An intelligent flow field regulation method for coupling driving of ground liquid treatment, efficiency improvement and capacity expansion

Abstract

This invention relates to an intelligent flow field control method driven by coupled improvement and capacity expansion of surface fluid treatment, comprising: constructing an objective function based on selected preset optimization objectives, wherein the preset optimization objectives include: maximizing cumulative oil production in a stage, maximizing net present value of development, and optimizing the balance of injection and production flow fields; constructing a set of collaborative decision variables, including: wellhead injection and production parameters, surface fluid treatment parameters, and profile control operation parameters; constructing a set of constraints based on engineering and physical constraints, including: single-well constraints, water quality constraints, treatment capacity constraints, and platform resource constraints; establishing a constrained nonlinear mixed-integer optimization mathematical model based on the aforementioned constructed objective function, set of collaborative decision variables, and set of constraints; solving the constrained nonlinear mixed-integer optimization mathematical model using a heuristic intelligent evolutionary algorithm; and generating and executing structured control commands based on the solution results of the constrained nonlinear mixed-integer optimization mathematical model.

Description

Technical Field

[0001] This invention relates to the field of oil and gas development engineering technology, and in particular to an intelligent flow field control method for improving efficiency and expanding capacity of surface fluid treatment through coupled driving. Background Technology

[0002] Offshore oilfield development has long been constrained by strict limitations on platform space, load-bearing capacity, and energy supply, with produced fluid treatment being a key bottleneck restricting capacity improvement.

[0003] Traditional production fluid treatment equipment is bulky, time-consuming, and inefficient in removing oil, making it difficult to meet the demands of large-volume production fluid treatment during periods of high water cut, and severely consuming valuable platform resources. In recent years, technological breakthroughs in next-generation produced fluid treatment devices have led to a leap forward in production fluid treatment capacity, with equipment weight reduced by 95% and floor space reduced by over 90%, achieving a daily processing capacity exceeding 10,000 cubic meters. This technology enables the treated produced fluid to have the water quality conditions for direct reinjection into the formation, providing unprecedented flexibility in the injection-production system.

[0004] However, current injection and production optimization methods are still based on traditional processing capacity assumptions. They fail to take into account the systematic modeling of the closed-loop chain of "efficiency improvement and capacity expansion of surface fluid treatment - water quality compliance and reinjection - matching of injection and production systems - intervention of profile control measures". They also fail to incorporate the lightweight deployment characteristics of processing equipment and platform resource constraints into the optimization framework. As a result, they are unable to support rapid decision-making and automatic execution in offshore oilfields and cannot meet the actual needs of efficient development of offshore oilfields. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to propose an intelligent flow field control method driven by the coupling of efficiency improvement and capacity expansion of surface fluid treatment. By constructing a unified optimization framework that integrates efficiency improvement and capacity expansion of surface fluid treatment, profile control measures, and physical constraints, the method achieves intelligent generation of injection and production schemes with the objectives of maximizing cumulative oil production, maximizing economic net present value, or optimizing the balance of injection and production flow fields. This fully leverages the control potential of new surface fluid treatment technologies and enhances the development efficiency of offshore oilfields.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, this application provides a smart flow field control method coupled with ground-based liquid treatment efficiency improvement and capacity expansion, comprising: Step (1): Construct an objective function based on the selected preset optimization objective. The preset optimization objective includes: maximizing the cumulative oil production in the stage, maximizing the net present value of development, and optimizing the balance of the injection and production flow field. Step (2) Construct a set of collaborative decision-making variables, including: wellhead injection and production parameters, surface fluid treatment parameters, and profile control operation parameters; Step (3): Based on engineering and physical constraints, construct a set of constraints, including: single well constraints, water quality constraints, treatment capacity constraints, and platform resource constraints; Step (4): Based on the objective function, collaborative decision-making variable set, and constraint set constructed above, establish a constrained nonlinear mixed integer optimization mathematical model; Step (5): The constrained nonlinear mixed integer optimization mathematical model is solved using a heuristic intelligent evolutionary algorithm. Step (6): Based on the solution results of the constrained nonlinear mixed integer optimization mathematical model, generate and execute structured control instructions.

[0008] In one implementation, in step (1), when the selected preset optimization objective is to maximize the net present value of development, the objective function... The calculation is based on discounted cash flows over the entire lifecycle, and the formula is:

[0009] In the formula, For the development cycle; For the first The annual cash inflow is determined by the projected crude oil production and crude oil price; For the first The annual cash outflow covers the cost of purchasing and installing processing modules, energy consumption costs, profile control agent costs, and maintenance costs; The benchmark rate of return; When the selected preset optimization objective is to achieve optimal injection-production flow field uniformity, the objective function is... Defined as minimizing the standard deviation of the recovery rate in each injection-production direction, the calculation process includes: dividing the reservoir into... Each injection and production unit is used to calculate the recovery rate of each unit. The standard deviation is calculated using the following formula:

[0010] In the formula, This represents the average recovery rate across all injection and production units.

[0011] In one implementation, the set of collaborative decision variables in step (2) includes: (1) Wellhead injection and production parameters: injection volume of injection well and production volume of production well; (2) Surface liquid treatment parameters: including the activation status and dynamic flux allocation value of the surface liquid treatment module; the activation status represents the capacity expansion deployment decision of the standby module; the dynamic flux allocation value is used to adjust the processing speed of the surface liquid treatment module and supports continuous adjustment within the allowable range of the equipment. (3) Profile control operation parameters: including whether profile control is carried out in each well, profile control injection rate, profile control concentration and profile control duration.

[0012] In one implementation, in step (3), the single-well constraints include upper and lower limits of injection and production rates, upper and lower limits of bottom hole pressure, and upper and lower limits of profile control parameters for each well; the water quality constraints include upper limits of oil content and suspended solids concentration in the treated reinjected water; the treatment capacity constraints are that the total treatment throughput does not exceed the upper limit of the operating capacity of the treatment equipment; and the platform resource constraints include upper limits of the total weight and total projected area of ​​the newly added treatment modules.

[0013] In one implementation, step (4) uses the collaborative decision variable set as the independent variable and the constraint condition set as the constraint condition to construct a constrained nonlinear mixed integer optimization mathematical model. The objective function value (predicted crude oil production) and the constraint condition (bottom hole pressure) are obtained based on an improved reservoir seepage equation and solved through numerical simulation. Specifically, this includes: (1) Map the set of collaborative decision variables to the grid properties of the reservoir numerical model; (2) Application of the sealing coefficient of the profile control agent Improved seepage equation (taking values ​​[0, 1]):

[0014] (3) The above equations are solved by numerical discretization method, and the pressure field and saturation field are calculated iteratively; (4) Calculate the oil production, water cut, and bottom pressure of each well to obtain the objective function value.

[0015] In one implementation, when using a heuristic intelligent evolutionary algorithm to solve the problem in step (5), a dual convergence determination mechanism is set up: Mechanism 1: Stability determination of the objective function, when continuous The change in the optimal objective function value of the population is less than a preset accuracy threshold. When the objective function converges, it is determined that the objective function has converged. Mechanism 2: Constraint Satisfaction Determination. Calculate the degree to which the current optimal solution violates all engineering and physical constraints. If the maximum violation is less than a preset feasibility threshold... The solution is deemed feasible. The optimization process terminates and the result is output only when both Mechanism 1 and Mechanism 2 are satisfied simultaneously, or when the maximum number of iterations preset by the algorithm is reached.

[0016] In one implementation, in step (6), a structured control command is output based on the solution result, including injection and production volume settings for each well, start and stop commands for the processing module, flux settings, and profile control operation plan.

[0017] In one implementation, a stepped transition logic is used. When the adjustment range of the injection / production volume or throughput setting exceeds the set threshold, the system automatically generates a stepped transition command for the platform control system to execute.

[0018] The present invention has the following advantages due to the adoption of the above technical solutions: By constructing a unified optimization framework that integrates surface fluid treatment capacity enhancement and expansion, profile control measures, and physical constraints, this invention breaks through the limitations of traditional injection and production optimization and surface system operation being separate. It achieves dynamic matching between the capacity expansion potential of the surface fluid treatment system and the reservoir flow field control requirements, significantly improving the overall system operating efficiency and development benefits.

[0019] By incorporating wellhead injection and production parameters, the activation status and dynamic flux allocation of surface fluid treatment modules, and profile control operation parameters into a unified decision-making system, the entire process from wellbore to surface and from conventional injection and production to profile control operations has been optimized collaboratively, significantly enhancing the precision and systematic nature of the control scheme.

[0020] It offers three optimization objectives: maximizing cumulative oil production, maximizing net economic present value, and optimizing the balance between injection and production flow fields. These objectives can be flexibly selected based on the actual development stage and management goals of the oilfield, taking into account development efficiency, economic benefits, and flow field balance. It has good engineering adaptability and strategic guidance.

[0021] By establishing a multi-dimensional constraint model covering single wells, water quality, treatment capacity, and platform resources, we ensure that the optimization scheme conforms to the actual situation of the on-site project. Furthermore, we design a stepped transition logic and a command deviation verification mechanism to avoid sudden changes in equipment operation and drastic fluctuations in operating conditions, and ensure the smooth and safe execution of control commands on-site.

[0022] By integrating functional units such as data acquisition, constraint modeling, optimization solution and command push, a closed-loop intelligent decision-making link from data to command is formed, supporting real-time or near-real-time optimization and control, and significantly improving the intelligence level and response speed of offshore oilfield production management. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of an intelligent flow field control system provided in an embodiment of the present invention.

[0024] Figure 2 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0025] 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0026] This application provides a smart flow field control method for improving and expanding the efficiency of ground-based liquid treatment through coupled driving, including: Step (1): Construct an objective function based on the selected preset optimization objective. The preset optimization objective includes: maximizing the cumulative oil production in the stage, maximizing the net present value of development, and optimizing the balance of the injection and production flow field. Step (2) Construct a set of collaborative decision-making variables, including: wellhead injection and production parameters, surface fluid treatment parameters, and profile control operation parameters; Step (3): Based on engineering and physical constraints, construct a set of constraints, including: single well constraints, water quality constraints, treatment capacity constraints, and platform resource constraints; Step (4): Based on the objective function, collaborative decision-making variable set, and constraint set constructed above, establish a constrained nonlinear mixed integer optimization mathematical model; Step (5): The constrained nonlinear mixed integer optimization mathematical model is solved using a heuristic intelligent evolutionary algorithm. Step (6): Based on the solution results of the constrained nonlinear mixed integer optimization mathematical model, generate and execute structured control instructions.

[0027] The methods and systems of this application are described below with reference to more accompanying drawings, in one or more detailed embodiments.

[0028] Detailed Implementation Examples exist Figure 1 In a detailed embodiment, a system is illustrated.

[0029] The intelligent flow field control system includes: a data acquisition unit, a constraint modeling unit, an optimization solution unit, and an instruction generation and delivery unit, wherein: The data acquisition unit is configured to acquire wellhead production data, processing module operating status, and effluent water quality data in real time. More specifically, this unit includes a data preprocessing module to ensure data quality: this unit can perform outlier removal, using the 3σ criterion to remove obvious data noise; and perform smoothing processing, using the moving average method to denoise the data, and the size of the moving average window is dynamically adjusted according to the data acquisition frequency (for example, a large window is used for high-frequency acquisition, and a small window is used for low-frequency acquisition).

[0030] The constraint modeling unit is configured to construct a numerical model that includes reservoir geological features and a water quality-damage criterion model, and to set engineering and physical constraints, including: single well constraints, water quality constraints, processing capacity constraints and platform resource constraints. The optimization unit is configured to call the optimization solver, receive a set of collaborative decision variables, calculate the objective function value using numerical simulation, and execute a convergence determination mechanism to solve the constrained nonlinear mixed integer optimization problem. The instruction generation and push unit is configured to convert optimization results into control instructions and has an instruction deviation verification function: when the deviation between the newly generated instruction and the current platform operating parameters exceeds the preset safety threshold, a secondary solution is triggered; when the deviation is within the safety threshold but the adjustment range exceeds the set threshold, a step transition instruction is generated.

[0031] Based on the above system, it is possible to achieve Figure 2 The method and specific process described in the detailed embodiments include: 1. Determine the objective function. Choose one of the following three objectives as the objective function for this optimization, and calculate it as follows: (1) Maximize cumulative oil production in a phase: The goal is to maximize the total amount of crude oil produced by all production wells in an offshore oilfield within a specific time period; (2) Maximizing Net Present Value (NPV): Calculating the objective function based on discounted cash flows over the entire lifecycle. The formula is:

[0032] In the formula, For the development cycle; For the first The annual cash inflow is determined by the projected crude oil production and crude oil price; For the first The annual cash outflow covers the cost of purchasing and installing processing modules, energy consumption costs, profile control agent costs, and maintenance costs; The benchmark rate of return; (3) Optimal uniformity of injection and production flow field: defined as minimizing the standard deviation of the production degree in each injection and production direction, with the objective function being... The calculation is as follows:

[0033] In the formula, the reservoir is divided into Each injection and extraction unit For the first The extraction level of each unit, This represents the average recovery rate across all injection and production units.

[0034] 2. Construct a set of collaborative decision-making variables Define a set of collaborative decision-making variables, including the following key variables: (1) Wellhead injection and production parameters: injection volume of injection well and production volume of production well; (2) Surface liquid treatment parameters: including the activation status and dynamic flux allocation value of the surface liquid treatment module; the activation status represents the capacity expansion deployment decision of the standby module; the dynamic flux allocation value is used to adjust the processing speed of the surface liquid treatment module and supports continuous adjustment within the allowable range of the equipment (e.g., 40%-120% of the rated flux). (3) Profile control operation parameters: including whether profile control is carried out in each well, profile control injection rate, profile control concentration and profile control duration.

[0035] 3. Establish engineering and physical constraint models Construct a set of constraints that cover the following aspects: (1) Single well constraints: including the upper and lower limits of injection and production rates, the upper and lower limits of bottom hole pressure (for fracturing or degassing) and the upper and lower limits of profile control parameters (rate, concentration, duration) for each injection well and production well. (2) Water quality constraints: The oil content of the treated reinjected water must be lower than the set upper limit (e.g., ≤10mg / L) and the suspended solids concentration must be lower than the set upper limit (e.g., ≤2mg / L). (3) Processing capacity constraint: The total processing throughput shall not exceed the upper limit of the operating capacity of the currently activated processing equipment; (4) Platform resource constraints: If new processing modules are involved, their total weight and total projected area shall not exceed the upper limit allowed by the platform.

[0036] 4. Construct a constrained nonlinear mixed-integer optimization mathematical model Using the optimization objective selected in step 1 as the objective function, the collaborative decision variable set described in step 2 as the independent variables, and the constraint condition set described in step 3 as the constraint conditions, an optimization mathematical model is constructed. The acquisition of predicted crude oil production in the objective function value and the acquisition of bottom hole pressure in the constraint conditions are solved based on an improved reservoir seepage equation through numerical simulation, specifically including: (1) Map the set of collaborative decision variables to the grid properties of the reservoir numerical model; (2) Application of the sealing coefficient of the profile control agent Improved seepage equation (taking values ​​[0, 1]):

[0037] (3) The above equations are solved by numerical discretization method, and the pressure field and saturation field are calculated iteratively; (4) Calculate the oil production, water cut, and bottom pressure of each well to obtain the objective function value.

[0038] 5. Solve the optimal mathematical model The above-mentioned optimization mathematical model is solved using heuristic intelligent evolutionary algorithms (such as improved genetic algorithms, particle swarm optimization, etc.).

[0039] (1) Constraint handling strategy: In the solution process, a boundary repair strategy is adopted for variables that violate direct constraints (such as upper and lower limits of variables) to force them to return to the boundary of the feasible region; for parameters that violate state constraints (such as water quality exceeding the standard or pressure exceeding the limit), an adaptive penalty function method is adopted to add a penalty term to the objective function to guide the population to evolve towards the feasible region.

[0040] (2) Dual convergence determination mechanism: Mechanism 1 (Objective Function Stability): When continuous The change in the optimal objective function value of the population is less than the preset accuracy threshold. When convergence is reached, it is determined that the convergence has occurred. Mechanism 2 (Constraint Satisfaction): Calculate the degree to which the current optimal solution violates all engineering and physical constraints. If the maximum violation is less than a preset feasibility threshold... The solution is deemed feasible. The solution is terminated only when both Mechanism 1 and Mechanism 2 are satisfied simultaneously, or when the maximum number of iterations is reached.

[0041] 6. Generate and execute structured control instructions. Based on the solution results, structured control commands are output, including injection and production rate settings for each well, start and stop commands for the processing module, flux settings, and profile control operation plans.

[0042] Stepped transition logic: When the adjustment range of the injection / production volume or throughput setpoint exceeds the set threshold (e.g., 30%), the system automatically generates a stepped transition command for the platform control system to execute. That is, at least one intermediate setpoint is inserted between the current value and the target value to construct a stepped control sequence executed according to the time step. The holding time of the intermediate setpoint is determined according to the platform response characteristic parameters to avoid sudden changes in equipment operation.

[0043] The following practical application scenario illustrates the above method flow and verifies its effectiveness.

[0044] Implementation Target: This embodiment involves the W offshore oilfield platform (hereinafter referred to as "W platform"). The oilfield is in the mid-to-late stage of high water-cut development, and the well network includes 3 water injection wells (numbered I-1 to I-3) and 8 production wells (numbered P-1 to P-8).

[0045] The surface liquid treatment system is configured in a "one main, one standby" mode. The main treatment module (M-1) has a rated processing capacity of 10,000 m³ / d and operates under normal conditions; the standby treatment module (M-2) has a rated processing capacity of 10,000 m³ / d and is in hot standby mode. In addition, the platform has reserved installation interfaces and space for the surface liquid treatment equipment module (M-3).

[0046] The intelligent flow field control system is configured as follows: Data acquisition and preprocessing unit: Data collection targets: Real-time collection of wellhead fluid volume (range 800~2,500 m³ / d), water cut (current average value approximately 92%), wellhead pressure of each production well from P-1 to P-8, as well as operating parameters such as inlet and outlet flow rates and oil content of the produced water of the M-1 module.

[0047] Data cleaning mechanism: Addressing the fluctuating operating conditions of ESPs (Electric Submersible Pumps) on offshore platforms, the data preprocessing module employs an adaptive window moving average method. Specifically: The standard deviation σ of the data is calculated in real time; when σ exceeds a set threshold (e.g., σ > 50 m³ / d for a certain well's fluid volume data, indicating unstable flow), the smoothing window is automatically expanded to 20-30 minutes to filter out noise caused by transient fluctuations such as slug flows; when σ returns to normal, the window is reduced to 5-10 minutes to maintain data real-time performance.

[0048] Constraint modeling unit: Reservoir Model: A detailed three-dimensional geological model is established, with mesh attributes including porosity distribution (0.15~0.28), permeability distribution (50~10000mD), and fluid PVT parameters.

[0049] Physical constraint settings: 1. Processing capacity constraint: Total liquid production Q total ≤ΣQ rated ×Status. Currently, with only M-1 running, the maximum liquid production capacity across the entire platform is 10,000 m³ / d; 2. Water quality constraints: Based on formation damage test data, the oil content of the reinjected water is set to ≤15mg / L (stricter than the industry standard of 20mg / L), and the suspended solids content is set to ≤2mg / L; 3. Single-well constraints: Production wells: Maximum daily fluid production of 2,200 m³ / d (subject to ESP rated displacement and casing size limitations); Water injection well: Maximum daily water injection capacity 3,000 m³ / d (calculated based on a fracture pressure gradient of 0.023 MPa / m) Optimize the solution unit: An integrated reservoir numerical simulator has been developed, which incorporates an improved seepage equation (as described in claim 3) that incorporates a profile control and plugging coefficient β, enabling it to simulate the plugging effect of profile control agents on high-permeability channels.

[0050] The methods and execution flow in this application scenario include: Operating Context: Platform W currently produces 9,800 m³ / d of fluid (close to full capacity of M-1) and approximately 780 m³ / d of oil. Reservoir dynamic analysis shows a high-permeability band (permeability > 5000 mD) between injection well I-2 and production wells P-4 and P-5, resulting in ineffective circulation of injected water. 75% of the injected water from I-2 flows to P-4 and P-5. To tap into the remaining lateral oil, profile control is necessary to regulate the flow field.

[0051] Step 1: Determine the objective function Based on the current state of oilfield development and economic considerations, the optimization objective was set as "maximizing the cumulative oil production in a given phase," with the optimization timeframe set at the next 180 days.

[0052] Step 2: Construct a set of collaborative decision-making variables Define a collaborative decision variable vector X It covers two levels: "surface treatment" and "underground flow field". (1) Injection well profile control parameters: X prof =[loc prof ,β target ] where loc prof Target well selection for profile control; β target The strength of profile control and plugging for the target well.

[0053] (2) Injection and production flow field parameters: Injection volume of 3 injection wells (range 0-3000 m³ / d); Production fluid distribution for 8 production wells (range 800-2200 m³ / d).

[0054] (3) Ground treatment parameters: The activation status of the M-2 module (binary variable: 0 indicates off, 1 indicates on). Flux allocation values ​​for M-1 and M-2 (continuous variables, must meet the rated range of a single unit).

[0055] Step 3: Construct a constrained nonlinear mixed-integer optimization mathematical model 1. Definition of the objective function To address the need for improved flow field control efficiency on the W platform, an objective function is defined. J ( X To maximize the cumulative oil production of the entire oilfield over the next 180 days:

[0056] In the formula: T = 180 days (optimization period); N p =8 (number of production wells); q o,i Let Δt be the oil production rate (m³ / d) of the i-th production well at time step t, which is calculated using a numerical simulator; Δt is the time step (usually 1 day).

[0057] 2. Structured definition of collaborative decision-making variable set The decision variable vector X contains 14 dimensions and is divided into three categories of sub-variables: (1) Injection-production flow field subset X flow : x I1 ,x I2 ,x I3 : Daily water injection rate of 3 injection wells, a continuous variable, ranging from [0, 3000] m³ / d; x P1 ,…,x P8 : Daily fluid production of 8 production wells, a continuous variable, ranging from [800, 2200] m³ / d.

[0058] (2) Ground processing subset X surf : S M2 : The activation status of the backup module M-2, an integer variable {0,1} (0 indicates off, 1 indicates on); the flux allocation is automatically balanced within the algorithm based on the total liquid volume and is not treated as an independent variable to reduce the search dimension.

[0059] (3) Profile parameter subset X prof : loc prof : The well locations for profile control are integer variables {1,2,3}, corresponding to I-1, I-2, and I-3 respectively; β target : Target blocking strength coefficient, continuous variable, range [0.1, 0.9].

[0060] 3. Set of Coupling Constraints (1) Hard constraints on ground processing capacity

[0061] This constraint ensures that the total daily liquid production of the entire platform does not exceed the sum of the rated throughput of the enabled modules.

[0062] (2) Water quality-flux nonlinear constraints Based on experimental fitting of equipment performance curves, the oil content C of reinjected water was determined. oil With total processing throughput Q totalThe relationship is:

[0063] Where: C0 = 5.0 mg / L (steady-state oil content at rated flux); k = 0.0005 - 0.001 mg·d / (L·m³) (overload growth coefficient, determined based on actual equipment); C limit =15mg / L (upper limit for water quality control). This constraint reflects the physical law that the separation effect is stable when the treatment capacity is within the rated range, and the oil content increases when the load is exceeded.

[0064] (3) Injection-production balance constraint To preserve formation energy, the injection-production ratio is limited:

[0065] This range is based on reservoir engineering experience. When the injection-production ratio is <0.9, the risk of formation deficit increases, and when the injection-production ratio is >1.2, it may cause excessive formation pressure.

[0066] Step 4: Solve the optimal mathematical model An improved genetic algorithm (GA-Sim) with nested numerical simulations was used to solve the problem. This algorithm enables interactive iteration between surface decision-making and subsurface response.

[0067] 1. Population Initialization and Encoding An initial population (size N=30-50, adjusted according to problem complexity) is generated using a mixture of real number encoding and binary encoding. Feasibility repair mechanism: If the randomly generated individuals satisfy Σx Pi >10000 but s M2 =0, the algorithm automatically performs logical repair: Option 1: s M2 Set to 1 (Enable standby module); Option 2: Reduce all x proportionally Pi Make the sum ≤ 10000.

[0068] 2. Fitness assessment (core computational loop) For each individual in the population (i.e., one control scheme), perform the following evaluation process: (1) Parameter mapping and geological modeling Read the individual's profile parameters (e.g., loc) prof =2,β target =0.7), locate the grid connection region where well I-2 is located in the numerical model.

[0069] (2) Improved solution of seepage equation The built-in reservoir simulator (based on IMPES or a fully implicit method) is invoked, and time-step iterations are performed using an improved seepage equation that incorporates a profile control coefficient.

[0070] The simulator calculated the water cut change curves of each well over 180 days. w (t).

[0071] (3) Calculation of oil production Combining the liquid production rate x in the decision variable Pi Calculate oil production:

[0072] The objective function value J is obtained by accumulating the results.

[0073] 3. Constraint Handling Mechanism During the evaluation process, a dual-constraint check is performed: (1) Boundary repair strategy (for C1 direct constraint) If a certain scheme has a total liquid volume of 16,000 m³ / d and M-2 is in the "off" state (capacity limit 10,000 m³ / d), it is judged as severely infeasible. Algorithm execution repair: s M2 Flip to "open" (capacity limit becomes 20000m³ / d); bring the solution back to the feasible region and mark the individual as having been repaired.

[0074] (2) Adaptive penalty function method (for C2 and C3 state constraints) If the simulation results show that the injection-production ratio drops to 0.8 at a certain moment (violating the lower limit of C3, 0.9), or the oil content of the reinjected water reaches 18 mg / L (violating the upper limit of C2, 15 mg / L), a penalty is imposed on the objective function:

[0075] In the formula, λ is an adaptive penalty coefficient that grows linearly or exponentially with the number of generations (e.g., λ = λ0 × (1 + 0.1 × generation)), forcing the population to strictly adhere to the constraints in the later stages.

[0076] 4. Dual convergence determination mechanism The algorithm enters an iterative loop until one of the following conditions is met: Mechanism 1 (Target Stability): The increase in optimal oil production over K consecutive generations (K is usually 8-15) is less than the preset accuracy threshold ε1 (e.g., cumulative oil production change < 50 tons, which accounts for about 0.5%-1% of total production).

[0077] Mechanism 2 (Constraint Feasibility): The number of all constraint violations in the current optimal solution is less than the feasibility threshold ε2 (e.g., the satisfaction rate of all constraints is >99%).

[0078] Typical operation process: Early stage (generations 1-15): The algorithm explored the combination of "I-2 well profile adjustment + M-2 opening + full platform strong production", which significantly improved oil production, but some schemes caused VRR alarms due to insufficient water injection; Mid-term (generations 16-30): The algorithm automatically adjusts the water injection volume of I-1 and I-3 to replenish formation energy, and the solution gradually becomes feasible; Later stages (generations 31-50): The solution is finely optimized within the feasible region, and the fluctuation of the objective function is narrowed; Finally, it converges in the 40th-50th generation, outputting the optimal solution vector X. .

[0079] Example of convergence result: The optimal solution includes: Profile control instruction: Implement deep profile control on injection well I-2, injecting a polymer system with a target plugging coefficient of 0.65-0.75; Capacity expansion command: Activate the M-2 backup module; Flow field control: Increase fluid volume across the entire platform to 14,500-15,500 m³ / d. Typical well examples: Increase fluid volume in well P-4 from 1200 to 1900-2000 m³ / d, and in well P-5 from 1100 to 1800-1900 m³ / d. Ground flux distribution: M-1 processes 7500-8500 m³ / d, M-2 processes 6500-7500 m³ / d (ensuring single unit load rate is within the optimal separation efficiency range of 70%-85%).

[0080] Step 5: Generate and execute structured control instructions 1. Generation of step transition instructions The system detected that the target total liquid volume (approximately 15,000 m³ / d) has increased by about 50% compared to the current value (9,800 m³ / d) (exceeding the preset threshold of 30%), and M-2 needs to be started from cold standby mode. The system automatically generates a tiered execution sequence: T+0h: Start M-2 for self-circulation preheating and equipment self-check; T+12h: Conduct profile control operations on well I-2; T+24h (end of profile control): Increase fluid volume of production wells P-4 and P-5 by 15%-20%, start process at M-2, initial diversion of 2,500-3,000 m³ / d; T+48-72h: The production well fluid volume is gradually increased to the target value, and the M-2 flow rate is simultaneously increased to 6500-7500 m³ / d; Subsequent monitoring period: Continuously monitor the reinjected water quality and formation pressure response, and make minor adjustments based on actual conditions.

[0081] 2. Command Push and Verification The instruction is issued after undergoing the following verifications: Load rate verification: The single-unit load rates of M-1 and M-2 are 75%-85% and 65%-75% respectively, which are within the optimal separation efficiency range; Water quality prediction: Based on the C2 constraint formula, the predicted oil content of the reinjected water is 8~12 mg / L, which meets the requirement of ≤15 mg / L; Pressure safety verification: The predicted formation pressure value is lower than the 90% safety factor of the fracture pressure gradient. After the verification is passed, the command is sent to the platform DCS (Distributed Control System) through OPCServer or dedicated communication protocol.

[0082] Implementation effect verification The effectiveness of the plan will be evaluated after 180 days of implementation. (1) Liquid processing capacity The platform's daily liquid processing capacity is stable at 14,800-15,200 m³ / d. The parallel operation of the M-1 and M-2 units is stable, with single-unit load rate fluctuations of <5%.

[0083] (2) Water quality compliance rate Due to reasonable load distribution, the separation efficiency was reduced due to single-unit overload. The average oil content of the reinjected water was 9-11 mg / L, achieving a 100% compliance rate (based on 180 days of continuous monitoring data).

[0084] (3) Oil-enhancing effect Through the enhanced effect of profile control in well I-2 and significant fluid extraction in production wells, the water cut of wells P-4 and P-5 decreased by 5% and 4%, respectively. The cumulative oil production during this period increased by 15% compared to the baseline forecast value based on the original "single unit operation + low liquid volume" model.

[0085] (4) Verification conclusion This embodiment verifies that, under the condition of limited surface processing capacity (10,000 m³ / d per unit), the coupling drive mechanism of the method of this invention enables intelligent decision-making to activate the backup module and cooperate with the underground profile adjustment and strong production strategy, successfully breaking through the constraint of surface bottleneck on reservoir potential tapping and realizing the synergistic flow field regulation and efficiency enhancement of surface and underground.

[0086] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0087] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0088] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart flow field control method for improving efficiency and expanding capacity in ground liquid treatment through coupled driving, characterized in that, include: Step (1): Construct an objective function based on the selected preset optimization objective. The preset optimization objective includes: maximizing the cumulative oil production in the stage, maximizing the net present value of development, and optimizing the balance of the injection and production flow field. Step (2) Construct a set of collaborative decision-making variables, including: wellhead injection and production parameters, surface fluid treatment parameters, and profile control operation parameters; Step (3): Based on engineering and physical constraints, construct a set of constraints, including: single well constraints, water quality constraints, treatment capacity constraints, and platform resource constraints; Step (4): Based on the objective function, collaborative decision-making variable set, and constraint set constructed above, establish a constrained nonlinear mixed integer optimization mathematical model; Step (5): The constrained nonlinear mixed integer optimization mathematical model is solved using a heuristic intelligent evolutionary algorithm. Step (6): Based on the solution results of the constrained nonlinear mixed integer optimization mathematical model, generate and execute structured control instructions.

2. The method according to claim 1, characterized in that, In step (1), when the selected preset optimization objective is to maximize the net present value of development, the objective function is... The calculation is based on discounted cash flows over the entire lifecycle, and the formula is: In the formula, For the development cycle; For the first The annual cash inflow is determined by the projected crude oil production and crude oil price; For the first The annual cash outflow covers the cost of purchasing and installing processing modules, energy consumption costs, profile control agent costs, and maintenance costs; The benchmark rate of return; When the selected preset optimization objective is to achieve optimal injection-production flow field uniformity, the objective function is... Defined as minimizing the standard deviation of the recovery rate in each injection-production direction, the calculation process includes: dividing the reservoir into... Each injection and production unit is used to calculate the recovery rate of each unit. The standard deviation is calculated using the following formula: In the formula, This represents the average recovery rate across all injection and production units.

3. The method according to claim 2, characterized in that, In step (2), the set of collaborative decision-making variables includes: (1) Wellhead injection and production parameters: injection volume of injection well and production volume of production well; (2) Surface liquid treatment parameters: including the activation status and dynamic flux allocation value of the surface liquid treatment module; the activation status represents the capacity expansion deployment decision of the standby module; the dynamic flux allocation value is used to adjust the processing speed of the surface liquid treatment module and supports continuous adjustment within the allowable range of the equipment. (3) Profile control operation parameters: including whether profile control is carried out in each well, profile control injection rate, profile control concentration and profile control duration.

4. The method according to claim 3, characterized in that, In step (3), the single-well constraints include the upper and lower limits of injection and production rates, bottom hole pressure, and profile control parameters for each well; the water quality constraints include the upper limit of oil content in the treated reinjected water and the upper limit of suspended solids concentration; the treatment capacity constraints are that the total treatment throughput does not exceed the upper limit of the operating capacity of the treatment equipment; and the platform resource constraints include the upper limit of the total weight and total projected area of ​​the newly added treatment modules.

5. The method according to claim 4, characterized in that, In step (4), the collaborative decision-making variable set is used as the independent variable, and the constraint condition set is used as the constraint condition to construct a constrained nonlinear mixed integer optimization mathematical model. The objective function value for predicting crude oil production and the constraint condition for obtaining bottom hole pressure are obtained based on the improved reservoir seepage equation and solved through numerical simulation. Specifically, this includes: (1) Map the set of collaborative decision variables to the grid properties of the reservoir numerical model; (2) Application of the sealing coefficient of the profile control agent Improved seepage equation (taking values ​​[0, 1]): (3) The above equations are solved by numerical discretization method, and the pressure field and saturation field are calculated iteratively; (4) Calculate the oil production, water cut, and bottom pressure of each well to obtain the objective function value.

6. The method according to claim 5, characterized in that, In step (5), when using the heuristic intelligent evolutionary algorithm to solve the problem, a dual convergence determination mechanism is set up: Mechanism 1: Stability determination of the objective function, when continuous The change in the optimal objective function value of the population is less than a preset accuracy threshold. When the objective function converges, it is determined that the objective function has converged. Mechanism 2: Constraint Satisfaction Determination. Calculate the degree to which the current optimal solution violates all engineering and physical constraints. If the maximum violation is less than a preset feasibility threshold... The solution is deemed feasible. The optimization process terminates and the result is output only when both Mechanism 1 and Mechanism 2 are satisfied simultaneously, or when the maximum number of iterations preset by the algorithm is reached.

7. The method according to claim 6, characterized in that, In step (6), structured control commands are output based on the solution results, including injection and production volume settings for each well, start and stop commands for the processing module, flux settings, and profile control operation plans.

8. The method according to claim 7, characterized in that, The system employs a stepped transition logic. When the adjustment range of the injection / production volume or throughput setting exceeds the set threshold, the system automatically generates a stepped transition command for the platform control system to execute.

9. A computer storage medium, characterized in that, The device contains a computer program that is executed by a processor to implement the method described in claims 1-8.

10. A smart flow field control system for improving efficiency and expanding capacity in ground-based liquid treatment, characterized in that, It includes a processor and a memory, the memory storing a computer program, which is executed by the processor to implement the method of claims 1-8.

Images