Dynamic evaluation method for water injection risk of high-flow-rate reinjection well based on fuzzy logic reasoning

Abstract

The invention discloses a high-flow-rate reinjection well water injection risk dynamic assessment method based on fuzzy logic reasoning, which comprises the following steps: S1, acquiring monitoring data of a high-flow-rate reinjection well, and carrying out normalization processing; s2, constructing a fuzzy logic reasoning model based on a Kailey manifold fuzzy reasoning algorithm, and determining an initial reasoning path; s3, based on a seepage Lagrange optimization model, optimizing a reasoning path in a high-dimensional Kailey manifold space; s4, calculating a seepage critical value, adjusting a fuzzy inference rule, and dynamically updating an inference path based on time sequence analysis; s5, performing historical data projection, and calculating the evolution trend of the seepage risk based on a time scale transformation method; s6, updating the reasoning path calculation weight according to different water injection pressures and permeability; and S7, based on a calculation result of the reasoning path, adjusting a calculation structure of the reasoning model. The problems that in the prior art, water injection risk assessment of the high-flow-speed reinjection well is inaccurate, and a risk prediction model cannot be dynamically adjusted are effectively solved.

Description

Technical Field

[0001] The invention relates to the technical field of fuzzy logic reasoning, and in particular to a method for dynamically evaluating water injection risks of a high-flow rate reinjection well based on fuzzy logic reasoning. Background Art

[0002] With the continuous growth of energy demand, fields such as oil and gas extraction and geothermal energy utilization have put forward higher requirements for efficient and safe water injection and reinjection technology. High-flow reinjection wells play a key role in oil field production, geothermal energy reinjection and wastewater reinjection. Their main purpose is to maintain formation pressure, improve recovery or achieve energy recycling by injecting water or other fluids. However, in actual applications, the reinjection process involves complex seepage mechanisms, changes in formation permeability, fluctuations in water injection pressure and non-uniform seepage characteristics of fluids, making accurate assessment and dynamic optimization of reinjection risks a challenge.

[0003] Traditional seepage simulation methods have poor adaptability to complex formations. Formation seepage models based on Darcy's law are generally applicable to homogeneous formations, but in the actual reinjection process, the formations are often highly heterogeneous, including multi-layer permeability structures, fracture networks, and local high-permeability areas. These complex formation characteristics lead to highly nonlinear seepage processes. Traditional numerical simulation methods are difficult to accurately characterize the dynamic changes of pressure, permeability, and fluid viscosity during the reinjection process, which in turn affects the accuracy of risk assessment.

[0004] The existing technology lacks an effective dynamic optimization mechanism, making it difficult to achieve adaptive adjustment of the reinjection process. Existing reinjection optimization strategies mainly adjust the water injection scheme based on historical data, but cannot dynamically optimize the risk assessment model during the reinjection process. For example, fixed water injection pressure and flow settings may cause unstable seepage when formation conditions change, thereby increasing the risk of formation rupture or reduced injection efficiency. Therefore, the lack of real-time risk assessment and adaptive optimization capabilities makes it difficult for reinjection strategies to maintain optimal conditions under different environments.

[0005] In summary, the existing reinjection risk assessment technology has many limitations under complex formation conditions, including difficulty in handling highly heterogeneous formations, lack of reasoning ability for uncertain factors, lack of real-time dynamic optimization mechanism, and insufficient calculation stability under high flow rate conditions. Therefore, how to provide a dynamic assessment method for water injection risk of high flow rate reinjection wells based on fuzzy logic reasoning is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0006] One purpose of the present invention is to propose a dynamic assessment method for water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning, combining Cayley manifold fuzzy reasoning algorithm, seepage Lagrangian optimization model and time scale dynamic adjustment, to accurately model and optimize the seepage characteristics, water injection pressure and formation permeability of high-velocity reinjection wells in real time. The present invention makes full use of fuzzy logic reasoning for risk assessment, combines high-dimensional manifold space to optimize the reasoning path, and uses time series analysis to achieve dynamic weight adjustment, which effectively improves the accuracy and adaptability of risk assessment.

[0007] A method for dynamically assessing water injection risk of a high-velocity reinjection well based on fuzzy logic reasoning according to an embodiment of the present invention comprises the following steps:

[0008] S1. Obtain monitoring data of high-flow rate reinjection wells, extract formation permeability, injection pressure, reinjection flow rate, formation temperature and fracture distribution, perform normalization processing, and construct an input data set;

[0009] S2. Based on the Cayley-manifold fuzzy reasoning algorithm, a fuzzy logic reasoning model is established, a fuzzy rule set is constructed, a fuzzy membership function is defined, the nonlinear association weights of risk factors are calculated, and the initial reasoning path is determined;

[0010] S3, construct a seepage-driven Lagrangian optimization model, calculate the effect of fluid seepage behavior on the fuzzy reasoning path, and adjust the reasoning path in the high-dimensional Cayley manifold space;

[0011] S4, calculate the critical value of seepage, adjust the fuzzy reasoning rules according to the changes of fluid velocity, formation pressure and permeability, adjust the fuzzy membership mapping based on time series analysis, and dynamically update it on the reasoning path;

[0012] S5. Project historical data in high-dimensional Cayley manifold space, calculate the evolution trend of seepage risk based on the time scale transformation method, and adjust the key variables of the reasoning model according to the historical status;

[0013] S6, calculating key parameters under different water injection pressures and permeabilities, and updating calculation weights in the reasoning path;

[0014] S7. Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model.

[0015] Optionally, S2 includes the following steps:

[0016] S21. Setting the input variable set in the high-dimensional Cayley manifold space ,in Represent the factors that affect the water injection risk of the reinjection well, and construct variables into manifold space The mapping satisfies the embedding constraint:

[0017] ;

[0018] in, For variables Embedding function in the manifold, is the parameter that controls the distribution range of the variable;

[0019] S22. Setting fuzzy reasoning rules , using fuzzy reasoning relationship , Representation variables The fuzzy set of is the fuzzy set of output variables, is the fuzzy weight, which is calculated as:

[0020] ;

[0021] in, For variables In fuzzy sets The membership degree in Fuzzy inference rules Correlation factor;

[0022] S23. Exponential membership function using high-dimensional Cayley manifold:

[0023] ;

[0024] in, For variables On high-dimensional Cayley manifolds Up to the center of the fuzzy set The geodesic distance ; is a high-dimensional Cayley manifold The metric tensor of Adjust parameters for blur;

[0025] S24. Calculate the nonlinear association weights of risk factors and construct the association matrix of fuzzy inference rules:

[0026] ;

[0027] in, is a regularization parameter that controls the degree of correlation between different variables and is calculated using the geodesic distance on a high-dimensional Cayley manifold: ; is the inverse matrix of the metric tensor;

[0028] S25. Based on the Cayley-manifold fuzzy inference algorithm, set the energy functional of the inference path:

[0029] ;

[0030] in, is the state function of the reasoning path, Representation variables The membership degree in the fuzzy set is solved by Lagrangian optimization method; the optimal state of the reasoning path is calculated and satisfied ; To optimize the weight factor, is the Laplace-Beltrami operator on the manifold;

[0031] S26. Calculate the risk assessment value using the optimized reasoning path:

[0032] ;

[0033] in, For the risk assessment results, the weights of variables in the fuzzy rule set are adjusted according to different percolation states;

[0034] S27. Store the optimized fuzzy reasoning path to provide input for subsequent dynamic risk assessment.

[0035] Optionally, S3 includes the following steps:

[0036] S31. Set the fluid penetration state variable set ,in Representing the formation permeability, injection pressure and reinjection velocity, in the high-dimensional Cayley manifold space The seepage dynamics constraint conditions are established, the seepage Lagrangian optimization model is constructed to constrain the reasoning path to satisfy the seepage dynamics equation, and the functional :

[0037] ;

[0038] in, is the inverse matrix of the metric tensor on the high-dimensional Cayley manifold, is the inference path state function, represents the seepage potential energy field, is the Lagrange multiplier;

[0039] S32, functional Perform variational differentiation to derive the control equation for inference path optimization: ;in, is the Laplace-Beltrami operator, represents the critical value of seepage, and the inference path optimization is adjusted according to the dynamic changes of the seepage state;

[0040] S33. Set the constraints for the reasoning path optimization to satisfy the continuity equation of seepage dynamics: ;in, is the fluid density, For the fluid velocity field, the inference path optimization follows the physical constraints of the seepage motion;

[0041] S34, calculate the inference path optimization weights under different seepage states to satisfy: ;in, is the optimized fuzzy weight matrix, is the dynamic adjustment factor;

[0042] S35, calculating the variation range of key parameters based on the optimized reasoning path, and combining the time scale transformation method to adaptively adjust the reasoning path to maintain stability at different time scales;

[0043] S36. Store the optimized reasoning path and dynamically adjust it according to different formation structures and fluid states to provide input for subsequent risk assessment, and iterate during the optimization process of the reasoning path calculation structure.

[0044] Optionally, S4 includes the following steps:

[0045] S41. Set the seepage state variable set ,in Represents the formation permeability, injection pressure and reinjection velocity on the historical time scale, in the high-dimensional Cayley manifold space Calculate the time evolution trend of the variables and set the data projection mapping:

[0046] ;

[0047] in, Representation variables Cumulative changes in time scale, mapped to a high-dimensional Cayley manifold To make dynamic reasoning adjustments;

[0048] S42. Calculate the critical value of seepage based on the seepage Lagrangian optimization model , and derive the optimization equation:

[0049] ;

[0050] in, For variables The degree of membership in the fuzzy set is is the fuzzy weight;

[0051] S43, based on the critical value of seepage , adjust the fuzzy weight on the reasoning path , dynamically adjust the inference path weight matrix according to the changes in the infiltration state at different historical time scales;

[0052] S44, combined with time series analysis method, based on the seepage critical value Changes in membership mapping are adjusted during the fuzzy logic reasoning process;

[0053] S45, calculating the stability of the optimized reasoning path at different time scales, and dynamically adjusting the reasoning path according to the change of the seepage state;

[0054] S46. Store the optimized reasoning path and adjusted fuzzy reasoning rules to provide input for subsequent risk assessment and perform iterative adjustments at different time scales.

[0055] Optionally, S5 includes the following steps:

[0056] S51. Set the historical data set of percolation in the high-dimensional Cayley manifold space. A data projection mapping is established to make the historical data satisfy the time evolution constraint in the manifold space;

[0057] S52, based on the time scale transformation method, time mapping the historical data, and setting the fluid permeation state at different time scales;

[0058] S53. In the high-dimensional Cayley manifold space, based on the seepage Lagrangian optimization model, the seepage risk change rate at different time scales is calculated, and the dynamic adjustment relationship of the reasoning path on the time scale is established;

[0059] S54. According to the historical status, adjust the key variables of the reasoning model and calculate the changing trend of risk factors at different time scales;

[0060] S55. Based on the fuzzy logic reasoning method, fuzzy clustering analysis is performed on key variables in historical data to adjust the weights of fuzzy rules;

[0061] S56. Calculate the stability of the inference path at different time scales;

[0062] S57. Store the calculated risk evolution trend data, and perform iterative optimization of the reasoning path in combination with the historical time data, so that the time evolution process of the reasoning model conforms to the changes in the seepage state under different time scales.

[0063] Optionally, the S6 comprises the following steps:

[0064] S61. Set a set of fluid penetration risk factors and calculate the The associated weights of the variables are calculated internally and the optimized weight matrix under different fluid states is constructed;

[0065] S62, calculating the variation range of key parameters based on the reasoning path, and optimizing the reasoning path according to the seepage state;

[0066] S63. Based on the weight adjustment strategy on high-dimensional Cayley manifold, the dynamic weight of the reasoning path under different percolation states is calculated, and the optimization factor is adjusted according to the time scale;

[0067] S64, calculating the risk distribution state under different water injection pressures and permeabilities, adjusting the key variables of the fuzzy reasoning rules, and adjusting the calculation weights on the reasoning path according to the seepage state;

[0068] S65. In the high-dimensional Cayley manifold space, the rate of change of the reasoning path is calculated according to the risk distribution state, the calculation parameters of the reasoning path under different pressure conditions are adjusted, and the convergence speed of the reasoning path is optimized;

[0069] S66. Calculate key parameters under different fluid states based on the optimized reasoning path, dynamically adjust the optimization factor of the reasoning path, and update the calculation weight based on historical data;

[0070] S67. Store the optimized reasoning path and perform iterative optimization under different water injection pressure and permeability conditions to provide input for subsequent risk assessment.

[0071] Optionally, the S7 includes the following steps:

[0072] S71, set the inference path calculation result set, in the high-dimensional Cayley manifold space Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model;

[0073] S72, according to the optimized reasoning path, adjusting the calculation weight of the reasoning rule, optimizing the calculation structure, and calculating the weight matrix according to the optimized reasoning path, and optimizing the calculation parameters of the reasoning rule under different stratum structure conditions;

[0074] S73. In the process of optimizing the calculation structure of the inference model, a seepage state adjustment factor is introduced to optimize the dynamic allocation of the inference rules, and the key variables in the calculation structure are optimized based on historical data;

[0075] S74, based on the calculation results of the reasoning path, adjusting the calculation weight of the reasoning model, optimizing the calculation structure, making the calculation structure adapt to the permeability state of different formations, and calculating the distribution of key variables according to the optimized reasoning path;

[0076] S75. According to the optimized calculation structure, the calculation stability of the reasoning model under different stratum structures is adjusted, and dynamically updated in the calculation process of the reasoning path;

[0077] S76. Store the optimized calculation structure and perform iterative optimization under different formation permeability states to provide input for subsequent risk assessment and optimize the updating mechanism of the inference rules in the inference path.

[0078] The beneficial effects of the present invention are:

[0079] The present invention proposes a dynamic assessment method for water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning. The fuzzy reasoning rule set is constructed through the Cayley manifold fuzzy reasoning algorithm, the nonlinear association weight of the risk factor is calculated, and the risk assessment model is optimized using the manifold geodesic distance. This method can effectively handle the uncertainty of seepage data under complex formation conditions, improve the accuracy of reinjection risk prediction, and make risk assessment more accurate and stable.

[0080] The present invention introduces a seepage Lagrangian optimization model, optimizes the reasoning path in the high-dimensional Cayley manifold space, calculates the impact of fluid permeability behavior on the fuzzy reasoning path, and optimizes the risk reasoning path through the Laplace-Beltrami operator. Compared with the traditional seepage simulation method, this method can adaptively adjust the reasoning path to meet the permeability conditions of different formations, reduce the evaluation error caused by formation heterogeneity and changes in reinjection rate, and improve the adaptability of reinjection risk assessment.

[0081] Based on time series analysis and time scale transformation methods, the present invention dynamically calculates the evolution trend of seepage risk, projects historical data in high-dimensional Cayley manifold space, and optimizes and adjusts key variables through inference path optimization. This method can capture the changing laws of formation permeability characteristics during long-term reinjection, effectively improve the accuracy of long-term risk assessment, and provide support for intelligent water injection management in complex reinjection scenarios.

[0082] Based on the feedback mechanism of the inference path calculation results, the present invention realizes the dynamic optimization of the risk assessment model, and adaptively adjusts the fuzzy rules and optimization weights under different water injection pressures, flow rates and permeability conditions. Compared with the traditional fixed parameter model, this method can continuously optimize the risk assessment strategy based on real-time monitoring data, improve the reliability of prediction, and ensure the safety and long-term stable operation of the reinjection well under complex formation conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0084] Figure 1 A flow chart of a method for dynamic assessment of water injection risk in high-velocity reinjection wells based on fuzzy logic reasoning proposed by the present invention;

[0085] Figure 2A flow chart for constructing a risk assessment model based on Cayley manifold fuzzy inference algorithm proposed in the present invention;

[0086] Figure 3 This is a schematic diagram of the inference path optimization based on the seepage Lagrangian optimization model proposed in the present invention. DETAILED DESCRIPTION

[0087] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0088] refer to Figure 1-3 The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning includes the following steps:

[0089] S1. Obtain monitoring data of high-flow rate reinjection wells, extract formation permeability, injection pressure, reinjection flow rate, formation temperature and fracture distribution, perform normalization processing, and construct an input data set;

[0090] S2. Based on the Cayley-manifold fuzzy reasoning algorithm, a fuzzy logic reasoning model is established, a fuzzy rule set is constructed, a fuzzy membership function is defined, the nonlinear association weights of risk factors are calculated, and the initial reasoning path is determined;

[0091] S3, construct a seepage-driven Lagrangian optimization model, calculate the effect of fluid seepage behavior on the fuzzy reasoning path, and adjust the reasoning path in the high-dimensional Cayley manifold space;

[0092] S4, calculate the critical value of seepage, adjust the fuzzy reasoning rules according to the changes of fluid velocity, formation pressure and permeability, adjust the fuzzy membership mapping based on time series analysis, and dynamically update it on the reasoning path;

[0093] S5. Project historical data in high-dimensional Cayley manifold space, calculate the evolution trend of seepage risk based on the time scale transformation method, and adjust the key variables of the reasoning model according to the historical status;

[0094] S6, calculating key parameters under different water injection pressures and permeabilities, and updating calculation weights in the reasoning path;

[0095] S7. Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model.

[0096] In this implementation, S2 includes the following steps:

[0097] S21. Setting the input variable set in the high-dimensional Cayley manifold space ,in Represent the factors that affect the water injection risk of the reinjection well, and construct variables into manifold space The mapping satisfies the embedding constraint:

[0098] ;

[0099] in, For variables Embedding function in the manifold, is the parameter that controls the distribution range of the variable;

[0100] S22. Setting fuzzy reasoning rules , using fuzzy reasoning relationship , Representation variables The fuzzy set of is the fuzzy set of output variables, is the fuzzy weight, which is calculated as:

[0101] ;

[0102] in, For variables In fuzzy sets The membership degree in Fuzzy inference rules Correlation factor;

[0103] S23. Exponential membership function using high-dimensional Cayley manifold:

[0104] ;

[0105] in, For variables On high-dimensional Cayley manifolds Up to the center of the fuzzy set The geodesic distance ; is a high-dimensional Cayley manifold The metric tensor of Adjust parameters for blur;

[0106] S24. Calculate the nonlinear association weights of risk factors and construct the association matrix of fuzzy inference rules:

[0107] ;

[0108] in, is a regularization parameter that controls the degree of correlation between different variables and is calculated using the geodesic distance on a high-dimensional Cayley manifold: ; is the inverse matrix of the metric tensor;

[0109] S25. Based on the Cayley-manifold fuzzy inference algorithm, set the energy functional of the inference path:

[0110] ;

[0111] in, is the state function of the reasoning path, Representation variables The membership degree in the fuzzy set is solved by Lagrangian optimization method; the optimal state of the reasoning path is calculated and satisfied ; To optimize the weight factor, is the Laplace-Beltrami operator on the manifold;

[0112] S26. Calculate the risk assessment value using the optimized reasoning path:

[0113] ;

[0114] in, For the risk assessment results, the weights of variables in the fuzzy rule set are adjusted according to different percolation states;

[0115] S27. Store the optimized fuzzy reasoning path to provide input for subsequent dynamic risk assessment.

[0116] In this implementation, S3 includes the following steps:

[0117] S31. Set the fluid penetration state variable set ,in Representing the formation permeability, injection pressure and reinjection velocity, in the high-dimensional Cayley manifold space The seepage dynamics constraint conditions are established, the seepage Lagrangian optimization model is constructed to constrain the reasoning path to satisfy the seepage dynamics equation, and the functional :

[0118] ;

[0119] in, is the inverse matrix of the metric tensor on the high-dimensional Cayley manifold, is the inference path state function, represents the seepage potential field, is the Lagrange multiplier;

[0120] S32, functional Perform variational differentiation to derive the control equation for inference path optimization: ;in, is the Laplace-Beltrami operator, represents the critical value of seepage, and the inference path optimization is adjusted according to the dynamic changes of the seepage state;

[0121] S33. Set the constraints for the reasoning path optimization to satisfy the continuity equation of seepage dynamics: ;in, is the fluid density, For the fluid velocity field, the inference path optimization follows the physical constraints of the seepage motion;

[0122] S34, calculate the inference path optimization weights under different seepage states to satisfy: ;in, is the optimized fuzzy weight matrix, is the dynamic adjustment factor;

[0123] S35, calculating the variation range of key parameters based on the optimized reasoning path, and combining the time scale transformation method to adaptively adjust the reasoning path to maintain stability at different time scales;

[0124] S36. Store the optimized reasoning path and dynamically adjust it according to different formation structures and fluid states to provide input for subsequent risk assessment, and iterate during the optimization process of the reasoning path calculation structure.

[0125] In this implementation, S4 includes the following steps:

[0126] S41. Set the seepage state variable set ,in Represents the formation permeability, injection pressure and reinjection velocity on the historical time scale, in the high-dimensional Cayley manifold space Calculate the time evolution trend of the variables and set the data projection mapping:

[0127] ;

[0128] in, Representation variables Cumulative changes in time scale, mapped to a high-dimensional Cayley manifold To make dynamic reasoning adjustments;

[0129] S42. Calculate the critical value of seepage based on the seepage Lagrangian optimization model , and derive the optimization equation:

[0130] ;

[0131] in, For variables The degree of membership in the fuzzy set is is the fuzzy weight;

[0132] S43, based on the critical value of seepage , adjust the fuzzy weight on the reasoning path , dynamically adjust the inference path weight matrix according to the changes in the infiltration state at different historical time scales;

[0133] S44, combined with time series analysis method, based on the seepage critical value Changes in membership mapping are adjusted during the fuzzy logic reasoning process;

[0134] S45, calculating the stability of the optimized reasoning path at different time scales, and dynamically adjusting the reasoning path according to the change of the seepage state;

[0135] S46. Store the optimized reasoning path and adjusted fuzzy reasoning rules to provide input for subsequent risk assessment and perform iterative adjustments at different time scales.

[0136] In this implementation, S5 includes the following steps:

[0137] S51. Set the historical data set of percolation in the high-dimensional Cayley manifold space. A data projection mapping is established to make the historical data satisfy the time evolution constraint in the manifold space;

[0138] S52, based on the time scale transformation method, time mapping the historical data, and setting the fluid permeation state at different time scales;

[0139] S53. In the high-dimensional Cayley manifold space, based on the seepage Lagrangian optimization model, the seepage risk change rate at different time scales is calculated, and the dynamic adjustment relationship of the reasoning path on the time scale is established;

[0140] S54. According to the historical status, adjust the key variables of the reasoning model and calculate the changing trend of risk factors at different time scales;

[0141] S55. Based on the fuzzy logic reasoning method, fuzzy clustering analysis is performed on key variables in historical data to adjust the weights of fuzzy rules;

[0142] S56. Calculate the stability of the inference path at different time scales;

[0143] S57. Store the calculated risk evolution trend data, and perform iterative optimization of the reasoning path in combination with the historical time data, so that the time evolution process of the reasoning model conforms to the changes in the seepage state under different time scales.

[0144] In this implementation, S6 includes the following steps:

[0145] S61. Set a set of fluid penetration risk factors and calculate the The associated weights of the variables are calculated internally and the optimized weight matrix under different fluid states is constructed;

[0146] S62, calculating the variation range of key parameters based on the reasoning path, and optimizing the reasoning path according to the seepage state;

[0147] S63. Based on the weight adjustment strategy on high-dimensional Cayley manifold, the dynamic weight of the reasoning path under different percolation states is calculated, and the optimization factor is adjusted according to the time scale;

[0148] S64, calculating the risk distribution state under different water injection pressures and permeabilities, adjusting the key variables of the fuzzy reasoning rules, and adjusting the calculation weights on the reasoning path according to the seepage state;

[0149] S65. In the high-dimensional Cayley manifold space, the rate of change of the reasoning path is calculated according to the risk distribution state, the calculation parameters of the reasoning path under different pressure conditions are adjusted, and the convergence speed of the reasoning path is optimized;

[0150] S66. Calculate key parameters under different fluid states based on the optimized reasoning path, dynamically adjust the optimization factor of the reasoning path, and update the calculation weight based on historical data;

[0151] S67. Store the optimized reasoning path and perform iterative optimization under different water injection pressure and permeability conditions to provide input for subsequent risk assessment.

[0152] In this implementation, S7 includes the following steps:

[0153] S71, set the inference path calculation result set, in the high-dimensional Cayley manifold space Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model;

[0154] S72, according to the optimized reasoning path, adjusting the calculation weight of the reasoning rule, optimizing the calculation structure, and calculating the weight matrix according to the optimized reasoning path, and optimizing the calculation parameters of the reasoning rule under different stratum structure conditions;

[0155] S73. In the process of optimizing the calculation structure of the inference model, a seepage state adjustment factor is introduced to optimize the dynamic allocation of the inference rules, and the key variables in the calculation structure are optimized based on historical data;

[0156] S74, based on the calculation results of the reasoning path, adjusting the calculation weight of the reasoning model, and optimizing the calculation structure, so that the calculation structure adapts to the permeability state of different formations, and calculating the distribution of key variables according to the optimized reasoning path;

[0157] S75. According to the optimized calculation structure, the calculation stability of the reasoning model under different stratum structures is adjusted, and dynamically updated in the calculation process of the reasoning path;

[0158] S76. Store the optimized calculation structure and perform iterative optimization under different formation permeability states to provide input for subsequent risk assessment and optimize the updating mechanism of the inference rules in the inference path.

[0159] Embodiment 1:

[0160] The present invention is applied to the high-flow rate reinjection well management system in an oil field block in a basin. The formation in this area is highly heterogeneous, with large permeability changes and complex fracture development. In the high-flow rate reinjection process, the traditional risk assessment method is not responsive to the change of formation permeability in time, which easily leads to the following problems: excessive reinjection pressure causes formation rupture, affecting production safety; the seepage path is blocked, reducing the reinjection efficiency and causing uneven water injection; it is impossible to accurately predict the critical seepage value, making it difficult to optimize the water injection strategy.

[0161] In order to overcome the above problems, the present invention proposes a dynamic assessment method for water injection risk of high-flow rate reinjection wells based on fuzzy logic reasoning, which utilizes Cayley manifold fuzzy reasoning, seepage Lagrangian optimization, time scale transformation and other technologies to improve the accuracy and adaptability of reinjection risk assessment and optimize the stability of the reinjection process.

[0162] Thirty high-flow rate reinjection wells were selected in the oil field block for testing, and the monitoring period was 6 months. High-precision sensors were deployed in the reinjection wells to collect key parameters such as formation permeability, injection pressure, reinjection flow rate, temperature, and fracture development in real time. The data were recorded once an hour, and the method of the present invention was used for risk assessment and optimization.

[0163] First, the Cayley manifold fuzzy inference algorithm is used to process 129,600 historical monitoring data, calculate the nonlinear association weights of each risk factor, and construct a fuzzy inference rule set. In traditional methods, the permeability fluctuation range is large (3-30mD), while this method calculates the fuzzy weights through the manifold geodesic distance, making the risk prediction more accurate and effectively improving the adaptability of risk assessment.

[0164] Based on the seepage Lagrangian optimization model, the optimal reinjection path under different water injection pressures was solved. After applying this method, the reinjection pressure fluctuation range was reduced from 6.2~7.9MPa to 6.7~7.3MPa, reducing the risk of formation rupture caused by abnormal pressure fluctuations. In addition, in areas with high heterogeneity, the injection pressure prediction error was reduced by 38.4%, optimizing the stability of risk assessment.

[0165] During the reinjection process, the evolution trend of seepage risk was calculated based on the time scale transformation method, and compared with historical data for analysis. In one block, four potential high-risk events (risk probability > 85%) were identified within one month, and an early warning was issued 60 minutes in advance, allowing on-site technicians to adjust the injection pressure in time to avoid sudden formation instability accidents.

[0166] Table 1 Comparison of risk assessment performance between the present invention and traditional methods

[0167] Evaluation Metrics Traditional methods Method of the present invention Improvement effect Risk prediction accuracy 72.30% 90.50% ↑18.2% High risk identification rate 79.80% 94.20% ↑14.4% Reinjection pressure fluctuation range (MPa) 6.2~7.9 6.7~7.3 ↓23.1% Calculation error of critical value of seepage 30.60% 16.30% ↓46.7% Reinjection stability improvement rate - 35.20% - Warning lead time (minutes) 10 60 ↑500%

[0168] Table 2 Reinjection risk assessment data under different water injection pressure conditions

[0169] Reinjection well number Formation permeability (mD) Water injection pressure (MPa) Flow rate (m³ / h) Predicted risk probability (%) Is there an abnormality? Well A01 15.6 6.8 35.2 60.2 no Well A02 9.2 7.2 30.8 72.3 no Well A03 24.1 7.3 40.1 85.7 yes Well A04 6.5 6.9 29.7 51.6 no Well A05 18.4 7.1 37.9 79.4 yes Well A06 21.7 7.3 38.5 91.2 yes

[0170] The risk prediction accuracy of the present invention is increased from 72.3% to 90.5%, and the high-risk identification rate is increased to 94.2%, effectively reducing the risk misjudgment rate. Under the traditional method, the reinjection pressure fluctuates greatly, which can easily lead to formation rupture. After the optimization of this method, the fluctuation of the reinjection pressure is reduced by 23.1%, effectively improving the stability of system operation. Under complex formation conditions, the calculation error of the traditional method is as high as 30.6%, while the calculation error of the method of the present invention is reduced to 16.3%, improving the reliability of the prediction. The traditional method can only warn 10 minutes before the risk occurs, while this method can predict the risk 60 minutes in advance, providing on-site technicians with sufficient time to adjust the reinjection parameters and avoid formation instability.

[0171] The present invention has been successfully applied in the management system of high-velocity reinjection wells in an oil field in a certain basin, significantly improving the accuracy of risk assessment and the stability of reinjection pressure control. Compared with traditional methods, the present invention optimizes risk assessment through fuzzy logic reasoning, adjusts the reinjection path through seepage Lagrangian optimization, and improves prediction ability through time scale transformation, effectively reducing reinjection anomalies caused by changes in formation structure and flow rate fluctuations. Experimental results show that this method can be widely used in complex seepage environments such as oil and gas production, geothermal reinjection, and groundwater reinjection, and improves the safety and long-term stability of reinjection wells.

[0172] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A dynamic assessment method for water injection risk in high-velocity reinjection wells based on fuzzy logic reasoning, characterized in that: The steps include: S1. Obtain monitoring data of high-flow rate reinjection wells, extract formation permeability, injection pressure, reinjection flow rate, formation temperature and fracture distribution, perform normalization processing, and construct an input data set; S2. Based on the Cayley-manifold fuzzy reasoning algorithm, a fuzzy logic reasoning model is established, a fuzzy rule set is constructed, a fuzzy membership function is defined, the nonlinear association weights of risk factors are calculated, and the initial reasoning path is determined; S3, construct a seepage-driven Lagrangian optimization model, calculate the effect of fluid seepage behavior on the fuzzy reasoning path, and adjust the reasoning path in the high-dimensional Cayley manifold space; S4, calculate the critical value of seepage, adjust the fuzzy reasoning rules according to the changes of fluid velocity, formation pressure and permeability, adjust the fuzzy membership mapping based on time series analysis, and dynamically update it on the reasoning path; S5. Project historical data in high-dimensional Cayley manifold space, calculate the evolution trend of seepage risk based on the time scale transformation method, and adjust the key variables of the reasoning model according to the historical status; S6, calculating key parameters under different water injection pressures and permeabilities, and updating calculation weights in the reasoning path; S7. Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model.

2. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S2 comprises the following steps: S21. Setting the input variable set in the high-dimensional Cayley manifold space ,in Represent the factors that affect the water injection risk of the reinjection well, and construct variables into manifold space The mapping satisfies the embedding constraint: ； in, For variables Embedding function in the manifold, is the parameter that controls the distribution range of the variable; S22. Setting fuzzy reasoning rules , using fuzzy reasoning relationship , Representation variables The fuzzy set of is the fuzzy set of output variables, is the fuzzy weight, which is calculated as: ； in, For variables In fuzzy sets The membership degree in Fuzzy inference rules Correlation factor; S23. Exponential membership function using high-dimensional Cayley manifold: ； in, For variables On high-dimensional Cayley manifolds Up to the center of the fuzzy set The geodesic distance ; is a high-dimensional Cayley manifold The metric tensor of Adjust parameters for blur; S24. Calculate the nonlinear association weights of risk factors and construct the association matrix of fuzzy inference rules: ； in, is a regularization parameter that controls the degree of correlation between different variables and is calculated using the geodesic distance on a high-dimensional Cayley manifold: ; is the inverse matrix of the metric tensor; S25. Based on the Cayley-manifold fuzzy inference algorithm, set the energy functional of the inference path: ； in, is the state function of the reasoning path, Representation variables The membership degree in the fuzzy set is solved by Lagrangian optimization method; the optimal state of the reasoning path is calculated and satisfied ; To optimize the weight factor, is the Laplace-Beltrami operator on the manifold; S26. Calculate the risk assessment value using the optimized reasoning path: ； in, For the risk assessment results, the weights of variables in the fuzzy rule set are adjusted according to different percolation states; S27. Store the optimized fuzzy reasoning path to provide input for subsequent dynamic risk assessment.

3. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S3 comprises the following steps: S31. Set the fluid penetration state variable set ,in Representing the formation permeability, injection pressure and reinjection velocity, in the high-dimensional Cayley manifold space The seepage dynamics constraint conditions are established, the seepage Lagrangian optimization model is constructed to constrain the reasoning path to satisfy the seepage dynamics equation, and the functional : ； in, is the inverse matrix of the metric tensor on the high-dimensional Cayley manifold, is the inference path state function, represents the seepage potential field, is the Lagrange multiplier; S32, functional Perform variational differentiation to derive the control equation for inference path optimization: ;in, is the Laplace-Beltrami operator, represents the critical value of seepage, and the inference path optimization is adjusted according to the dynamic changes of the seepage state; S33. Set the constraints for the reasoning path optimization to satisfy the continuity equation of seepage dynamics: ;in, is the fluid density, For the fluid velocity field, the inference path optimization follows the physical constraints of the seepage motion; S34, calculate the inference path optimization weights under different seepage states to satisfy: ;in, is the optimized fuzzy weight matrix, is the dynamic adjustment factor; S35, calculating the variation range of key parameters based on the optimized reasoning path, and combining the time scale transformation method to adaptively adjust the reasoning path to maintain stability at different time scales; S36. Store the optimized reasoning path and dynamically adjust it according to different formation structures and fluid states to provide input for subsequent risk assessment, and iterate during the optimization process of the reasoning path calculation structure.

4. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S4 comprises the following steps: S41. Set the seepage state variable set ,in Represents the formation permeability, injection pressure and reinjection velocity on the historical time scale, in the high-dimensional Cayley manifold space Calculate the time evolution trend of the variables and set the data projection mapping: ； in, Representation variables Cumulative changes in time scale, mapped to a high-dimensional Cayley manifold To make dynamic reasoning adjustments; S42. Calculate the critical value of seepage based on the seepage Lagrangian optimization model , and derive the optimization equation: ； in, For variables The degree of membership in the fuzzy set is is the fuzzy weight; S43, based on the critical value of seepage , adjust the fuzzy weight on the reasoning path , dynamically adjust the inference path weight matrix according to the changes in the infiltration state at different historical time scales; S44, combined with time series analysis method, based on the seepage critical value Changes in membership mapping are adjusted during the fuzzy logic reasoning process; S45, calculating the stability of the optimized reasoning path at different time scales, and dynamically adjusting the reasoning path according to the change of the seepage state; S46. Store the optimized reasoning path and adjusted fuzzy reasoning rules to provide input for subsequent risk assessment and perform iterative adjustments at different time scales.

5. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S5 comprises the following steps: S51. Set the historical data set of percolation in the high-dimensional Cayley manifold space. A data projection mapping is established to make the historical data satisfy the time evolution constraint in the manifold space; S52, based on the time scale transformation method, time mapping the historical data, and setting the fluid permeation state at different time scales; S53. In the high-dimensional Cayley manifold space, based on the seepage Lagrangian optimization model, the seepage risk change rate at different time scales is calculated, and the dynamic adjustment relationship of the reasoning path on the time scale is established; S54. According to the historical status, adjust the key variables of the reasoning model and calculate the changing trend of risk factors at different time scales; S55. Based on the fuzzy logic reasoning method, fuzzy clustering analysis is performed on key variables in historical data to adjust the weights of fuzzy rules; S56. Calculate the stability of the inference path at different time scales; S57. Store the calculated risk evolution trend data, and perform iterative optimization of the reasoning path in combination with the historical time data, so that the time evolution process of the reasoning model conforms to the changes in the seepage state under different time scales.

6. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S6 comprises the following steps: S61. Set a set of fluid penetration risk factors and calculate the The associated weights of the variables are calculated internally and the optimized weight matrix under different fluid states is constructed; S62, calculating the variation range of key parameters based on the reasoning path, and optimizing the reasoning path according to the seepage state; S63. Based on the weight adjustment strategy on high-dimensional Cayley manifold, the dynamic weight of the reasoning path under different percolation states is calculated, and the optimization factor is adjusted according to the time scale; S64, calculating the risk distribution state under different water injection pressures and permeabilities, adjusting the key variables of the fuzzy reasoning rules, and adjusting the calculation weights on the reasoning path according to the seepage state; S65. In the high-dimensional Cayley manifold space, the rate of change of the reasoning path is calculated according to the risk distribution state, the calculation parameters of the reasoning path under different pressure conditions are adjusted, and the convergence speed of the reasoning path is optimized; S66. Calculate key parameters under different fluid states based on the optimized reasoning path, dynamically adjust the optimization factor of the reasoning path, and update the calculation weight based on historical data; S67. Store the optimized reasoning path and perform iterative optimization under different water injection pressure and permeability conditions to provide input for subsequent risk assessment.

7. The method for dynamic assessment of water injection risk of high-velocity reinjection wells based on fuzzy logic reasoning according to claim 1 is characterized in that: The S7 comprises the following steps: S71, set the inference path calculation result set, in the high-dimensional Cayley manifold space Based on the calculation results of the reasoning path, adjust the calculation structure of the reasoning model; S72, according to the optimized reasoning path, adjusting the calculation weight of the reasoning rule, optimizing the calculation structure, and calculating the weight matrix according to the optimized reasoning path, and optimizing the calculation parameters of the reasoning rule under different stratum structure conditions; S73. In the process of optimizing the calculation structure of the inference model, a seepage state adjustment factor is introduced to optimize the dynamic allocation of the inference rules, and the key variables in the calculation structure are optimized based on historical data; S74, based on the calculation results of the reasoning path, adjusting the calculation weight of the reasoning model, and optimizing the calculation structure, so that the calculation structure adapts to the permeability state of different formations, and calculating the distribution of key variables according to the optimized reasoning path; S75. According to the optimized calculation structure, the calculation stability of the reasoning model under different stratum structures is adjusted, and dynamically updated in the calculation process of the reasoning path; S76. Store the optimized calculation structure and perform iterative optimization under different formation permeability states to provide input for subsequent risk assessment and optimize the updating mechanism of the inference rules in the inference path.

Images