Intelligent well location configuration method based on multi-objective particle swarm optimization
By using a multi-objective particle swarm optimization algorithm to optimize well location configuration, the problem of low computational efficiency of traditional particle swarm optimization algorithms is solved, achieving efficient well location optimization, which is particularly suitable for oil and gas field development.
Patent Information
- Application Number
- CN202511473595.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-15
- Publication Date
- 2026-01-20
AI Technical Summary
Existing particle swarm optimization algorithms have low computational efficiency in well location optimization and are difficult to achieve real-time optimization, especially in the development of complex oil and gas fields where a large number of computational steps are required, and traditional methods have long computation times.
A multi-objective particle swarm optimization algorithm is adopted. By determining the key parameters for well location optimization, the particle velocity and position are updated using the particle swarm optimization algorithm. Parallel computation and assignment are performed. Combined with multi-objective function optimization, a repair mechanism and dynamic weight adjustment are introduced to optimize the well location configuration.
It improves the computational efficiency and concurrent computing performance of well location optimization, enabling efficient well location configuration in a short time, adapting to complex oil and gas field optimization problems, significantly shortening computation time and improving economic benefits.
Smart Images
Figure CN121365585A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of well site configuration, and particularly relates to an intelligent well site configuration method based on multi-objective particle swarm optimization. BACKGROUND
[0002] Particle swarm optimization (PSO) finds the optimal solution by simulating the cooperation and information exchange between individuals, and particle swarm optimization has been widely applied to multiple optimization problems, including well site optimization, equipment scheduling, etc. However, the existing particle swarm algorithm still has some deficiencies in the application of well site optimization, especially in terms of computational efficiency. For example, in the traditional particle swarm algorithm, the next round of iteration cannot be continued until the slower particles are calculated, which leads to long calculation time and difficulty in efficient parallel execution. Existing group evolution optimization algorithms also face similar problems, especially when dealing with complex oil and gas field development optimization problems, which require a large number of calculation steps.
[0003] In the well site optimization problem, due to the combination of numerous well sites and parameters, the calculation time of the traditional optimization method is usually long, making it difficult to achieve real-time optimization. SUMMARY
[0004] The purpose of the present application is to overcome the above technical problems and provide an intelligent well site configuration method based on multi-objective particle swarm optimization.
[0005] To achieve the above purpose, the present application provides the following technical solutions: An intelligent well site configuration method based on multi-objective particle swarm optimization, comprising the following steps: S1. Determine the key parameters of well site optimization, including the well sites to be optimized, the parameter range of each well site, and the target function, the target function including the net present value, environmental impact and resource utilization rate corresponding to the target function; select a certain number of particles, each particle representing a potential well site configuration and its corresponding optimization parameters, and assign an initial position and velocity to each particle, and initialize the initial value of the optimization parameter; S2. Calculate and analyze the matching degree value of the particles in step S1, calculate the matching degree value of each particle according to the multi-objective function, and analyze the matching degree of all particles as a whole; S3. Update the position and velocity of each particle according to the analysis result of step S2, after calculating the matching degree value of each particle, update the velocity and position of the particle using the particle swarm optimization algorithm according to the velocity update formula; S4. Parallel computing and assignment, in the optimization process, the particle that completes the particle calculation result is calculated first, and the particle is directly assigned a new value and recalculated its corresponding well location scheme; for other particles, after the calculation is completed, the parameters of the particle are assigned to the particle that completes the calculation first, and the corresponding scheme is recalculated; the well spacing of the updated particle is detected, and if the well spacing is less than the safety threshold, the repair mechanism is started, and the repair mechanism is: the coordinates of the conflict well location are translated to the high recovery sub-area; if it cannot be translated, the particle is eliminated and reinitialized; S5. Repeat step S4 until the set maximum number of iterations is reached or the predetermined convergence condition is met, output the solution of the optimal particle, and call the drilling cost model to verify the economic feasibility before outputting the final scheme.
[0006] Optionally, in step S1, the net present value maximization objective function is: The environmental impact minimization objective function is: The resource utilization maximization objective function is: Wherein, X is the well location configuration decision variable vector, which includes well location coordinates, well depth and well type; n is the number of wells; T is the production time, years; is the crude oil production of the ith well in the t year, barrels / year; is the natural gas production of the ith well in the t year, cubic meters / year; is the water production of the ith well in the t year, barrels / year; is the crude oil price in the t year, US dollars / barrel; is the natural gas price in the t year, US dollars / cubic meter; is the water treatment cost in the t year, US dollars / barrel; r: is the discount rate; is the drilling cost of the ith well, US dollars; is the completion cost of the ith well, US dollars; are environmental impact weight coefficients; is the land area of the ith well, square meters; is the carbon emission of the ith well, tons; is the waste production of the ith well, tons; GOR is the gas-oil ratio; OOIP is the original geological reserves, barrels.
[0007] Optionally, in step S2, the computer first calculates the net present value, carbon emission and recovery ratio of the well location configuration according to the position of each particle, wherein the carbon emission is calculated by the formula: Wherein, Total carbon emission, tons; Carbon emission of the i th well drilling stage, tons; Carbon emission of the i th well per unit time, tons / year; recovery factor is calculated by the following formula: Wherein, RF: recovery factor, %; Then, the three objective function values are standardized, and the net present value standardized value is: The environmental impact standardized value is: The resource utilization rate standardized value is: Finally, the comprehensive matching degree formula is: Wherein, , The minimum and maximum values of the net present value are represented by min (NPV) and max (NPV) respectively; , The minimum and maximum values of the environmental impact are represented by min (EI) and max (EI) respectively; , The minimum and maximum values of the resource utilization rate are represented by min (RU) and max (RU) respectively; F (X) is the comprehensive matching degree value, and the value range is [0, 1]; w1 is the net present value weight, w2 is the environmental impact weight, and w3 is the resource utilization rate weight; The matching degree value of each particle is calculated and recorded; all particles are sorted according to the matching degree value, the optimal particle is selected for updating, the top 20% particles are selected as elite particles for guiding the evolution direction of other particles; the computer adopts multi-core parallel computing mode.
[0008] Optionally, w1 is 0.6, w2 is 0.25, and w3 is 0.15.
[0009] Optionally, in step S3, global optimization is performed according to the matching degree values of all particles, and particles with higher matching degree are preferentially selected for value assignment and updating.
[0010] Optionally, in well site optimization, the matching degree of the particle is weighted and evaluated according to the dynamic response of the geological model, and the comprehensive matching degree calculation formula is dynamically adjusted: Wherein, , , The net present value weight, the environmental impact weight and the resource utilization rate weight at the t th iteration are represented by w1, w2 and w3 respectively; The net present value weight The following formula is used to dynamically adjust the oil and gas price fluctuation model: in, The baseline weight is 0.6; The price of oil and gas at time t is relative to the benchmark price. The rate of change; Environmental impact weight Adjustments should be made according to the following formula, taking into account the ecological sensitivity classification of the drilling area: in, The baseline weight is 0.25; Ecological sensitivity coefficient, water source protection area Take 2, general area Take 1; Resource utilization rate weight The reservoir recovery rate prediction is optimized according to the following formula: in, With a baseline weight of 0.15, To predict the recovery rate, The average recovery rate of the block; the adjusted weights must meet the normalization condition. , Normalization is performed using the following formula: During each update, particle parameters whose well spacing exceeds the safety threshold are adjusted first to ensure that the scheme complies with geological engineering constraints.
[0011] Optionally, the convergence condition in step S5 is that the maximum or minimum value of the matching degree no longer changes, or the preset number of iterations is reached.
[0012] In summary, the technical effects and advantages of this invention are as follows: By improving the particle swarm optimization algorithm and considering the characteristics of well location optimization problems, this invention focuses on multiple optimization objectives such as net present value (NPV), environmental impact, and resource utilization. It maximizes economic benefits by optimizing well location configuration and improves computational efficiency and concurrent computing performance. This method can adapt to complex well location optimization problems. Through the improved particle swarm optimization algorithm, it can complete efficient well location configuration optimization in a short time, demonstrating high application prospects and being particularly suitable for well location optimization in oil and gas fields. Attached Figure Description
[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only represent some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0014] Figure 1 A flowchart of an intelligent well location configuration method based on multi-objective particle swarm optimization in an embodiment of the present application; Figure 2 A schematic diagram of a mechanism model in an embodiment of the present application; Figure 3 An optimization effect schematic diagram based on a multi-objective particle swarm optimization algorithm in an embodiment of the present application. DETAILED DESCRIPTION
[0015] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments only represent some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort fall within the scope of the present application.
[0016] In addition, the technical features involved in different embodiments of the present application described below can be combined with each other as long as there is no conflict.
[0017] The embodiment provides an intelligent well location configuration method based on multi-objective particle swarm optimization, and a specific algorithm is shown in Figure 1 The specific steps include the following steps: S1: Determine the basic parameters of well location optimization. According to the actual demand, first determine the key parameters in the well location optimization problem, including the well location to be optimized, the parameter range of each well location, the objective function, etc. The objective function includes the net present value (NPV), the objective function corresponding to the environmental impact and the resource utilization rate. A certain number of particles are selected, each particle represents a potential well location configuration and the corresponding optimization parameters. Assign an initial position and velocity to each particle, and initialize the initial value of the optimization parameter.
[0018] S2: Calculation and analysis of particle matching degree value. A computer program is used to calculate the matching degree of each particle according to the multi-objective function, and the matching degree value comprehensively reflects the pros and cons of the well location configuration scheme corresponding to each particle. The matching degree value is obtained by comprehensively considering the NPV, environmental impact and resource utilization rate after simulating the well location configuration. Then, the matching degree analysis is carried out, the overall evaluation of the matching degree value of all particles is carried out, and the position information and speed of the particles are updated according to the evaluation result. Among them, the overall evaluation adopts the weighted standardized multi-objective decision method, and the evaluation principle is: first, the standardization of each objective function is carried out, and the net present value objective function is standardized according to: positive standardization; the environmental impact objective function is standardized according to: negative standardization; the resource utilization rate objective function is standardized according to: positive standardization; make the value of each objective function normalized to the interval [0, 1]; then calculate the comprehensive matching degree F(X) according to the preset weight coefficient In this embodiment, w1 takes 0.6 as the weight of net present value, w2 takes 0.25 as the weight of environmental impact, and w3 takes 0.15 as the weight of resource utilization rate; then the non-dominated solution set is identified through Pareto optimization (Pareto) front analysis, and the solution that achieves the best balance between multiple objectives is found out; finally, all particles are arranged in descending order according to the comprehensive matching degree F(X), and the top 20% of particles are selected as elite particles to guide the evolution direction of other particles, so as to ensure that the well location configuration scheme achieves comprehensive optimization in economic benefit, environmental protection and resource utilization efficiency. Through the multi-objective optimization mechanism, the overall solution quality of the particle group can be effectively improved each time the update is performed. The matching degree value calculation calls the solver of the well location configuration optimization problem.
[0019] Net present value (NPV) maximization objective function: Environmental impact minimization objective function: Resource utilization maximization objective function: Symbol explanation: X: well location decision variable vector (including well location coordinates, well depth, well type, etc.), n: well number, T: production time (year), : the oil production of the ith well in the t year (bar / year), : The natural gas production of the ith well in the tth year (m3 / year), : The water production of the ith well in the tth year (bbl / year), : The crude oil price in the tth year (USD / bbl), : The natural gas price in the tth year (USD / m3), : The water disposal cost in the tth year (USD / bbl), r: the discount rate, : The drilling cost of the ith well (USD), : The completion cost of the ith well (USD), : The environmental impact weight coefficient, : The land occupation of the ith well (m2), : The carbon emission of the ith well (ton), : The waste production of the ith well (ton), GOR: gas-oil ratio, OOIP: original geological reserves (bbl).
[0020] The computer first calculates the net present value (NPV), carbon emission, and recovery ratio of each particle's position and well configuration simultaneously, where the carbon emission is calculated by the formula: is the carbon emission of the ith well in the drilling stage, is the carbon emission per unit time of the ith well; The recovery ratio is calculated by the formula: Then, the three objective function values are standardized. The standardized value of the net present value is: The standardized value of the environmental impact is: The standardized value of the resource utilization is: Finally, the matching degree value of each particle is calculated and recorded by the formula: The computer then sorts the matching degree values and selects the optimal particle for updating. The computer uses multi-core parallel computing and multi-thread processing to further improve the computing efficiency and reduce the iteration waiting time.
[0021] Symbol explanation: : Total carbon emission (ton), : Carbon emission of the ith well in the drilling stage (ton), Carbon emission of ith well unit time (tons / year), Production time (years), RF: Recovery factor (%), , Minimum and maximum of net present value, , Minimum and maximum of environmental impact, , Minimum and maximum of resource utilization, F(X): Comprehensive matching degree value, range [0, 1].
[0022] S3: Update of particle position and velocity. After calculating the matching degree value of each particle, the particle velocity and position are updated using the particle swarm optimization algorithm according to the velocity update formula. Each particle will move towards the direction of the current optimal solution, so as to constantly search and converge to the global optimal solution. The update of the particle includes the reference of the individual optimal position and the global optimal position, so that each particle gradually approaches the optimal well location configuration. According to the matching degree value of all particles, global optimization is performed, and particles with higher matching degree are preferentially selected for value assignment and update.
[0023] S4: Parallel calculation and assignment process. In the optimization process, to improve the calculation efficiency, particles whose calculation results are completed are preferentially calculated, and the particles are directly assigned new values and their corresponding well location schemes are recalculated. For other particles, after the calculation is completed, the parameters of the particles are assigned to the preferentially calculated particles, and the corresponding schemes are recalculated. In this way, through parallel calculation, the particles are gradually assigned and calculated instead of waiting for all particles to be calculated before assignment, which greatly saves the calculation time and improves the optimization efficiency. The well spacing of the updated particle is detected, and if the well spacing is less than the safety threshold (for example, 300 meters), a repair mechanism is started, which translates the coordinates of the conflicting well location to the high recovery factor sub-area. If it cannot be translated, the particle is eliminated and reinitialized.
[0024] The assignment process of other particles is: analyzing the calculation results of the current particle, and adjusting the value of the particle according to the difference in matching degree with other particles. If the particle calculation result is good, the parameters of the particle are directly assigned. If the expected result is not achieved, the parameters are re-adjusted until convergence.
[0025] In well location optimization, in some embodiments, the matching degree of the particle can be weighted and evaluated according to the dynamic response of the geological model, and the comprehensive matching degree calculation formula is dynamically adjusted by adjusting the weight coefficients in the formula: , where , , are the net present value weight, the environmental impact weight and the resource utilization weight at the tth iteration, respectively; the net present value weight According to the oil and gas price fluctuation model, the dynamic adjustment is performed according to: wherein is the benchmark weight 0.6, is the change rate of the oil and gas price at the t th moment relative to the benchmark price; is the environmental impact weight is adjusted in combination with the ecological sensitivity classification of the drilling area according to: wherein is the benchmark weight 0.25, is the ecological sensitivity coefficient, the water source protection area takes 2, and the general area takes 1; is the resource utilization rate weight is optimized based on the predicted value of the reservoir recovery rate according to: wherein is the benchmark weight 0.15, is the predicted recovery rate, is the average recovery rate of the block; the adjusted weight needs to meet the normalization condition , and is normalized by: The particle parameters of the well spacing exceeding the safety threshold (such as ≥ 300 meters) are preferentially adjusted each time the update is performed, to ensure that the scheme meets the geological engineering constraints.
[0026] Symbol explanation: , , : the dynamic weight coefficient at the t th iteration, , , : the benchmark weight coefficient, : the change rate of the oil and gas price at the t th moment, : the benchmark oil and gas price, : the ecological sensitivity coefficient, : the predicted recovery rate, : the average recovery rate of the block.
[0027] S5: Iterative optimization and ending condition. Repeat step four above until the stopping condition is met. The stopping condition is that the matching degree in the particle swarm no longer significantly improves, or after a certain number of iterations, the optimal solution is converged. Finally, the solution of the optimal particle, i.e. the optimal well location configuration and the corresponding parameters, is output. Before outputting the final scheme, the drilling cost model is called to verify the economic feasibility.
[0028] In the actual case, the method shortens the well location configuration optimization time from 48 hours of the traditional PSO algorithm to 9.5 hours through dynamic weight adjustment (NPV weight: 60%, environmental impact weight: 25%, resource utilization rate weight: 15%), and increases the NPV by 12.3%. The optimized well spacing all meets the safety drilling specification (≥300 meters). Compared with the traditional optimization algorithm, the application can quickly converge to the optimal solution, and has strong scalability and applicability, and is suitable for optimization problems of complex oil and gas field development. In addition, the application also has high calculation efficiency, can effectively reduce the calculation time and improve the parallel computing performance, thereby providing a more efficient technical scheme for solving the actual well location optimization problem.
[0029] Embodiment 1: An intelligent well location configuration method based on multi-objective particle swarm optimization, comprising the following steps: Step one: determine the basic parameters of well location optimization, in the embodiment, the particle parameters include: well location coordinates: 3D porosity grid generated based on reservoir geology modeling software (such as Petrel), randomly sample high porosity area (>15%) as initial well location; well depth: according to the oil-water interface depth constraint, the value range is 2500±300 meters; well type: mixed coding of horizontal well and vertical well (horizontal well is marked as 1, and vertical well is 0).
[0030] Step two: when matching degree is calculated, call the reservoir numerical simulator (Eclipse) to perform 20-year production dynamic prediction, and output NPV, carbon emission and recovery rate data. Carbon emission reflects environmental impact, and recovery rate reflects environmental utilization rate. According to the demand, the computer analyzes the matching degree of all particles as a whole, and determines the value of the to-be-optimized parameter again.
[0031] In this step, for example, the target of the particle swarm is to maximize the net present value (NPV). Assuming that the initial well location parameters of 3 particles are: [wellhead position 1, wellhead position 2, wellhead position 3], that is, a set of initial values of well location are set respectively. The computer first calculates the NPV of the well location configuration according to the position of each particle, and records the matching degree value of each particle. Then, the computer sorts the matching degree values and selects the optimal particle for updating.
[0032] Step three: for the calculated particles, directly assign and recalculate the well location configuration corresponding to the particles.
[0033] Assuming particle 2 (i.e., corresponding to well site configuration [wellhead location 2]) completes the calculation first, with an NPV value of -16. At this point, the computer adjusts the well site parameters of this particle according to its update strategy (such as inertia weight, individual optimal solution, and global optimal solution in particle swarm optimization algorithm). Without the completion of calculations by other particles, particle 2 can only choose to fine-tune in the current direction, resulting in a new well site configuration (such as wellhead location 2 adjusted to a new location) and recalculate the NPV.
[0034] Step four: When the second particle completes the calculation, its well site configuration is adjusted to approach the optimal solution of the previous particle, its parameters are updated, and the matching degree of this particle is recalculated.
[0035] When particle 3 (i.e., corresponding to well site configuration [wellhead location 3]) completes the calculation and obtains an NPV of -81, according to the evolution rules of the multi-objective particle swarm optimization algorithm, particle 3 will choose to approach the optimal solution of particle 3 (assuming the optimal solution of particle 3 is a new location of wellhead location 3). At this point, the new well site configuration of particle 3 is that wellhead location 3 is adjusted to the optimal location, and the NPV is recalculated.
[0036] Step five: Repeat step four until the matching degree calculated by each particle meets the requirements, and the calculation is ended to obtain the optimization result.
[0037] Step six: Well site conflict detection and repair, detect the well spacing of the updated particle, if the well spacing < safety threshold (set to 300 meters according to the region), start the repair mechanism: translate the coordinates of the conflicting well site to the high recovery sub-region; if it cannot be translated, eliminate the particle and reinitialize; before outputting the final scheme, call the drilling cost model to verify the economic feasibility (single well cost ≤ budget threshold).
[0038] As each round of calculation proceeds, the multi-objective particle swarm optimization algorithm continuously updates the well site configuration of the particles, gradually approaching the optimal solution. After multiple iterations, the final particle swarm converges to an optimal solution, which corresponds to the maximum NPV value of the optimal well site configuration. The final output result is the optimal well site configuration and its corresponding parameters, such as wellhead location.
[0039] Figure 1 The evolution process of the multi-objective particle swarm optimization algorithm in this method is shown, where the black solid arrows represent the evolution process of the standard particle swarm optimization algorithm, and the dashed arrows represent the additional optimization brought by parallel computing and multiple rounds of updates in the method of the present invention. Through parallel evolution of the particle swarm, not only the calculation time is significantly saved, but also the calculation efficiency and convergence speed are improved.
[0040] Figure 2The parameter distribution of the mechanism model established in the application under different optimization scenarios is shown. The 2D and 3D models in the figure show the distribution of the porosity field, which is optimized by random sampling. These optimization scenarios reflect how the evolution process of the multi-objective particle swarm optimization algorithm searches for the optimal solution in the multi-dimensional space and how to adjust the well location parameters according to different geological conditions. Through this method, the optimization process can fully consider different underground conditions, so as to realize more efficient and more accurate well location configuration.
[0041] Figure 3 The application effect of the multi-objective particle swarm optimization algorithm in well location optimization is shown. The figure shows the well location configuration and drilling sequence optimized by the method of the application under a specific scenario, and is compared with the general well placement method. The results show that the method can significantly improve the net present value (NPV), thereby optimizing the development plan of the oil and gas field. Through the optimization of the particle swarm optimization algorithm, the well location configuration and drilling sequence are effectively adjusted, so that the resource development efficiency is maximized, and compared with the traditional method, the well location configuration obtained by the application has higher economic benefits.
[0042] The purpose of the application is to provide an intelligent well location configuration method based on multi-objective particle swarm optimization, especially in the application method in well location optimization. The application improves the particle swarm algorithm, combines the characteristics of well location optimization problems, and takes the net present value (NPV), environmental impact and resource utilization rate as the optimization target, optimizes the well location configuration to maximize the economic benefit, and improves the calculation efficiency and concurrent computing performance.
[0043] The intelligent well location configuration method based on multi-objective particle swarm optimization of the application can converge to the optimal solution more quickly through multiple iterations. In addition, the multi-objective optimization mechanism introduced by the application greatly improves the calculation efficiency and significantly reduces the calculation time. The method can adapt to complex well location optimization problems, and through the improved particle swarm optimization algorithm, efficient well location configuration optimization can be completed in a short time, and has high application prospect, especially suitable for well location optimization of oil and gas fields.
[0044] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same and similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part.
[0045] The principles and implementation manners of the present application are described by using specific examples in the present application, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for the general technical personnel in the art, the specific implementation manners and application ranges will be changed according to the idea of the present application. In conclusion, the content of the present specification should not be understood as the limitation of the present application.
Claims
1. A multi-objective particle swarm optimization based intelligent well placement method, characterized in that, The method comprises the following steps: S1. Determine the key parameters of well site optimization, including the well sites to be optimized, the parameter ranges of each well site, and the objective functions, including the objective functions corresponding to the net present value, environmental impact, and resource utilization rate; select a certain number of particles, each particle representing a potential well site configuration and its corresponding optimization parameters, and assign an initial position and velocity to each particle, and initialize the initial values of the parameters to be optimized; S2. Calculate and analyze the matching degree value of the particles in step S1, and calculate the matching degree value of each particle according to the multi-objective function, and perform overall analysis on the matching degrees of all particles; S3. Update the position and velocity of each particle according to the analysis result of step S2, and update the velocity and position of each particle according to the velocity update formula after calculating the matching degree value of each particle; S4. Parallel calculation and assignment, during the optimization process, the particles whose calculation results are completed are preferentially calculated, and the new values of the particles are directly assigned and the corresponding well site schemes are recalculated; for other particles, after the calculation is completed, the parameters of the particles are assigned to the particles whose calculation results are completed, and the corresponding schemes are recalculated; detect the well spacing of the updated particles, if the well spacing is less than the safety threshold, start the repair mechanism, the repair mechanism is: translate the coordinates of the conflict well site to the high recovery sub-area; if it cannot be translated, eliminate the particle and reinitialize; S5. Repeat step S4 until the maximum number of iterations is reached or the predetermined convergence condition is met, output the solution of the optimal particle, and call the drilling cost model to verify the economic feasibility before outputting the final scheme.
2. The intelligent well placement method based on multi-objective particle swarm optimization according to claim 1, characterized in that, In step S1, the net present value maximization objective function is: The environmental impact minimization objective function is: The resource utilization rate maximization objective function is: where X is the well location configuration decision variable vector, which contains well location coordinates, well depth, and well type; n is the number of wells; T is the production time, years; Qi,t is the crude oil production of the ith well in the tth year, barrels / year; Qi,t is the natural gas production of the ith well in the tth year, cubic meters / year; Qi,t is the water production of the ith well in the tth year, barrels / year; Pt is the crude oil price in the tth year, dollars / barrel; Pt is the natural gas price in the tth year, dollars / cubic meter; Ct is the water treatment cost in the tth year, dollars / barrel; r: is the discount rate; Ci is the drilling cost of the ith well, dollars; Ci is the completion cost of the ith well, dollars; are the environmental impact weight coefficients; Ai is the land area of the ith well, square meters; Ci is the carbon emission of the ith well, tons; Ci is the waste generation of the ith well, tons; GOR is the gas-oil ratio; OOIP is the original geological reserves, barrels.
3. The intelligent well placement method based on multi-objective particle swarm optimization according to claim 2, characterized in that, In step S2, the computer first calculates the net present value, carbon emissions, and recovery rate of the well site configuration according to the position of each particle, wherein the carbon emissions are calculated by the formula: wherein, is the total carbon emissions, tons; is the carbon emissions for the drilling phase of the ith well, tons; is the carbon emissions per unit of time produced by the ith well, tons / year; the recovery factor is calculated by the following formula: Wherein, RF: recovery rate, %; Then, the three objective function values are standardized, the net present value standardized value is: The environmental impact standardized value is: The resource utilization rate standardized value is: Finally, the matching degree value of each particle is calculated and recorded by the comprehensive matching degree formula: wherein, , respectively represent the minimum and maximum values of the net present value; , respectively represent the minimum and maximum values of the environmental impact; , respectively represent the minimum and maximum values of the resource utilization rate; F(X) is a comprehensive matching degree value, with a value range of [0, 1]; w1 is a net present value weight, w2 is an environmental impact weight, and w3 is a resource utilization rate weight; Sort all particles according to the matching degree value, select the optimal particle for updating, select the top 20% of particles as elite particles to guide the evolution direction of other particles; the computer uses multi-core parallel computing.
4. The intelligent well placement method based on multi-objective particle swarm optimization according to claim 3, characterized in that, w1 is 0.6, w2 is 0.25, and w3 is 0.
15.
5. The intelligent well placement method based on multi-objective particle swarm optimization of claim 1, wherein, In step S3, the matching degree value of all particles is globally optimized, and the particles with higher matching degree are preferentially selected for value updating.
6. The intelligent well placement method based on multi-objective particle swarm optimization of claim 1, wherein, In well site optimization, the matching degree of the particles is weighted and evaluated according to the dynamic response of the geological model, and the comprehensive matching degree calculation formula is dynamically adjusted as follows: wherein, , , are the net present value weight, the environmental impact weight and the resource utilization rate weight at the tth iteration, respectively; Net present value weight According to the oil and gas price fluctuation model, dynamic adjustment is made according to the following formula: wherein, is the reference weight 0.6; is the change rate of the oil and gas price at the t time relative to the reference price ; Environmental impact weight The environmental impact weight is adjusted in combination with the drilling area ecological sensitivity classification according to the following formula: wherein, is a baseline weight of 0.25; is an ecological sensitivity coefficient, water source protection zone is taken as 2, general area is taken as 1; resource utilization rate weight Based on the reservoir recovery factor prediction value, optimization is performed according to the following formula: wherein, is the reference weight 0.15, is the predicted recovery factor, is the block average recovery factor; the adjusted weight needs to satisfy the normalization condition , The normalization processing is performed by the following formula: Each time the parameters of the particles whose well spacing exceeds the safety threshold are preferentially adjusted to ensure that the scheme meets the geological engineering constraints.
7. The intelligent well placement method based on multi-objective particle swarm optimization of claim 1, wherein, The convergence condition in step S5 is that the maximum or minimum value of the matching degree does not change, or the preset number of iterations is reached.