A single-pipeline double-layer optimized layout method inside an aircraft fuel tank

Abstract

The present invention provides a method for optimizing the layout of a single pipeline with a double layer inside an aircraft fuel tank, which relates to the technical field of aircraft pipeline layout design. This method generates a point cloud set based on the three-dimensional model of the aircraft fuel tank. According to the three-dimensional coordinates of the two endpoints of the pipeline, the pipeline laying space is extracted from the point cloud set and converted into a three-dimensional grid map. The ant colony algorithm is used to generate an initial solution set of paths that meet the constraints based on the three-dimensional grid map. An initial arm layout scheme is generated for each initial path and the length of each arm is calculated. The genetic algorithm is used to iteratively optimize to obtain the optimal solution of the arm layout for each path. The scheme with the minimum sum of the total length of the arm layout and the pipeline path length is used to guide the next round of search, and finally an optimal layout scheme with the minimum sum of the pipeline length and the arm length is obtained. The present invention adopts a double-layer optimization framework, using the ant colony algorithm to calculate the pipeline path in the outer layer and the genetic algorithm to calculate the pipeline arm layout scheme in the inner layer, greatly improving the engineering feasibility of the pipeline laying result.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft pipeline layout design, and particularly relates to a method for optimizing the layout of a single pipeline in a double layer in an aircraft fuel tank. Background Art

[0002] The layout design and assembly work of pipelines account for a relatively large proportion in the research and development of complex products, and it is a complicated and time-consuming task. The layout design of pipelines is usually carried out on the basis of the design of product structural parts. In the design process, not only the functional connection of pipelines needs to be considered, but also a reasonable route needs to be determined according to the structural parts and layout space to which the pipelines are attached, and at the same time, the requirements in terms of process, flow resistance, reliability, etc. need to be met. Any unreasonable pipeline design may cause quality problems of the product and may also trigger design changes of a series of other components. After the pipeline layout design is completed, the assembly scheme of the pipeline usually needs to be determined after repeated trial assembly and modification.

[0003] Computer-aided pipeline layout design software usually takes the three-dimensional model of the product as the basis and completes the pipeline layout design through a man-machine interaction method. However, for the layout design work of a large number of pipelines in complex products, the layout design efficiency is still relatively low, which affects the research and development cycle of the product. At the same time, due to the lack of consideration of pipeline support constraints in the existing pipeline layout auxiliary tools, many problems in the pipeline assembly process are difficult to be found in the assembly design stage, resulting in problems such as more pipeline assembly rework and poor reliability.

[0004] Currently, the methods for dealing with pipeline path design and pipeline support arm layout are mainly to carry out support arm layout on the basis of the shortest pipeline path by establishing a certain constraint processing mechanism or rule, or to make the pipeline path adhere to the inner wall or outer surface of the structure of the pipeline laying object as much as possible. However, the pipelines in the aircraft fuel tank generally need to ensure that the sum of the pipeline path length and the support arm length is the shortest and do not always adhere to the inner surface of the fuel tank. And due to the large structural size and complex structure of the aircraft fuel tank, the solution time required by the traditional pipeline layout algorithm is relatively long. In summary, it can be seen that there is no perfect theory and method for the aircraft fuel tank pipeline layout considering the total length of the pipeline path and the support arm at present. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a single-pipeline double-layer optimization layout method for an aircraft fuel tank in view of the deficiencies of the above-mentioned prior art. The method can comprehensively consider the comprehensive influence of the pipeline length and the total length of the pipeline arms on the total mass of the pipeline, which is more in line with the actual engineering requirements. A double-layer optimization framework is adopted. The ant colony algorithm is used in the outer layer to calculate the pipeline path, and the genetic algorithm is used in the inner layer to calculate the pipeline arm layout scheme. Then, the comprehensive results of the path and the arms are fed back to the ant colony algorithm in the outer layer to guide the ant colony to optimize and obtain the comprehensive optimization solution of the pipeline and arm layout. This method considers the influence of the pipeline arm length and greatly improves the engineering feasibility of the pipeline laying result.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0007] A single-pipeline double-layer optimization layout method for an aircraft fuel tank. According to the three-dimensional model of the aircraft fuel tank, a point cloud set of the three-dimensional model of the aircraft fuel tank is generated. According to the three-dimensional coordinates of the two ends of the pipeline, the pipeline laying space is extracted from the point cloud set and converted into a three-dimensional grid map. According to the three-dimensional grid map, the ant colony algorithm is used to generate an initial solution set with the shortest path that satisfies the obstacle avoidance constraint, the minimum bending radius constraint, and the minimum straight line segment constraint. According to the initial solution set, an initial arm layout scheme that satisfies the span constraint is generated for each path. According to the initial scheme, the length of each arm is calculated. Then, the genetic algorithm is used to iteratively optimize to obtain the optimal solution of the arm layout for each path. Then, the scheme with the minimum sum of the total length of the arm layout and the pipeline path length, that is, the scheme with the lightest total weight of the pipeline and the arms, is used as the optimal scheme for this iteration to guide the next round of search, and finally an optimal layout scheme with the minimum sum of the pipeline length and the arm length is obtained. The specific steps are as follows:

[0008] Step 1: Establish a three-dimensional model of the aircraft fuel tank, generate a point cloud set according to the three-dimensional model, and set the endpoint coordinates of the pipeline to be laid;

[0009] Step 2: According to the point cloud set generated by the three-dimensional model and the input pipeline endpoint coordinates, extract a local point cloud subset from the point cloud set and convert it into a three-dimensional grid map;

[0010] Step 3: Generate M groups of path initial solutions including the shortest path initial solution that satisfies the obstacle avoidance constraint, the minimum bending radius constraint, and the minimum straight line segment constraint according to the three-dimensional grid map;

[0011] Step 4: Generate N groups of initial arm layout schemes that satisfy the span constraint for each initial path, and calculate the length of each arm;

[0012] Step 5: Merge all N groups of initial arm layout schemes into a group of initial populations, sort the initial populations according to the arm layout optimization goal, and impose a penalty value on the fitness of the individuals that do not meet the pipeline span constraint;

[0013] Step 6: Cross, mutate, and sort the population according to the population sorting result to obtain a new population, and obtain the optimal pipeline path and arm layout scheme through κ iterations;

[0014] Step 7: Repeat Steps 5 - 6 for M groups of initial path solutions. Calculate the fitness value of each path according to the fitness function of the bidirectional single - objective ant colony algorithm and sort them according to the global optimization goal; Update the pheromone of the population according to the population sorting result to guide the ant colony to perform a new round of optimization, and obtain the comprehensive optimal scheme of the optimal pipeline path and arm layout through T iterations, where T is the preset maximum number of iterations;

[0015] Step 8: Layout the pipeline in the aircraft fuel tank according to the optimal pipeline path obtained in Step 7 to generate a 3D model, which is used as the best path scheme for the pipeline path - arm layout in the fuel tank.

[0016] The beneficial effects of adopting the above - mentioned technical solution are as follows: The single - pipeline double - layer optimization layout method for the aircraft fuel tank provided by the present invention generates a point cloud set according to the 3D model of the aircraft fuel tank, extracts the pipeline laying space in the point cloud set and converts it into a 3D grid map according to the 3D spatial positions of the pipeline endpoints (starting and ending points), uses the ant colony algorithm to generate an initial solution set with the shortest path that meets the obstacle - avoidance constraint, minimum bending radius constraint, and minimum straight - line segment constraint according to the 3D grid map, generates an initial arm layout scheme that meets the span constraint for each path according to the initial solution set, calculates the length of each arm according to the initial scheme, then uses the genetic algorithm to iteratively optimize to obtain the optimal solution of the arm layout for each path, and then takes the scheme with the minimum sum of the total arm layout length and the pipeline path length as the optimal scheme for this iteration to guide the next round of iteration, and finally obtains the layout scheme with the optimal sum of the pipeline length and the arm length to guide the layout of the pipeline path in the fuel tank. The present invention is essentially different from the traditional pipeline layout method. This method can comprehensively consider the comprehensive influence of the pipeline length and the total length of the pipeline arms on the total pipeline mass, and better meets the requirements of engineering practice. It adopts a double - layer optimization framework, uses the ant colony algorithm to calculate the pipeline path in the outer layer, uses the genetic algorithm to calculate the pipeline arm layout scheme in the inner layer, and then feeds back the comprehensive result of the path and the arm to the ant colony algorithm in the outer layer to guide the ant colony to find the comprehensive optimization solution of the pipeline and arm layout. This method considers the influence of the pipeline arm length and greatly improves the engineering feasibility of the pipeline laying result. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is the flow chart of the single - pipeline double - layer optimization layout method for the aircraft fuel tank in the embodiment of the present invention;

[0018] Figure 2 is the visualization image of the 3D grid map extracted and converted by using the local point cloud subset method in the embodiment of the present invention; wherein,

[0019] Figure 3 It is the sub - flowchart for generating the boom layout and calculating the boom length in the embodiment of the present invention;

[0020] Figure 4 It is the visualization result diagram generated by Siemens NX software for the optimal path - boom layout scheme obtained by using the method of the present invention in the embodiment of the present invention. Specific embodiments

[0021] The following combines the accompanying drawings and embodiments to further describe in detail the specific embodiments of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0022] To solve the technical problem of the aircraft fuel tank pipeline layout regarding the comprehensive pipeline length and boom length, the present invention proposes a single - pipeline double - layer optimization layout method for the aircraft fuel tank. The specific principle is described as follows: Based on the physical structure of the aircraft fuel tank, a 3D model is established using 3D drawing software (such as Siemens NX software), and then the point cloud information of the 3D model of the aircraft fuel tank is generated using Geomagic Wrap software. The local sub - point cloud participating in the solution in the 3D point cloud data is converted into a 3D grid map using MATLAB software and input information. The ant colony algorithm is used to obtain the initial path, and an initial solution set of boom layouts for each path is randomly generated. On this basis, the multi - objective genetic algorithm is used with all the initial boom layout solutions as the initial population. Then, according to the pre - designed fitness function, the initial population is sorted, and in this order, the boom solutions in the population are iteratively updated using the crossover and mutation methods of the multi - objective genetic algorithm to complete the boom layout optimization process. Then, the boom optimization solution and the corresponding path length are combined to evaluate the fitness of the current pipeline in the population. All the initial paths obtained by the aforementioned ant colony algorithm are calculated in turn and the fitness is evaluated, and the population is updated to finally complete the iteration, which is used to guide the layout of the pipeline path in the fuel tank.

[0023] As Figure 1 shown, the method of this embodiment generates a point cloud set according to the 3D model of the aircraft fuel tank, extracts the pipeline laying space in the point cloud set and converts it into a 3D grid map according to the 3D coordinates of the two endpoints (start and end points) of the pipeline. According to the 3D grid map, the ant colony algorithm is used to generate an initial solution set with the shortest path that satisfies the obstacle - avoidance constraint, the minimum bending radius constraint, and the minimum straight - line segment constraint. For each path in the initial solution set, an initial boom layout solution that satisfies the span constraint is generated, and the length of each boom is calculated. Then, the genetic algorithm is used to iteratively optimize to obtain the optimal boom layout solution for each path. Then, the solution with the minimum sum of the total boom layout length and the pipeline path length (i.e., the solution with the lightest total weight of the pipeline and the boom) is used as the optimal solution for this iteration to guide the next - round search, and finally, the layout solution with the optimal sum of the pipeline length and the boom length is obtained, which is specifically described as follows.

[0024] Step 1: Establish a 3D model of the aircraft fuel tank, generate a point cloud set based on the 3D model, and set the endpoint coordinates of the pipeline to be laid.

[0025] Step 2: According to the point cloud set generated from the 3D model and the input pipeline endpoint coordinates, extract a local point cloud subset within the point cloud set and convert it into a 3D grid map, as Figure 2 shown. The specific method is as follows:

[0026] Step 2.1: Calculate the maximum value P max and minimum value P min in the three directions of the X, Y, and Z axes of the 3D space required for pipeline layout according to formula (1);

[0027]

[0028] In the formula, x s and x e are the coordinates of the two endpoints of the pipeline in the X-axis direction, y s and y e are the coordinates of the two endpoints of the pipeline in the Y-axis direction, z s and z e are the coordinates of the two endpoints of the pipeline in the Z-axis direction;

[0029] Step 2.2: Obtain the number of grids s in the three directions of the X, Y, and Z axes of the 3D grid map according to formula (2);

[0030] s = round(max(P i ) - min(P min )) / ψ (2)

[0031] In the formula, ψ is the ratio between the size of a single grid and the actual size, P i is any coordinate value in the 3D point cloud set of the aircraft fuel tank that is less than or equal to P max and greater than or equal to P min , and round() is the rounding operation function;

[0032] Step 2.3: Calculate the 3D grid coordinates Grid i of P k in the 3D grid map in sequence using formula (3);

[0033] Grid k = round(P i - P min ) / ψ (3)

[0034] Step 2.4: If point P iIf the point coordinates after proportional conversion belong to a certain cell, then set the cell to 1, otherwise set it to 0. The obtained grid map result is as Figure 2 shown.

[0035] Step 3: Generate M sets of initial path solutions that include the initial solution of the shortest path, satisfy the obstacle avoidance constraint, the minimum bending radius constraint, and the minimum straight line segment constraint according to the three-dimensional grid map;

[0036] For the three-dimensional grid map, use the bidirectional single-object ant colony algorithm to obtain a set of initial paths. The heuristic function of the bidirectional single-object ant colony algorithm is:

[0037]

[0038] where ED l is the Euclidean distance from the l-th candidate point to the pipeline end point, MD l is the Manhattan distance from the l-th candidate point to the pipeline end point, and sigmod() is the normalization function.

[0039] Step 4: Generate N sets of initial boom layout solutions that satisfy the constraints according to each initial path, and calculate the length of each boom. The specific method is as follows:

[0040] Step 4.1: For the M sets of initial path solutions generated in Step 3, sequentially select a set of paths from the M sets of initial paths as the input and set the maximum span value L max_span between boom nodes;

[0041] Step 4.2: Extract all the nodes that can be used as booms from the path nodes and sort them. Let the first and last path nodes that can be used as boom nodes starting from the starting point be P s , P e ;

[0042] Step 4.3: Randomly select a point P s from the entire range of optional boom nodes {P e , P k};

[0043] Step 4.4: Calculate the pipeline lengths L k from P s to the first and last nodes P e , P s , L e respectively;

[0044] Step 4.5: If L s > L max_span and L e > L max_span , then return to Step 4.2 to reselect a point; if L s <= L max_span and Le <= L max_span , then go to step 4.6; if L s <= L max_span and L e > L max_span , then save P k to the boom layout scheme, and narrow the selection range of the boom nodes in the next round, that is, let P s = P k ; if L s > L max_span and L e <= L max_span , then save P k to the boom layout scheme, and narrow the selection range of the boom nodes in the next round, that is, let P e = P k ;

[0045] Step 4.6: Repeat steps 4.3 to 4.5 until all optional boom nodes are judged, and obtain an initial boom layout scheme that satisfies the span constraint;

[0046] Step 4.7: According to the initial boom layout scheme, calculate the length of each boom and sum them up;

[0047] Step 4.8: Loop and execute steps 4.1 to 4.7, loop N times to obtain all N groups of initial boom layout schemes.

[0048] In this embodiment, M = 50, N = 100.

[0049] Step 5: Combine all N groups of initial boom layout schemes into a group of initial populations, sort the initial populations according to the boom layout optimization goal, and impose a penalty value on the fitness corresponding to the individuals that do not meet the pipeline span constraint; the specific method is as follows:

[0050] Step 5.1: Combine all N groups of initial boom layout schemes into a group of initial populations as the initial solution of the genetic algorithm;

[0051] Step 5.2: Establish the optimization goal f'(x) of the total boom length and the optimization goal f''(x) of minimizing the span constraint penalty, as shown in the following formula:

[0052]

[0053] where, L i is the length of the i-th section of the pipeline, and n is the number of sections of the pipeline; J j is the boom length of the j-th node in the pipeline boom layout scheme, and m is the total number of boom nodes on the pipeline;

[0054] Step 5.3: Calculate the fitness of each solution in the initial population according to the optimization objectives f'(x) and f”(x), and perform non-dominated sorting on the fitness values;

[0055] Step 5.4: For individuals that do not meet the pipeline span constraint, impose a penalty factor M on the fitness of the optimization objective with the minimum span constraint penalty. k , and the penalty factor is shown in the following formula:

[0056]

[0057] In the formula, g i (x) is the difference between the span of two adjacent arm nodes in each arm layout scheme and the maximum span allowed between two arm nodes of the pipeline; h i (x) is the difference between the span of two adjacent arm nodes in each arm layout scheme and the minimum span allowed between two arm nodes of the pipeline.

[0058] Step 6: According to the population sorting results, perform crossover, mutation, and sorting on the population to obtain a new population, and obtain the optimal pipeline path arm layout scheme through κ iterations; the specific method is as follows:

[0059] Step 6.1: Determine whether the length of one of the two individuals in the population is 1. If not, continue to Step 6.2; otherwise, go to Step 6.5;

[0060] Step 6.2: For the first individual in the population, randomly select a node; for the second individual in the population, randomly select a node;

[0061] Step 6.3: According to the nodes obtained in Step 6.2, perform crossover on the first individual and the second individual in the population to generate new individuals;

[0062] Step 6.4: Determine whether the nodes in the new individuals generated after crossover according to Step 6.3 are all unique. If they are, execute Step 6.6; if not, filter out the duplicate values in the new individuals, and then execute Step 6.5;

[0063] Step 6.5: If the length of the first individual is 1, randomly select a node from the second individual to perform a crossover operation with the first individual; if the length of the second individual is 1, randomly select a node from the first individual to perform a crossover operation with the second individual;

[0064] Step 6.6: According to the new individuals obtained in Step 6.3, randomly select a node to perform a mutation operation. If there is a segment in the mutated new individual that does not meet the span constraint, force the mutated node to be selected from the segment that does not meet the span constraint;

[0065] Step 6.7: Repeat Steps 6.1 to 6.6 until all individuals in the initial population have completed crossover and mutation operations, i.e., the population update is completed;

[0066] Step 6.8: Calculate the fitness of the updated population according to formula (5) and perform non-dominated sorting;

[0067] Step 6.9: Use the population after non-dominated sorting in Step 6.8 as the initial population for the next iteration. Repeat Steps 6.1 to 6.8, perform κ iterations. Take the individual combination corresponding to the minimum value of the optimization objective of the evaluation function, i.e., formula (5), as the optimal pipeline path and arm layout scheme.

[0068] Step 7: Repeat Steps 5 - 6 for the M initial path solutions. Calculate the fitness value of each path according to the fitness function of the two-way single-objective ant colony algorithm based on the global optimization objective and sort them; Update the pheromone of the population according to the current population sorting result to guide the ant colony to perform a new round of optimization. Obtain the comprehensive optimal scheme of the optimal pipeline path and arm layout through T iterations, where T is the preset maximum number of iterations;

[0069] The fitness function of the two-way single-objective ant colony algorithm is:

[0070]

[0071] In the formula, L i is the length of the i-th section of the pipeline, and n is the number of pipeline sections; J j is the length of the arm at the j-th node in the pipeline arm layout scheme, and m is the total number of arm nodes on the pipeline. ω1 and ω2 are weight coefficients respectively.

[0072] Step 8: Layout the pipeline in the aircraft fuel tank according to the optimal pipeline path in Step 7 to generate a 3D model, which is used as the best path scheme for the pipeline path - arm layout in the fuel tank. As Figure 4 shown, it is the visualization result diagram generated by Siemens NX software for the optimal path - arm layout scheme obtained by the method of the present invention.

[0073] The present invention is essentially different from the traditional pipeline layout method. This method can comprehensively consider the comprehensive influence of the pipeline length and the total length of the pipeline arms on the total mass of the pipeline, which better meets the requirements of engineering practice. It adopts a double-layer optimization framework. The ant colony algorithm is used in the outer layer to calculate the pipeline path, and the genetic algorithm is used in the inner layer to calculate the pipeline arm layout scheme. Then, the comprehensive result of the path and the arm is fed back to the ant colony algorithm in the outer layer to guide the ant colony to optimize and obtain the comprehensive optimization solution of the pipeline and arm layout. This method considers the influence of the pipeline arm length, greatly improving the engineering feasibility of the pipeline laying result.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope defined by the claims of the present invention.

Claims

1. A single - pipeline double - layer optimized layout method in an aircraft fuel tank, characterized in that: The method includes the following steps: Step 1: Establish a three-dimensional model of the aircraft fuel tank, generate a point cloud set based on the three-dimensional model, and set the endpoint coordinates of the pipeline to be laid. Step 2: According to the point cloud set generated from the three-dimensional model and the input pipeline endpoint coordinates, extract a local point cloud subset within the point cloud set and convert it into a three-dimensional grid map. Step 3: Generate M sets of initial path solutions that include the initial solution of the shortest path, satisfy the obstacle avoidance constraint, the minimum bending radius constraint, and the minimum straight segment constraint according to the three-dimensional grid map. Step 4: Generate N sets of initial boom layout solutions that satisfy the span constraint according to each initial path, and calculate the length of each boom. The specific method is as follows: Step 4.1: For the generated M groups of initial path solutions, sequentially select one group of paths from the M groups of initial paths as the input and set the maximum span value L between the boom nodes max_span ; Step 4.2: Extract all the nodes that can serve as the support arm from the path nodes and sort them. Let the first and the last path nodes that can serve as the support arm nodes starting from the starting point be P s , P e ; Step 4.3: Randomly select a point P from the entire range of optional arm nodes {P s , P e}; k ; Step 4.4: Calculate P respectively k to the start and end nodes P s 、P e of the pipeline length L s 、L e ; Step 4.5: If L s > L max_span and L e > L max_span , then return to Step 4.2 to reselect points; if L s <= L max_span and L e <= L max_span , then go to Step 4.6; if L s <= L max_span and L e > L max_span , then save P k to the boom layout plan, and narrow the range of boom node selection for the next round, that is, let P s = P k ; if L s > L max_span and L e <= L max_span , then save P k to the boom layout plan, and narrow the range of boom node selection for the next round, that is, let P e = P k ; Step 4.6: Repeat Steps 4.3 to 4.5 until all optional boom nodes are judged, and obtain a set of initial boom layout solutions that satisfy the span constraint. Step 4.7: Calculate the length of each boom according to the initial boom layout solution and sum them up. Step 4.8: Loop through Steps 4.1 to 4.7, loop N times to obtain all N sets of initial boom layout solutions. Step 5: Combine all N sets of initial boom layout solutions into a set of initial populations, sort the initial populations according to the boom layout optimization goal, and impose a penalty value on the fitness of the individuals that do not meet the pipeline span constraint. Step 6: Cross, mutate, and sort the population according to the population sorting result to obtain a new population, and obtain the optimal pipeline path boom layout solution through κ iterations. Step 7: Repeat Steps 5 - 6 for the M sets of initial path solutions, calculate the fitness value of each path according to the global optimization goal through the fitness function of the bidirectional single-objective ant colony algorithm and sort them; update the pheromone of the population according to the population sorting result, guide the ant colony to perform a new round of optimization, and obtain the comprehensive optimal solution of the optimal pipeline path and boom layout through T iterations, where T is the preset maximum number of iterations. Step 8: Layout the pipeline in the aircraft fuel tank according to the optimal pipeline path obtained in Step 7 to generate a three-dimensional model, which is used as the best path solution for the pipeline path - boom layout in the fuel tank.

2. The single - pipeline double - layer optimized layout method in an aircraft fuel tank according to claim 1, characterized in that: The specific method of Step 2 is as follows: Step 2.1: Calculate the maximum value P and minimum value P of the three-dimensional space required for the pipeline layout in the three directions of the X, Y, and Z axes according to formula (1). max and minimum value P min . where x s , x e are the coordinates of the two endpoints of the pipeline in the X-axis direction, y s , y e are the coordinates of the two endpoints of the pipeline in the Y-axis direction, z s , z e are the coordinates of the two endpoints of the pipeline in the Z-axis direction; Step 2.2: Obtain the grid number s in the three directions of the X, Y, and Z axes of the three-dimensional grid map according to formula (2). s = round(max(P i ) - min(P min )) / ψ(2) where ψ is the ratio between the size of a single grid and the actual size, and P i is any coordinate value in the 3D point cloud concentration of the aircraft fuel tank that is less than or equal to P max and greater than or equal to P min is a point, and round() is a rounding operation function; Step 2.3: Calculate P successively using formula (3) i for the three-dimensional grid coordinates Grid in the three-dimensional grid map k ; Grid k = round(P i - P min ) / ψ(3) Step 2.4: If the point coordinates of point P i belong to a certain cell after proportional conversion, set the cell to 1; otherwise, set it to 0.

3. The single - pipeline double - layer optimized layout method in an aircraft fuel tank according to claim 2, characterized in that: In Step 3, for the three-dimensional grid map, use the bidirectional single-objective ant colony algorithm to obtain a set of initial paths, and obtain M sets of path initial solutions as the initial solution set. The heuristic function of the bidirectional single-objective ant colony algorithm is: where ED l is the Euclidean distance from the l-th candidate point to the pipeline end point, MD l is the Manhattan distance from the l-th candidate point to the pipeline end point, and sigmod() is the normalization function.

4. The single - pipeline double - layer optimized layout method in an aircraft fuel tank according to claim 3, characterized in that: The specific method of Step 5 is as follows: Step 5.1: Combine all N sets of initial boom layout solutions into a set of initial populations as the initial solution of the genetic algorithm. Step 5.2: Establish the optimization goal f'(x) of the total boom length and the optimization goal f”(x) of the minimum span constraint penalty, as shown in the following formula: where L i is the length of the i-th section of the pipeline, and n is the number of pipeline sections; J j is the length of the j-th node's support arm in the pipeline support arm layout scheme, and m is the total number of support arm nodes on the pipeline; Step 5.3: Calculate the fitness of each solution in the initial population according to the optimization goals f'(x) and f”(x), and perform non-dominated sorting on the fitness values. Step 5.4: For individuals that do not meet the pipeline span constraint, apply a penalty factor M to the fitness of the minimum optimization objective of the span constraint penalty k , and the penalty factor is shown in the following formula: where g i (x) is the difference between the span of two adjacent arm nodes in each arm layout scheme and the maximum span allowed between two arm nodes of the pipeline; h i (x) is the difference between the span of two adjacent arm nodes in each arm layout scheme and the minimum span allowed between two arm nodes of the pipeline.

5. The single-pipeline double-layer optimized layout method in the aircraft fuel tank according to claim 4, wherein: The specific method of Step 6 is as follows: Step 6.1: Judge whether one of the lengths of the two individuals in the population is 1. If not, continue to Step 6.2; otherwise, go to Step 6.

5. Step 6.2: For the first individual in the population, randomly select a node; for the second individual in the population, randomly select a node; Step 6.3: Based on the nodes obtained in Step 6.2, perform crossover on the first and second individuals in the population to generate new individuals; Step 6.4: Determine whether the nodes in the new individuals generated after crossover according to Step 6.3 all satisfy uniqueness. If they do, execute Step 6.6; if not, screen out the duplicate values in the new individuals, and then execute Step 6.5; Step 6.5: If the length of the first individual is 1, randomly select a node from the second individual to perform a crossover operation with the first individual; if the length of the second individual is 1, randomly select a node from the first individual to perform a crossover operation with the second individual; Step 6.6: Based on the new individuals obtained in Step 6.3, randomly select a node to perform a mutation operation. If there is a segment in the mutated new individual that does not satisfy the span constraint, force the mutated node to be selected from the segment that does not satisfy the span constraint; Step 6.7: Repeat Steps 6.1 to 6.6 until all individuals in the initial population have completed crossover and mutation operations, i.e., complete population update; Step 6.8: Calculate the fitness of the updated population according to formula (5) and perform non-dominated sorting; Step 6.9: Use the population after non-dominated sorting in Step 6.8 as the initial population for the next iteration. Repeat Steps 6.1 to 6.8 and perform κ iterations. Take the individual combination corresponding to the minimum value of the optimization objective of the evaluation function, i.e., formula (5), as the optimal pipeline path support arm layout scheme.

6. The single-pipeline double-layer optimized layout method in the aircraft fuel tank according to claim 5, wherein: The fitness function of the two-way single-objective ant colony algorithm in Step 7 is as follows: where L i is the length of the i-th section of the pipeline, and n is the number of pipeline sections; J j is the length of the arm of the j-th node in the pipeline arm layout scheme, m is the total number of arm nodes on the pipeline, and ω1 and ω2 are weight coefficients respectively.

Images