Intelligent design method and system for well track of high-density cluster well

Abstract

The invention discloses an intelligent design method and system for a well track of a high-density cluster well. The method comprises the steps that a three-dimensional collision detection route map is constructed; performing path search and intelligent obstacle avoidance on the three-dimensional collision detection route map by using a path search algorithm, judging whether the search paths meet the dog-leg degree constraint in the search process, performing path cost calculation on the paths meeting the dog-leg degree constraint in sequence, and sorting and storing the search paths meeting the dog-leg degree constraint according to the path cost; and according to a sorting result, selecting an optimal path, and smoothing the segment of which the dog-leg degree exceeds a preset threshold value. According to the method, a route map from a wellhead position to a target spot for three-dimensional collision detection is automatically constructed through a probability route map algorithm, then an automatic path search algorithm is utilized to perform path search under the condition that the set maximum dog-leg degree is met until an optimal feasible path is found, and local smoothing processing is performed on the found path, so that the optimal path is obtained. The design time can be effectively saved, and the collision risk of the borehole track is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of wellbore trajectory design, and particularly to an intelligent design method and system for wellbore trajectories of high-density cluster wells. Background Art

[0002] Wellbore trajectory design plays a crucial role in the field of oil and gas exploration and production. Its core objective is to plan a spatial curve extending from the surface wellhead position to the target point, while taking into account various constraints and achieving specific goals, such as minimizing drilling costs, effectively avoiding collisions with adjacent wells, strictly adhering to the dogleg severity constraint, and ensuring full compliance with relevant safety regulations.

[0003] In recent years, with the continuous advancement of onshore and offshore oil and gas field development, the technology of dense cluster drilling has emerged, which undoubtedly further increases the complexity of wellbore trajectory design. Especially in offshore trajectory design work, the distance between wellheads is often only a few meters, while the wellbore trajectory can extend several kilometers in the lateral section. Although this design significantly reduces the surface footprint and greatly improves the operation efficiency, with the continuous increase in well pattern density, the risk of wellbore trajectory collision also rises sharply. It can be seen that the anti-collision design of wellbore trajectories is a highly challenging and significant task in high-density cluster well operations. Especially in offshore dense well clusters and unconventional oilfields, the number of wells in each block can reach 60 or more, and at this time, the anti-collision work in trajectory design is particularly crucial.

[0004] Currently, in the design of wellbore trajectories for dense cluster wells, the main method is to plan the wellbore trajectory based on past experience or by referring to the design of adjacent platforms, and then carry out anti-collision scanning. If the anti-collision index meets the requirements, the trajectory design is considered feasible; if the anti-collision index does not meet the standard, the wellbore trajectory needs to be redesigned until the relevant requirements are met.

[0005] Meanwhile, some domestic and foreign scholars have proposed various wellbore trajectory design methods. However, most of these methods focus on the design of single-well wellbore trajectories, aiming to achieve multi-objective optimization such as shortening the trajectory length, reducing frictional resistance and torque, and decreasing the well inclination or azimuth change rate. Moreover, most research results have not been widely promoted and applied. Although some intelligent algorithms can realize the automatic design or optimization of wellbore trajectories, there are many problems such as difficult solution of the optimization model, high operation difficulty in on-site practical applications, and lack of anti-collision constraints for multiple adjacent wells. Summary of the Invention

[0006] The object of the present invention is to overcome the above technical deficiencies, and propose an intelligent design method and system for the wellbore trajectory of high-density cluster wells, so as to solve the technical problems in the prior art that the wellbore trajectory design relies on experience, the single-well design method is difficult to adapt to the complex scenarios of high-density cluster wells, the intelligent algorithm optimization model is difficult to solve, and there is a lack of multi-adjacent well anti-collision constraints.

[0007] To achieve the above technical object, the present invention adopts the following technical solutions:

[0008] The present invention provides an intelligent design method and system for the wellbore trajectory of high-density cluster wells, including:

[0009] S1. Obtain the data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, and mark the wellbore trajectories of the drilled wells as obstacles in the three-dimensional block, and construct a three-dimensional collision detection roadmap. Among them, the three-dimensional collision detection roadmap includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles.

[0010] S2. Use the path search algorithm to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection roadmap. During the search process, it is necessary to judge whether the search path meets the dogleg constraint, and calculate the path cost for the paths that meet the dogleg constraint in turn, and sort and store the search paths that meet the dogleg constraint according to the path cost.

[0011] S3. According to the sorting result, select the optimal path, calculate the full-track dogleg for this path, smooth the local segments of this path where the dogleg exceeds the preset threshold, and then evaluate the smoothed optimal path to judge whether it meets both the dogleg constraint requirements and can avoid all obstacles. If it meets the dogleg constraint and can avoid the obstacles, then determine this smoothed path as the target wellbore trajectory. Otherwise, if the above conditions are not met, according to the sorting result, select the next path and repeat the dogleg calculation and smoothing operation until a smoothed path that meets both the dogleg constraint and can avoid the obstacles is found.

[0012] In some embodiments, in step S1, the method for constructing the three-dimensional collision detection roadmap includes:

[0013] S11. Automatically generate a number of random points in the three-dimensional block using the probabilistic roadmap algorithm.

[0014] S12. Judge whether each random point collides with the obstacle. If a certain random point does not collide with the obstacle, then retain this random point. Otherwise, delete it and continue to generate new random points until the number of random points meets a certain sampling quantity requirement.

[0015] S13. Starting from the wellhead of the new well and ending at the target target point, traverse all the random points, connect each random point to its adjacent random points, and during the connection process, determine whether the connection route between the current random point and the adjacent random point collides with an obstacle. If a collision occurs, remove the connection; if no collision occurs, retain the connection to obtain a three-dimensional collision detection route map.

[0016] In some embodiments, in step S12, the method for determining whether each random point collides with an obstacle is:

[0017] Determine whether the three-dimensional coordinates of the random point are within the three-dimensional coordinate range of the obstacle. If so, determine that the random point collides with the obstacle; otherwise, determine that the random point does not collide with the obstacle.

[0018] In some embodiments, in step S13, the method for determining whether the connection route between the current random point and the adjacent random point collides with an obstacle includes:

[0019] S131. Calculate the direction vector of the line segment between the current random point and the adjacent random point;

[0020] S132. Starting from the current random point, accumulate in steps of the unit vector of the direction vector. If the point reached during the increase of the line segment is within the coordinate range of the obstacle, determine that the connection route between the current random point and the adjacent random point collides with the obstacle. If all the points reached during the process of increasing from the current random point to the adjacent random point are outside the coordinate range of the obstacle, determine that the connection route between the current random point and the adjacent random point does not collide with the obstacle.

[0021] In some embodiments, in step S2, the method for calculating the dogleg severity of each path in the three-dimensional collision detection route map is:

[0022] S21. Traverse each node of the current path, calculate the well inclination angle and azimuth angle of the previous node of the current node, and then calculate the well inclination angle and azimuth angle of the current node;

[0023] S22. Calculate the dogleg severity of the current node according to the well inclination angles and azimuth angles of the current node and the previous node.

[0024] In some embodiments, in step S22, the calculation formula for the dogleg severity DLS is:

[0025]

[0026] Δθ = arccos{cos(α′2 - α′1) - sin(α′1)sin(α′2)[1 - cos(φ′2 - φ′1)]}

[0027]

[0028] Among them, P1(x1, y1, z1) is the previous node, P2(x2, y2, z2) is the current node, α1 is the well deviation angle of the previous node, is the azimuth angle of the previous node, α2 is the well deviation angle of the P2 node, is the azimuth angle of the P2 node.

[0029] In some embodiments, in step S3, if all the paths cannot meet the dogleg severity constraint and avoid obstacles after the path smoothing process, a three-dimensional collision detection roadmap is regenerated.

[0030] The present invention also provides a high-density cluster wellbore trajectory intelligent design system, including:

[0031] A three-dimensional collision detection roadmap generation module, which is used to obtain the data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, mark the wellbore trajectories of the drilled wells as obstacles in the three-dimensional block, and construct a three-dimensional collision detection roadmap, where the three-dimensional collision detection roadmap includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles;

[0032] An automatic path search module, which is used to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection roadmap using a path search algorithm. During the search process, it is necessary to judge whether the search path meets the dogleg severity constraint, calculate the path cost for the paths that meet the dogleg severity constraint in sequence, and sort and store the search paths that meet the dogleg severity constraint according to the path cost;

[0033] A path smoothing process module, which is used to select the optimal path according to the sorting result, calculate the dogleg severity of the entire orbit for this path, smooth the local segments where the dogleg severity exceeds the preset threshold in this path, and then evaluate the smoothed optimal path to judge whether it meets both the dogleg severity constraint requirements and can avoid all obstacles. If it meets the dogleg severity constraint and can avoid the obstacles at the same time, the smoothed path is determined as the target wellbore trajectory. Otherwise, if the above conditions are not met, according to the sorting result, the next path is selected to repeat the dogleg severity calculation and smoothing process operation, and this cycle is continuously carried out until a smoothed path that meets both the dogleg severity constraint and can avoid obstacles is found.

[0034] The present invention also provides a computer-readable storage medium, and the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the high-density cluster wellbore trajectory intelligent design method.

[0035] Compared with the prior art, the beneficial effects of the intelligent design method and system for the wellbore trajectory of high-density cluster wells provided by the present invention are as follows: By using the probabilistic roadmap algorithm to automatically construct a roadmap from the wellhead position for three-dimensional collision detection to the target target point, and then using the automatic path search algorithm to perform path search under the condition of meeting the set maximum dogleg severity, until the optimal feasible path is found, and the found path is locally smoothed to ensure that the path meets the maximum dogleg severity constraint. This work content is of great significance in the intelligent anti-collision trajectory design, especially when adding an additional anti-collision well in high-density cluster wells, which will effectively save a large amount of design time and reduce the risk of wellbore trajectory collision. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic flowchart of the intelligent design method for the wellbore trajectory of high-density cluster wells provided by an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of the geometric configuration process of the two-dimensional space for two-dimensional trajectory design;

[0038] Figure 3 is a schematic diagram of the random point generation process of the two-dimensional space for two-dimensional trajectory design;

[0039] Figure 4 is a schematic diagram of the automatic path search process for two-dimensional trajectory design;

[0040] Figure 5 is a schematic diagram of path smoothing for two-dimensional trajectory design;

[0041] Figure 6 is a schematic diagram of the two-dimensional anti-collision path for two-dimensional trajectory design;

[0042] Figure 7 is Figure 1 a schematic overall design flowchart of the intelligent design method for the wellbore trajectory of high-density cluster wells in

[0043] Figure 8 is Figure 7 a schematic diagram of the three-dimensional initial parameter setting process in

[0044] Figure 9 is Figure 7 a schematic diagram of the three-dimensional automatic path search process in

[0045] Figure 10 is Figure 7 a schematic diagram of the intelligent anti-collision wellbore trajectory in DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0047] In order to solve the technical problems in the prior art that the wellbore trajectory design relies on experience, the single-well design method is difficult to adapt to the complex scenarios of high-density cluster wells, the intelligent algorithm optimization model is difficult to solve, and there is a lack of multi-adjacent well anti-collision constraints.

[0048] Please refer to Figure 1 , Figure 1 which is a schematic flow chart of the intelligent design method for the wellbore trajectory of high-density cluster wells in an embodiment of the present invention. Since the present invention involves three-dimensional wellbore trajectory design and has a certain degree of difficulty, in this application, first, the technical concept of the present invention is described in a relatively simple manner through two-dimensional trajectory design.

[0049] As Figures 2 - 6 shown, the steps of two-dimensional trajectory design include:

[0050] Step 1: Set the starting point and the target point, and add obstacles to perform automatic feasible path search (as Figure 2 shown);

[0051] Step 2: Use the PRM algorithm to sample random points in the two-dimensional space, and automatically detect whether the random points collide with the obstacles. If a collision occurs, delete the random point and continue to generate random points. Otherwise, retain the random point until the number of random points reaches a certain requirement (as Figure 3 shown);

[0052] Step 3: Traverse all random points and connect nodes with their adjacent random points. If the connection route between the current random point and the adjacent random point collides with the obstacle, eliminate the connection. Otherwise, retain the connection, and finally automatically generate a two-dimensional collision detection route map (as Figure 3 shown);

[0053] Step 4: Use the automatic path search algorithm to search for paths in the two-dimensional space, and perform intelligent obstacle avoidance within the set dogleg range, automatically search for feasible paths, and sort and store the structures according to the path cost (as Figure 4 shown);

[0054] Step 5: Select the optimal path, check its dogleg degree, select the segment whose dogleg degree exceeds the predetermined threshold, and perform local path smoothing on this segment (as Figure 5 shown);

[0055] Step 6: If the smoothed path does not meet the dogleg severity constraint, remove the current path and select the next one. If the storage sequence is empty, increase the random point parameter and reconstruct the roadmap; if the storage sequence is not empty, select the next path for smoothing until the requirements are met, and finally automatically generate a two-dimensional anti-collision path (as Figure 6 shown).

[0056] Similarly, for the intelligent design of wellbore trajectories in three-dimensional space, the following technical solutions can be referred to.

[0057] Please refer to Figure 1 and Figure 7 , the intelligent design method for wellbore trajectories of high-density cluster wells includes:

[0058] S1. Obtain the data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, and mark the wellbore trajectories of the drilled wells as obstacles in the three-dimensional block to construct a three-dimensional collision detection roadmap, where the three-dimensional collision detection roadmap includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles.

[0059] It should be understood that in actual drilling operations, the existing wellbore trajectories are areas that need to be avoided in subsequent drilling, because if a new wellbore collides with an existing wellbore, it will cause serious safety accidents and economic losses. Therefore, marking them as obstacles is the basis for subsequent collision detection.

[0060] In step S1, the method for constructing a three-dimensional collision detection roadmap includes:

[0061] S11. Automatically generate several random points in the three-dimensional block using the probabilistic roadmap algorithm.

[0062] The probabilistic roadmap algorithm (PRM algorithm) is a sampling algorithm for path planning. It constructs a roadmap by randomly sampling in the environment. The spacing of the random points can be set according to actual needs. For example, if more accurate path planning is required, the spacing can be set smaller; if the accuracy requirement is not high, the spacing can be appropriately increased.

[0063] In this embodiment, taking the drilled dense cluster wells as an example, the east direction and north direction of the block are set as the positive x-axis and y-axis respectively, the vertical depth downward is set as the positive z-axis, the maximum and minimum values of the three-dimensional coordinate axes are set, and the three-dimensional coordinates of the wellhead, target point, and the wellbore trajectories of the drilled wells in the block are added, as Figure 8 shown.

[0064] S12. Determine whether each random point collides with an obstacle. If a certain random point does not collide with the obstacle, retain the random point; otherwise, delete it and continue to generate new random points until the number of random points meets a certain sampling quantity requirement.

[0065] In this embodiment, the wellbore trajectories of the densely drilled cluster wells are marked as obstacles. The PRM algorithm is used to automatically generate three-dimensional random points and automatically detect whether the random points collide with the obstacles. Continuously delete the collision points. Next, automatically traverse and connect the neighbor nodes of the random points. During the connection process, automatic collision detection needs to be performed on the connecting line segments. If the line segment collides with the obstacle during the increasing process, that is, the line segment collides with the trajectory of the drilled well, then eliminate the connecting line segment. If there is no collision when reaching the neighbor node during the increasing process from the current node, then connect the line segment. Finally, automatically generate a three-dimensional collision detection route map.

[0066] In step S12, the method for determining whether each random point collides with an obstacle is as follows:

[0067] Determine whether the three-dimensional coordinates of the random point are within the three-dimensional coordinate range of the obstacle. If so, determine that the random point collides with the obstacle; otherwise, determine that the random point does not collide with the obstacle.

[0068] S13. Starting from the wellhead of the new well and ending at the target target point, traverse all random points and connect each random point with its adjacent random points. During the connection process, determine whether the connection route between the current random point and the adjacent random point collides with the obstacle. If there is a collision, eliminate the connection; if there is no collision, retain the connection to obtain a three-dimensional collision detection route map.

[0069] In step S13, the methods for determining whether the connection route between the current random point and the adjacent random point collides with the obstacle include:

[0070] S131. Calculate the direction vector of the line segment between the current random point and the adjacent random point;

[0071] S132. Starting from the current random point, accumulate with the unit vector step length of the direction vector. If the point reached during the increasing process of the line segment is within the coordinate range of the obstacle, determine that the connection route between the current random point and the adjacent random point collides with the obstacle. If all the points reached during the process of increasing from the current random point to the adjacent random point are outside the coordinate range of the obstacle, determine that the connection route between the current random point and the adjacent random point does not collide with the obstacle.

[0072] The specific collision detection method is as follows:

[0073] Suppose the three-dimensional environment is represented by a discrete grid model, and the grid values are marked as obstacles or free space. Define two nodes as A(x0, y0, z0) and B(x1, y1, z1), and the steps to detect whether the straight-line path from A to B is collision-free are as follows:

[0074] (1) Calculate the direction vector v = (v x , v y , v z ), where:

[0075] v x = x1 - x0, v y = y1 - y0, v z = z1 - z0

[0076] Vector modulus Unit vector

[0077] (2) Let the initial step size δ = 1, the maximum distance D max = ∥v∥, and the initial detection point P δ coordinates are:

[0078] P δ = (x0 + δu x , y0 + δu y , z0 + δu z )

[0079] (3) Loop through the following steps until P δ exceeds point B:

[0080] a. Grid coordinate indices (i, j, k), that is:

[0081] i = P δ，x , j = P δ，y , k = P δ，z

[0082] b. Obstacle determination: If the grid (i, j, k) is within the obstacle, return path collision.

[0083] c. Step size increase: Update δ = δ + 1 and calculate the new check point:

[0084] P δ = (x0 + δu x , y0 + δu y , z0 + δu z )

[0085] d. Termination condition: Terminate the loop when ||P δ - A|| ≥ D max .

[0086] (4) If the loop does not detect a collision with an obstacle, it is determined that the path has no collision.

[0087] Specifically, the specific process of generating a three-dimensional collision detection roadmap is as follows:

[0088] Let the three-dimensional workspace be The obstacle area is The feasible region F = W / O. The steps for constructing a three-dimensional collision detection roadmap are as follows:

[0089] (1) Node collision-free sampling

[0090] Generate a set of candidate nodes V candidate ={q i =(x i , y i , z i )|q i ~U(W), i = 1, 2,..., N}, where U(W) is a random distribution.

[0091] (2) Collision detection

[0092] For nodes q i , q j ∈V, if the adjacent edge condition ||q i -q j ||≤d max is satisfied, then the nodes are connected to generate an edge e ij .

[0093] Path collision detection determination: Use the collision detection algorithm (see above for details),

[0094] CollisionFree(q i , q j ) = False

[0095] The path that has not collided with the obstacle is added to the set E.

[0096] e ij ∈E

[0097] (3) Construction of a three-dimensional collision detection roadmap

[0098] Generate a collision-free graph G = (V, E), where: the node set V satisfies q∈F, and the edge set E satisfies

[0099] S2. Use the path search algorithm to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection roadmap. During the search process, it is necessary to judge whether the search path satisfies the dogleg degree constraint, calculate the path cost for the paths that satisfy the dogleg degree constraint in turn, and sort and store the search paths that satisfy the dogleg degree constraint according to the path cost (such asFigure 9 (as shown in); in this embodiment, path cost calculation refers to calculating the length of the path. During actual drilling, the path length is closely related to the drilling cost. Therefore, when designing the wellbore trajectory, it should reach the target point with the shortest path as much as possible.

[0100] In step S2, the method for calculating the dogleg severity of each path in the three-dimensional collision detection roadmap includes:

[0101] S21. Traverse each node of the current path, calculate the well inclination angle and azimuth angle of the previous node of the current node, and then calculate the well inclination angle and azimuth angle of the current node;

[0102] S22. Calculate the dogleg severity of the current node according to the well inclination angles and azimuth angles of the current node and the previous node.

[0103] In step S22, the calculation formula for the dogleg severity is:

[0104]

[0105] Δθ = arccos{cos(α′2 - α′1) - sin(α′1)sin(α′2)[1 - cos(φ′2 - φ′1)]}

[0106]

[0107] where, P1(x1, y1, z1) is the previous node, P2(x2, y2, z2) is the current node, α1 is the well inclination angle of the previous node, is the azimuth angle of the previous node, α2 is the well inclination angle of the P2 node, is the azimuth angle of the P2 node.

[0108] S3. According to the sorting result, select the optimal path, calculate the full-track dogleg severity of this path, smooth the local segments in this path where the dogleg severity exceeds the preset threshold, and then evaluate the smoothed optimal path to determine whether it not only meets the dogleg severity constraint requirements but also can avoid all obstacles. If it meets the dogleg severity constraint and can avoid the obstacles, then determine this smoothed path as the target wellbore trajectory. Otherwise, if the above conditions are not met, according to the sorting result, select the next path and repeat the dogleg severity calculation and smoothing operation, and continue this loop until a smoothed path that meets both the dogleg severity constraint and can avoid the obstacles is found. (as Figure 10 shown)

[0109] In step S3, if all paths cannot meet the conditions of the dogleg severity constraint and avoiding obstacles after smoothing, then regenerate the three-dimensional collision detection roadmap.

[0110] The present invention also provides an intelligent design system for the wellbore trajectory of high-density cluster wells, comprising:

[0111] A three-dimensional collision detection route map generation module, which is used to obtain the data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, mark the wellbore trajectories of the drilled wells as obstacles in the three-dimensional block, and construct a three-dimensional collision detection route map, wherein the three-dimensional collision detection route map includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles;

[0112] An automatic path search module, which is used to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection route map by using a path search algorithm. During the search process, it is necessary to judge whether the searched path meets the dogleg severity constraint, calculate the path cost for the paths that meet the dogleg severity constraint in sequence, and sort and store the searched paths that meet the dogleg severity constraint according to the path cost;

[0113] A path smoothing processing module, which is used to select the optimal path according to the sorting result, calculate the dogleg severity of the entire orbit for this path, smooth the local segments of this path where the dogleg severity exceeds the preset threshold, and then evaluate the smoothed optimal path to judge whether it meets both the dogleg severity constraint requirements and can avoid all obstacles. If it meets the dogleg severity constraint and can avoid the obstacles, the smoothed path is determined as the target wellbore trajectory. Otherwise, if the above conditions are not met, according to the sorting result, the next path is selected to repeat the dogleg severity calculation and smoothing processing operations, and this loop is continuously carried out until a smoothed path that meets both the dogleg severity constraint and can avoid obstacles is found.

[0114] The present invention also provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the intelligent design method for the wellbore trajectory of high-density cluster wells.

[0115] In summary, the present invention automatically constructs a route map from the wellhead position to the target target point for three-dimensional collision detection through a probabilistic roadmap algorithm, and then uses an automatic path search algorithm to perform path search under the condition of meeting the set maximum dogleg severity until the optimal feasible path is found, and performs local smoothing processing on the found path to ensure that the path meets the maximum dogleg severity constraint. This work content has important significance in the intelligent anti-collision trajectory design, especially when adding an additional anti-collision well in high-density cluster wells, which will effectively save a large amount of design time and reduce the risk of wellbore trajectory collision.

[0116] The specific embodiments of the present invention described above do not limit the protection scope of the present invention. Any other corresponding changes and deformations made according to the technical concept of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. An intelligent design method for the wellbore trajectory of high-density cluster wells, characterized in that Including: S1. Obtain data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, mark the wellbore trajectories of the drilled wells as obstacles in a three-dimensional block, and construct a three-dimensional collision detection roadmap, where the three-dimensional collision detection roadmap includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles; S2. Use a path search algorithm to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection roadmap. During the search process, it is necessary to judge whether the search path meets the dogleg severity constraint, calculate the path cost for the paths that meet the dogleg severity constraint in sequence, and sort and store the search paths that meet the dogleg severity constraint according to the path cost; S3. According to the sorting result, select the optimal path, calculate the full trajectory dogleg severity for this path, smooth the local segments where the dogleg severity exceeds the preset threshold in this path, and then evaluate the smoothed optimal path to judge whether it meets both the dogleg severity constraint requirements and can avoid all obstacles. If it meets the dogleg severity constraint and can avoid the obstacles, determine this smoothed path as the target wellbore trajectory. Otherwise, if the above conditions are not met, select the next path according to the sorting result and repeat the dogleg severity calculation and smoothing operation until a smoothed path that meets both the dogleg severity constraint and can avoid the obstacles is found.

2. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 1, characterized in that, In step S1, the method for constructing the three-dimensional collision detection roadmap includes: S11. Automatically generate several random points in the three-dimensional block using the probabilistic roadmap algorithm; S12. Judge whether each random point collides with an obstacle. If a certain random point does not collide with an obstacle, retain the random point. Otherwise, delete it and continue to generate new random points until the number of random points meets a certain sampling quantity requirement. S13. Take the wellhead of the new well as the starting point and the target target point as the ending point, traverse all random points, and connect each random point with its adjacent random points. During the connection process, judge whether the connection route between the current random point and the adjacent random point collides with an obstacle. If a collision occurs, eliminate the connection. If no collision occurs, retain the connection to obtain the three-dimensional collision detection roadmap.

3. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 2, wherein, In step S12, the method for judging whether each random point collides with an obstacle is: Judge whether the three-dimensional coordinates where the random point is located are within the three-dimensional coordinate range of the obstacle. If so, judge that the random point collides with the obstacle. Otherwise, judge that the random point does not collide with the obstacle.

4. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 2, wherein In step S13, the method for judging whether the connection route between the current random point and the adjacent random point collides with an obstacle includes: S131. Calculate the direction vector of the line segment between the current random point and the adjacent random point; S132. Accumulate from the current random point with the unit vector step of the direction vector. If the point reached during the increase of the line segment is within the coordinate range of the obstacle, it is determined that the connection route between the current random point and the adjacent random point collides with the obstacle. If all the points reached during the process of increasing from the current random point to the adjacent random point are outside the coordinate range of the obstacle, it is determined that the connection route between the current random point and the adjacent random point does not collide with the obstacle.

5. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 1, wherein In step S2, the method for calculating the dogleg severity of each path in the three-dimensional collision detection route map is as follows: S21. Traverse each node of the current path, calculate the well inclination angle and azimuth angle of the previous node of the current node, and then calculate the well inclination angle and azimuth angle of the current node. S22. Calculate the dogleg severity of the current node according to the well inclination angles and azimuth angles of the current node and the previous node.

6. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 5, characterized in that, In step S22, the calculation formula for the dogleg severity DLS is: Δθ = arccos{cos(α′2 - α′1) - sin(α′1)sin(α′2)[1 - cos(φ′2 - φ′1)]} Among them, P1(x1, y1, z1) is the previous node, P2(x2, y2, z2) is the current node, α1 is the well deviation angle of the previous node, is the azimuth angle of the previous node, α2 is the well deviation angle of the P2 node, is the azimuth angle of the P2 node.

7. The intelligent design method for the wellbore trajectory of high-density cluster wells according to claim 1, wherein In step S3, if all the paths cannot meet the dogleg severity constraint and the condition of avoiding obstacles after smoothing, a new three-dimensional collision detection route map is regenerated.

8. An intelligent design system for wellbore trajectories of high-density cluster wells, characterized in that, Including: A three-dimensional collision detection route map generation module, which is used to obtain the data of the wellbore trajectories of the drilled wells within a preset distance range around the new well, mark the wellbore trajectories of the drilled wells as obstacles in the three-dimensional block, and construct a three-dimensional collision detection route map, where the three-dimensional collision detection route map includes several paths from the wellhead of the new well to the target target point and can avoid the obstacles. An automatic path search module, which is used to perform path search and intelligent obstacle avoidance on the three-dimensional collision detection route map using a path search algorithm. During the search process, it is necessary to judge whether the search path meets the dogleg severity constraint, calculate the path cost of the paths that meet the dogleg severity constraint in sequence, and sort and store the search paths that meet the dogleg severity constraint according to the path cost. A path smoothing processing module, which is used to select the optimal path according to the sorting result, calculate the full-orbit dogleg severity of this path, smooth the local segments of this path where the dogleg severity exceeds the preset threshold, and then evaluate the smoothed optimal path to judge whether it meets both the dogleg severity constraint requirements and can avoid all obstacles. If it meets the dogleg severity constraint and can avoid the obstacles, this smoothed path is determined as the target wellbore trajectory. Otherwise, if the above conditions are not met, according to the sorting result, select the next path and repeat the operations of dogleg severity calculation and smoothing processing, and continue this cycle until a smoothed path that meets both the dogleg severity constraint and can avoid obstacles is found.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the intelligent design method for the wellbore trajectory of high-density cluster wells described in any one of claims 1 - 7.

Images