Track profile dynamic optimization method for well drilling

Abstract

The invention relates to the technical field of petroleum and natural gas engineering drilling, and discloses a track profile dynamic optimization method and system for drilling, and the system comprises a track model library, an initial value generation unit, an optimization solving unit, a constraint evaluation unit and a result output unit. The orbit model library constructs three-section and five-section orbit section models based on a constant tool face angle, and the optimization solving unit fuses a quasi-Newton method and an artificial bee colony algorithm to realize orbit parameter optimization under multiple variables and multiple constraints. According to the method, an adaptive track profile type is selected through real-time drilling data, and optimized parameters are output after initial value solving, algorithm optimization and constraint evaluation. According to the method, the feedback of the drilling process can be dynamically responded, the target window drilling rate is remarkably improved, the track complexity and the dragging pressure risk of a long horizontal section are reduced, the drilling requirements of directional wells and horizontal wells under complex stratums are met, and technical support is provided for efficient and safe development of drilling operation.

Description

Technical Field

[0001] This invention relates to the field of oil and gas engineering technology, specifically to a method for dynamic optimization of wellbore profiles. Background Technology

[0002] As oil and gas extraction extends to areas with complex geological conditions, the application of directional and horizontal well drilling is becoming increasingly widespread. The design and optimization of the trajectory profile has become a core factor affecting drilling success rate, operational efficiency, and safety risks.

[0003] Traditional drilling trajectory design methods often rely on static models or empirical formulas, failing to incorporate real-time data during the drilling process for dynamic feedback and adjustment. This leads to issues such as trajectory deviation from the target point and high trajectory complexity in complex formations. Furthermore, existing technologies for trajectory design optimization either employ a single algorithm, resulting in difficulties in solving nonlinear constraint equations and low convergence efficiency, or lack efficient solution schemes for multi-objective optimization. This makes it difficult to simultaneously address multiple constraints such as drilling extension limits, target point accuracy, trajectory complexity, and the risk of dragging pressure in long horizontal sections, leading to low drilling success rates, high safety risks, and an inability to meet drilling requirements in complex formations.

[0004] Although existing technologies have made some progress in the design of spatial circular arc tracks, the track design model based on constant tool face angles remains imperfect. Its constraint equations contain multiple parameters with complex interrelationships, making them difficult to solve. Furthermore, it lacks a suitable multi-algorithm fusion optimization scheme, hindering flexible, efficient, and dynamic optimization of track parameters. Therefore, we propose a dynamic optimization method for track profiles in drilling. Summary of the Invention

[0005] The purpose of this invention is to provide a dynamic optimization method for wellbore trajectory profiles to solve the problems mentioned in the background art.

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

[0007] A dynamic optimization system for wellbore trajectory profiles includes a trajectory model library, an initial value generation unit, an optimization solution unit, a constraint evaluation unit, and a result output unit.

[0008] The track model library provides two track profile types based on the constant tool face angle track model: a three-segment system and a five-segment system. The three-segment system consists of a straight segment - a constant tool face angle segment - a straight segment, and the five-segment system consists of a straight segment - a constant tool face angle segment - a straight segment - a constant tool face angle segment - a straight segment. The length of the straight segment can be zero, which meets the track design requirements of limiting the target position and simultaneously limiting the target position and direction.

[0009] The initial value generation unit solves for the initial values ​​of the orbit parameters based on the spatial circular arc orbit model, providing a convergence basis for the optimization solution unit;

[0010] The optimization solution unit uses a combination of the quasi-Newton method and the artificial bee colony algorithm to solve and optimize the orbital parameters. The quasi-Newton method is used to solve the nonlinear constraint equations, and the artificial bee colony algorithm guides variable optimization through an iterative update strategy of hired bees, observation bees, and scout bees.

[0011] The constraint evaluation unit evaluates the compliance of the optimized parameters based on constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length.

[0012] The result output unit outputs the trajectory profile optimization parameters evaluated by constraints, including well inclination angle, azimuth angle, build-up rate, length of each segment, tool face angle, etc.

[0013] Preferably, the constant tool face angle trajectory model in the trajectory model library is constructed based on the northeast-northeast coordinate system OXYZ, where O is the wellhead, the X-axis points due north, the Y-axis points due east, and the Z-axis points vertically. The model contains a set of trajectory constraint equations with trigonometric function numerical integration, which is suitable for the trajectory design requirements of directional wells, horizontal wells, and sidetracking wells.

[0014] Preferably, the quasi-Newton method of the optimization solution unit avoids complex differentiation processes and improves the convergence efficiency of solving the nonlinear constraint equation system by using the initial value of the spatial circular arc orbit model.

[0015] It supports nine solution combinations for the three-stage model and six solution combinations for the five-stage model, enabling flexible solutions for design variables.

[0016] Preferably, the artificial bee colony algorithm of the optimization solution unit constructs the orbit optimization variable as a d-dimensional nectar source vector, and achieves iterative updates through local random search by hired bees, fitness ratio selection search by observation bees, and global re-search by scout bees. It uses nectar source direction information and optimization gradient information to guide the variable to update in the optimal direction, reducing the risk of getting trapped in local optima.

[0017] A method for dynamic optimization of wellbore trajectory profiles includes the following steps:

[0018] S1. Data Acquisition: Acquire real-time data of drilling operations, including the coordinates of the current drilling point, inclination angle, azimuth angle, target point coordinates, target point inclination / azimuth requirements, formation strike, depth range of the build-up point, build-up rate limits, and other constraints.

[0019] S2. Profile type selection: Based on the real-time data from step S1 and drilling requirements, select a three-segment or five-segment track profile type from the track model library.

[0020] S3. Initial value solution: The initial value generation unit solves the corresponding initial values ​​of the track parameters based on the selected track profile type and the spatial circular arc track model, including the length of each segment, the inclination angle, the azimuth angle, etc.

[0021] S4. Parameter Optimization: The optimization solution unit substitutes the initial value from step S3 into the solution and first solves the nonlinear constraint equations using the quasi-Newton method to obtain preliminary optimization parameters. Then, iterative optimization of the preliminary optimization parameters is performed using the artificial bee colony algorithm. With the goal of maximizing the drilling encounter rate and minimizing the trajectory complexity, variables such as build-up rate, well inclination angle, azimuth angle, and length of each trajectory segment are optimized.

[0022] S5. Constraint Evaluation: The constraint evaluation unit compares the optimized parameters of step S4 with constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length. If all constraints are met, proceed to step S6; otherwise, return to step S4 to re-optimize.

[0023] S6. Results Output: The results output unit outputs the complete parameters of the optimized trajectory profile, including well depth increment, inclination angle, azimuth angle, vertical depth increment, north / east increment, tool face angle, build-up rate, etc., to guide drilling operations.

[0024] Preferably, in step S4, the iterative optimization process of the artificial bee colony algorithm includes:

[0025] S41. Initialization: Set the number of nectar sources, the dimension of optimization variables, the maximum number of trials, and the maximum number of iterations. Initialize the orbit optimization variables to construct the nectar source vector and determine the upper and lower bounds of the variable search.

[0026] S42, Hired Bee Stage: For each nectar source, perform a local random search to generate new nectar sources, and retain nectar sources with better fitness through greedy selection;

[0027] S43. Observation stage: Calculate the selection probability based on the nectar source adaptability, select nectar sources according to the probability for searching, and retain high-quality nectar sources again through greedy selection.

[0028] S44, Scout Bee Phase: If the nectar source is not improved within the maximum number of trials, then the nectar source is searched globally again and replaced with a new nectar source;

[0029] S45. Convergence Judgment: Determine whether the maximum number of iterations has been reached or the objective function has converged. If so, end the optimization; otherwise, return to S42 to continue iterating.

[0030] Preferably, in step S4, the objective function for trajectory optimization comprehensively considers factors such as target window encounter rate, trajectory complexity, total drilling footage, and drag risk, so that each objective achieves an optimal balance.

[0031] Preferably, during the drilling operation, real-time drilling data is continuously collected and steps S1-S6 are repeated to achieve dynamic iterative optimization of the trajectory profile parameters and adapt to formation changes and drilling deviations during the drilling process in real time.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] This invention breaks through the limitations of traditional static models. By collecting drilling data in real time and performing dynamic iterative optimization, it can adapt to geological changes in complex formations and trajectory deviations during the drilling process, greatly improving the adaptability of trajectory design.

[0034] This invention constructs three-segment and five-segment trajectory models based on constant tool face angles. The length of the straight segment can be zero, supporting design requirements that limit the target position and simultaneously limit the target position and direction. It can also degenerate into a two-dimensional circular arc well segment, adapting to various drilling scenarios such as directional wells, horizontal wells, and sidetracking wells.

[0035] This invention integrates the quasi-Newton method and the artificial bee colony algorithm. The quasi-Newton method efficiently solves the nonlinear constraint equations, while the artificial bee colony algorithm achieves multivariable global optimization, avoiding the local optimum problem of a single algorithm. At the same time, the initial value of the spatial circular arc model is used to improve the convergence efficiency, thus achieving efficient and accurate optimization of orbital parameters under multiple constraints.

[0036] The optimization process of this invention comprehensively considers multiple objectives such as target window drilling rate, trajectory complexity, total drilling footage, and dragging pressure risk, to achieve the optimal balance of each objective. Experimental verification shows that it can significantly improve the target window drilling rate, reduce the dragging pressure risk in long horizontal sections, and reduce trajectory complexity.

[0037] This invention sets comprehensive constraints such as drilling extension limit, target accuracy, build-up rate range, and formation complexity. Through constraint evaluation, it ensures the compliance of optimized parameters, reduces the safety risks of drilling operations from the design level, and meets the drilling needs under complex formation conditions. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the single constant tool face corner segment track (three-segment track) of the present invention;

[0039] Figure 2 For the present invention Figure 2 Schematic diagram of a dual constant tool face corner section track (five-segment track);

[0040] Figure 3 This is a schematic diagram of the three-stage design process of the present invention;

[0041] Figure 4 This is a schematic diagram of the five-stage design process of the present invention;

[0042] Figure 5This is a schematic diagram showing the optimization results of the three profile types of this invention;

[0043] Figure 6 This is a schematic diagram illustrating the three different profile track optimization scenarios of the present invention;

[0044] Figure 7 This is a schematic diagram of the module connection of the present invention;

[0045] Figure 8 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] like Figure 1-8 As shown, a dynamic optimization system for wellbore trajectory profiles includes a trajectory model library, an initial value generation unit, an optimization solution unit, a constraint evaluation unit, and a result output unit.

[0048] The orbit model library provides two orbit profile types based on the constant toolface angle orbit model: a three-segment system and a five-segment system. The three-segment system consists of a straight segment - a constant toolface angle segment - a straight segment, while the five-segment system consists of a straight segment - a constant toolface angle segment - a straight segment - a constant toolface angle segment - a straight segment. The length of the straight segment can be zero, meeting the orbit design requirements of limiting the target position and simultaneously limiting the target position and direction. The constant toolface angle orbit model in the orbit model library is constructed based on the northeast-east coordinate system OXYZ, where O is the wellhead, the X-axis points due north, the Y-axis points due east, and the Z-axis points vertically. The model contains a set of orbit constraint equations with trigonometric function numerical integration, adapting to the orbit design requirements of directional wells, horizontal wells, and sidetracking wells.

[0049] The initial value generation unit solves for the initial values ​​of the orbit parameters based on the spatial circular arc orbit model, providing a convergence basis for the optimization solution unit;

[0050] The optimization unit employs a combination of the quasi-Newton method and the artificial bee colony algorithm to solve and optimize orbital parameters. The quasi-Newton method is used to solve the nonlinear constraint equations, while the artificial bee colony algorithm guides variable optimization through an iterative update strategy involving hired bees, observer bees, and scout bees. The quasi-Newton method in the optimization unit avoids complex differentiation processes and improves the convergence efficiency of solving the nonlinear constraint equations by using the initial values ​​of the spatial circular arc orbit model. It supports nine solution combinations for the three-segment model and six solution combinations for the five-segment model, enabling flexible solutions to design variables. The artificial bee colony algorithm in the optimization unit constructs the orbital optimization variables as a d-dimensional nectar source vector and achieves iterative updates through local random search by hired bees, fitness ratio selection search by observer bees, and global re-search by scout bees. It utilizes nectar source direction information and optimization gradient information to guide the variables towards the optimal direction, reducing the risk of getting trapped in local optima.

[0051] The constraint assessment unit evaluates the compliance of the optimized parameters based on constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length.

[0052] The output unit outputs the optimized trajectory profile parameters that have passed the constraint evaluation, including well inclination angle, azimuth angle, build-up rate, length of each segment, tool face angle, etc.

[0053] The device of this invention can quickly adjust trajectory parameters and replan the trajectory based on real-time drilling data, ensuring trajectory accuracy, reducing trajectory complexity, optimizing various constraints during the drilling process, and significantly improving drilling efficiency and safety. Furthermore, the device can select appropriate trajectory profile types based on different formation conditions and target requirements, further optimizing drilling operations. Through the technical solution of this invention, the drilling success rate can be effectively improved, the complexity and risk of drilling operations reduced, and the application of intelligent drilling technology promoted.

[0054] like Figure 8 As shown, a method for dynamic optimization of wellbore trajectory profiles includes the following steps:

[0055] S1. Data Acquisition: Acquire real-time data of drilling operations, including the coordinates of the current drilling point, inclination angle, azimuth angle, target point coordinates, target point inclination / azimuth requirements, formation strike, depth range of the build-up point, build-up rate limits, and other constraints.

[0056] S2. Profile type selection: Based on the real-time data from step S1 and drilling requirements, select a three-segment or five-segment track profile type from the track model library.

[0057] S3. Initial value solution: The initial value generation unit solves the corresponding initial values ​​of the track parameters based on the selected track profile type and the spatial circular arc track model, including the length of each segment, the inclination angle, the azimuth angle, etc.

[0058] S4. Parameter Optimization: The optimization solution unit substitutes the initial value from step S3 into the solution and first solves the nonlinear constraint equations using the quasi-Newton method to obtain preliminary optimization parameters. Then, iterative optimization of the preliminary optimization parameters is performed using the artificial bee colony algorithm. With the goal of maximizing the drilling encounter rate and minimizing the trajectory complexity, variables such as build-up rate, well inclination angle, azimuth angle, and length of each trajectory segment are optimized.

[0059] S41. Initialization: Set the number of nectar sources, the dimension of the optimization variables, the maximum number of trials, and the maximum number of iterations. Initialize the trajectory optimization variables to construct the nectar source vector, determine the upper and lower bounds of the variable search, and comprehensively consider factors such as the target window drilling rate, trajectory complexity, total drilling footage, and drag risk to achieve the optimal balance of each objective.

[0060] S42, Hired Bee Stage: For each nectar source, perform a local random search to generate new nectar sources, and retain nectar sources with better fitness through greedy selection;

[0061] S43. Observation stage: Calculate the selection probability based on the nectar source adaptability, select nectar sources according to the probability for searching, and retain high-quality nectar sources again through greedy selection.

[0062] S44, Scout Bee Phase: If the nectar source is not improved within the maximum number of trials, then the nectar source is searched globally again and replaced with a new nectar source;

[0063] S45. Convergence check: Determine whether the maximum number of iterations has been reached or the objective function has converged. If yes, end the optimization; otherwise, return to S42 to continue iterating.

[0064] S5. Constraint Evaluation: The constraint evaluation unit compares the optimized parameters of step S4 with constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length. If all constraints are met, proceed to step S6; otherwise, return to step S4 to re-optimize.

[0065] S6. Results Output: The results output unit outputs the complete parameters of the optimized trajectory profile, including well depth increment, inclination angle, azimuth angle, vertical depth increment, north / east increment, tool face angle, build-up rate, etc., to guide drilling operations.

[0066] During drilling operations, real-time drilling data is continuously collected and steps S1-S6 are repeated to achieve dynamic iterative optimization of the trajectory profile parameters and adapt to formation changes and drilling deviations during the drilling process in real time.

[0067] Example 1:

[0068] The wellbore trajectory design adopts the northeast-northeast coordinate system OXYZ, where O is the wellhead, the X-axis points due north, the Y-axis points due east, and the Z-axis points vertically. The basic calculation formula is derived based on the characteristics of the constant toolface angle curve.

[0069]

[0070] Wherein, the subscript is the starting point, the subscript j is the ending point, the length unit is m, the angle unit is rad, and the curvature unit is rad / m. In equation (3), "+" is used for increasing azimuth and "-" is used for decreasing azimuth.

[0071] To meet the trajectory design requirements of directional and horizontal wells, two typical trajectory design models were constructed using spatial circular arcs. By modifying the circular arc segment into a constant toolface angle segment, two design models were established: "straight segment - constant toolface angle segment - straight segment" and "straight segment - constant toolface angle segment - straight segment - constant toolface angle segment - straight segment". The straight segment length can be zero, satisfying the requirements of two typical trajectory designs: one with a defined target location, and the other with both target location and direction defined. The two design models are as follows: Figure 1 , Figure 2 As shown.

[0072] The constraint equations for the constant tool face angle track design model are as follows:

[0073]

[0074] Where n=1 and 2 represent the three-segment and five-segment tracks, respectively. ,and

[0075]

[0076] When the initial well inclination angle of the constant tool face corner segment is 0, the segment degenerates into a two-dimensional circular arc segment, and the above constraint equations need to be modified accordingly, which will not be elaborated here. As can be seen from the equations (7), there are many design variables, complex interrelationships, and great difficulty in solving them. Therefore, a model solution method is studied.

[0077] The track design model contains three independent constraint equations, which can be solved for three unknowns. To meet the need for flexible solution in the design, the lengths of straight segments L1, L2, and L3 or curvatures K1 and K2 are used as one basic solution variable, and two other variables are selected as unknowns, while the remaining parameters are knowns. Since the constraint equations involve numerical integration calculations containing trigonometric functions, the quasi-Newton numerical method is used to avoid complex differentiation calculations. When applying the quasi-Newton method, the selection of initial values ​​for the solution variables is extremely important. Choosing initial values ​​near the feasible solution allows for rapid convergence to the correct result. After determining the combination of the three solution variables, the initial values ​​are first solved according to the corresponding circular arc track model, and then the quasi-Newton method is used to solve the problem, thereby resolving the complex nonlinear equation system.

[0078] For the three-segment track design model, given any two parameters among variables L1, L2 and K1, the other parameter can be solved. And given any one of the three variables, the other two variables can be solved, for a total of nine solution combinations, as shown in the figure.

[0079] For the design model of a double constant tool face angle track, given any four parameters from variables L1, L2, L3, K1, and K2, the remaining parameter can be solved. Also, given variables L1, L2, and L3, K1 = K2 can be solved, resulting in a total of six solution combinations. For example... Figure 3 , Figure 4 As shown.

[0080] Taking the solution of L2 given L1, L3, K1, and K2 as an example, the solution steps of the design model are introduced:

[0081] Using L2 as the basic solution variable, and then selecting α2 and φ2 as unknowns, as the three solution parameters of the system of equations (1);

[0082] Based on the coordinates and directions of the design starting point and target point, and given the known parameters L1, L3, K1 and K2, find the solution of the corresponding spatial circular arc track model, and use the results L2, α2 and φ2 as the initial values ​​of the solution of the corresponding constant tool face angle model.

[0083] Find K α1 K α2 S1 and S2 determine the signs of the increase or decrease of the azimuth angles of the two curve segments;

[0084]

[0085] In equation (8), "+" is used for increasing the slope and "-" is used for decreasing the slope. When the slope is stable, ,when hour, .

[0086]

[0087] Similarly, K can be obtained. α2 and S 2。

[0088] Solve the system of equations using the quasi-Newton method (7);

[0089] If the system of equations has a solution, then calculate the orbital node parameters.

[0090] Based on the model solution method, 15 solution combination calculation programs for two typical design models were developed. The solution is flexible and fast, and can meet the needs of directional design, horizontal well, sidetracking well and drilling trajectory design.

[0091] Design Example 1: For a directional well, the inclination angle and azimuth angle at the current point are α1 = 10° and φ1 = 315°, respectively. The coordinates of the target point are: X... T =200m, Y T =-1100m, Z T =4000m. If drilling continues in the current direction for L1=200m, then the wellbore trajectory is adjusted using a tool with a build-up rate of K1=3° / 30m. The design result is that the build-up tool drills for S1=235.54m at a constant tool face angle ω=-33.28°, and then the stabilizing drill string drills for L2=1821.75m to accurately reach the design target point. Wellbore trajectory data are shown in the table.

[0092] Design Example 2: The endpoint coordinates X of a three-dimensional horizontal well T =31.70m, Y T =-1288.80m, Z T =869.42m, horizontal section inclination angle α3=90.36°, azimuth angle φ3=270°, section length L3=941.72m. Start-up from the vertical section, start-up point depth L1=532m, start-up rate K1=K2=6° / 30m, wellbore trajectory designed using constant tool face angle method.

[0093] Using the dual constant tool face angle track design model, the length of the stable inclination section, the inclination angle, and the azimuth angle of the corresponding circular arc track were calculated as initial values ​​for solving the constant tool face angle track constraint equation using the quasi-Newton method. After two numerical iterations, the results were obtained as L2=80.706m, α2=46.164°, and φ2=277.686°. The wellbore track data are shown in the table.

[0094] Design Example 3: The landing point of a horizontal well is designed with an inclination and azimuth of α3=88° and φ3=50°. The current well inclination and azimuth angles are α1=76° and φ1=54°, and the coordinates of the distance from the landing point are ΔX=52.70m, ΔY=57.51m, and ΔZ=12m. How should the trajectory be designed to ensure accurate landing?

[0095] Since the vertical distance between the current point and the landing point is only 12m, a simple double constant tool face angle track is designed, namely a constant tool face angle segment + a constant tool face angle segment (K1=K2), so that a set of tilting tools can accurately land the object. The design result is K1=K2=8.327° / 30m, and the track data is shown in the table.

[0096] Table 1. Directional Well Track Design Data (Three-Section Track)

[0097]

[0098] Table 2 Horizontal Well Track Design Data (Five-Segment Track)

[0099]

[0100] Table 3 Horizontal Well Track Design Data (Permanent Tool Face Corner Section + Permanent Tool Face Corner Section)

[0101]

[0102] Design Example 4: The coordinates of the current point of a directional well are: X S =100m, Y S =―100m, Z S =2000m; the inclination angle and azimuth angle are 10° and 315° respectively. The coordinates of the target point are: X T =200m, Y T =―1100m, Z T =4000m. If drilling continues in the current direction for L1=200m, then a tool with a build-up rate of K=3° / 30m is used to adjust the wellbore trajectory. The design result is that after drilling S=230.80m with the build-up tool, drilling L2=1826.25m with the stabilizing drill string can accurately reach the design target point. Wellbore trajectory data are shown in Table 4.

[0103] Design Example 5: A horizontal well is designed with a landing point inclination and azimuth of 88° and 50°, respectively. The current well inclination and azimuth angles are 76° and 54°, and the coordinates of the distance from the landing point are ΔX=52.70m, ΔY=57.51m, and ΔZ=12m. How should the trajectory be designed to ensure accurate landing?

[0104] Since the vertical distance between the current point and the landing point is only 12m, we designed a simple double-circular-arc track, namely a circular arc segment + a circular arc segment track (K1=K2), so that a set of tilt-building tools can enable it to land accurately. The design result is K1=K2=8.317° / 30m, and the track data is shown in Table 5.

[0105] Table 4. Directional well trajectory design data (stable inclination section + circular arc section + stable inclination section)

[0106]

[0107] Table 5 Horizontal Well Track Design Data (Circular Arc Segment + Circular Arc Segment, K1=K2=8.317° / 30m)

[0108]

[0109] Example 2:

[0110] The wellbore profile optimization design model utilizes nectar source direction information to guide the optimization of decision variables. When adjusting the trajectory design, variables such as the build-up point, build-up rate, well inclination / azimuth, and target segment length need to be initialized. Subsequently, the model is updated sequentially using hired bees, observation bees, and reconnaissance bees. By setting a maximum number of iterations, the trajectory design variables converge to the optimal value of the objective function.

[0111] Here, the set of all candidate solutions in the search space can be represented as a d-dimensional vector, i.e.:

[0112]

[0113] in, Represents the quantity of nectar sources; X i This represents a candidate solution to the optimization problem, i.e., a nectar source. In the orbit optimization problem, each row X... i The corresponding function value f(X) i Both represent the fitness of a track; x represents the number of optimization variables for a single trajectory. i，j This represents the j-th optimization variable within the i-th nectar source, encompassing all orbital optimization variables:

[0114]

[0115] Suppose that in an optimization problem, all variables to be optimized have the same dimension, and the upper and lower bounds of the search in each dimension are X. max X min ,Right now:

[0116]

[0117] Here, rand(0,1) represents a random number in the range of 0 to 1.

[0118] In each iteration, each hired bee selects a nectar source and performs a random search within its internal dimensions according to equation (3) to generate an updated nectar source. The new nectar source is compared with the original nectar source, and the one with better fitness is retained as the optimal nectar source for the current iteration, i.e.:

[0119]

[0120] Where, r ij A random real number in the range [-1, 1] This represents the updated value of the j-th optimization variable under the i-th honey source.

[0121] Considering the disorder of equation (4) during the optimization process, the optimization variables are further updated towards a better search direction based on the optimization gradient information, that is:

[0122]

[0123] Where k is a randomly selected nectar source location. And k≠i; Gbest j Let represent the j-th optimization variable of the current globally optimal honey source.

[0124] After the hiring bee stage, the observation bees will search according to the solution of equation (4), select high-quality nectar sources based on fitness ratios, and retain nectar sources with high fitness based on greedy selection. This process effectively utilizes nectar sources with high fitness, realizing further development within a small search space, that is:

[0125]

[0126] Wherein, P(X) i Let be the probability that the observed bee chooses the i-th nectar source.

[0127] During the algorithm iteration process, if the fitness f(X) of the current honey source is... i If no improvement is achieved within the maximum number of trials limit, the current honey source may be trapped in a local optimum. Therefore, the following formula is used:

[0128]

[0129] Where r1 and r2 are random real numbers in the range [0, 1].

[0130] Based on the above, the optimization framework of the ABC algorithm is as follows:

[0131]

[0132] Based on the model and algorithm testing scheme, trajectory profile type-parameter optimization tests were conducted. The parameters for the current drilling points are as follows: well depth =964.17m; Coordinates: X S =306221m, Y S =283927m, Z S =956m; Inclination and azimuth: α S =4.04°, φ S =285.69°; Based on the geological conditions of the drilled wells, the depth range of the build-up point is 900m to 1000m; the strike of the formation is between 60° and 80°; target coordinates: X T =306739m, Y T =285823m, Z T=2200m; the target section length L2 is controlled between 1200-1600m to maximize production capacity and avoid serious dragging problems caused by long horizontal sections. The current algorithm parameters are shown in Table 6, and the constraints are shown in Table 7.

[0133] Table 6 DABC Parameter Settings

[0134]

[0135] Table 7 Constraints

[0136]

[0137] The following three profile types are examples: (I) Straight line segment ( (I) + arc segment (S) + straight segment (L2); (II) straight segment ( ) + arc segment (S1) + straight segment (L) h (III) Linear segment (S2) + Circular arc segment (S2) + Linear segment (L2) The three cross-sections are: S1 + S2 + L2; The optimization results of Algorithm 1 on the above three cross-sections are as follows: Figure 5 , Figure 6 As shown.

[0138] Overall results show that type (II) achieves the best balance between total advance and track complexity, and has the best optimization effect. The variable results for the three profiles are shown in Table 8.

[0139] Table 8 Results of Profile Variable Optimization

[0140] .

[0141] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dynamic optimization system for wellbore trajectory profiles, characterized in that, It includes a track model library, initial value generation unit, optimization solution unit, constraint evaluation unit, and result output unit; The track model library provides two track profile types based on the constant tool face angle track model: a three-segment system and a five-segment system. The three-segment system consists of a straight segment - a constant tool face angle segment - a straight segment, and the five-segment system consists of a straight segment - a constant tool face angle segment - a straight segment - a constant tool face angle segment - a straight segment. The length of the straight segment can be zero, which meets the track design requirements of limiting the target position and simultaneously limiting the target position and direction. The initial value generation unit solves for the initial values ​​of the orbit parameters based on the spatial circular arc orbit model, providing a convergence basis for the optimization solution unit; The optimization solution unit uses a combination of the quasi-Newton method and the artificial bee colony algorithm to solve and optimize the orbital parameters. The quasi-Newton method is used to solve the nonlinear constraint equations, and the artificial bee colony algorithm guides variable optimization through an iterative update strategy of hired bees, observation bees, and scout bees. The constraint evaluation unit evaluates the compliance of the optimized parameters based on constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length. The result output unit outputs the optimized trajectory profile parameters evaluated by constraints, including well inclination angle, azimuth angle, build-up rate, length of each segment, tool face angle, etc.

2. The dynamic optimization system for wellbore trajectory profiles according to claim 1, characterized in that, The constant tool face angle trajectory model in the trajectory model library is constructed based on the northeast-northeast coordinate system OXYZ, where O is the wellhead, the X-axis points due north, the Y-axis points due east, and the Z-axis points vertically. The model contains a set of trajectory constraint equations with trigonometric function numerical integration, which is suitable for the trajectory design requirements of directional wells, horizontal wells, and sidetracking wells.

3. The dynamic optimization system for wellbore trajectory profiles according to claim 1, characterized in that, The quasi-Newton method of the optimization solution unit avoids complex differentiation processes and improves the convergence efficiency of solving nonlinear constraint equations by using the initial values ​​of the spatial circular arc orbit model. It supports nine solution combinations for the three-stage model and six solution combinations for the five-stage model, enabling flexible solutions for design variables.

4. The dynamic optimization system for drilling trajectory profiles according to claim 1, characterized in that, The artificial bee colony algorithm of the optimization solution unit constructs the orbit optimization variable as a d-dimensional nectar source vector. It achieves iterative updates through local random search by hired bees, fitness ratio selection search by observation bees, and global re-search by scout bees. It uses nectar source direction information and optimization gradient information to guide the variable to update in the optimal direction, reducing the risk of getting trapped in local optima.

5. A method for dynamic optimization of wellbore trajectory profiles, characterized in that, The application of the dynamic optimization system for drilling trajectory profiles according to any one of claims 1-4 includes the following steps: S1. Data Acquisition: Acquire real-time data of drilling operations, including the coordinates of the current drilling point, inclination angle, azimuth angle, target point coordinates, target point inclination / azimuth requirements, formation strike, depth range of the build-up point, build-up rate limits, and other constraints. S2. Profile type selection: Based on the real-time data from step S1 and drilling requirements, select a three-segment or five-segment track profile type from the track model library. S3. Initial value solution: The initial value generation unit solves the corresponding initial values ​​of the track parameters based on the selected track profile type and the spatial circular arc track model, including the length of each segment, the inclination angle, the azimuth angle, etc. S4. Parameter Optimization: The optimization solution unit substitutes the initial value from step S3 into the solution and first solves the nonlinear constraint equations using the quasi-Newton method to obtain preliminary optimization parameters. Then, iterative optimization of the preliminary optimization parameters is performed using the artificial bee colony algorithm. With the goal of maximizing the drilling encounter rate and minimizing the trajectory complexity, variables such as build-up rate, well inclination angle, azimuth angle, and length of each trajectory segment are optimized. S5. Constraint Evaluation: The constraint evaluation unit compares the optimized parameters of step S4 with constraints such as drilling extension limit, target accuracy, formation complexity, build-up rate range, and track length. If all constraints are met, proceed to step S6; otherwise, return to step S4 to re-optimize. S6. Results Output: The results output unit outputs the complete parameters of the optimized trajectory profile, including well depth increment, inclination angle, azimuth angle, vertical depth increment, north / east increment, tool face angle, build-up rate, etc., to guide drilling operations.

6. The method for dynamic optimization of wellbore trajectory profiles according to claim 5, characterized in that, In step S4, the iterative optimization process of the artificial bee colony algorithm includes: S41. Initialization: Set the number of nectar sources, the dimension of optimization variables, the maximum number of trials, and the maximum number of iterations. Initialize the orbit optimization variables to construct the nectar source vector and determine the upper and lower bounds of the variable search. S42, Hired Bee Stage: For each nectar source, perform a local random search to generate new nectar sources, and retain nectar sources with better fitness through greedy selection; S43. Observation stage: Calculate the selection probability based on the nectar source adaptability, select nectar sources according to the probability for searching, and retain high-quality nectar sources again through greedy selection. S44, Scout Bee Phase: If the nectar source is not improved within the maximum number of trials, then the nectar source is searched globally again and replaced with a new nectar source; S45. Convergence Judgment: Determine whether the maximum number of iterations has been reached or the objective function has converged. If so, end the optimization; otherwise, return to S42 to continue iterating.

7. The method for dynamic optimization of wellbore trajectory profiles according to claim 5, characterized in that, In step S4, the objective function of trajectory optimization comprehensively considers factors such as target window encounter rate, trajectory complexity, total drilling footage, and drag risk, so as to achieve the optimal balance of each objective.

8. The method for dynamic optimization of wellbore trajectory profiles according to claim 5, characterized in that, During drilling operations, real-time drilling data is continuously collected and steps S1-S6 are repeated to achieve dynamic iterative optimization of the trajectory profile parameters and adapt to formation changes and drilling deviations during the drilling process in real time.

Images