Well pattern and well location optimization method based on flow field area and equilibrium utilization evaluation

Abstract

The invention discloses a well network and well location optimization method based on a flow field area and equilibrium utilization evaluation, and relates to the technical field of petroleum engineering. On the basis of streamline numerical simulation, a water-drive reservoir streamline numerical simulation well location optimization model is constructed, a target function based on a production system, optimization variables, constraint variables and flight time balance is defined, an initial population is constructed on the basis of Latin hypercube sampling, and a water-drive reservoir streamline numerical simulation well location optimization agent model is constructed; and based on a Q-Learning enhanced marine predator algorithm, performing simulation by using the water-drive reservoir streamline numerical simulation well location optimization agent model, obtaining streamline data and derivative data obtained by simulation of the water-drive reservoir streamline numerical simulation well location optimization agent model, and performing well network and well location optimization. According to the method, the well position optimization speed and optimization effect are effectively improved, the evaluation period of the well position arrangement scheme is remarkably shortened, and the comprehensive economic benefits of oilfield development are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of petroleum engineering, and in particular to a well pattern and well location optimization method based on flow field area and balanced production evaluation. Background Art

[0002] During oil and gas field exploration and development, the spatial heterogeneity of reservoirs and fluids has a crucial impact on development efficiency. Constrained by multiple factors, including reservoir permeability distribution, fluid viscosity differences, and fault structure, well production rates vary significantly across different spatial locations within a reservoir, with single-well production fluctuations in local areas reaching orders of magnitude. This strong heterogeneity makes the scientific formulation of well pattern deployment under complex geological conditions and the maximization of recovery through spatially optimized well placement a core challenge in improving the economic efficiency of oilfield development.

[0003] At present, there are two main methods for well location design: one is the traditional design method based on reservoir engineering experience. This method relies on geological modeling and numerical simulation, and by screening reservoir physical parameters, it prioritizes well placement in areas with high oil saturation and structural highs. Although this method can avoid the risk of edge and bottom water coning, its decision-making mechanism has significant flaws: First, the generation of well placement plans relies on engineers' subjective interpretation of static geological models, with a high weight of experience, and randomness in the exploration of the plan space. Second, when analyzing multi-source heterogeneous data coupling, the efficiency of manual data processing is limited, and the final well location often deviates from the theoretical optimal position because the cumulative error of the ambiguity of key parameters exceeds the allowable threshold.

[0004] Another approach is an automated well location optimization method based on optimization theory. Its technical process includes three core steps: establishing a multi-constraint optimization model with economic net present value (NPV) or stage recovery as the objective function; using intelligent optimization algorithms such as differential evolution to drive iterative updates of well location coordinate parameters; and quantitatively evaluating the production dynamic response of each candidate solution through a reservoir numerical simulator. Although this method reduces human intervention by introducing an automated optimization mechanism, it faces severe computational efficiency bottlenecks, especially when applied to reservoir models with a million-grid scale. A single full-cycle numerical simulation is time-consuming, and the optimization algorithm typically requires evaluating a large number of candidate solutions before convergence, resulting in a long overall optimization cycle. More seriously, some iterative solutions become invalid solutions due to violations of inter-well interference suppression criteria or deviations from the laws of seepage dynamics, resulting in a significant waste of computing resources and severely limiting the engineering applicability of this method.

[0005] Therefore, it is urgent to propose a well network and well location optimization method based on flow field area and balanced production evaluation to improve the well network and well location optimization efficiency, shorten the well network and well location optimization cycle, reduce the time and money spent on oil field well location deployment, and improve the comprehensive economic benefits of oil field development. Summary of the Invention

[0006] The present invention aims to overcome the defects of the existing well position design method. carry A well pattern and well position optimization method based on flow field region and balanced production evaluation is provided. By using streamline numerical simulation to obtain the streamline derivative information of the well position optimization model for waterflooding reservoirs, a streamline feature objective function based on flight time balance is constructed, and the meta-heuristic QQLMPA algorithm is used to optimize the well position. During the well pattern and well position optimization process, a better well position distribution plan can be obtained with fewer numerical simulation times, significantly shortening the evaluation cycle of the well position layout plan.

[0007] To achieve the above object, the present invention adopts the following technical solutions: A well pattern and well position optimization method based on flow field region and balanced production evaluation, comprising the following steps: Step 1, construct a well position optimization model for streamline numerical simulation of waterflooding reservoirs, clarify the production system, determine the optimization variables, constraint variables and objective function, and construct an initial population based on Latin hypercube sampling; Step 2, construct an intelligent agent model for well position optimization of streamline numerical simulation of waterflooding reservoirs based on the Q-Learning enhanced marine predator algorithm; Step 3, use the intelligent agent model for well position optimization of streamline numerical simulation of waterflooding reservoirs to perform simulation, obtain and process the streamline data and derivative data obtained from the simulation of the intelligent agent model for well position optimization of streamline numerical simulation of waterflooding reservoirs, and establish a streamline data set; Step 4, combine the historical data in the elite data retention area, update the Q-table to optimize the phase selection strategy through Q-Learning, generate candidate solutions using QOL quasi-oppositional learning to expand the search range, and use the updated strategy to guide MPA for the next generation of well position optimization; Step 5, based on the reservoir streamline numerical simulator, continuously interact the intelligent agent model for well position optimization of streamline numerical simulation of waterflooding reservoirs with the reservoir numerical simulation, evaluate the fitness of the new population, and screen out elite samples and store them in the elite data retention area; Step 6, repeat Steps 4 to 5 until the preset maximum number of iterations is reached, and determine the optimal well position arrangement.

[0008] Preferably, Step 1 includes the following steps: Step 1.1, construct a well position optimization model for streamline numerical simulation of waterflooding reservoirs; Step 1.2, use the two-dimensional coordinates of injection wells and production wells as optimization variables; the constraint variables are the minimum well position spacing constraint and the non-repetitive well position constraint; the objective function is the well position optimization objective function for streamline numerical simulation , and the goal is to maximize the cumulative oil production under the conditions of meeting the well position spacing constraint and the non-repetitive well position constraint; Step 1.3, construct the initial population based on Latin hypercube sampling; Latin hypercube sampling is used to sample from a multivariate parameter distribution, and the number of injection well location plans to be generated is Determine the number of samples to be drawn and divide the (0,1) interval into Segments are randomly sampled in each segment, the sampled samples are mapped to standard normal distribution samples through the inverse function of the standard normal distribution and randomly sorted to obtain the initial population.

[0009] Preferably, the water drive reservoir streamline numerical simulation well location optimization model is: (1) The constraints are: ; ; ; ; In the formula, is the well location coordinate set; Optimize the objective function for well location in streamline numerical simulation; is the minimum function; is the streamline displacement potential coefficient; is the coefficient of variation of streamline flight time; is the penalty term between injection-production well pairs; It is the internal penalty item for injection and production wells; and are the upper and lower bounds of the boundary constraints of the well location coordinate set respectively; is the well spacing constraint; , All are numbered; , All are single well location coordinates; is the number of injection wells, is the number of producing wells.

[0010] Preferably, the step 2 comprises the following steps: Step 2.1: Determine the well location coordinates of the expected production cycle based on the particle coordinate values in the initial population and obtain the well location coordinate set : (2); Step 2.2, set the simulation time step according to the time step of the expected production cycle and perform streamline numerical simulation; Step 2.3: Determine the streamline displacement potential coefficient at the current time step based on the proportion of the number of grids that the streamline passes through to the number of grids in the water flooding reservoir streamline numerical simulation well location optimization model. ; Step 2.4, based on the coefficient of variation of streamline flight time at the current time step Set the balance degree of each injection and production well on the average flight time; Step 2.5: Penalty term between injection and production wells at the current time step Set the positive difference between the average flight time of each injection-production well pair and the expected production cycle; Step 2.6, based on the internal penalty item of the injection and production well at the current time step Set the average value of the positive difference between the flight time of the streamlines contained in each injection-production well pair and the expected production period; Step 2.7, determine the initial parameters of the learning environment of the intelligent agent model for well location optimization in the numerical simulation of water flooding reservoir streamlines.

[0011] Preferably, the streamline displacement potential coefficient at the current time step is for: (3) In the formula, is the total number of grids in the well location optimization model for the numerical simulation of the water drive reservoir streamline that the streamline passes through; The total number of grids in the well location optimization model for numerical simulation of waterflood reservoir flow lines; The coefficient of variation of streamline flight time at the current time step for: (4) In the formula, is the sample standard deviation of the flight time of the injection and production wells; is the average flight time of injection and production well pairs; is the number of injection-production well pairs; For the Flight time of each injection-production well pair; The penalty term between injection and production wells at the current time step for: (5) In the formula, is the number of injection-production well pairs whose flight time is less than the expected production cycle, ; is the penalty intensity power series, ; The internal penalty item of the injection and production well at the current time step for: (6) In the formula, is the sequence number of the streamline, For the the number of streamlines within the group; is the streamline water breakthrough index, , where is the end flight time of the th streamline in the th group, is the predicted production cycle.

[0012] Preferably, the initial parameters of the learning environment include population size, maximum number of iterations, particle dimension, number of states in Q-Learning, number of actions in Q-Learning, number of training rounds in Q-Learning, step size coefficient, discount factor, Levy random factor, Brownian random factor, and adaptive adjustment factor.

[0013] Preferably, step 3 includes the following steps: Step 3.1, obtain the streamline data file simulated by the streamline numerical simulation well position optimization agent model for the water drive reservoir, perform decoding processing on the streamline data file, and obtain the streamline data; Obtain the streamline data files at each time step during the simulation process of the streamline numerical simulation well position optimization agent model for the water drive reservoir, divide the stored content of the streamline data file into string-type data and numerical-type data, obtain the string-type data through ASCII decoding, directly obtain the numerical-type data through binary decoding according to the IEEE rule, and obtain the streamline attributes at the time step from the streamline data file, including the positions, saturations, and flight times of each streamline in the flow field; Step 3.2, preprocess the streamline data; The preprocessing of the streamline data includes deleting the characteristic indication coordinates in the streamline data and averaging the streamline attribute data of the repeated streamline points; Step 3.3, obtain other attributes of the streamline field; Calculate the derivative data according to the streamline data, obtain the grid pressure and grid number of the grid passed by the streamline, establish the pressure mapping relationship to obtain the pressure at the current streamline point, and then use the interpolation algorithm to determine the pressure at any position in the streamline field; determine the injection-production well pair number of the streamline according to the starting well number and ending well number of the streamline, and calculate the streamline water breakthrough index according to the flight time of the streamline; According to Darcy's law and the grid pressure, the fluid velocity within any grid interface is determined as: (7) where is the average velocity of the fluid in the grid ; is the grid permeability; is the fluid viscosity; is the grid pressure of the th grid; is the The grid pressure of a grid; is the grid length; Step 3.4: Group the streamlines according to the injection-production well pairs, remove the abnormal boundary streamlines, and construct a streamline dataset.

[0014] Preferably, the said step 4 includes the following steps: Step 4.1: Based on the initialized Q-Learning parameters, set the state space and the action space ; Initialize the Q table as a zero matrix ; Initialize the reward table as: (8) In the formula, is the state space, , where , , correspond to the three search phases of the Marine Predators Algorithm MPA, is the Brownian phase, used for local exploitation, is the Levy-Brownian phase, used for hybrid exploration; is the Levy phase, used for global exploration; is the action space, , is the switching strategy of the Brownian phase, is the switching strategy of the Levy-Brownian phase, is the switching strategy of the Levy phase; Step 4.2: According to the current state , use the greedy strategy to select an action , and after executing the action, switch the search phase and update the individual position: (9) In the formula, is the position of the rd individual at the th time step, is the position of the th individual at the th time step; is the step size coefficient; is a random number factor between (0, 1); is the th individual using the elite solution in the current iteration; is the Brownian random factor; is the adaptive adjustment factor; Step 4.3, determine the immediate reward according to the difference in fitness before and after optimization , and update the Q-table using the Bellman equation: (10) Wherein, (11) In the formula, is the learning decay rate function; is the discount factor; is the immediate reward at the th time step; is the state at the th time step; is the action selected for the corresponding state ; is the time step, is the maximum number of iterations; Step 4.4, expand the search range based on the quasi-oppositional learning mechanism, and construct its quasi-opposite solution for each current individual : : (12) In the formula, is the random number function; Compare the fitness of the individual with its opposite solution , and retain the optimal solution; Step 4.5, judge whether to enter the perturbation mechanism based on the fish swarm aggregation mechanism : (13) In the formula, and are the positions of two groups of random individuals; is the upper boundary position vector of the well position coordinates, is the lower boundary position vector of the well position coordinates; Step 4.6, store the individual with the optimal fitness in the current iteration round into the elite retention area, and use it as the vector for the next generation update, and complete the search iteration of the current iteration round.

[0015] Preferably, the said step 5 includes the following steps: Step 5.1, perform a single simulation on the individuals of the current population based on the reservoir streamline numerical simulator, process the streamline information through the intelligent agent model for well position optimization of the waterflooding reservoir streamline numerical simulation, and calculate the fitness of each individual in the current population in combination with the objective function; Step 5.2: For each individual in the current population, the well location coordinates are checked for legitimacy to ensure that they meet the distance approximation and repeatability filtering requirements. If not, the nearest neighbor feasible coordinate search function is called to make corrections. Step 5.3, compare the current optimal fitness with the previous generation fitness, retain the better individual, update the position coordinates of the best individual, and record the current optimal fitness in the convergence curve.

[0016] The beneficial technical effects brought about by the present invention are: Among the well network and well location optimization methods, the traditional well network and well location optimization method uses cumulative oil production (COP) or net present value (NPV) as the objective function. When using a reservoir numerical simulator to complete the full-time simulation of the expected production cycle, it takes a long time and the objective function has poor convergence, which cannot meet the time cost requirements in the actual optimization process.

[0017] The present invention proposes a well pattern and well location optimization method based on flow field area and balanced production evaluation, which overcomes the shortcomings of traditional well pattern and well location optimization methods. First, by constructing a streamline numerical simulation well location optimization model for water-flooding reservoirs with the goals of flight time balance, streamline coverage maximization, and water inrush penalty minimization, the well location optimization control is closer to the dynamic response of the reservoir, thereby improving the physical consistency and engineering operability of the well pattern well location optimization model. Second, the method of the present invention utilizes streamline simulation and only needs to complete the simulation of the first time step of the expected production cycle to quickly obtain the injection-production connectivity and its changing law, effectively reducing the high computational cost brought by traditional fully implicit numerical simulation in the optimization process and significantly improving the optimization efficiency of the well pattern well location. Third, by introducing the Q-Learning reinforcement learning mechanism, the method of the present invention dynamically learns the advantages and disadvantages of each phase in the search process and guides strategy switching with the Q value, thereby improving the algorithm's adaptability and global search performance in multi-peak optimization problems. At the same time, the method combines the quasi-adversarial learning strategy to expand the search space and avoid local convergence, and integrates the FADs aggregation disturbance mechanism to simulate ecological group intelligence, thereby enhancing population diversity. Finally, by constructing an elite individual retention mechanism, the method achieves continuous tracking and inheritance of high-quality solutions, making the optimization process more robust and reliable.

[0018] It can be seen that the method of the present invention significantly reduces the optimization cycle of the well location layout plan, and the optimized well location distribution plan is better, providing a basis for improving the comprehensive economic benefits of oil field development. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The present invention is a flow chart of a well pattern and well location optimization method based on flow field area and balanced production evaluation.

[0020] Figure 2 This is the permeability field map of the reservoir model in the embodiment of the present invention.

[0021] Figure 3 It is the porosity field map of the reservoir model in the embodiment of the present invention.

[0022] Figure 4 It is the optimization convergence graph with the cumulative oil production COP as the objective function in the embodiment of the present invention.

[0023] Figure 5 It is the optimization objective function for well location optimization using streamline numerical simulation in the embodiment of the present invention of the optimization convergence graph.

[0024] Figure 6 It is the comparison graph of the method of the present invention and the traditional well pattern and well location optimization method in this embodiment.

[0025] Figure 7 It is the optimized well location coordinate graph with the cumulative oil production COP as the objective function in the embodiment of the present invention.

[0026] Figure 8 It is the oil saturation graph with the cumulative oil production COP as the objective function in the embodiment of the present invention.

[0027] Figure 9 It is the streamline distribution graph with the cumulative oil production COP as the objective function in the embodiment of the present invention.

[0028] Fig.10 It is the optimization objective function for well location optimization using streamline numerical simulation in the embodiment of the present invention as the objective function of the optimized well location coordinate graph.

[0029] Fig.11 It is the optimization objective function for well location optimization using streamline numerical simulation in the embodiment of the present invention as the objective function of the oil saturation graph.

[0030] Figure 12 It is the optimization objective function for well location optimization using streamline numerical simulation in the embodiment of the present invention as the objective function of the oil saturation graph.

[0031] In the figure, P1, P2, P3, P4, P5, P6, P7, P8, and P9 are all production wells, and INJ1, INJ2, INJ3, and INJ4 are injection wells. Specific implementation manner

[0032] The present invention will be further described in detail below in conjunction with the drawings and embodiments.

[0033] Embodiment 1 In this embodiment, a well pattern and well location optimization method based on the flow field region and balanced production evaluation is proposed, as Figure 1 shown, which specifically includes the following sub-steps: Step 1: Build a well location optimization model for numerical simulation of waterflooding reservoir streamlines, clarify the production system, determine the optimization variables, constraint variables, and objective function, and construct the initial population based on Latin hypercube sampling, which includes the following sub-steps:

[0034] Step 1.1: Construct a well location optimization model for numerical simulation of water drive reservoir streamlines.

[0035] The water drive reservoir streamline numerical simulation well location optimization model is: (1) The constraints are: ; ; ; ; In the formula, is the well location coordinate set; Optimize the objective function for well location in streamline numerical simulation; is the minimum function; is the streamline displacement potential coefficient; is the coefficient of variation of streamline flight time; is the penalty term between injection-production well pairs; It is the internal penalty item for injection and production wells; and are the upper and lower bounds of the boundary constraints of the well location coordinate set respectively; is the well spacing constraint; , All are numbered; , All are single well location coordinates; is the number of injection wells, is the number of producing wells.

[0036] Step 1.2: The two-dimensional coordinates of the injection well and production well are used as optimization variables; the constraint variables are the minimum well spacing constraint and the well location non-repetition constraint; the objective function is the streamline numerical simulation well location optimization objective function The goal is to maximize the cumulative oil production under the constraints of well spacing and well non-repetition, ensure the balance of flight time between each injection and production well pair, maximize the streamline coverage, and minimize the occurrence of water inrush to maximize the cumulative oil production.

[0037] Step 1.3: Construct the initial population based on Latin hypercube sampling.

[0038] Latin hypercube sampling is used to sample from a multivariate parameter distribution, and the number of injection well location plans to be generated is Determine the number of injection well location plans that need to be generated , according to the number of planned injection well positions to be generated determine the number of samples to be extracted, and divide the interval (0, 1) into segments, randomly extract in each segment, map the extracted samples to standard normal distribution samples through the inverse function of the standard normal distribution and shuffle them to obtain the initial population.

[0039] Step 2, based on the Q-Learning enhanced Marine Predators Algorithm (QQLMPA), construct an intelligent agent model for optimizing the well positions in streamline numerical simulation of waterflooding reservoirs. The Q-Learning enhanced Marine Predators Algorithm aims to improve the optimization performance of the traditional Marine Predators Algorithm (MPA). The introduction of Q-Learning enables MPA to utilize the information generated in historical iterations, adaptively select the optimal strategy (Lévy movement, Brownian movement, or a combination of both), balance exploration and exploitation, and accelerate convergence. The introduction of Quasi-Oppositional Learning (QOL) can generate quasi-oppositional solutions of the current solution, expand the search space, increase population diversity, and thus reduce the risk of falling into local optima. QQLMPA significantly improves the convergence speed and global search ability of MPA through the dual improvements of quantum behavior to improve QL and quasi-oppositional learning QOL, specifically including the following sub-steps:

[0040] Step 2.1, according to the particle coordinate values in the initial population, determine the well position coordinates for the expected production period to obtain a set of well position coordinates : (2).

[0041] Step 2.2, based on the characteristics of the streamline numerical simulation method, the attribute information of the streamline can be obtained after the numerical simulation in the first time step. Therefore, set the simulation time step according to the time step of the expected production period and conduct streamline numerical simulation.

[0042] Step 2.3, according to the ratio of the number of grids passed by the streamline to the total number of grids in the streamline numerical simulation well position optimization model of the waterflooding reservoir, determine the streamline displacement potential coefficient at the current time step as: (3) In the formula, is the total number of grids in the streamline numerical simulation well position optimization model of the waterflooding reservoir passed by the streamline, which is determined by extracting and removing duplicates and counting the grid numbers passed by all streamlines; is the total number of grids in the streamline numerical simulation well position optimization model of the waterflooding reservoir, which is determined according to the model parameters of the streamline numerical simulation well position optimization model of the waterflooding reservoir.

[0043] Step 2.4, according to the streamline flight time variation coefficient at the current time step Set the balance degree of each injection and production well on the average flight time; the coefficient of variation of the streamline flight time at the current time step for: (4) In the formula, is the sample standard deviation of the flight time of the injection and production wells; is the average flight time of injection and production well pairs; is the number of injection-production well pairs; For the The flight time of an injection-production well pair is used to represent the average flight time of all streamlines contained in the injection-production well pair.

[0044] Step 2.5: Penalty term between injection and production wells at the current time step Set the positive difference between the average flight time of each injection-production well pair and the expected production cycle; the penalty term between the injection-production well pairs at the current time step for: (5) In the formula, is the number of injection-production well pairs whose flight time is less than the expected production cycle, ; is the penalty intensity power series, .

[0045] Step 2.6, based on the internal penalty item of the injection and production well at the current time step Set the average value of the positive difference between the flight time of the streamlines contained in each injection-production well pair and the expected production cycle; the penalty item within the injection-production well pair at the current time step is for: (6) In the formula, is the sequence number of the streamline, For the the number of streamlines within the group; is the streamline water inrush index, ,in, For the No. The flight time at the end of the root streamline, Estimated production cycle.

[0046] Step 2.7, determining the initial parameters of the learning environment of the intelligent agent model for well location optimization in numerical simulation of waterflooding reservoir streamlines; The initial parameters of the learning environment include the population size , maximum number of iterations , particle dimensions , the number of states of Q-Learning 、The number of actions in Q-Learning 、The number of training rounds in Q-Learning 、Step size coefficient 、Discount factor 、Levy random factor 、Brownian random factor 、Adaptive adjustment factor 。

[0047] Step 3: Use the streamline numerical simulation well position optimization agent model for water drive reservoirs to perform simulations, obtain and process the streamline data and derivative data obtained from the simulation of the streamline numerical simulation well position optimization agent model for water drive reservoirs, and establish a streamline dataset, including the following sub-steps: Step 3.1: Obtain the streamline data file obtained from the simulation of the streamline numerical simulation well position optimization agent model for water drive reservoirs, perform decoding processing on the streamline data file, and obtain the streamline data; During each simulation process, obtain the streamline data file at each time step of the streamline numerical simulation well position optimization agent model for water drive reservoirs, divide the stored content of the streamline data file into string type and numerical type. Among them, the string type data is obtained through ASCII decoding, and the numerical type data is directly decoded from binary according to the specific numerical type by the IEEE rule. Obtain the streamline attributes at the time step from the streamline data file, including the positions, saturations, flight times, and other relevant streamline attributes of each streamline in the flow field.

[0048] Step 3.2: Preprocess the streamline data; The preprocessing of the streamline data includes deleting the feature indication coordinates in the streamline data and averaging the streamline attribute data of duplicate streamline points. Align the relevant data dimensions through the feature indication coordinates in the streamline data. Averaging the streamline attribute data of duplicate streamline points can effectively avoid repeated calculations and delete redundant duplicate streamline points, thus ensuring that each streamline point data is unique.

[0049] Step 3.3: Obtain other attributes of the streamline field; Calculate the derivative data according to the streamline data, obtain the grid pressure and grid number of the grid through which the streamline passes, establish a pressure mapping relationship, obtain the pressure of the current streamline point, and then use the interpolation algorithm to determine the pressure at any position in the streamline field; Determine the injection-production well pair number of the streamline according to the starting well number and ending well number of the streamline, and calculate the streamline water breakthrough index according to the flight time of the streamline.

[0050] According to Darcy's law and the grid pressure, determine the fluid velocity within any grid interface as: (7) In the formula, For Grid The average velocity of the internal fluid; is the grid permeability, in units of ; is the fluid viscosity, in units of ; For the The grid pressure of the grid, in units of ; For the The grid pressure of the grid, in units of ; is the grid length in units of .

[0051] In step 3.4, all streamlines are divided into groups corresponding to their injection-production well pair numbers and abnormal boundary streamlines are removed to ensure that the streamline data is not contaminated, and a streamline dataset is constructed.

[0052] Step 4: Combine historical data from the elite data reserve, update the Q-table through Q-Learning to optimize the phase selection strategy, and use quasi-adversarial learning (QOL) to generate candidate solutions, expand the search scope, and use the updated strategy to guide MPA to perform next-generation well location optimization. This includes the following sub-steps: Step 4.1: Set the state space based on the initialized Q-Learning parameters and action space ; Initialize the Q table to a zero matrix ; Initialize reward table for: (8) In the formula, is the state space, ,in, 、 、 Corresponding to the three search phases of the marine predator algorithm MPA, is the Brownian phase, used for local development, is the Levy-Brownian phase, used for mixed exploration; is the Levy phase, used for global exploration; is the action space, , is the switching strategy of the Brownian phase, is the switching strategy of the Levy-Brownian phase, is the switching strategy of the Levy phase.

[0053] Step 4.2, based on the current status , using a greedy strategy to select actions , after executing the action, switch the search phase and update the individual position: (9) In the formula, For the Time step The location of an individual, For the Time step the location of each individual; is the step length coefficient; is a random number factor between (0,1); For the Individual The elite solution used in the current iteration; is the Brownian random factor; is the adaptive adjustment factor.

[0054] Step 4.3: Determine the immediate reward based on the fitness difference before and after optimization , and update the Q table using the Bellman equation: (10) in, (11) In the formula, is the learning decay rate function; is the discount factor; For the The immediate reward of time steps; For the The state of the time step; For the corresponding status the action chosen; is the time step, is the maximum number of iterations.

[0055] Step 4.4, based on the quasi-adversarial learning mechanism, expand the search range and Construct its quasi-opposite solution : (12) In the formula, is a random number function.

[0056] Comparison individuals Its opposite solution fitness, and retain the optimal solution.

[0057] Step 4.5, based on the fish aggregating mechanism (Fish Aggregating Devices), determine whether to enter the perturbation mechanism : (13) In the formula, and are the positions of two groups of random individuals; is the upper boundary position vector of the well location coordinates, is the lower boundary position vector of the well location coordinates.

[0058] Step 4.6, store the individual with the optimal fitness in the current iteration round into the elite reservation area and use it as the updated vector for the next generation to complete the search iteration of the current iteration round.

[0059] Step 5, based on the reservoir streamline numerical simulator, conduct continuous interaction between the waterflood reservoir streamline numerical simulation well location optimization agent model and the reservoir numerical simulation, evaluate the fitness of the new population, screen the elite samples and store them in the elite data reservation area, including the following sub-steps: Step 5.1, conduct a single simulation on the individuals of the current population based on the reservoir streamline numerical simulator, process the streamline information through the waterflood reservoir streamline numerical simulation well location optimization agent model and calculate the fitness of each individual in the current population in combination with the objective function.

[0060] Step 5.2, for each individual in the current population, judge the legality of the well location coordinates to ensure compliance with distance approximation and repeatability filtering. If not satisfied, call the nearest neighbor feasible coordinate search function for correction.

[0061] Step 5.3, compare the current optimal fitness with the fitness of the previous generation, retain the better individuals, update the position coordinates of the top predator, that is, update the position coordinates of the optimal individual, and record the current optimal fitness in the convergence curve.

[0062] Step 6, repeat Steps 4 to 5 until the preset maximum number of iterations is reached to determine the optimal well location arrangement.

[0063] Example 2 Taking the reservoir in a certain research area as an example, using the well pattern and well location optimization method based on the flow field region and balanced utilization evaluation described in Example 1, conduct waterflood reservoir well location optimization for this research area. In this example, the reservoir model uses a two-dimensional reservoir model, and the grid system of the reservoir model is 60 220 1. The grid size is 15m 5m 5m, and the permeability field of the reservoir model is as Figure 2As shown, the porosity field of the reservoir model is as Figure 3 shown; the well location coordinates that need to be optimized in this reservoir include 9 production wells and 4 injection wells. Among them, the production wells are numbered P1 - P9, and the injection wells are numbered INJ1 - INJ4. The production time of the reservoir is 1000 days in total, divided into 20 time steps, with each time step being 50 days. However, when using the well pattern and well location optimization method based on flow field region and balanced production evaluation proposed in Example 1 for water - drive reservoir well location optimization, only one time step needs to be selected. The well location spacing constraint is based on the grid distance, and the distance between adjacent well locations is not less than 10m.

[0064] Using the well pattern and well location optimization method based on flow field region and balanced production evaluation proposed in Example 1 to optimize the well pattern and well location of this reservoir, the following steps are included: Step 1, construct a streamline numerical simulation well location optimization model for the water - drive reservoir, clarify the production regime, determine the optimization variables, constraint variables, and objective function, and construct an initial population based on Latin hypercube sampling.

[0065] Step 2, construct an intelligent agent model for streamline numerical simulation well location optimization of the water - drive reservoir based on the Q - Learning enhanced marine predator algorithm.

[0066] Step 3, use the intelligent agent model for streamline numerical simulation well location optimization of the water - drive reservoir to conduct simulations, obtain and process the streamline data and derivative data obtained from the simulation of the intelligent agent model for streamline numerical simulation well location optimization of the water - drive reservoir, and establish a streamline dataset grouped by injection - production well pairs.

[0067] Step 4, combine the historical data in the elite data retention area, update the Q - table to optimize the phase selection strategy through Q - Learning, generate candidate solutions using QOL quasi - opposition learning to expand the search range, and use the updated strategy to guide MPA for the next - generation well location optimization.

[0068] Step 5, based on the reservoir streamline numerical simulator, conduct continuous interaction between the intelligent agent model for streamline numerical simulation well location optimization of the water - drive reservoir and the reservoir numerical simulation, evaluate the fitness of the new population, and screen out elite samples and store them in the elite data retention area.

[0069] Step 6, repeat Steps 4 - 5 until the preset maximum number of iterations is reached, and determine the optimal well location arrangement.

[0070] To verify the superiority of the streamline numerical simulation well location optimization objective function in the method of the present invention compared with the traditional well pattern and well location optimization method, use the same well location optimization model as the present invention with the cumulative oil production COP as the objective function to optimize the well pattern and well location, and obtain the convergence situation within a total of 20 time steps in the entire predicted production cycle, as Figure 4 shown. Figure 5It is an optimized convergence graph of the method of the present invention. The cumulative oil production and optimization duration of the well positions obtained by optimizing with the cumulative oil production COP as the objective function are compared with those of the well positions obtained by optimizing with the well position optimization objective function of streamline numerical simulation proposed by the present invention as the objective function. As Figure 6 shown, it is found that the method of the present invention is better than the case of using the cumulative oil production COP as the objective function in terms of time consumption and oil production;

[0071] Then, the optimal well position maps, oil saturation maps and streamline distribution maps obtained by optimizing the well positions with the cumulative oil production COP as the objective function and the well position optimization objective function of streamline numerical simulation as the objective function are compared. As Figures 7 to 12 shown, among them, Figure 7 is the optimized well position coordinate map with the cumulative oil production COP as the objective function in the embodiment of the present invention, Figure 8 is the oil saturation map with the cumulative oil production COP as the objective function in the embodiment of the present invention, Figure 9 is the streamline distribution map with the cumulative oil production COP as the objective function in the embodiment of the present invention, Fig.10 is the optimized well position coordinate map with the well position optimization objective function of streamline numerical simulation as the objective function in the embodiment of the present invention, Fig.11 is the oil saturation map with the well position optimization objective function of streamline numerical simulation as the objective function in the embodiment of the present invention, Figure 12 is the oil saturation map with the well position optimization objective function of streamline numerical simulation as the objective function in the embodiment of the present invention.

[0072] Thus, it is further verified that the method of the present invention can better formulate the well position distribution scheme compared with the traditional well position optimization method with the cumulative oil production COP as the objective function, significantly reducing the evaluation period of the well position layout scheme, and the effect of the well position distribution scheme is better.

[0073] In summary, the method of the present invention can obtain a better well position distribution scheme in a shorter optimization time, significantly reducing the optimization period of the well position layout scheme. The optimization process is more stable and reliable, which is beneficial to improving the comprehensive economic benefits of oilfield development.

[0074] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the essence of the present invention should also fall within the protection scope of the present invention.

Claims

1. A well pattern and well location optimization method based on flow field region and balanced production evaluation, characterized in that, Including the following steps: Step 1: Construct a streamline numerical simulation well position optimization model for waterflooding reservoirs, clarify the production regime, determine the optimization variables, constraint variables, and objective function, and construct an initial population based on Latin hypercube sampling; Step 2: Based on the marine predator algorithm enhanced by Q-Learning, construct an intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs; Step 3: Use the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs to perform simulations, obtain and process the streamline data and derivative data obtained from the simulation of the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs, and establish a streamline dataset; Step 4: Combine the historical data in the elite data retention area, update the Q-table to optimize the phase selection strategy through Q-Learning, generate candidate solutions using QOL quasi-oppositional learning to expand the search range, and use the updated strategy to guide MPA for the next generation of well position optimization; Step 5: Based on the reservoir streamline numerical simulator, conduct continuous interaction between the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs and reservoir numerical simulation, evaluate the fitness of the new population, and screen out elite samples and store them in the elite data retention area; Step 6: Repeat Steps 4 to 5 until the preset maximum number of iterations is reached, and determine the optimal well position arrangement.

2. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 1, wherein The said Step 1 includes the following steps: Step 1.1: Construct a streamline numerical simulation well position optimization model for waterflooding reservoirs; Step 1.2, taking the two-dimensional coordinates of the injection well and the production well as the optimization variables; the constraint variables are the minimum well spacing constraint and the non-repetitive well location constraint; the objective function is the streamline numerical simulation well location optimization objective function , with the goal of maximizing the cumulative oil production under the conditions of meeting the well spacing constraint and the non-repetitive well location constraint; Step 1.3: Construct an initial population based on Latin hypercube sampling; Sampling is carried out from a multivariate parameter distribution using Latin hypercube sampling, and according to the number of planned injection well locations to be generated the number of samples to be drawn is determined, and the interval (0, 1) is evenly divided into segments, randomly drawn in each segment, and the drawn samples are mapped to standard normal distribution samples through the inverse function of the standard normal distribution and randomly sorted to obtain the initial population.

3. The well pattern and well location optimization method based on flow field region and balanced utilization evaluation according to claim 2, characterized in that The streamline numerical simulation well position optimization model for waterflooding reservoirs is as follows: （1） The constraint conditions are: ； ； ； ； Wherein, is the well location coordinate set; is the streamline numerical simulation well location optimization objective function; is the minimum value function; is the streamline displacement potential coefficient; is the streamline flight time variation coefficient; is the penalty term between injection-production well pairs; is the penalty term within injection-production well pairs; and are respectively the upper and lower bounds of the boundary constraints of the well location coordinate set; is the well spacing constraint; 、 are both numbers; 、 are both single well location coordinates; is the number of injection wells, is the number of production wells.

4. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 3, characterized in that The said Step 2 includes the following steps: Step 2.1: Determine the well location coordinates for the expected production period based on the particle coordinate values in the initial population to obtain a set of well location coordinates : （2）； Step 2.2: Set the simulation time step according to the time step of the expected production cycle and conduct streamline numerical simulation; Step 2.3: Determine the streamline displacement potential coefficient at the current time step according to the proportion of the number of grids passed by the streamline to the number of grids in the streamline numerical simulation well position optimization model of the waterflood reservoir. ; Step 2.4, according to the coefficient of variation of the streamline flight time at the current time step Set the balance degree of the average flight time for each injection-production well pair; Step 2.5, according to the injection-production well pair penalty term at the current time step Set the positive difference degree between the average flight time of each injection-production well pair and the expected production cycle; Step 2.6, according to the injection-production wells at the current time step, set the internal penalty term for each injection-production well, set the average value of the positive difference degree between the flight time of the streamline contained therein and the predicted production cycle; Step 2.7: Determine the initial parameters of the learning environment of the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs.

5. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 4, wherein The streamline displacement potential coefficient at the current time step is as follows: （3） In the formula, is the total number of grids of the streamline numerical simulation well position optimization model for the water drive reservoir through which the streamline passes; is the total number of grids of the streamline numerical simulation well position optimization model for the water drive reservoir; The coefficient of variation of streamline flight time at the current time step is as follows: （4） Wherein, is the sample standard deviation of the flight time of the injection-production well pair; is the average value of the flight time of the injection-production well pair; is the number of injection-production well pairs; is the flight time of the The penalty term between the injection-production well pairs at the current time step is as follows: （5） Wherein, is the number of injection-production well pairs with flight time less than the expected production cycle, ; is the penalty intensity power series, ; The internal penalty term of the injection-production well at the current time step is as follows: （6） In the formula, is the serial number of the streamline, is the number of streamlines in the th group; is the streamline water breakthrough index, , where is the end flight time of the th streamline in the th group, is the predicted production cycle.

6. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 4, characterized in that The said initial parameters of the learning environment include population size, maximum number of iterations, particle dimension, number of states of Q-Learning, number of actions of Q-Learning, number of training rounds of Q-Learning, step size coefficient, discount factor, Levy random factor, Brownian random factor, and adaptive adjustment factor.

7. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 5, wherein The said Step 3 includes the following steps: Step 3.1: Obtain the streamline data file obtained from the simulation of the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs, decode the streamline data file to obtain streamline data; Obtain the streamline data files at each time step during the simulation of the intelligent agent model for streamline numerical simulation well position optimization in waterflooding reservoirs, divide the stored content of the streamline data file into string-type data and numerical-type data, obtain the string-type data through ASCII decoding, directly obtain the numerical-type data according to the IEEE rule through binary decoding, and obtain the streamline attributes at the time step from the streamline data file, including the positions, saturations, and flight times of each streamline in the flow field; Step 3.2: Preprocess the streamline data; The said preprocessing of the streamline data includes deleting the characteristic indication coordinates in the streamline data and averaging the streamline attribute data of duplicate streamline points; Step 3.3: Obtain other attributes of the streamline field Calculate derived data based on streamline data, obtain the grid pressure and grid number of the grid through which the streamline passes, establish a pressure mapping relationship, obtain the pressure at the current streamline point, and then use the interpolation algorithm to determine the pressure at any position in the streamline field; determine the streamline injection and production well pair number based on the streamline starting and ending well numbers, and calculate the streamline water inrush index based on the streamline flight time; According to Darcy's law and grid pressure, the fluid flow velocity within any grid interface is determined as: （7） In the formula, is the average velocity of the fluid in the grid; is the grid permeability; is the fluid viscosity; is the grid pressure of the th grid; is the grid pressure of the th grid; is the grid length; is the grid length; In step 3.4, streamlines are grouped according to the injection-production well pair numbers, abnormal boundary streamlines are eliminated, and a streamline dataset is constructed.

8. The well pattern and well location optimization method based on flow field region and balanced production evaluation according to claim 7, wherein The step 4 comprises the following steps: Step 4.1, based on the initialized Q-Learning parameters, set the state space and the action space ; Initialize the Q-table as a zero matrix ; Initialize the reward table as follows: （8） In the formula, is the state space, , where , , correspond to three search phases of the Marine Predators Algorithm (MPA), is the Brownian phase for local exploitation, is the Levy-Brownian phase for hybrid exploration; is the Levy phase for global exploration; is the action space, , is the switching strategy of the Brownian phase, is the switching strategy of the Levy-Brownian phase, is the switching strategy of the Levy phase; Step 4.2, according to the current state , select an action using the greedy strategy , switch the search phase after executing the action, and update the individual position: （9） In the formula, is the position of the -th individual at the -th time step, is the position of the -th individual at the -th time step; is the step size coefficient; is a random number factor between (0, 1); is the elite solution used by the -th individual in the current iteration; is the Brownian random factor; is the adaptive adjustment factor; Step 4.3, determine the immediate reward based on the fitness difference before and after optimization , and update the Q-table using the Bellman equation: （10） in, （11） wherein, is the learning decay rate function; is the discount factor; is the immediate reward at the -th time step; is the state at the -th time step; is the action selected for the corresponding state ; is the time step, is the maximum number of iterations; Step 4.4, expand the search range based on the quasi-opposition learning mechanism, and construct the quasi-opposite solution for each current individual to construct its quasi-opposite solution : （12） In the formula, is a random number function; Comparative individual and its opposing solution for fitness, and retain the optimal solution; Step 4.5, determine whether to enter the perturbation mechanism based on the fish school aggregation mechanism : （13） In the formula, and are two sets of random individual positions; is the upper boundary position vector of the well position coordinates, is the lower boundary position vector of the well position coordinates; Step 4.6, store the individual with the optimal fitness in the current iteration round into the elite reservation area and use it as the vector for the update of the next generation, completing the search iteration of the current iteration round. ​ 9. The well pattern and well location optimization method based on flow field region and balanced utilization evaluation according to claim 8, wherein The step 5 comprises the following steps: Step 5.1, a single simulation of the individuals in the current population is performed based on the reservoir streamline numerical simulator. The water flooding reservoir streamline numerical simulation well location optimization intelligent agent model processes the streamline information and calculates the fitness of each individual in the current population in combination with the objective function. Step 5.2: For each individual in the current population, the well location coordinates are checked for legitimacy to ensure that they meet the distance approximation and repeatability filtering requirements. If not, the nearest neighbor feasible coordinate search function is called to make corrections. Step 5.3, compare the current optimal fitness with the previous generation fitness, retain the better individual, update the position coordinates of the best individual, and record the current optimal fitness in the convergence curve.

Images