j (P) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n P ; j = 1,2,···,m P .

[0042] The aforementioned raw data matrix is ​​used for indicator standardization, CRITIC objective weighting calculation, and TOPSIS comprehensive evaluation analysis in subsequent steps.

[0043] Step S3: Perform single-well index attribute determination, positive conversion, and standardization processing; Based on the comprehensive evaluation index system for injection wells and production wells constructed in step S2, in order to eliminate the differences in attribute direction, dimension and order of magnitude of different evaluation indicators, and to provide a unified data input basis for subsequent CRITIC objective weighting and TOPSIS comprehensive evaluation, the original data matrix of injection wells and production wells are respectively subjected to index attribute determination, homogenization and standardization processing.

[0044] It should be noted that injection well samples and production well samples should be processed independently and not mixed for standardization, so as to avoid interference with the evaluation results caused by different well types and different indicator systems.

[0045] Step S3-1: Perform indicator attribute determination and homogenization processing; Based on the direction of each evaluation indicator's effect on the target evaluation attribute, the indicators are divided into positive indicators and negative indicators. Among them: The positive index refers to an index whose larger value is more conducive to characterizing reservoir connectivity, porosity, permeability, fracture conductivity, or CO2 migration capacity; the negative index refers to an index whose larger value is less conducive to characterizing the above-mentioned target evaluation attributes.

[0046] Porosity, permeability, perforation thickness, sand addition volume, sand ratio, discharge rate, and influent volume are typically used as positive indicators; clay content and rock density are typically used as negative indicators. To ensure that all indicators have a consistent direction of comparison in the subsequent evaluation process, all indicators are uniformly converted into a unidirectional form where a larger value indicates a stronger target evaluation attribute.

[0047] Let the original data matrix of any single well evaluation object be: ; in, N Indicates the number of evaluation objects. M Indicates the number of evaluation indicators. x i,j Indicates the first i The evaluation object is in the first j The original values ​​of the indicators.

[0048] The matrix after the unification process is denoted as: ; In the above processing, the homogenization process involves converting inverse indicators into evaluation directions consistent with positive indicators for indicators with different attribute directions. This ensures that all converted indicators meet the unified principle that "the larger the indicator value, the better the evaluation performance." Determine as follows: ; in, ε This represents a small constant (taken as 10) to prevent the denominator from being zero. -6 ).

[0049] After the above-mentioned homogenization process, all evaluation indicators are converted to a unified direction, that is, the larger the indicator value, the stronger the corresponding evaluation object performs in the target evaluation attribute, thereby meeting the calculation requirements of subsequent objective weighting and comprehensive evaluation.

[0050] Step S3-2: Use the Z-score standardization method to perform dimensionless processing on the homogenized index data; Since different evaluation indicators have different dimensions, orders of magnitude and value ranges, in order to eliminate the differences in dimensions and retain the sample dispersion and the correlation structure of indicators, this invention uses the Z-score standardization method to perform dimensionless processing on the homogenized indicator data.

[0051] Based on the above processing, the evaluation matrix after homogenization is obtained as follows: X Based on the principle of sample statistics in mathematical statistics, the first... j Statistical analysis was performed on the values ​​of the first indicator across all evaluation objects to obtain the second indicator. j Sample mean of each indicator and standard deviation They are respectively: ; ; The standardized index based on the Z-score standardization method is: ; in, z i,j For the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator.

[0052] Based on the common notation conventions for matrix representation, the standardized matrix is ​​denoted as . Z : ; in, ZThis represents the foundational input matrix for subsequent calculations of CRITIC index information, determination of objective weights, and TOPSIS proximity evaluation. For comprehensive evaluation of a single injection well, the matrix is ​​taken as... N = n I , M = m I For comprehensive evaluation of a single oil well, take... N = n P , M = m P Injection wells and production wells were each independently processed for alignment and standardization using the methods described above.

[0053] After the Z-score standardization process described above, each indicator has a mean of approximately 0 and a standard deviation of approximately 1 on the sample set. This effectively eliminates the dimensional differences between different indicators and preserves the original degree of variation and correlation characteristics of each indicator, providing a unified and reliable data foundation for subsequent CRITIC weighting and TOPSIS comprehensive evaluation.

[0054] Step S4: Perform objective weighting of single-well formations using CRITIC and comprehensive evaluation using TOPSIS; Based on the standardized injection well matrix and oil production well matrix obtained in step S3, to objectively reflect the differences in sample discrimination ability and information complementarity of each evaluation index, this embodiment uses the CRITIC method to determine the weight of each index. On this basis, the TOPSIS method is combined to comprehensively evaluate the injection wells and oil production wells separately, outputting the corresponding single-well comprehensive scores and ranking results. In step S4, the injection well samples and oil production well samples are calculated independently to form their respective single-well evaluation results.

[0055] Step S4-1: Determine the objective weights of single-well evaluation indicators based on the CRITIC method; The CRITIC method comprehensively considers the degree of variability and inter-indicator conflict among various evaluation indicators to characterize the amount of information contained in the indicators and determine the objective weights accordingly. The standardized matrix for any type of single well obtained in step S3 is... Z .

[0056] Step S4-1-1: Calculate the degree of variation of the index; No. j The sample standard deviation of the evaluation index is denoted as . The calculation formula is as follows: ; in, Indicates the first j The sample mean of each evaluation indicator across all evaluation objects.

[0057] ; Standard deviation It is used to characterize the degree of dispersion of this indicator among different evaluation objects. Generally speaking The larger the value, the stronger the indicator's ability to distinguish between different evaluation objects, and the richer the effective information it contains.

[0058] Step S4-1-2: Calculate the conflict of indicators; To characterize the information redundancy among indicators, the Pearson correlation coefficient matrix among indicators is calculated. R =[ r j,k ]: ; in: r j,k Indicates the first j Evaluation indicators and the first k Pearson correlation coefficient among the evaluation indicators; z i,j For the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first j The sample mean of each evaluation indicator across all evaluation objects. z i,k For the first i The evaluation object is in the first k The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first k The sample mean of each evaluation indicator across all evaluation objects; M This indicates the number of evaluation indicators.

[0059] Using the CRITIC method to measure indicators j Other indicators k Conflictality is defined as: ; Among them, the conflict coefficient H j The larger the value, the more likely it is to be the first. j The lower the correlation between an indicator and other indicators, the less overlap there is in the information it provides with other indicators, and the stronger the information complementarity.

[0060] Step S4-1-3: Calculate the amount of information in the indicator; The CRITIC method will use indicators j Comprehensive information content C jDefined as the product of its standard deviation and its conflict rate: ; in, C j The larger the value, the higher the value. j The higher the amount of effective information contained in an evaluation indicator, the more important its role in the comprehensive evaluation.

[0061] Step S4-1-4: Determine the objective weights; Normalize the information content of each indicator to obtain the first... j CRITIC objective weight of the evaluation indicators for: ; Therefore, the weight vector is obtained as follows: ; In this embodiment, the weight vector of the evaluation index for injection wells and the weight vector of the evaluation index for oil production wells are calculated independently to meet the single-well evaluation requirements of different well types.

[0062] Step S4-2: After obtaining the CRITIC weights in step S4-1, the TOPSIS method is used to comprehensively evaluate and rank the individual wells; Step S4-2-1: Construct a weighted evaluation matrix; Construct the weighted evaluation matrix V: ; ; in: V This represents the weighted evaluation matrix; v i,j Indicates the first i The weighted value of each evaluation object on the j-th indicator; N Indicates the number of evaluation objects; M Indicates the number of evaluation indicators; w j For the first j CRITIC objective weights of the evaluation indicators; z i,j For the first i The evaluation object is in the first j The dimensionless Z-score value of the indicator.

[0063] Step S4-2-2: Determine the positive ideal solution and the negative ideal solution; For the weighted evaluation matrix V Then, determine the positive and negative ideal solutions for each indicator. Positive ideal solution V +This represents an ideal evaluation object where all indicators are at their optimal levels; a negative ideal solution. V - This represents the ideal evaluation object where all indicators are at their worst level. The TOPSIS method defines the positive ideal solution. V + and negative ideal solution V - as follows: ; ; Step S4-2-3: Calculate the distance measure; Calculate the first i The Euclidean distances from each evaluation object to the positive and negative ideal solutions are: ; ; in, D i + The smaller the value, the higher the value. i The closer an evaluation object is to its optimal state; D i - The larger the value, the further away it is from the worst-case scenario.

[0064] Step S4-2-4: Calculate the proximity coefficient and overall score; The TOPSIS method is used to map the degree of superiority or inferiority to a similarity coefficient. C i : ; Among them, the proximity coefficient C i The larger the value, the closer the evaluated object is to the positive ideal solution and the further away it is from the negative ideal solution, and the better its overall evaluation result.

[0065] In this embodiment, the proximity coefficient C i This serves as the comprehensive evaluation score for the corresponding single well. For injection wells, C i This can characterize the comprehensive impact of the injection well on CCUS injection migration and potential crossflow formation under geological and engineering conditions; for oil production wells, C i This comprehensive evaluation score can characterize the sensitivity of the oil well to gas channeling under static geological conditions. Based on the comprehensive evaluation score, injection wells and production wells can be ranked separately, serving as an important basis for subsequently constructing an index system for evaluating the gas channeling risk of injection and production wells.

[0066] Step S5: Output the comprehensive evaluation score for a single well; Step S5-1: Based on the TOPSIS comprehensive evaluation scores of the injection wells and production wells obtained in Step S4, output the comprehensive evaluation result table for the injection wells and the comprehensive evaluation result table for the production wells, respectively. The result tables shall include at least the well number and its corresponding comprehensive evaluation score.

[0067] The comprehensive evaluation score of injection wells characterizes the combined influence of formation properties, reservoir connectivity, and injection conditions on CO2 migration and potential gas channeling. The comprehensive evaluation score of production wells characterizes their sensitivity to injected gas breakthrough and gas channeling responses under static geological constraints. A higher comprehensive evaluation score indicates that the corresponding well is closer to the positive ideal solution, and its comprehensive evaluation result is better.

[0068] Step S5-2: Output the comprehensive evaluation results of the single well obtained in step S5-1 in tabular form to provide basic input data for subsequent gas channeling risk identification of injection and production wells.

[0069] Step S6: Establishing injection-production relationship, fusing multi-source data, and constructing a well group layer evaluation index system; Based on the well location distribution, well network deployment and reservoir correspondence obtained in step S5, injection-production well pairs are established. By integrating multi-source data such as single well evaluation results, well-to-well distance, injection-production dynamic parameters and derived indicators, a well group layer evaluation index system is established to characterize well-to-well connectivity, the degree of development of dominant channels and CO2 gas channeling risk.

[0070] Step S6-1: First, based on the well location distribution, well network deployment, and reservoir correspondence of injection wells and production wells in the target block, establish the injection-production relationship between injection wells and production wells.

[0071] The injection-production correlation is used to characterize the potential displacement effect, pressure transmission relationship, and potential gas channeling connections between a given injection well and adjacent or affected production wells. In practice, an injection well can be paired with one or more production wells within its influence range to form an injection-production well pair; or an injection well and its corresponding multiple production wells can form an injection-production well pair, which serves as the basic evaluation object for subsequent gas channeling risk identification.

[0072] Step S6-2: After establishing the injection-production correlation, perform multi-source data fusion on each injection-production well pair or injection-production well combination. The multi-source data includes at least the following indicators: (1) The comprehensive score of injection well physical properties is used to characterize the comprehensive influence of the corresponding injection well on CO2 injection migration and gas channeling under static geological and reservoir physical property conditions; (2) Comprehensive score of oil well physical properties, used to characterize the sensitivity of the corresponding oil well to the response of injected gas breakthrough and gas channeling under static reservoir conditions; (3) Injection-production well distance, used to characterize the spatial distance relationship between injection wells and production wells; (4) Average daily injection volume, used to characterize the average injection intensity of injection wells during the evaluation period; (5) Average daily fluid production, used to characterize the production capacity and fluid response characteristics of the oil wells during the evaluation period; (6) Water content is used to characterize the water content level or fluid production characteristics of oil wells, and to provide auxiliary dynamic information for identifying the risk of injected gas channeling.

[0073] The comprehensive scores of the physical properties of the injection well and the comprehensive scores of the physical properties of the production well are derived from the comprehensive evaluation results of the single well in step S5; the injection-production well distance can be calculated based on the coordinates of the injection well and the production well; the average daily injection volume, average daily fluid production volume and water cut can be obtained from the field production data statistics during the target evaluation period.

[0074] Specifically, for any injection-production well pair i The basic data index vector of the well group layer is: ; in: Indicates the first i The comprehensive physical property score of the corresponding injection well in each injection-production well pair; Indicates the first i The comprehensive physical property score of each injection-production well relative to the corresponding production well; Indicates the first i The well spacing between injection and production wells in a single injection-production well pair; Indicates the first i The average daily injection volume of the corresponding injection well in each injection-production well during the evaluation period; Indicates the first i The average daily fluid production of each injection-production well and its corresponding production well during the evaluation period; Indicates the first i The water cut parameters of the corresponding production wells in each injection-production well pair.

[0075] Step S6-3: The above basic water cut parameters are the average water cut of the production wells during the evaluation period. For any injection-production well pair... i , with the first i The water cut parameters of the injection and production wells are denoted as follows: f i ,and f i When expressed in decimal form, its derived index can be calculated using the following formula: ; ; ; ; Indicates the first i Average daily water production of each injection-production well pair; Indicates the first i Average daily oil production of each injection-production well pair; Indicates the first i The average daily injection volume of each injection-production well relative to the corresponding injection well during the evaluation period; Indicates the first i The average daily fluid production of each injection-production well relative to the corresponding production well during the evaluation period; Indicates the first i The production / injection ratio of each injection-production well to its corresponding production well during the evaluation period; Indicates the first i The water production / injection ratio of each injection-production well to the corresponding production well during the evaluation period.

[0076] For the i For each injection-production well pair, the original evaluation index vector of its well group is constructed as follows:

[0077] If the target block is constructed P For each injection-production well pair, a raw evaluation data matrix of the well group can be further formed: ; By constructing the above-mentioned index system, the comprehensive evaluation results of static physical properties of a single well layer are organically combined with the spatial relationship between wells, injection dynamic parameters, production dynamic parameters and derived response parameters. This allows for a comprehensive characterization of the development degree of the dominant flow channels between injection and production wells and the gas channeling risk during CO2 flooding from the perspective of combining static and dynamic characteristics and synergistic characterization of oil and water phases. This provides a data foundation for the subsequent attribute determination, homogenization and standardization of well group layer evaluation indicators, objective weighting of CRITIC, and comprehensive evaluation of TOPSIS.

[0078] Step S7: Perform attribute determination and standardization on the well group layer evaluation indicators to obtain a standardized matrix; Based on the evaluation index system for the advantageous channels and gas channeling risks of injection and production wells constructed in step S6, in order to eliminate the differences in the attribute direction, dimension and order of magnitude of evaluation indexes of different well groups, and to provide a unified data input basis for the subsequent objective weighting of CRITIC and comprehensive evaluation of well groups and TOPSIS, the original evaluation data matrix of well groups is subjected to index attribute determination, positive processing and standardization.

[0079] It should be noted that the processing principle of step S7 is basically the same as that of step S3, but the evaluation object is the injection-production well pair rather than a single well sample. Therefore, it is no longer necessary to perform independent standardization processing for injection wells and production wells separately. Instead, the attribute determination, positiveization and standardization calculations are completed uniformly for all injection-production well pairs within the same evaluation matrix.

[0080] Step S7-1: Determine and positively process the index attributes of the well group; Step S7-1-1: Based on the direction of each evaluation index's effect on the development of the dominant flow channel between injection and production wells and the risk of gas channeling / channelization, each index is divided into positive and negative indices.

[0081] The positive indicators refer to those with larger values, which indicate stronger inter-well connectivity, higher flow guidance, more developed dominant channels, or higher risk of gas channeling / channelization. The negative indicators refer to those with larger values, which are less conducive to characterizing the above-mentioned target evaluation attributes.

[0082] In the well group gas channeling risk identification of this embodiment, the comprehensive score of injection well physical properties, the comprehensive score of production well physical properties, the average daily injection volume, the average daily production volume, the basic water cut parameters, the average daily water production volume, and the water production / injection ratio can be used as positive indicators; the injection-production well spacing and the oil production / injection ratio can be used as negative indicators.

[0083] For ease of description, the original evaluation data matrix of the well group obtained in step S7 is denoted as: ; in, P Indicates the total number of injection and production wells. Q This represents the total number of evaluation indicators for the well group. y i,j Indicates the first i The evaluation well group in the first j The original values ​​of the indicators.

[0084] Step S7-1-2: To ensure a consistent comparison direction for all indicators in the subsequent comprehensive evaluation process, all indicators are uniformly converted into a positive form where larger values ​​indicate more developed dominant channels and higher gas channeling risk. The matrix after positive conversion is denoted as: ; in, Determine as follows: ; in, ε To prevent the use of tiny constants with a denominator of zero (taken as 10). -6 ).

[0085] After the above positive transformation process, all well group layer evaluation indicators are unified into unidirectional indicators. That is, the larger the indicator value, the more developed the dominant flow channel of the corresponding injection-production well pair, and the higher the risk of gas channeling or channelization.

[0086] Step S7-2: Use the Z-score standardization method to perform dimensionless processing on the homogenized index data; The Z-score normalization method was used to normalize the evaluation matrix. Y Processing is performed to obtain the first... j Sample mean of each indicator and standard deviation They are respectively: ; ; The standardized indicators are: ; Let the standardized matrix be: ; in, Z This serves as the foundational input matrix for subsequent calculations of CRITIC index information, determination of objective weights, and evaluation of TOPSIS relevance.

[0087] After the above positive transformation and Z-score standardization processes, the evaluation indicators of each well group layer obtain a data expression form with consistent direction and comparable dimensions within a unified sample space, thus providing a unified and reliable data foundation for subsequent objective weighting of well group layer CRITIC and comprehensive evaluation of TOPSIS.

[0088] Step S8: Use the CRITIC method to assign overall weights to all injection-production well pairs, and combine the TOPSIS method to calculate the proximity coefficient of each injection-production well pair as the comprehensive evaluation score of the well group layer; Based on the standardized evaluation matrix of well groups obtained in step S7, in order to objectively reflect the differences in the distinguishing ability and information complementarity of the evaluation indicators of each well group between different injection and production well pairs, this embodiment uses the CRITIC method to determine the objective weight of each evaluation indicator of the well group, and on this basis, combines the TOPSIS method to conduct a comprehensive evaluation of each injection and production well pair, obtain the comprehensive score of the degree of development of the dominant channel and the gas channeling risk corresponding to each well group sample, and form a ranking result accordingly.

[0089] Unlike step S4, step S8 evaluates injection-production well pairs, using a unified evaluation matrix comprised of all well group samples within the same evaluation space. Therefore, it eliminates the need for separate calculations by well type; instead, it directly assigns overall weights and performs a comprehensive ranking of all injection-production well pairs. The comprehensive evaluation score characterizes the development level of dominant flow channels between injection and production wells and the risk level of gas channeling, crossflow, or channelization. A higher score indicates that the corresponding injection-production well pair is closer to a state of high connectivity, high channelization, and high gas channeling risk. Specifically, this includes: Step S8-1: Determine the objective weights of well group layer evaluation indicators based on the CRITIC method; The CRITIC method comprehensively considers the degree of variation and conflict between various evaluation indicators to characterize the amount of information contained in the indicators and determines the objective weights accordingly. The standardized matrix obtained in step S7 is Z.

[0090] Step S8-1-1: Calculate the degree of variation of the index; No. j The sample standard deviation of the evaluation index is denoted as . σ j The calculation formula is as follows:

[0091] in, No. j The sample mean of each evaluation indicator across all evaluation objects: ; Standard deviation σ j It is used to characterize the degree of dispersion of this indicator among different evaluation objects. Generally speaking σ j The larger the value, the stronger the indicator's ability to distinguish between different evaluation objects, and the richer the effective information it contains.

[0092] Step S8-1-2: Calculate the conflict of indicators; To characterize the information redundancy among indicators, the Pearson correlation coefficient matrix among indicators is calculated. R =[ r j,k ]: ; Using the CRITIC method to measure indicators j Other indicators k Conflictality is defined as: ; Among them, the conflict coefficient H j The larger it is, the more likely it is to be the first jThe lower the correlation between an indicator and other indicators, the less overlap there is in the information it provides with other indicators, and the stronger the information complementarity.

[0093] Step S8-1-3: Calculate the amount of indicator information; Using the CRITIC method to measure indicators j Comprehensive information content C j Defined as the product of its standard deviation and its conflict rate: ; in, C j The larger the value, the higher the value. j The higher the amount of effective information contained in an evaluation indicator, the more important its role in the comprehensive evaluation.

[0094] Step S8-1-4: Determine the objective weights; Normalize the information content of each indicator to obtain the first... j CRITIC objective weight of the evaluation indicators for: ; Therefore, the weight vector is obtained as follows: ; The weight vector is used to reflect the relative importance of the evaluation indicators of each well group in the identification of advantageous channels and the judgment of gas channeling risk, and serves as the weight input for the subsequent TOPSIS comprehensive evaluation.

[0095] Step S8-2: After obtaining the CRITIC objective weights of the well group evaluation indicators, the TOPSIS method is used to comprehensively evaluate each injection-production well pair, thereby obtaining the comprehensive score of the dominant channel development and gas channeling risk for each well group sample. Specifically, this includes: Step S8-2-1: Construct a weighted evaluation matrix; Construct a weighted evaluation matrix V : ; ; Step S8-2-2: Determine the positive ideal solution and the negative ideal solution; For the weighted evaluation matrix V The positive and negative ideal solutions for each indicator are determined separately. Since this invention has unified all well group layer evaluation indicators into a positive form in step S7, where larger values ​​indicate more developed dominant channels and higher gas channeling risk, therefore: Positive Ideal Solution V +An ideal evaluation object indicates that all indicators are in a state of high risk, high connectivity, and high channelization; Negative ideal solution V - This indicates an ideal evaluation object where all indicators are in a state of low risk, low connectivity, and weak channelization.

[0096] TOPSIS defines the positive ideal solution V + and negative ideal solution V - as follows: ; ; Step S8-2-3: Calculate the distance measure; Calculate the first i The Euclidean distances from each evaluation object to the positive and negative ideal solutions are: ; ; in, D i + The smaller the value, the closer the injection-production well is to the ideal high-risk state; D i - The larger the value, the further it is from the ideal low-risk state.

[0097] Step S8-2-4: Calculate the proximity coefficient and overall score; The TOPSIS method is used to map the degree of superiority or inferiority to a similarity coefficient. C i : ; Among them, the proximity coefficient C i The larger the value, the closer the injection-production well is to the positive ideal solution and the further away it is from the negative ideal solution. In other words, the more developed its dominant flow channels are, the stronger the inter-well connectivity is, and the higher the risk of gas channeling, crossflow, or channelization.

[0098] In this embodiment, the proximity coefficient C i This serves as the comprehensive evaluation score for the corresponding injection-production well pair. Based on this comprehensive evaluation score, all injection-production well pairs can be ranked, thereby identifying key well groups with strong advantageous channel characteristics and high gas channeling risk. This provides a quantitative basis for subsequent risk classification, injection-production structure optimization, stratified and zoned channeling control measures, and injection parameter adjustment.

[0099] Step S9: Identify the dominant CO2 pathways and classify the risk level of gas channeling; Based on the TOPSIS comprehensive evaluation scores of each injection-production well pair obtained in step S8, this embodiment further sorts the injection-production well pairs and identifies the CO2 dominant channels and their corresponding gas channeling risk levels by combining the spatial relationship of well locations, thereby providing a basis for subsequent injection-production structure optimization and governance measures formulation.

[0100] Step S9-1: Dominant channel identification; Based on the TOPSIS comprehensive evaluation score obtained in step S8, all injection-production well pairs are ranked. A higher comprehensive evaluation score indicates a more significant advantage in terms of dominant channel development, inter-well connectivity, and gas channeling potential. Therefore, injection-production well pairs with higher scores can be identified as CO2 dominant channels. Spatially overlaying the comprehensive evaluation score of the injection-production well pairs with the planar coordinates of the injection and production wells creates highly connected zones or dominant seepage paths on the well network plan, further identifying the dominant channels where CO2 may channel, break through, or preferentially migrate within the target block.

[0101] Step S9-2: Quantify and classify gas channeling risks; The TOPSIS comprehensive evaluation score obtained in step S8 is directly used as the quantitative score for CO2 gas channeling risk. Since a higher comprehensive evaluation score indicates stronger connectivity between the corresponding injection and production well pairs, more developed dominant channels, and a higher probability of CO2 gas channeling, the gas channeling risk can be classified and managed based on this score. A fixed threshold is used to divide CO2 gas channeling risk into five levels: high-risk zone (0.8–1.0), relatively high-risk zone (0.6–0.8), medium-risk zone (0.4–0.6), relatively low-risk zone (0.2–0.4), and low-risk zone (0–0.2).

[0102] By overlaying and displaying the risk level with the spatial coordinates of the well location and the relationship between injection and production through the above steps, the spatial zoning results of CO2 gas channeling risk in the target block can be obtained. This allows for the intuitive identification of high-risk channels and key treatment areas, providing a quantitative basis for formulating targeted measures such as adjusting injection and production, blocking water and gas, stratified injection and production, and optimizing injection parameters.

[0103] Step S10: Verification of evaluation results and suggestions for output control; Based on the CO2 dominant channel identification results and gas channeling risk level classification results obtained in step S9, this embodiment further verifies the consistency of the evaluation results on-site, and outputs targeted injection and production control and risk management suggestions in combination with different risk levels to improve the practicality of the identification results and their value for on-site application.

[0104] Step S10-1: Verification of evaluation results; The CO2 dominant channels and high-risk injection-production well pairs identified in step S9 are compared and analyzed with existing development dynamic data, field monitoring data and historical knowledge of the target block to verify the rationality and reliability of the gas channeling risk identification results in this embodiment.

[0105] The data used for verification preferably includes, but is not limited to: wellhead gas volume, CO2 concentration, gas-oil ratio variation characteristics, abnormal gas breakthrough time, daily liquid production variation, water cut variation, injection-production response relationship, as well as tracer monitoring results, profile test results, inter-well interference test results, or other available field data.

[0106] When direct on-site monitoring data is limited, the evaluation results can be supplemented by considering anomalies in well group production dynamics, known distribution of high connectivity zones, historical gas channeling knowledge, and geological connectivity characteristics. If the identified high-risk injection-production well pairs are substantially consistent with the locations of actual gas-bearing wells, abnormal response wells, or known dominant channels, the identification results of this embodiment can be considered to have high reliability. If there is a significant discrepancy between the two, the evaluation indicators, evaluation period, indicator weights, or sample data can be corrected by incorporating new data, and steps S6 to S9 can be re-executed to achieve iterative optimization of the evaluation results.

[0107] Step S10-2: Output control recommendations; Based on the verified CO2 dominant channel identification results and gas channeling risk levels, differentiated control recommendations are made for output from different injection and production wells.

[0108] (1) For injection-production well pairs with high comprehensive evaluation scores and classified as high-risk areas and relatively high-risk areas, they should be given priority for treatment. It is recommended to take measures such as optimizing injection-production structure, adjusting injection-production intensity, optimizing single-well injection allocation, implementing layered injection-production, profile adjustment to block water and gas, blocking high-permeability advantageous channels, improving swept volume, or other control and regulation measures to reduce abnormal connectivity between wells and the risk of CO2 crossflow.

[0109] (2) For injection-production well pairs that are classified as medium-risk areas, it is recommended to strengthen dynamic tracking and on-site monitoring, appropriately optimize injection parameters and production regime, and continuously monitor changes in the inter-well connectivity status to prevent them from further evolving into high-risk channels.

[0110] (3) For injection-production well pairs that are classified as lower risk areas and low risk areas, the main focus is on routine monitoring, rolling evaluation and development dynamic tracking. On the basis of maintaining the basic stability of the existing development system, risk reviews are carried out regularly.

[0111] The above-mentioned control recommendations can be linked with well number, well group number, comprehensive evaluation score, risk level and spatial location to form a risk management list for injection and production well pairs for field application, providing a basis for block injection and production optimization and management measure deployment.

[0112] Example 1: This embodiment, in conjunction with a CCUS injection development block in a low-permeability reservoir, further illustrates the CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS proposed in the above embodiment.

[0113] This embodiment selects a low-permeability oil reservoir CCUS injection development block as the research object. This block has 23 CO2 injection wells and 45 production wells. The method of this invention is used to identify and evaluate potential CO2 dominant migration channels and gas channeling risks during the CCUS injection development process of this block.

[0114] First, following step S1 above, multi-source data related to CCUS injection gas channeling risk identification within the target block are collected and preprocessed. This data includes static data for individual wells, such as geological data, logging data, and engineering stimulation data for both injection and production wells, as well as dynamic injection-production parameters at the well group level. After performing integrity checks, outlier screening, missing value handling, unit standardization, duplicate data removal, and injection-production relationship verification on the collected data, original datasets for individual injection wells, original datasets for individual production wells, and evaluation datasets for injection-production well pairs are generated.

[0115] Secondly, following steps S2 to S5 above, a comprehensive evaluation index system for injection wells and production wells is constructed, and a comprehensive evaluation of the single-well formation is completed. In this embodiment, 9 static indicators are used for the evaluation of injection wells, and 13 static indicators are used for the evaluation of production wells. No dynamic production data is introduced into the evaluation of the single-well formation. Subsequently, attribute determination, homogenization, and standardization are performed on the single-well evaluation indicators. The CRITIC method is used to calculate the objective weights of each indicator, and the TOPSIS method is used to obtain the comprehensive evaluation scores for each injection well and production well, thus obtaining the comprehensive evaluation results for each single well. Based on the evaluation scores, the injection wells and production wells can be ranked separately. The higher the score, the stronger the reservoir conditions and channel formation potential of the corresponding single well. The comprehensive evaluation results for each single well are shown in Table 1.

[0116] Table 1: Comprehensive Evaluation Results of Injection Wells and Production Wells

[0117] After completing the single-well formation evaluation, following step S6 above, injection-production well pairs were established as evaluation targets based on the actual injection-production correspondence within the block. According to the well network deployment relationship, well spacing conditions, and formation correspondence, a total of 99 injection-production well pair relationships were selected, and a well group formation advantage channel and gas channeling risk evaluation index system was constructed. In this embodiment, the comprehensive physical property score of the injection well, the comprehensive physical property score of the production well, the injection-production well spacing, the average daily injection volume, the average daily fluid production, water cut, the average daily water production, the oil production / injection ratio, and the water production / injection ratio were selected as well group formation evaluation indicators, with the evaluation period from January 2024 to June 2025. These indicators reflect both the reservoir conditions at both ends of the injection and production wells and comprehensively embody the inter-well connectivity conditions and the dynamic response characteristics of injection and production.

[0118] Then, following steps S7 and S8 above, the evaluation indicators for the well group layers are subjected to attribute determination, homogenization, and standardization. The CRITIC method is used to determine the objective weights of the evaluation indicators for each well group layer, and the TOPSIS method is used to calculate the comprehensive evaluation score for each injection-production well pair, which is then used as the gas channeling risk score for the corresponding injection-production well pair. The higher the score, the more likely the corresponding injection-production well pair is to form a dominant CO2 migration channel, and the higher its risk of gas channeling.

[0119] Further, following step S9 above, the injection-production well pairs are sorted and classified according to their gas channeling risk scores. These pairs can be categorized into five levels: high risk, relatively high risk, medium risk, relatively low risk, and low risk, thereby identifying potential CO2 dominant channels and key control well groups within the block. The gas channeling risk scores and classification results are shown in Table 2. Combined with the spatial distribution of well locations, a gas channeling risk map between wells in the block can be further drawn to visually demonstrate the spatial distribution characteristics of dominant channels and the distribution of high-risk areas.

[0120] Table 2: Gas Channeling Risk Scores and Classification of Injection and Production Wells

[0121] Based on the results in Table 2, a bar chart of well-to-well gas channeling risk scores, a distribution chart of gas channeling risk scores, and a spatial distribution chart of gas channeling risk in the block can be further generated. This will allow for a direct representation of the distribution characteristics of potential advantageous channels and key control targets within the block, providing a basis for developing targeted control measures on-site.

[0122] Finally, following step S10 above, the evaluation results were verified on-site, and control recommendations were output. Field application verification showed that for well pairs identified as having high gas channeling risk, their corresponding injection wells ZR4-9, ZR4-11, and ZR1-4 were shut down; for well groups identified as having relatively high risk, dynamic monitoring and production tracking were strengthened.

[0123] The on-site verification results show that the identification results obtained by the above embodiments of the present invention are in good consistency with the actual development of the block, verifying the applicability and effectiveness of the method proposed in the above embodiments in identifying gas channeling risks in CCUS injection development.

[0124] This embodiment demonstrates that, under conditions where direct monitoring data such as wellhead gas production, gas-oil ratio, and CO2 concentration in produced gas are relatively limited, the present invention can make full use of existing static geological data and conventional injection-production dynamic data to identify CO2 dominant channels and classify gas channeling risks during CCUS injection development, providing technical support for on-site risk warning, dynamic control, and optimal measures.

[0125] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for identifying CCUS injection gas channeling risk based on CRITIC-double-layer TOPSIS, characterized in that, Includes the following steps: Step S1: Perform multi-source data acquisition and preprocessing; Acquire multi-source data related to CCUS injected gas channeling risk identification within the target block, and after preprocessing the data, form raw datasets for single injection wells, raw datasets for single production wells, and evaluation datasets for injection-production well pairs. Step S2: Based on the original datasets of injection wells and production wells formed in Step S1, construct the comprehensive evaluation index system for injection wells and the comprehensive evaluation index system for production wells, respectively, and establish the original data matrix with each well as the evaluation object. Step S3: Determine the attributes of the single-well evaluation indicators obtained in Step S2, classify them into positive indicators and negative indicators, and convert all indicators into a uniform form, perform dimensionless processing, and establish a standardized matrix for injection wells and a standardized matrix for oil production wells. Step S4: Based on the standardized matrix of injection wells and oil production wells obtained in step S3, the weights of each evaluation index are determined using the CRITIC method, and the injection wells and oil production wells are comprehensively evaluated using the TOPSIS method. Step S5: Based on the TOPSIS comprehensive evaluation scores of the injection well and the production well obtained in Step S4, output the comprehensive evaluation result table of the injection well and the comprehensive evaluation result table of the production well respectively, so as to provide basic input data for subsequent gas channeling risk identification of the injection and production wells. Step S6: Based on the multi-source data obtained in Step S5, establish a well group layer evaluation index system to characterize inter-well connectivity, the degree of development of dominant channels, and the risk of CO2 gas channeling; Step S7: Perform attribute determination and standardization on the well group layer evaluation indicators to obtain the well group layer standardized evaluation matrix; Step S8: Based on the well group standardized evaluation matrix obtained in step S7, the CRITIC method is used to assign overall weight to all injection-production well pairs, and the proximity coefficient of each injection-production well pair is calculated as the comprehensive evaluation score of the well group. Step S9: Sort all injection and production well pairs according to the comprehensive evaluation score of the well group obtained in step S8, and classify the CO2 gas channeling risk level using a fixed threshold. Step S10: Compare and analyze the CO2 advantageous channels and high-risk injection-production wells identified in Step S9 with the development dynamic data, field monitoring data and historical understanding of the target block, and output differentiated control suggestions based on the verification results.

2. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S1, The static data for a single well includes geological data, well logging data, and engineering data; The geological data include at least reservoir thickness, perforation thickness, porosity, permeability, permeability gradient, hydrocarbon saturation, and clay content. The well logging data shall include at least resistivity, sonic transit time, and rock density; the engineering data shall include at least sand addition volume, sand ratio, displacement, and fluid injection volume. The dynamic parameters of well group injection and production include at least the injection-production well spacing, average daily injection volume, average daily fluid production, water cut, and corresponding parameters of well group.

3. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S1, The preprocessing process includes, in sequence, well name unification, coordinate unification, layer alignment, unit conversion, outlier identification, missing value completion, and duplicate data removal. The outlier identification process includes one or more of the statistical threshold method, box plot method, or 3σ method; the missing value completion process includes one or more of the mean substitution method, interpolation method, or regression estimation method.

4. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S2, The comprehensive evaluation index system for injection wells includes perforation thickness, porosity, permeability, permeability gradient, hydrocarbon saturation, resistivity, sonic transit time, rock density, and clay content. The comprehensive evaluation index system for a single oil well includes static evaluation indexes and parameters for well completion and fracturing stimulation engineering. The static evaluation indicators include perforation thickness, porosity, permeability, permeability gradient, oil and gas saturation, resistivity, sonic transit time, rock density, and clay content; the well completion and fracturing engineering parameters include sand addition volume, sand ratio, displacement, and fluid injection volume.

5. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S2, the process of establishing the original data matrix with a single well as the evaluation object includes: Assume there are a total of n I Injection wells, each injection well selected m I Based on these indicators, the original data matrix of the injection wells is as follows: ； Where, x i,j (I) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n I ; j =1,2,···, m I ; Assume there are a total of n P Each oil well is selected as a production well. m P Based on these indicators, the original data matrix of the oil wells is as follows: ； Where, x i,j (P) Indicates the first i Injection well in the first j The original values ​​of the evaluation indicators i = 1,2,···, n P ; j =1,2,···, m P .

6. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S3, the process of converting the positive and negative indicators of a single well evaluation index into a unified unidirectional form includes: Let the original data matrix of any single well evaluation object be: ； in, N Indicates the number of evaluation objects; M Indicates the number of evaluation indicators; x i,j Indicates the first i The evaluation object is in the first j The original values ​​of the indicators; The matrix after the unification process is denoted as: ； in, Determine as follows: ； in, ε This represents a small constant used to prevent the denominator from being zero.

7. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 6, characterized in that, In step S3, the Z-score normalization method is used to perform dimensionless processing on the homogenized index data, specifically including: Let the evaluation matrix after homogenization be... X , obtained the j Sample mean of each indicator and standard deviation They are respectively: ； ； The standardized index based on the Z-score standardization method is: ； Let the standardized matrix be: ； in, Z This represents the basic input matrix for subsequent calculation of CRITIC indicator information, determination of objective weights, and evaluation of TOPSIS closeness.

8. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S4, the process of determining the objective weights of single-well evaluation indicators based on the CRITIC method includes: For any type of single-well normalized matrix obtained in step S3 Z ; Step S4-1-1: Calculate the degree of variation of the index: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] j The sample standard deviation of the evaluation index is denoted as . The calculation formula is as follows: ； in, Indicates the first j The sample mean of each evaluation indicator across all evaluation objects; z i,j Indicates the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; N Indicates the number of evaluation objects; ； Step S4-1-2: Calculate the conflict of indices: Calculate the Pearson correlation coefficient matrix R between indices. r j,k ]: ； Using the CRITIC method to measure indicators j Other indicators k Conflictality is defined as: ； in, r j,k Indicates the first j Evaluation indicators and the first k Pearson correlation coefficient among the evaluation indicators; z i,j For the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first j The sample mean of each evaluation indicator across all evaluation objects; z i,k Indicates the first i The evaluation object is in the first k The standardized dimensionless Z-score value for each evaluation indicator; Indicates the first k The sample mean of each evaluation indicator across all evaluation objects; M Indicates the number of evaluation indicators; H j Indicates the conflict coefficient; Step S4-1-3: Calculate the information content of the indicators: [This step involves] calculating the information content of the indicators. j Comprehensive information content C j Defined as the product of its standard deviation and its conflict rate: ； Step S4-1-4: Determine objective weights: Normalize the information content of each indicator to obtain the first... j CRITIC objective weight of the evaluation indicators : ； Therefore, the weight vector is obtained as follows: 。 9. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 8, characterized in that, In step S4, the process of comprehensively evaluating the injection well and the production well using the TOPSIS method includes: Step S4-2-1: Construct the weighted evaluation matrix V: ； ； in: v i,j Indicates the first i The evaluation object is in the first j Weighted values ​​on the indicators; N Indicates the number of evaluation objects; M Indicates the number of evaluation indicators; w j Indicates the first j CRITIC objective weights of the evaluation indicators; z i,j Indicates the first i The evaluation object is in the first j The standardized dimensionless Z-score value for each evaluation indicator; Step S4-2-2: Determine the ideal solution V + and negative ideal solution V - : ； ； Step S4-2-3: Calculate the distance measure: Calculate the first... i Euclidean distance from each evaluation object to the positive and negative ideal solutions and They are respectively: ； ； Step S4-2-4: Calculate the proximity coefficient and overall score: Use the TOPSIS method to map the degree of excellence to the proximity coefficient. C i : 。 10. The CCUS injection gas channeling risk identification method based on CRITIC-double-layer TOPSIS as described in claim 1, characterized in that, In step S6, the comprehensive evaluation index system for well groups includes basic indexes and derived indexes, which are as follows: For the i For each injection-production well pair, the original evaluation index vector of its well group is constructed as follows: ； If a total of P injection-production well pairs are constructed within the target block, a raw evaluation data matrix of the well group layer can be further formed: ； in, Indicates the first i The comprehensive physical property score of the corresponding injection well in each injection-production well pair; Indicates the first i The comprehensive physical property score of each injection-production well relative to the corresponding production well; Indicates the first i The well spacing between injection and production well pairs; Indicates the first i The average daily injection volume of the corresponding injection well in each injection-production well during the evaluation period; Indicates the first i The average daily fluid production of each injection-production well and its corresponding production well during the evaluation period; Indicates the first i Water cut parameters of each injection-production well and its corresponding production well; Indicates the first i Average daily water production of each injection-production well pair; Indicates the first i Average daily oil production of each injection-production well pair; Indicates the first i The production / injection ratio of each injection-production well to its corresponding production well during the evaluation period; Indicates the first i The water production / injection ratio of each injection-production well to the corresponding production well during the evaluation period.

Images