Carbon dioxide foam fracturing fluid gas injection parameter multi-objective optimization method and device

Abstract

The application discloses a kind of carbon dioxide foam fracturing fluid gas injection parameter multi-objective optimization method and device, wherein, method includes: obtaining target formation geologic parameter and engineering constraint parameter and carrying out normalization processing;Normalize geologic parameter is converted into dynamic constraint condition associated with equipment pressure resistance capacity, to dynamically adjust gas injection pressure upper limit;Multi-objective function is constructed with fracture complexity, flowback rate and carbon emission as optimization target;With dynamic constraint and engineering constraint as boundary, improved multi-objective evolutionary algorithm is used to solve function, which generates uniformly distributed reference points in target space and uses reference point correlation strategy for population evolution, to generate Pareto optimal solution set;According to engineering constraint, the solution set is filtered to obtain the feasible optimal solution set, and the gas injection pressure, discharge capacity and carbon dioxide phase proportion are real-time closed-loop controlled according to the optimal solution set.The application realizes collaborative optimization, and significantly improves fracturing effect and low-carbon performance.

Description

Technical Field

[0001] This invention relates to the field of well site fracturing technology, specifically to a multi-objective optimization method and apparatus for carbon dioxide foam fracturing fluid injection parameters. Background Technology

[0002] Traditional carbon dioxide foam fracturing devices and methods suffer from poor multi-objective coordination capabilities. While conventional water-based fracturing fluids (such as slickwater) can form complex fractures through large-volume injection, they have low flowback rates and liquid residues that lead to secondary pollution. Although carbon dioxide foam fracturing fluids are water-saving and have high flowback rates, they are limited by low gas injection volume, resulting in insufficient fracture complexity, and they do not quantify carbon emission targets.

[0003] Existing algorithms struggle to dynamically balance geological parameters and engineering constraints when dealing with multiple objectives, and suffer from issues such as solution set offset and insufficient population diversity. Furthermore, the lack of real-time sensing and closed-loop control of formation property changes leads to lag in parameter adjustments, making it difficult to adapt to the demands of complex reservoirs.

[0004] In view of this, the present invention is hereby proposed. Summary of the Invention

[0005] To address the problems of limited fracture morphology, low flowback efficiency, and high carbon emissions in existing fracturing technologies, this invention proposes a multi-objective optimization method and apparatus for carbon dioxide foam fracturing fluid injection parameters. Specifically, the following technical solution is adopted:

[0006] A multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters includes:

[0007] Obtain the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation. The geological parameters include the movable water saturation and permeability of the target fractured formation, and the engineering constraint parameters include the equipment pressure resistance threshold, the gas injection displacement range, and the upper limit of the carbon dioxide liquid phase ratio.

[0008] The geological parameters and engineering constraint parameters are normalized to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters.

[0009] Normalized geological parameters are transformed into dynamic constraints related to the equipment's pressure resistance, so as to dynamically adjust the upper limit of the gas injection pressure.

[0010] Construct a multi-objective function with crack complexity, backflow rate, and carbon emissions as optimization objectives;

[0011] Using the dynamic constraints and normalized engineering constraint parameters as boundaries, a multi-objective evolutionary algorithm is used to solve the multi-objective function; wherein, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy to perform individual selection and population evolution to generate a Pareto optimal solution set;

[0012] The Pareto optimal solution set is filtered based on the normalized engineering constraint parameters to obtain a feasible optimal solution set;

[0013] The target gas injection parameter combination is determined from the set of feasible optimal solutions, and the gas injection pressure, gas injection rate and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid are adjusted in real time based on the combination.

[0014] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters of the present invention, dynamic data of fracture propagation obtained by monitoring the target fracturing formation through microseismic monitoring is used to quantify the complexity of the fracture.

[0015] In addition, data on the rheological properties and phase change patterns of the carbon dioxide foam fracturing fluid used to construct the multi-objective function or the dynamic constraint conditions, determined through laboratory testing.

[0016] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters, the step of converting normalized geological parameters into dynamic constraints related to the pressure resistance of the equipment, so as to dynamically adjust the upper limit of the injection pressure, includes:

[0017] Based on the movable water saturation in the normalized geological parameters, the upper limit of the gas injection pressure is dynamically calculated through a preset functional relationship.

[0018] The preset functional relationship satisfies: Pmax=α×P_pump_max×(1-β×Sw_norm), where Pmax is the upper limit of the injection pressure obtained by dynamic calculation, α is the safety factor, β is the reduction factor of the formation's sensitivity to water lock damage or uncontrolled fracture height, P_pump_max is the upper limit threshold of the pressure resistance of the fracturing equipment, and Sw_norm is the normalized mobile water saturation with a value range of [0,1].

[0019] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters, the construction of a multi-objective function with fracture complexity, flowback rate, and carbon emissions as optimization objectives includes:

[0020] Based on microseismic monitoring data, a first objective function characterizing the complexity of fractures is constructed by using the fractal dimension of fractures or the ratio of the surface area to the volume of the fracture network.

[0021] A second objective function characterizing flowback efficiency is constructed with the goal of minimizing the residual amount of liquid carbon dioxide after fracturing the target fracturing formation.

[0022] A third objective function characterizing carbon emissions is constructed based on the amount of carbon dioxide injected, the energy consumption for maintaining the supercritical state during fracturing, and the energy consumption of the flowback fluid treatment system.

[0023] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters, the step of solving the multi-objective function using a multi-objective evolutionary algorithm with the dynamic constraints and normalized engineering constraints as boundaries includes:

[0024] An initial population is generated, and each individual in the initial population represents a set of parameters including injection pressure, injection displacement, and carbon dioxide phase ratio.

[0025] Iterative calculations are performed starting from the initial population, and each iteration includes the following operations:

[0026] Simulated binary crossover and polynomial mutation operations are performed on the individuals in the population at the beginning of this iteration to generate offspring individuals;

[0027] The population at the start of this iteration is merged with the offspring individuals, and the merged individuals are sorted non-dominated based on the multi-objective function value;

[0028] Based on the uniformly distributed reference points and the reference point association strategy, individuals are selected from the sorted individuals to form the population for the next iteration.

[0029] When the preset termination condition is met, the iteration stops, and the resulting population constitutes the Pareto optimal solution set.

[0030] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters of the present invention, the real-time control of the injection pressure, injection volume and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid based on this combination is a closed-loop control, including: dynamically adjusting the injection pressure, injection volume and carbon dioxide phase ratio according to the real-time fracture propagation data obtained by downhole sensors and microseismic monitoring.

[0031] As an optional embodiment of the present invention, in a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters of the present invention, the normalization process adopts the Min-Max standardization method to map the geological parameters and engineering constraint parameters to a preset numerical range.

[0032] As an optional embodiment of the present invention, a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters of the present invention includes:

[0033] Based on historical data generated from completed fracturing operations, update the weight coefficients of each optimization objective in the multi-objective function, and / or adjust the range of model parameters of the multi-objective function for different geological structure types.

[0034] This invention also provides a multi-objective optimization device for carbon dioxide foam fracturing fluid injection parameters, comprising:

[0035] The parameter acquisition module is used to acquire the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation. The geological parameters include the movable water saturation and permeability of the target fractured formation, and the engineering constraint parameters include the equipment pressure resistance threshold, the gas injection displacement range, and the upper limit of the carbon dioxide liquid phase ratio.

[0036] The data processing module is used to normalize the geological parameters and engineering constraint parameters to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters, and to convert the normalized geological parameters into dynamic constraint conditions related to the pressure resistance of the equipment, so as to dynamically adjust the upper limit of the gas injection pressure.

[0037] A multi-objective function construction module is used to construct multi-objective functions with crack complexity, backflow rate, and carbon emissions as optimization objectives.

[0038] The multi-objective optimization algorithm processing module uses the dynamic constraints and the normalized engineering constraint parameters as boundaries to solve the multi-objective function using a multi-objective evolutionary algorithm. Specifically, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy for individual selection and population evolution to generate a Pareto optimal solution set. The Pareto optimal solution set is then filtered according to the normalized engineering constraint parameters to obtain a feasible optimal solution set, and the target gas injection parameter combination is determined from the feasible optimal solution set.

[0039] The real-time injection parameter control module is used to control the injection pressure, injection rate, and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid in real time according to the target injection parameter combination.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] 1. Multi-objective synergistic optimization has been achieved, resulting in a significant improvement in overall benefits.

[0042] Traditional fracturing parameter optimization methods often focus on a single objective, making it difficult to simultaneously consider fracture propagation, flowback efficiency, and environmental impact. This invention constructs a multi-objective optimization model for fracture complexity, flowback rate, and carbon emissions, and employs an advanced evolutionary algorithm to solve it. For the first time in engineering practice, it achieves simultaneous and coordinated optimization of these three key objectives. Practical application shows that this method can comprehensively improve fracturing performance without sacrificing the performance of any single objective: on average, it can increase fracture network complexity (in terms of fractal dimension) by 25%-40%, improve flowback rate by 15%-30%, and simultaneously reduce carbon emissions per unit fracturing operation by more than 18%. This effectively solves the long-standing problem of conflicting objectives and achieves a balance between increased production, improved efficiency, and reduced emissions.

[0043] 2. The optimized algorithm boasts excellent performance and strong decision support capabilities.

[0044] For high-dimensional multi-objective optimization problems, this invention employs an improved multi-objective evolutionary algorithm. By introducing uniformly distributed reference points and an association selection strategy, it significantly improves search efficiency and solution set quality. Compared to traditional algorithms such as NSGA-II, the algorithm described in this invention achieves approximately 33% faster convergence speed (convergence within 120 generations), a 30.8% improvement in solution set coverage (HV metric), and a 52% improvement in solution set uniformity (SP metric). This means the algorithm can find Pareto optimal solutions that are closer to the true global optimum and are widely distributed more quickly, providing rich, high-quality, and diverse alternative parameter schemes for engineering decision-making, greatly enhancing the practical value of the optimization results.

[0045] 3. It has the ability to dynamically adapt and control in real time, and is highly practical for engineering applications.

[0046] This invention is not an offline static optimization, but rather constructs a closed-loop system of "perception-optimization-execution". By converting key formation parameters (such as movable water saturation) into dynamic engineering constraints in real time and integrating downhole sensor and microseismic monitoring data, the system can adapt to changes in geological conditions and fluctuations in the construction process, dynamically adjusting core parameters such as injection pressure, displacement, and phase ratio. This closed-loop control mechanism improves the parameter adjustment response speed by more than 60% compared to traditional offline design, and controls the single optimization iteration time to within 3 minutes, thereby significantly improving the accuracy, safety, and operational efficiency of fracturing operations, truly realizing an intelligent upgrade from "experience-driven" to "data and model-driven".

[0047] 4. It has outstanding environmental protection and low carbon characteristics, which meet the requirements of green development.

[0048] This invention quantifies and manages "carbon emissions" as one of the core optimization objectives, optimizing the design from the source. By optimizing the carbon dioxide phase ratio (controlling the liquid phase proportion ≤33%) and combining it with efficient flowback fluid treatment processes such as CDOF+CDFU, it significantly reduces supercritical phase change energy consumption and processing energy consumption while stably controlling the residual liquid carbon dioxide in the flowback fluid at an extremely low level of ≤50ppm. This not only significantly reduces the carbon and water footprint of fracturing operations but also provides a reliable and complete technical solution for the green and efficient development of unconventional oil and gas resources in areas with strict environmental policies and water scarcity.

[0049] In summary, the multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters of this invention derives its technical effectiveness from its systematic technical solution of "multi-objective modeling + intelligent algorithm solution + dynamic constraints + closed-loop execution". These effects are not only reflected in the improvement of abstract algorithm performance indicators, but also directly transformed into quantifiable and verifiable engineering benefits (increased production, improved efficiency, reduced consumption, and reduced emissions), forming a complete technical advantage chain from method to device, from design to execution, possessing outstanding practical value and broad prospects for promotion and application. Attached Figure Description

[0050] Figure 1 This is a flowchart of a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0052] Therefore, the following detailed description of embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0053] It should be noted that, unless otherwise specified, the embodiments and features and technical solutions in the present invention can be combined with each other.

[0054] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0055] In the description of this invention, it should be noted that the terms "upper," "lower," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use, or the orientation or positional relationship commonly understood by those skilled in the art. These terms are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. In addition, the terms "first," "second," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0056] like Figure 1 As shown, this embodiment of the invention provides a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters, including:

[0057] Obtain the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation. Normalize the geological parameters and engineering constraint parameters to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters.

[0058] Normalized geological parameters are transformed into dynamic constraints related to the equipment's pressure resistance, so as to dynamically adjust the upper limit of the gas injection pressure.

[0059] Construct a multi-objective function with crack complexity, backflow rate, and carbon emissions as optimization objectives;

[0060] Using the dynamic constraints and normalized engineering constraint parameters as boundaries, a multi-objective evolutionary algorithm is used to solve the multi-objective function; wherein, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy to perform individual selection and population evolution to generate a Pareto optimal solution set;

[0061] The Pareto optimal solution set is filtered based on the normalized engineering constraint parameters to obtain a feasible optimal solution set;

[0062] The target gas injection parameter combination is determined from the set of feasible optimal solutions, and the gas injection pressure, gas injection rate and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid are adjusted in real time based on the combination.

[0063] This invention provides a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters. Its core lies in integrating geological conditions, engineering constraints, and low-carbon objectives into an optimization framework through an intelligent algorithm and control system, achieving dynamic and synergistic optimization of injection parameters (pressure, displacement, and phase ratio). The method begins with the acquisition and processing of formation and equipment data, performs global optimization through an improved multi-objective evolutionary algorithm, and ultimately controls fracturing operations in real-time using a closed-loop approach. Its working principle can be summarized as follows: using formation adaptability (dynamic constraints) and equipment safety (static constraints) as boundaries, within a three-dimensional objective space comprised of fracture complexity, flowback rate, and carbon emissions, a high-dimensional evolutionary algorithm based on a reference point mechanism is used to find the optimal equilibrium point (Pareto front). Through engineering constraint screening and real-time feedback, the injection system is driven to execute optimal or suboptimal parameter combinations, thereby achieving a synergistic effect of increased production and efficiency with low-carbon emission reduction.

[0064] The present invention provides a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters. Compared with traditional single-objective optimization, it can improve fracture complexity and flowback rate, while reducing carbon emissions per unit fracturing operation. It achieves synergistic optimization of oil and gas production enhancement and low-carbon processes, and provides intelligent technical support for the green development of unconventional oil and gas resources.

[0065] Specifically, the following describes in detail each step of a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to an embodiment of the present invention.

[0066] S101: Data acquisition and parameter preprocessing.

[0067] This step aims to prepare standardized, dimensionless input data for the optimization model. Geological parameters of the target fracturing formation are acquired through a downhole sensor network (such as temperature, pressure, and water saturation sensors), including but not limited to: mobile water saturation (Sw), permeability (K), pore pressure (Pp), and formation temperature (Tf). Simultaneously, engineering constraint parameters are obtained from the engineering design, primarily including: the pressure withstand threshold of the fracturing equipment (such as the upper limit of pump pressure P_pump_max), the feasible range of gas injection displacement [Q_min, Q_max], and the upper limit of the liquid carbon dioxide ratio (η_CO2_liquid_max) based on fluid performance and safety considerations.

[0068] To eliminate the impact of differences in the dimensions and numerical ranges of different parameters on the stability of the algorithm, the above parameters need to be normalized. In a preferred embodiment of the present invention, the Min-Max normalization method is used to linearly map each parameter to the interval [0, 1]. For geological parameters, such as mobile water saturation Sw, its normalized value Sw_norm = (Sw - Sw_min) / (Sw_max - Sw_min), where Sw_min and Sw_max are determined based on regional geological data. For engineering constraints, such as equipment pressure resistance thresholds, they can be mapped to constraint boundary weights. After normalization, normalized geological parameters and normalized engineering constraint parameters are obtained for use in subsequent models.

[0069] S102: Dynamic constraint modeling and multi-objective function construction.

[0070] This step transforms geological conditions into actionable engineering constraints and establishes optimization objectives.

[0071] Dynamic constraint modeling: Geological parameters, especially movable water saturation (Sw), are transformed into dynamic upper limit constraints on gas injection pressure. High water saturation formations are prone to water-locking damage or uncontrolled fracture height under high pressure, thus requiring dynamic suppression of construction pressure. The maximum allowable gas injection pressure in real time is dynamically calculated using a preset functional relationship: Pmax = f(Sw). In a specific embodiment, this function can be expressed as: Pmax = α × P_pump_max × (1 - β × Sw_norm), where α is a safety factor (e.g., 0.9), and β is a reduction factor (e.g., 0.3-0.5) for the formation's sensitivity to water-locking damage or uncontrolled fracture height. This dynamic constraint, together with static normalized engineering constraint parameters (e.g., displacement range, phase ratio upper limit), constitutes the boundary of the algorithm's search.

[0072] The coefficients α and β are key parameters predetermined based on engineering safety specifications and formation property analysis. α (safety factor, e.g., 0.9) is used to ensure a safety margin for construction pressure; β (reduction factor, e.g., 0.3-0.5) is pre-calibrated based on the formation's sensitivity to water-locking damage through prior experiments or historical data. This specific formula clarifies how to dynamically and reasonably calculate the upper pressure limit using movable water saturation, overcoming the shortcomings of traditional methods where static constraints do not match the real-time formation conditions.

[0073] Multi-objective function construction: Construct an objective function that includes three key performance indicators.

[0074] Fracture complexity objective function (first objective function f1, maximizing): Fracture complexity is key to evaluating reservoir stimulation effectiveness. Based on point cloud data of fracture events acquired through microseismic monitoring, fractal dimension (D_f) or the ratio of fracture network surface area to stimulated volume (SRV) is used for quantification. A higher fractal dimension or a larger specific surface area indicates a more complex fracture network. Therefore, f1 can be set as D_f or f1 = surface area / SRV.

[0075] The flowback rate objective function (second objective function f2, maximizing / minimizing residual volume): A high flowback rate means less liquid phase retention and higher gas recovery. The objective is to minimize the residual liquid carbon dioxide (C_residual) in the flowback fluid after fracturing. This objective function can be modeled as: f2 = 1 / (C_residual + ε), or directly as minimizing C_residual. The residual volume can be estimated through flowback fluid composition analysis or a mass transfer model based on the injected phase and formation conditions.

[0076] The objective function for carbon emissions (the third objective function f3, minimization): Total carbon emissions (E_total) mainly come from:

[0077] a) Energy consumption for carbon dioxide injection (E_inj);

[0078] b) Phase change energy consumption (E_phase) required to heat and pressurize carbon dioxide to a supercritical state;

[0079] c) Energy consumption of the flowback fluid recovery and treatment system (E_treat). Therefore, f3 = E_total = E_inj + E_phase + E_treat. E_phase can be obtained by monitoring and integrating the heating / insulation power with time to maintain downhole supercritical conditions (T > 31.1℃, P > 7.38 MPa); E_treat is calculated based on the energy consumption model of the flowback fluid treatment process (such as the CDOF / CDFU system).

[0080] S103: Solving and filtering based on an improved multi-objective evolutionary algorithm.

[0081] The embodiments of the present invention employ a multi-objective evolutionary algorithm with a built-in reference point association strategy (such as an improved algorithm based on the NSGA-III framework) to solve the above three objective functions.

[0082] Algorithm initialization and iteration: First, within the decision space consisting of injection pressure (P), injection displacement (Q), and carbon dioxide phase ratio (η), an initial population is randomly generated, with each individual representing a possible combination of parameters (P, Q, η). Algorithm parameters are set, such as population size (N=200-300), crossover probability (e.g., 0.8), and mutation probability (e.g., 0.2).

[0083] Iterative evolution process: The algorithm enters an iterative loop. In each iteration:

[0084] a) Genetic operations: For individuals in the current population (parent generation), simulated binary crossover (SBX) and polynomial mutation operations are used to generate offspring individuals in order to explore new parameter spaces.

[0085] b) Evaluation and ranking: Calculate the three objective function values ​​(f1, f2, f3) for all parent and offspring individuals, and perform non-dominated ranking of the merged population according to Pareto dominance, dividing individuals into multiple frontier levels.

[0086] c) Reference Point Selection: To maintain the diversity and uniformity of the solution set in the high-dimensional objective space, the algorithm pre-generates a set of uniformly distributed reference points in the three-dimensional objective space (f1, f2, f3). Subsequently, a reference point association strategy is employed to associate the sorted individuals with the nearest reference point. Based on association relationships and the niche preservation principle, individuals are selected from the high-level fronts to form the next generation population. This strategy effectively avoids the problems of uneven solution set distribution and easy convergence to local fronts in high-dimensional multi-objective optimization of traditional algorithms.

[0087] d) Repeat a)-c) until the preset termination condition is met (such as reaching the maximum number of iterations or the solution set converges).

[0088] Constraint Screening: After the algorithm iteration is complete, a Pareto optimal solution set is output, where each solution is a trade-off between the three objectives that cannot be improved simultaneously. Subsequently, based on the normalized engineering constraint parameters (mainly the equipment pressure threshold P_pump_max) and formation safety requirements (such as distance from the aquifer), this solution set is screened to remove all solutions that violate hard constraints, thus obtaining the feasible optimal solution set.

[0089] S104: Real-time parameter control and iterative effect evaluation.

[0090] Closed-loop real-time control: From the set of feasible optimal solutions, a set of target gas injection parameter combinations (P, Q, η*) is determined based on on-site engineering preferences (e.g., prioritizing flowback rate in the current operation). This combination is sent to the fracturing equipment through the closed-loop control system. The real-time gas injection parameter control module dynamically adjusts the gas injection pressure, flow rate, and carbon dioxide phase adjustment unit of the fracturing pump based on this combination and in conjunction with real-time feedback data from downhole sensors (pressure, temperature) and microseismic monitoring, achieving precise parameter control. If real-time monitoring shows that fracture propagation deviates from expectations (e.g., directional deviation >15° or excessively rapid propagation), the system can trigger parameter fine-tuning.

[0091] Performance Evaluation and Model Iteration: After a single fracturing operation, the actual fracture complexity, flowback rate, and carbon emissions are evaluated based on microseismic interpretation, flowback fluid detection, and energy consumption statistics. This historical operational data is stored and used for iterative model updates. Specifically, this includes updating the weight coefficients of each objective in the multi-objective function (reflecting priority changes under different blocks or policies), and adjusting the range of model parameters for different geological structures (such as shale and tight sandstone) to enable the model to have continuous adaptive learning capabilities.

[0092] This invention discloses a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters. Through a collaborative mechanism of "dynamic constraint modeling + three-objective Pareto optimization + reference point-guided evolutionary search + closed-loop real-time control," it achieves a significant improvement in technical performance. The improved multi-objective evolutionary algorithm (based on the NSGA-III framework) achieves a breakthrough in performance compared to traditional algorithms, as shown in the table below:

[0093] Evaluation indicators This invention (improved algorithm) Traditional algorithms (such as NSGA-II) Increase Convergence speed (number of iterations) Convergence within 120 generations Convergence within 180 generations Speed ​​increased by 33% Solution set coverage (HV value) 0.85 0.65 Increased by 30.8% Uniformity of the solution set (SP value, the smaller the better). 0.12 0.25 Optimized by 52% Single iteration computation time Approximately 480 seconds Approximately 520 seconds Reduced by 7.7% Engineering practical value (proportion of solutions that satisfy constraints) More than 90% 40%-60% Significant improvement

[0094] Based on the aforementioned algorithmic advantages and systematic design, the multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters in this embodiment of the invention achieves a significant improvement in technical performance:

[0095] The multi-objective synergistic optimization yields significant results: For the first time, three objectives—fracture complexity, flowback rate, and carbon emissions—are simultaneously quantified and optimized in fracturing parameter optimization. Field applications and simulation comparisons show that the method of this invention can achieve an average increase of 25-40% in fracture fractal dimension, a 15-30% increase in flowback rate, and a reduction of over 18% in carbon emissions per unit of fracturing operation, effectively solving the industry problem of conflicting objectives in traditional methods.

[0096] Superior Algorithm Performance: The improved multi-objective evolutionary algorithm employs a hypercube partitioning technique to generate uniform reference points and combines it with an association strategy, effectively ensuring the diversity and uniformity of high-dimensional Pareto front solutions. Compared to traditional algorithms such as NSGA-II, it achieves a 30.8% improvement in solution set coverage (HV metric), a 33% improvement in convergence speed, and a 52% optimization in solution set uniformity.

[0097] The system boasts outstanding engineering practicality and real-time performance: Geological parameters (such as Sw) are dynamically converted into engineering constraints (Pmax), ensuring the geological adaptability of the scheme. Combined with a closed-loop control system for real-time monitoring, the response speed of parameter adjustments is improved by more than 60% compared to traditional offline optimization, with a single algorithm iteration time of ≤3 minutes, significantly improving the accuracy and safety of construction control.

[0098] Excellent low-carbon and environmental benefits: By optimizing the carbon dioxide phase ratio (controlling the liquid phase ratio to ≤33%) and supporting efficient flowback fluid treatment processes (such as CDOF / CDFU), while ensuring the fracturing effect, it achieves an ultra-low emission standard of ≤50ppm liquid phase residue, and significantly reduces phase change and processing energy consumption, providing a reliable technical path for the green development of unconventional oil and gas resources.

[0099] This invention also provides a multi-objective optimization device for carbon dioxide foam fracturing fluid injection parameters, comprising:

[0100] The parameter acquisition module is used to acquire the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation;

[0101] The data processing module is used to normalize the geological parameters and engineering constraint parameters to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters, and to convert the normalized geological parameters into dynamic constraint conditions related to the pressure resistance of the equipment, so as to dynamically adjust the upper limit of the gas injection pressure.

[0102] A multi-objective function construction module is used to construct multi-objective functions with crack complexity, backflow rate, and carbon emissions as optimization objectives.

[0103] The multi-objective optimization algorithm processing module uses the dynamic constraints and the normalized engineering constraint parameters as boundaries to solve the multi-objective function using a multi-objective evolutionary algorithm. Specifically, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy for individual selection and population evolution to generate a Pareto optimal solution set. The Pareto optimal solution set is then filtered according to the normalized engineering constraint parameters to obtain a feasible optimal solution set, and the target gas injection parameter combination is determined from the feasible optimal solution set.

[0104] The real-time injection parameter control module is used to control the injection pressure, injection rate, and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid in real time according to the target injection parameter combination. Example 1

[0105] This embodiment of a multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters includes the following steps:

[0106] Step 1: Data acquisition and parameter preprocessing. Geological parameters are obtained by real-time acquisition of key parameters such as mobile water saturation (28%), permeability (0.05mD), and formation temperature (120℃) through downhole sensors. A three-dimensional reservoir model is constructed using historical microseismic data. Engineering constraints are set, such as equipment pressure resistance threshold (75MPa), upper limit of carbon dioxide liquid phase ratio (30%), and construction discharge range (3-12m³ / min).

[0107] Step 2: Use Min-Max standardization to map formation parameters and engineering constraints to the [0,1] interval, and perform data normalization to eliminate dimensional differences.

[0108] Step 3: Dynamic Constraint Modeling. First, normalize the movable water saturation: According to the geological data of the area involved in this embodiment, the minimum value of movable water saturation Sw_min is 20%, and the maximum value Sw_max is 40%. Therefore, the normalized value Sw_norm corresponding to Sw=28% is (Sw - Sw_min) / (Sw_max - Sw_min) = (28 - 20) / (40 - 20) = 0.4.

[0109] Then, Sw_norm is substituted into the dynamic constraint function Pmax = α × P_pump_max × (1 - β × Sw_norm) to calculate the upper limit of the gas injection pressure. In this embodiment, a safety factor α = 0.9 is taken (based on the equipment safety specifications, a 10% margin is reserved), and a reduction factor β = 0.25 is taken according to the formation's sensitivity to waterlock damage. The equipment pressure resistance threshold P_pump_max = 75 MPa.

[0110] The calculation yields: Pmax = 0.9 × 75 × (1 - 0.25 × 0.4) = 0.9 × 75 × 0.9 = 60.75 MPa.

[0111] In practical engineering applications, to facilitate operation and maintain a conservative approach, the upper limit of the gas injection pressure is rounded to 60 MPa. Simultaneously, the proportion of liquid carbon dioxide is limited to ≤30%. Step 4: Construct a multi-objective function, calculate the target value based on the spatial distribution density of microseismic events, and calculate the crack complexity.

[0112] Step 5: Calculate the return efficiency by combining the residual amount of liquid carbon dioxide (target ≤50ppm) with the total injected liquid volume, and calculate the return rate.

[0113] Step 6: Quantify the energy consumption of supercritical carbon dioxide phase change (temperature-pressure curve integration) and the energy consumption of backflow liquid treatment (power consumption of CDOF+CDFU system).

[0114] Step 7: Initialize the population size to 200, generate offspring using simulated binary crossover (SBX) and polynomial mutation strategies, and screen the Pareto optimal solution set through non-dominated sorting and reference point association.

[0115] Step 8: Select the optimal solution based on the model (injection pressure 58MPa, displacement 8 m³ / min, carbon dioxide liquid phase ratio 25%), and simultaneously inject liquid carbon dioxide (25%) and slickwater-based fluid through a fracturing pump truck.

[0116] Step 9: Trigger closed-loop control based on real-time microseismic data (when the crack propagation rate is ≥5m / s). If the crack extension direction deviates from the preset range (±15°), automatically reduce the discharge rate to 6 m³ / min to correct the path.

[0117] Step 10: Assess the complexity of the cracks and conduct a backflow efficiency test. Example 2

[0118] Step 1: Obtain formation parameters. Collect movable water saturation (35%), permeability (0.08mD), well temperature (120℃), and geostress distribution (σ_H / σ_h=1.8) using downhole fiber optic sensors. Limit the engineering equipment, with an upper limit of 70MPa for gas injection pressure (dynamically adjusted to 65MPa based on movable water saturation) and a discharge range of 3-12 m³ / min.

[0119] Step 2: Use Min-Max standardization to normalize the formation parameters and engineering constraints to eliminate dimensional differences.

[0120] Step 3: Construct a multi-objective function and calculate the crack complexity of the model; initialize the population size to 300, the crossover rate to 0.8 (SBX crossover operator), and the mutation rate to 0.2 (polynomial mutation).

[0121] Step 4: Quantify the energy consumption of carbon dioxide phase change (supercritical phase change temperature 31.1℃) and the energy consumption of backflow treatment (CDOF system power consumption ≤ 50kW·h / m), and calculate the carbon emissions.

[0122] Step 5: Set the liquid phase residue to ≤50ppm and the target return rate to ≥75%, and calculate the return rate.

[0123] Step 6: Generate Pareto solution set and remove solutions that do not meet the equipment pressure resistance (65MPa) and aquifer safety distance (≥50m).

[0124] Step 7: Select the optimal parameters for the Pareto solution set: injection pressure 65 MPa, displacement 10 m³ / min, and carbon dioxide foam mass 80%.

[0125] Step 8: Configure fracturing parameters, prepare carbon dioxide foam fracturing fluid with a carbon dioxide liquid phase ratio of 80% (phase adjusted to supercritical state, temperature 35℃), slickwater-based fluid containing 0.3% drag reducer, pump injection operation, initial discharge rate 10 m³ / min, sand concentration gradually increased (20%→40%), single-stage sand addition 105m.

[0126] Step 9: Adjust the microseismic monitoring. When the crack extension speed is >8m / s or the direction deviates from the preset trajectory (±15°), automatically reduce the discharge rate to 9 m³ / min. The FPGA unit adjusts the foam quality to 85% according to the real-time pressure fluctuation (±2MPa) to suppress excessive crack extension.

[0127] Step 10: Collect flowback fluid. Flowback should be initiated within 24 hours after fracturing. Initial flowback fluid should be desandered to remove proppant residue (particle size > 100 mesh).

[0128] Step 11: Conduct fracture complexity assessment. Microseismic monitoring shows that the fracture fractal dimension reaches 2.8 (33% higher than traditional hydraulic fracturing), and the fracture network coverage area is expanded by 45% (through proppant migration trajectory inversion).

[0129] Step 12: Calculate carbon emissions. The carbon emissions per unit of fracturing were reduced to 1.0 ton of carbon dioxide (a reduction of 23%), mainly due to the optimization of supercritical phase change energy consumption (a reduction of 18%) and the recycling of flowback fluid (a water saving of 66%).

[0130] Step 13: Verify the efficiency of the algorithm. The improved NSGA-III takes 2.5 minutes for a single iteration (the traditional NSGA-II takes 5 minutes), and the parameter adjustment lag throughout the fracturing process is ≤5 minutes (the traditional method takes >10 minutes).

[0131] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described herein. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or equivalent substitutions to the present invention, as well as all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.

Claims

1. A multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters, characterized in that, include: Obtain the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation. The geological parameters include the movable water saturation and permeability of the target fractured formation, and the engineering constraint parameters include the equipment pressure resistance threshold, the gas injection displacement range, and the upper limit of the carbon dioxide liquid phase ratio. The geological parameters and engineering constraint parameters are normalized to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters. Normalized geological parameters are transformed into dynamic constraints related to the equipment's pressure resistance, so as to dynamically adjust the upper limit of the gas injection pressure. Construct a multi-objective function with crack complexity, backflow rate, and carbon emissions as optimization objectives; Using the dynamic constraints and normalized engineering constraint parameters as boundaries, a multi-objective evolutionary algorithm is used to solve the multi-objective function; wherein, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy to perform individual selection and population evolution to generate a Pareto optimal solution set; The Pareto optimal solution set is filtered based on the normalized engineering constraint parameters to obtain a feasible optimal solution set; The target gas injection parameter combination is determined from the set of feasible optimal solutions, and the gas injection pressure, gas injection rate and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid are adjusted in real time based on the combination. The process of converting normalized geological parameters into dynamic constraints related to the equipment's pressure resistance, in order to dynamically adjust the upper limit of the gas injection pressure, includes: Based on the movable water saturation in the normalized geological parameters, the upper limit of the gas injection pressure is dynamically calculated through a preset functional relationship. The preset functional relationship satisfies: Pmax=α×P_pump_max×(1-β×Sw_norm), where Pmax is the upper limit of the dynamically calculated gas injection pressure, α is the safety factor, β is the reduction factor of the formation's sensitivity to water lock damage or uncontrolled fracture height, P_pump_max is the upper limit threshold of the pressure resistance of the fracturing equipment, and Sw_norm is the normalized mobile water saturation with a value range of [0,1]. The construction of the multi-objective function with crack complexity, backflow rate, and carbon emissions as optimization objectives includes: Based on microseismic monitoring data, a first objective function characterizing the complexity of fractures is constructed by using the fractal dimension of fractures or the ratio of the surface area to the volume of the fracture network. A second objective function characterizing flowback efficiency is constructed with the goal of minimizing the residual amount of liquid carbon dioxide after fracturing the target fracturing formation. A third objective function characterizing carbon emissions is constructed based on the amount of carbon dioxide injected, the energy consumption for maintaining the supercritical state during fracturing, and the energy consumption of the flowback fluid treatment system.

2. The multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to claim 1, characterized in that, Dynamic data on fracture propagation obtained through microseismic monitoring of the target fractured formation, used to quantify the complexity of the fractures; In addition, data on the rheological properties and phase change patterns of the carbon dioxide foam fracturing fluid used to construct the multi-objective function or the dynamic constraint conditions, determined through laboratory testing.

3. The multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to claim 1, characterized in that, The step of solving the multi-objective function using the dynamic constraints and normalized engineering constraint parameters as boundaries and employing a multi-objective evolutionary algorithm includes: An initial population is generated, wherein each individual in the initial population represents a combination of parameters, including injection pressure, injection displacement, and carbon dioxide phase ratio. Iterative calculations are performed starting from the initial population, and each iteration includes the following operations: Simulated binary crossover and polynomial mutation operations are performed on the individuals in the population at the beginning of this iteration to generate offspring individuals; The population at the start of this iteration is merged with the offspring individuals, and the merged individuals are sorted non-dominated based on the multi-objective function value. Based on the uniformly distributed reference points and the reference point association strategy, individuals are selected from the sorted individuals to form the population for the next iteration. When the preset termination condition is met, the iteration stops, and the resulting population constitutes the Pareto optimal solution set.

4. The multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to claim 1, characterized in that, Based on this combination, the injection pressure, injection rate, and carbon dioxide phase ratio of the carbon dioxide foam fracturing fluid are controlled in real time as a closed-loop control, including: dynamically adjusting the injection pressure, injection rate, and carbon dioxide phase ratio according to the real-time fracture propagation data obtained from downhole sensors and microseismic monitoring.

5. The multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to claim 1, characterized in that, The normalization process employs the Min-Max standardization method to map the geological parameters and engineering constraint parameters to a preset numerical range.

6. The multi-objective optimization method for carbon dioxide foam fracturing fluid injection parameters according to claim 1, characterized in that, include: Based on historical data generated from completed fracturing operations, update the weight coefficients of each optimization objective in the multi-objective function, and / or adjust the range of model parameters of the multi-objective function for different geological structure types.

7. A multi-objective optimization device for carbon dioxide foam fracturing fluid injection parameters, characterized in that, include: The parameter acquisition module is used to acquire the geological parameters of the target fractured formation and the engineering constraint parameters set for the target fractured formation. The geological parameters include the movable water saturation, permeability and pore pressure of the target fractured formation. The engineering constraint parameters include the equipment pressure resistance threshold, the gas injection displacement range and the upper limit of the carbon dioxide liquid phase ratio. The data processing module is used to normalize the geological parameters and engineering constraint parameters to obtain the corresponding normalized geological parameters and normalized engineering constraint parameters, and to convert the normalized geological parameters into dynamic constraint conditions related to the pressure resistance of the equipment, so as to dynamically adjust the upper limit of the gas injection pressure. A multi-objective function construction module is used to construct multi-objective functions with crack complexity, backflow rate, and carbon emissions as optimization objectives. The multi-objective optimization algorithm processing module uses the dynamic constraints and the normalized engineering constraint parameters as boundaries to solve the multi-objective function using a multi-objective evolutionary algorithm. Specifically, the multi-objective evolutionary algorithm generates uniformly distributed reference points in the objective space composed of the optimization objectives, and uses a reference point association strategy for individual selection and population evolution to generate a Pareto optimal solution set. The Pareto optimal solution set is then filtered according to the normalized engineering constraint parameters to obtain a feasible optimal solution set, and the target gas injection parameter combination is determined from the feasible optimal solution set. The real-time injection parameter control module is used to control the injection pressure, injection volume and carbon dioxide phase ratio of carbon dioxide foam fracturing fluid in real time according to the target injection parameter combination. The process of converting normalized geological parameters into dynamic constraints related to the equipment's pressure resistance, in order to dynamically adjust the upper limit of the gas injection pressure, includes: Based on the movable water saturation in the normalized geological parameters, the upper limit of the gas injection pressure is dynamically calculated through a preset functional relationship. The preset functional relationship satisfies: Pmax=α×P_pump_max×(1-β×Sw_norm), where Pmax is the upper limit of the dynamically calculated gas injection pressure, α is the safety factor, β is the reduction factor of the formation's sensitivity to water lock damage or uncontrolled fracture height, P_pump_max is the upper limit threshold of the pressure resistance of the fracturing equipment, and Sw_norm is the normalized mobile water saturation with a value range of [0,1]. The construction of the multi-objective function with crack complexity, backflow rate, and carbon emissions as optimization objectives includes: Based on microseismic monitoring data, a first objective function characterizing the complexity of fractures is constructed by using the fractal dimension of fractures or the ratio of the surface area to the volume of the fracture network. A second objective function characterizing flowback efficiency is constructed with the goal of minimizing the residual amount of liquid carbon dioxide after fracturing the target fracturing formation. A third objective function characterizing carbon emissions is constructed based on the amount of carbon dioxide injected, the energy consumption for maintaining the supercritical state during fracturing, and the energy consumption of the flowback fluid treatment system.

Images