Low-permeability reservoir flooding and burying collaborative optimization method based on swarm intelligence algorithm

Abstract

A low-permeability reservoir flooding and burying collaborative optimization method based on a swarm intelligence algorithm is characterized in that a mathematical model (optimization variables, constraint conditions and objective functions) under carbon dioxide flooding and storage collaborative optimization and construction of related solving methods are defined on the basis of an optimization theory, and the swarm intelligence algorithm is combined, so that the reservoir flooding and burying collaborative optimization of the low-permeability reservoir is realized. And establishing a carbon dioxide flooding and storage collaborative optimization theoretical system, and performing instance application analysis. According to the CO2 oil displacement and storage collaborative optimization method for the low-permeability reservoir, the swarm intelligence algorithm and the numerical simulation technology are coupled, and a theoretical basis and a technical support are provided for actual CO2 oil displacement and storage projects.

Description

Technical Field

[0001] The present invention belongs to the field of carbon dioxide enhanced oil recovery technology and carbon dioxide storage technology in oil field development, and specifically relates to a low-permeability oil reservoir flooding and burial collaborative optimization method based on a swarm intelligence algorithm. Background Art

[0002] China has a high proportion of low-permeability crude oil reserves. In recent years, low-permeability reserves have accounted for an average of over 70% of CNPC's proven reserves. The world is facing severe global warming, with carbon dioxide being the primary driver. Carbon dioxide flooding in low-permeability reservoirs can not only stabilize and increase production, improving oil recovery, but also achieve geological storage of carbon dioxide, reducing CO2 emissions.

[0003] Therefore, when adjusting CO2 flooding development plans in low-permeability reservoirs, both recovery efficiency and storage efficiency must be considered. In CO2 flooding EOR (enhanced oil recovery) applications, the goal is to maximize cumulative oil production by injecting less CO2, while CO2 geological storage aims to maximize CO2 storage within the geological reservoir. However, the relationship between CO2 flooding and storage is not linear. Maintaining CO2 recovery capacity while maximizing CO2 storage efficiency, while also balancing economic benefits, constitutes a multi-objective collaborative optimization problem. A comprehensive optimization theory framework is needed for application in the field of CO2 flooding and storage. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, the present invention aims to provide a swarm intelligence-based collaborative optimization method for flooding and storage in low-permeability reservoirs, addressing the theoretical framework for the collaborative optimization of CO2 flooding and storage. This approach provides a theoretical basis and practical examples for CO2 flooding and storage in low-permeability reservoirs, enabling collaborative optimization of CO2 flooding, CO2 storage, and economic benefits.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A collaborative optimization method for flooding of low permeability oil reservoirs based on swarm intelligence algorithm, characterized by comprising the following steps:

[0007] Step 1: Establish a numerical simulation model for CO2 flooding and storage;

[0008] First, based on the reservoir parameters (structural map, formation thickness, porosity, permeability, etc.) of the CO2 flooding and storage study area, fluid parameters (PVT data of crude oil, natural gas, and water, including density, viscosity, compressibility, relative permeability, etc.), production data (historical pressure, production, water content, injection rate, etc.), and well data (well location, completion data, well type, well pattern, etc.), a commercial numerical simulator is used to establish a CO2 flooding and storage numerical simulation model. This numerical simulation model is then used to achieve coordinated optimization of CO2 flooding and storage in actual reservoirs.

[0009] Step 2: Establish a mathematical model related to the coordinated optimization of CO2 flooding and storage;

[0010] CO2 flooding and storage technology aims to improve CO2 recovery while storing as much CO2 as possible. Based on the optimization theory, a mathematical model under the relevant conditions of CO2 flooding and storage is established. The steps are as follows:

[0011] (1) Design of optimization variables;

[0012] The effectiveness of oil recovery and storage is influenced by reservoir conditions and the oilfield development plan. Reservoir conditions include reservoir physical properties (reservoir geological structure, reservoir porosity and permeability characteristics, oil, gas, and water saturation field distribution in the reservoir, pressure distribution, etc.) and fluid characteristics (crude oil composition, oil-gas-water relative permeability, etc.). These parameters are fixed in each reservoir and are difficult to change. The oilfield development plan includes the well pattern layout (well pattern type, well location, well spacing, number of wells) and the design of the injection-production plan (CO2 flooding method, injection-production parameters, etc.). These parameters can be adjusted to achieve optimal CO2 recovery and storage.

[0013] (2) Establishment of constraints;

[0014] Due to the constraints of the actual economic and technical conditions of the reservoir and actual field experience, it is necessary to constrain the optimization variables and other conditions in the collaborative process to achieve the purpose of simulating the real situation;

[0015] (3) Construction of objective function;

[0016] CO2 flooding and storage technology aims to increase CO2 recovery while storing as much CO2 as possible. At the same time, as an enterprise unit, the oil field needs to achieve greater economic benefits while reducing costs. Evaluation indicators should be constructed from three aspects: flooding effect, storage efficiency, and economic benefits, as the objective function of the project.

[0017] CO2 flooding evaluation (recovery factor RF):

[0018] (1);

[0019] CO2 storage evaluation indicators (CO2 storage efficiency CE):

[0020] (2);

[0021] Economic benefit evaluation indicators (internal rate of return IRR):

[0022] (3);

[0023] (4);

[0024] in: Indicates the Annual cash flow, represents the initial investment, Represents the life cycle of a project; is the revenue from producing crude oil, is the income from carbon emission trading, is the cost of capturing, compressing and transporting CO2, is the CO2 produced gas separation cost, is the cost of injecting water, Produced water treatment costs, is the cost of supercritical CO2 injection;

[0025] Step 3: Collaborative optimization solution method;

[0026] The Pareto front indicates that in a multi-objective optimization problem, the solution obtained according to the Pareto front is an optimal state that cannot be further optimized and improved. It makes one or more objective functions or preference criteria better and no individual or any preference criterion becomes worse. The Pareto front indicates that in a multi-objective optimization problem, the solution obtained according to the Pareto front is an optimal state that cannot be further optimized and improved. It makes one or more objective functions or preference criteria better and no individual or any preference criterion becomes worse. The following relationship exists:

[0027] (1) Dominance Relationship: Given two feasible solutions and , can be regarded as Dominate In this case, the solution Significantly better than , which can be written as , the mathematical expression is as follows:

[0028] (5);

[0029] (2) Non-dominated Relationship: When the following conditions are met, it is called a solution. and The relationship of right and wrong:

[0030] (6);

[0031] (3) Pareto optimal solution: given solution The Pareto optimal solution of the target optimization problem does not exist. satisfy hour, It is called the Pareto optimal solution of the multi-objective optimization problem;

[0032] (4) Pareto Set: Design Space The set of all Pareto optimal solutions in ;

[0033] (5) Pareto Front: The curve or surface formed by the vectors corresponding to the Pareto optimal solution set in the target space, that is, ;

[0034] In the coordinated optimization of CO2 flooding and storage, the purpose of the Pareto front is to find more injection and production parameter solution sets that meet actual needs. Based on the Pareto front of the multi-objective optimization solution model and combined with the swarm intelligence algorithm, the coordinated optimization solutions of CO2 flooding and storage under different multi-objective solution methods and different objective functions are obtained to facilitate reservoir workers to choose among them.

[0035] Swarm intelligence algorithms, as an emerging solution, are inspired by natural phenomena such as biological evolution, plant reproduction, and animal predation. They are a class of random search algorithms designed by mimicking these natural processes. Due to their simple mechanisms, limited variables, ease of implementation, and excellent generalization, they have attracted widespread attention from researchers across various fields and have seen significant development and application. Currently, research in swarm intelligence theory primarily includes genetic algorithms (GAs), ant colony algorithms (ACOs), particle swarm algorithms (PSOs), artificial fish swarm algorithms (AFSAs), artificial bee colony algorithms (ABCs), artificial firefly algorithms (GSOs), frog leaping algorithms (SFLAs), grey wolf algorithms (GWOs) (Mirjalili et al., 2014), whale algorithms (WOAs), and raccoon algorithms (COAs).

[0036] The beneficial effects of the present invention are:

[0037] The collaborative optimization method for CO2 flooding and storage in low permeability oil reservoirs based on swarm intelligence algorithm provided by the present invention has certain reference value in CO2 flooding and storage collaborative projects, and provides a theoretical basis for CO2 flooding and storage projects in low permeability oil reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of the coordinated optimization method for CO2 flooding and storage in low permeability reservoirs;

[0039] Figure 2 3D permeability distribution diagram of low permeability reservoir in an embodiment of the present invention;

[0040] Figure 3 The Pareto frontier diagram of the three objectives (recovery rate, CO2 storage efficiency, and economic net present value) in the embodiment of the present invention;

[0041] Figure 4 The mapping of the three-dimensional Pareto front under different objective functions in the embodiment of the present invention (the economic net present value is the evaluation indicator);

[0042] Figure 5 The Pareto frontier diagram of the three objectives (recovery factor, CO2 storage efficiency, and internal rate of return) in the embodiment of the present invention;

[0043] Figure 6 The mapping of the three-dimensional Pareto front under different objective functions in an embodiment of the present invention (internal rate of return is the evaluation indicator);

[0044] Table 1 is the definition of the mathematical model in the coordinated optimization problem of CO2 flooding and storage in the embodiment of the present invention;

[0045] Table 2 is a Pareto optimal solution set table (recovery factor, CO2 storage efficiency, economic net present value) in the embodiment of the present invention;

[0046] Table 3 is a Pareto optimal solution set table (recovery factor, CO2 storage efficiency, internal rate of return) in the embodiment of the present invention;

[0047] Table 4 shows the optimization results of injection-production parameter combinations under different economic evaluation indicators in the embodiment of the present invention. DETAILED DESCRIPTION

[0048] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0049] The test reservoir is based on the typical reservoir and fluid physical properties of a low-permeability reservoir. A 41×31×10 inverse seven-point horizontal well mechanism model was established using numerical simulation technology. The grid size on the plane of the model is 15×15 m, and the vertical grid size is 3.08 m. The model is divided into four small layers and simulated using a three-phase component model with five pseudo-components. The state equation is Peng-Robinson. The specific three-dimensional distribution of reservoir permeability is shown in the attached figure. Figure 2 shown.

[0050] Example:

[0051] As described in the patent claim method, Figure 1 As shown in the figure, based on optimization theory and combined with the actual model of CO2 flooding and storage, the optimization variables for the coordinated optimization of CO2 flooding and storage are established, the constraints are clarified, and the relevant CO2 flooding and storage coordinated optimization indicators are established and integrated with the multi-objective solution model. On this basis, the appropriate swarm intelligence algorithm is selected, and the Matlab numerical simulator is combined with commercial numerical simulation software to obtain the CO2 flooding and storage coordinated optimization solution / solution set required by reservoir workers, completing the application of the CO2 flooding and storage coordinated optimization method under low permeability reservoir conditions.

[0052] As described in the patent claim method, based on the actual reservoir physical properties, well pattern development mode and other conditions in the study area, a numerical simulation model for the coordinated optimization of CO2 flooding and storage in the study area is first established. Figure 2 Test reservoir shown.

[0053] As described in the patent's claimed method, a method for synergistic optimization of CO2 flooding and storage was applied to a test reservoir. While the well pattern layout in the field development plan is fixed, the injection-production scheme design is the variable to be optimized. This paper uses five injection-production parameters under the continuous gas injection mode as the variables to be optimized. In combination with the aforementioned constraint formulation rules, the relevant optimization variables and their constraints were established, as shown in Table 1. Based on the desired results from the synergistic optimization of CO2 flooding and storage, evaluation indicators related to recovery factor, CO2 storage efficiency, and internal rate of return / economic net present value were established.

[0054] As described in the patent claim method, based on the multi-objective optimization solution model Pareto front, and combined with the swarm intelligence algorithm (this paper uses the particle swarm algorithm, with 30 populations and 30 iterations), the CO2 flooding and storage collaborative optimization solutions / solution sets under different multi-objective solution methods and different objective functions are as follows Figure 3-Figure 6As shown in Appendix Tables 2 and 3. Under the same optimization conditions, on the Pareto front, because the dimensionless objective function eliminates optimization instabilities caused by scaling issues with the dimensional objective function, the internal rate of return (IRR) produces a larger set of solutions, a more concentrated distribution, and greater selectivity than the solution set consisting of the economic net present value (ENPV), recovery factor, and CO2 storage efficiency. Table 4 also shows that, under the same computing resources (optimization conditions), using the dimensionless IRR as the objective function can yield more injection-production parameter combinations that meet the practical requirements of CO2 flooding and storage coordinated optimization. In CO2 flooding and storage projects, decision makers typically ensure that multiple objective functions meet certain standards before selecting one or more solutions to guide their implementation. The dimensionless IRR not only provides an intuitive view of the actual investment returns of CO2 flooding and storage projects, but also allows for the generation of more suitable injection-production parameter combinations in multi-objective optimization, facilitating decision makers' selection based on their specific circumstances. Therefore, we believe that using the IRR as the objective function is superior to the ENPV in CO2 flooding and storage coordinated optimization.

[0055] In comprehensive comparison, the CO2 flooding and storage collaborative optimization method proposed in this patent is to first select the optimization variables based on the relevant conditions of the optimization variables and formulate reasonable constraints. On this basis, it is necessary to establish a dimensionless evaluation index and use the Pareto frontier solution method combined with the swarm intelligence algorithm to obtain the optimal solution set for the CO2 flooding and storage collaborative optimization.

[0056] The above description is only a specific embodiment of the present invention and is not intended to limit the scope of the present invention. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principle of the present invention should fall within the scope of protection of the present invention.

[0057]

[0058]

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] .

Claims

1. A collaborative optimization method for flooding in low permeability reservoirs based on swarm intelligence algorithm, characterized in that: The following steps are included: Step 1: Establish a numerical simulation model for CO2 flooding and storage; First, based on the reservoir parameters (structural map, formation thickness, porosity, permeability, etc.) of the CO2 flooding and storage study area, fluid parameters (PVT data of crude oil, natural gas, and water, including density, viscosity, compressibility, relative permeability, etc.), production data (historical pressure, production, water content, injection rate, etc.), and well data (well location, completion data, well type, well pattern, etc.), a commercial numerical simulator is used to establish a CO2 flooding and storage numerical simulation model. This numerical simulation model is then used to achieve coordinated optimization of CO2 flooding and storage in actual reservoirs. Step 2: Establish a mathematical model related to the coordinated optimization of CO2 flooding and storage; CO2 flooding and storage technology aims to improve CO2 recovery while storing as much CO2 as possible. Based on the optimization theory, a mathematical model under the relevant conditions of CO2 flooding and storage is established. The steps are as follows: (1) Design of optimization variables; The effectiveness of oil recovery and storage is influenced by reservoir conditions and the oilfield development plan. Reservoir conditions include reservoir physical properties (reservoir geological structure, reservoir porosity and permeability characteristics, oil, gas, and water saturation field distribution in the reservoir, pressure distribution, etc.) and fluid characteristics (crude oil composition, oil-gas-water relative permeability, etc.). These parameters are fixed in each reservoir and are difficult to change. The oilfield development plan includes the well pattern layout (well pattern type, well location, well spacing, number of wells) and the design of the injection-production plan (CO2 flooding method, injection-production parameters, etc.). These parameters can be adjusted to achieve optimal CO2 recovery and storage. (2) Establishment of constraints; Due to the constraints of the actual economic and technical conditions of the reservoir and actual field experience, it is necessary to constrain the optimization variables and other conditions in the collaborative process to achieve the purpose of simulating the real situation; (3) Construction of objective function; CO2 flooding and storage technology aims to increase CO2 recovery while storing as much CO2 as possible. At the same time, as an enterprise unit, the oil field needs to achieve greater economic benefits while reducing costs. Evaluation indicators should be constructed from three aspects: flooding effect, storage efficiency, and economic benefits, as the objective function of the project. CO2 flooding evaluation (recovery factor RF): （1）； CO2 storage evaluation indicators (CO2 storage efficiency CE): （2）； Economic benefit evaluation indicators (internal rate of return IRR): （3）； （4）； in: Indicates the Annual cash flow, represents the initial investment, Represents the life cycle of a project; is the revenue from producing crude oil, is the income from carbon emission trading, is the cost of capturing, compressing and transporting CO2, is the CO2 produced gas separation cost, is the cost of injecting water, Produced water treatment costs, is the cost of supercritical CO2 injection; Step 3: Collaborative optimization solution method; The Pareto front indicates that in a multi-objective optimization problem, the solution obtained according to the Pareto front is an optimal state that cannot be further optimized and improved. It makes one or more objective functions or preference criteria better and no individual or any preference criterion becomes worse. The Pareto front indicates that in a multi-objective optimization problem, the solution obtained according to the Pareto front is an optimal state that cannot be further optimized and improved. It makes one or more objective functions or preference criteria better and no individual or any preference criterion becomes worse. The following relationship exists: (1) Dominance Relationship: Given two feasible solutions and , can be regarded as Dominate In this case, the solution Significantly better than , which can be written as , the mathematical expression is as follows: （5）； (2) Non-dominated Relationship: When the following conditions are met, it is called a solution. and The relationship of right and wrong: （6）； (3) Pareto optimal solution: given solution The Pareto optimal solution of the target optimization problem does not exist. satisfy hour, It is called the Pareto optimal solution of the multi-objective optimization problem; (4) Pareto Set: Design Space The set of all Pareto optimal solutions in ; (5) Pareto Front: The curve or surface formed by the vectors corresponding to the Pareto optimal solution set in the target space, that is, ; In the coordinated optimization of CO2 flooding and storage, the purpose of the Pareto front is to find more injection and production parameter solution sets that meet actual needs. Based on the Pareto front of the multi-objective optimization solution model and combined with the swarm intelligence algorithm, the coordinated optimization solutions of CO2 flooding and storage under different multi-objective solution methods and different objective functions are obtained to facilitate reservoir workers to choose among them.

2. The collaborative optimization method for flooding of low permeability reservoirs based on swarm intelligence algorithm according to claim 1, characterized in that: The swarm intelligence algorithms are genetic algorithm (GA), ant colony algorithm (ACO), particle swarm algorithm (PSO), artificial fish swarm algorithm (AFSA), artificial bee colony algorithm (ABC), artificial firefly algorithm (GSO), frog leaping algorithm (SFLA), grey wolf algorithm (GWO) (Mirjalili et al., 2014), whale algorithm (WOA), and raccoon algorithm (COA).

Images