Method for complex flng ship cabin three-dimensional piping layout

Abstract

The application relates to the technical field of pipelines, in particular to a complex FLNG cabin three-dimensional pipeline layout method, which comprises the following steps: S1, initializing basic parameters; S2, calculating an initial solution by adopting an intelligent initial strategy based on obstacle density, and taking the optimal solution in the initial solution as a current global optimal solution; S3, entering an iteration process, calculating an energy factor, if the energy factor is greater than 1, performing long-distance migration with a probability of 30%, and performing hole digging with a probability of 70%; otherwise, randomly performing foraging or avoiding predators with equal probability; S4, periodically performing an elite reservation and neighborhood search strategy; S5, performing an adaptive orthogonal constraint relaxation strategy and boundary limiting operation; S6, calculating and determining a final global optimal solution, and outputting a final layout result. The application improves the calculation efficiency and convergence speed, and is in line with the actual engineering situation; on the basis of preferentially meeting the orthogonalization feasibility, the engineering cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of pipeline technology, specifically to a method for three-dimensional pipeline layout in complex FLNG ship compartments. Background Technology

[0002] Floating liquefied natural gas (FLNG) systems are a new type of offshore floating production system integrating natural gas liquefaction, storage, and loading / unloading, characterized by high integration, convenience, and reliability. However, compared to typical onshore facilities, liquefaction process equipment and piping systems need to be arranged in a more confined space. The complex internal structure of FLNG ship tanks is intricate, with numerous types of equipment and extremely high requirements for safety and reliability. This presents unprecedented challenges to three-dimensional piping design. In terms of spatial layout, the limited space within FLNG ship tanks must accommodate numerous pipelines with different functions, such as liquefied natural gas transmission pipelines, cooling water pipelines, and fuel oil pipelines. These pipelines not only need to be rationally planned to avoid collisions and interference, but also need to reserve sufficient space for the installation and maintenance of other equipment, making their spatial layout far more difficult than that of ordinary ship compartments. From the perspective of the complexity of the piping system, the piping system in FLNG ship tanks not only includes a large number of straight pipe sections, but also involves numerous bends and other complex structures. There are complex connections between different pipelines, and the pipe diameters, pressure ratings, and characteristics of the transported media vary among pipelines, further increasing the complexity of the design.

[0003] Currently, existing 3D piping design methods have many shortcomings when dealing with the complexities of FLNG ship compartments. Traditional design methods based on human experience are inefficient, rely entirely on the professional knowledge and experience of designers, making it difficult to guarantee layout quality, and resulting in significant differences in design outcomes among different designers. While some design methods employing intelligent algorithms, such as genetic algorithms and ant colony algorithms, have improved design efficiency to some extent, they struggle to fully consider the numerous constraints and complex piping connections in the complex spatial environment and piping systems of FLNG ship compartments. Furthermore, the low quality of orthogonalized piping solutions leads to low feasibility of the design schemes in practical applications. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for three-dimensional piping layout of complex FLNG ship compartments.

[0005] This invention is achieved through the following technical solution: a method for three-dimensional piping layout in complex FLNG ship compartments, comprising the following steps:

[0006] S1. Initialize basic parameters: Initialize the basic parameters of the artificial lemming algorithm, construct a three-dimensional space based on the FLNG compartment deck area and equipment location, and set the corresponding parameters.

[0007] S2. Calculate the initial solution using an intelligent initial strategy based on obstacle density, and take the optimal solution in the initial solution as the current global optimal solution;

[0008] S3. Enter the iteration process and calculate the energy factor. If the energy factor is greater than 1, there is a 30% probability of long-distance migration and a 70% probability of digging a hole; otherwise, there is an equal probability of randomly foraging for food or avoiding predators.

[0009] S4. Periodically employ elite retention and neighborhood search strategies to continuously update the current global optimal solution;

[0010] S5. Execute the adaptive orthogonal constraint relaxation strategy and boundary constraint operation;

[0011] S6. Calculate and determine the final global optimal solution. In the final global optimal solution, each three consecutive dimensions correspond to the coordinates of the pipeline node in three-dimensional space. Reconstruct the solution vector into a pipeline node coordinate matrix and output it.

[0012] The basic parameters for initializing the artificial lemming algorithm in S1 include setting the maximum number of iterations, population size, and number of waypoints.

[0013] A three-dimensional space is constructed based on the FLNG cabin deck area and equipment location, and the map size, start point, end point, and obstacle parameters of the three-dimensional space are set.

[0014] The specific calculation process for the global optimal solution is as follows:

[0015] The fitness value of each solution is calculated using a fitness function based on the pipe length, number of bends, obstacle collision penalty, orthogonality, and smoothness.

[0016] The fitness values ​​are compared, and the solution with the smallest fitness value is the global optimal solution.

[0017] The fitness function is expressed as follows:

[0018] ;

[0019] In the formula, fit is the fitness value of the current solution, and L, W, Obs, Ort and S are the pipe length, number of bends, obstacle collision penalty, orthogonality and smoothness, respectively.

[0020] The intelligent initial strategy based on obstacle density in S2 includes:

[0021] S2-1. An initial solution is generated by combining the generation of pipeline nodes along a straight line with the random generation of pipeline nodes.

[0022] S2-2. Perform orthogonality optimization after generating the initial solution;

[0023] S2-3. Dynamic adjustment based on obstacle density.

[0024] The specific process of generating pipeline nodes along a straight line in S2-1 is as follows: first, calculate the distance from the starting point to the ending point and the average length of each segment of the pipeline node;

[0025] Then, moving towards the endpoint dimension by dimension, and adding random perturbations;

[0026] The specific calculation formula for generating pipe nodes along a straight line is as follows:

[0027] ;

[0028] ;

[0029] In the formula, Let be the j-th dimension of the i-th solution, S and E be the pipeline start and end points respectively, r be the random disturbance, a be the control parameter, j be the current dimension, and norm be the vector norm.

[0030] The specific operation of the orthogonality optimization process in S2-2 is as follows: find the dimension that differs most from the previous point point by point, and keep that dimension the same as the dimension of the previous point.

[0031] In S2-3, the obstacle density is used to dynamically adjust the control parameters. If the obstacle density is greater than 0.1, then a equals 0.5; otherwise, a equals 0.2.

[0032] The predator avoidance phase in S3 employs a pheromone-guided Levy flight search strategy, specifically including:

[0033] The pheromone update formula is as follows:

[0034] ;

[0035] In the formula, For the updated pheromones, The pheromone level before the update is given, and e is the natural constant.

[0036] After the pheromone is updated, the average pheromone concentration in the pipeline is calculated to see if it is greater than the threshold. If it is, an offset operation is performed; otherwise, Levy flight continues.

[0037] In S4, every 10 iterations, the optimal solution in the current iteration is selected for neighborhood search. After 5 neighborhood searches, the optimal solution in the current iteration and the global optimal solution are updated according to the greedy criterion.

[0038] The greedy criterion is: calculate the fitness values ​​of two solutions, and keep and replace the one with the smaller fitness value.

[0039] The adaptive orthogonal constraint relaxation strategy in S5 is as follows: every even-numbered iteration, a dimension is randomly selected and movement is restricted to that dimension.

[0040] During movement, a certain proportion of the dimensions are kept constant by using a relaxation factor. The relaxation factor formula is as follows:

[0041] ;

[0042] Where t is the current iteration number, T max This represents the maximum number of iterations.

[0043] The boundary constraint operation process is as follows: compare the value of a certain dimension with the upper limit of that dimension, and keep the smaller of the two; compare the value of the certain dimension with the lower limit of that dimension, and keep the larger of the two.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] This application realizes the application of the artificial lemming algorithm for automatic pipeline layout in the complex three-dimensional FLNG ship compartment space. This application improves computational efficiency and convergence speed, conforms to actual engineering conditions, and reduces engineering costs while prioritizing the feasibility of orthogonalization.

[0046] This application employs an intelligent initialization strategy based on obstacle density, combining a hybrid approach of straight-line generation and random generation, and ensuring the basic engineering feasibility of the initial path through orthogonality processing. Simultaneously, control parameters are dynamically adjusted according to obstacle density to make the initial solution more adaptable to complex environments and reduce ineffective iterations.

[0047] This application introduces dynamic regulation of energy factors, focusing on long-distance migration and burrowing in the early stages of iteration, and shifting to foraging and predator avoidance in the later stages, achieving an adaptive balance between exploration and development. Specifically, the predator avoidance phase incorporates pheromone-guided levy flight, using pheromone markers to indicate high-quality areas and guide random searches toward the target, thus improving path quality.

[0048] This application designs an adaptive orthogonal constraint relaxation strategy, combined with boundary restrictions, to ensure that the final pipeline fully complies with the FLNG orthogonal layout specifications and spatial boundary requirements.

[0049] This application improves the robustness of the algorithm by using dynamic parameter adjustment and an elite retention strategy for FLNG ship cabin environments with different obstacle densities, and can adapt to the layout requirements of all scenarios from sparse to dense obstacles. Attached Figure Description

[0050] Figure 1 This is a flowchart of the method used in this application;

[0051] Figure 2This is a schematic diagram of the traditional ALA parallel cable layout in Environment 1 of Example 2;

[0052] Figure 3 This is a schematic diagram of the parallel cable layout result of the present invention in environment 1 of Example 2;

[0053] Figure 4 This is a comparison chart of the fitness of conventional ALA in Environment 1 and the method proposed in this invention in Example 2;

[0054] Figure 5 This is a schematic diagram of the traditional ALA parallel cable layout result in Environment 2 of Example 2;

[0055] Figure 6 This is a schematic diagram of the parallel cable layout result of the present invention in environment 2 of Example 2;

[0056] Figure 7 This is a comparison chart of the fitness of conventional ALA in Environment 2 and the method proposed in this invention in Example 2;

[0057] Figure 8 This is a schematic diagram of the traditional ALA parallel cable layout result in Environment 3 of Example 2;

[0058] Figure 9 This is a schematic diagram of the parallel cable layout result of the present invention in environment 3 of Example 2;

[0059] Figure 10 This is a comparison chart of the fitness of conventional ALA in Environment 3 of Example 2 and the method proposed in this invention;

[0060] Figure 11 This is a schematic diagram of the traditional ALA parallel cable layout result in Environment 4 of Example 2;

[0061] Figure 12 This is a schematic diagram of the parallel cable layout result of the present invention in environment 4 of Example 2;

[0062] Figure 13 This is a comparison chart of the fitness of conventional ALA in Environment 4 and the method proposed in this invention in Example 2. Detailed Implementation

[0063] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0064] Example 1

[0065] The Artificial Lemming Algorithm (ALA), an emerging intelligent optimization algorithm, is inspired by four behaviors of lemmings in nature: long-distance migration, burrowing, foraging, and predator avoidance. This algorithm mathematically models these four behaviors to optimize problem-solving, achieving a better balance between exploration and exploitation while maintaining computational efficiency. It effectively addresses challenges such as premature convergence, insufficient exploration, and lack of robustness in high-dimensional, non-convex search spaces.

[0066] Specifically, this algorithm solves complex optimization problems by simulating these behaviors: long-distance migration and burrowing are used for exploring the search space, while foraging and predator avoidance are used for developing and utilizing the already searched space. Furthermore, an energy reduction mechanism is employed to dynamically adjust the balance between exploration and development, enhancing the algorithm's ability to avoid local optima and converge to a global solution.

[0067] The energy factor is a parameter that gradually decreases with the number of iterations. It is used by the ALA algorithm to select between exploration (long-distance migration and burrowing) or exploitation (foraging and predator avoidance). Its expression is as follows:

[0068] ;

[0069] In the formula, rand is a random number between 0 and 1; t is the current iteration number; T max Let E(t) be the maximum number of iterations, and E(t) be the energy factor.

[0070] The Artificial Lemming Algorithm (ALA) includes the following stages:

[0071] In the initialization phase, ALA first generates a random solution within the search space as the initial solution, thus initiating the optimization process. The formula for obtaining the initial population is as follows:

[0072] ;

[0073] Where N represents the number of individuals, i.e., the population size; X i It is the i-th individual in the population; X l and X u These represent the lower and upper bounds of the search space, respectively.

[0074] During long-distance migration, when overpopulation leads to food scarcity, lemmings will randomly undertake long-distance migrations. At this time, lemmings will explore the search space based on their current location and the locations of random individuals within the population, seeking habitats rich in food resources to obtain better survival conditions and resources. It is worth noting that due to the influence of various factors such as the ecological environment, the direction and distance of lemming migration are not fixed. This behavior is modeled, and the following equation is proposed:

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] In the formula, X best The global optimal solution is represented by F, which is used to randomly select the migration direction; BM is the Brownian motion vector, which follows a standard normal distribution; and R is the step size control parameter, used to control the migration distance. and Let i and k be the i-th and k-th solutions in the t-th iteration; i and k are individual indices, and i and k are different. Let be the i-th solution in iteration t+1; N is the population size; dim is the upper limit of dimension, which in three-dimensional space is three times the number of nodes in the pipeline; and t is the current iteration number.

[0080] The burrowing stage involves excavating burrows within the habitat, forming complex tunnels that provide safe shelter and food storage. Lemmings randomly dig new burrows based on the location of their current burrow and the location of random individuals within the population. This design helps them quickly evade predators. Modeling this behavior, the following equation is proposed:

[0081] ;

[0082] ;

[0083] In the formula, L is a random number related to the current iteration number.

[0084] During the foraging phase, lemmings move widely and randomly within their burrows, relying on their keen sense of smell and hearing to locate food sources. Lemmings typically establish a relatively small foraging area within their habitat, the size of which depends on the abundance and availability of food. To maximize their food intake, lemmings will wander randomly within this foraging area. Based on the solution obtained during the burrowing phase and the global optimal solution, a spiral wrapping mechanism is used to model the process, leading to the following equation:

[0085] ;

[0086] ;

[0087] ;

[0088] In the formula, spiral is the spiral shape factor, radius is the foraging radius, and j is the dimension. Let be the j-th dimension of the global optimal solution in the t-th iteration. Let be the j-th dimension of the i-th solution in the t-th iteration.

[0089] In the predator avoidance phase, the modeling focuses on the lemming's avoidance and protective behaviors when encountering danger. Burrows serve as refuges for lemmings. Upon spotting an enemy, lemmings will use their excellent running ability to escape back to their burrows. Simultaneously, lemmings will also employ deceptive maneuvers to evade predators, the corresponding mathematical expressions of which are shown below:

[0090] ;

[0091] ;

[0092] In the formula, G is the escape coefficient; Levy is the Levy flight function.

[0093] Given the outstanding performance of the artificial lemming algorithm in high-dimensional problems and the urgent need for practical engineering, this application improves upon the artificial lemming algorithm by studying an improved artificial lemming algorithm based on orthogonal constraints. This improves the algorithm and enables its application in solving the complex three-dimensional pipeline layout problem of FLNG ship compartments. This is of great significance for improving layout efficiency and quality and reducing engineering costs.

[0094] Specifically, refer to Figure 1 This application proposes a method for the three-dimensional piping layout of complex FLNG ship compartments based on an improved artificial lemming algorithm, which specifically includes the following steps:

[0095] S1. Initialize the basic parameters of the artificial lemming algorithm, construct a three-dimensional space based on the FLNG compartment deck area and equipment location, and set the corresponding parameters.

[0096] The basic parameters for initializing the artificial lemming algorithm in S1 include setting the maximum number of iterations, population size, and number of path points (solution dimension).

[0097] Construct a three-dimensional space and set parameters such as map size, start point, end point, and obstacles based on the FLNG cabin deck area and equipment location.

[0098] S2. Calculate the initial solution using an intelligent initialization strategy based on obstacle density. The optimal solution in the initial solution is then taken as the current global optimal solution. Each solution corresponds to a path, and the current global optimal solution is the current optimal path. The global optimal solution is determined based on the fitness function. In this step, an intelligent initialization strategy based on obstacle density is used, and the data is stored as an N×dim matrix.

[0099] Intelligent initialization strategies based on obstacle density include:

[0100] S2-1. An initial solution is generated by combining the generation of pipeline nodes along a straight line with the random generation of pipeline nodes.

[0101] S2-2. Perform orthogonality optimization after generating the initial solution;

[0102] S2-3. Dynamic adjustment based on obstacle density.

[0103] To improve the breadth of exploration and ensure the orthogonality of the initial solution, this application adopts a combination of generating pipeline nodes along a straight line and randomly generating pipeline nodes to generate the initial solution. In addition, after generating the initial solution, an orthogonality guarantee process is added to ensure that the pipeline results corresponding to the initial solution have a certain degree of orthogonality.

[0104] The specific process of generating pipeline nodes along a straight line in S2-1 is as follows: First, calculate the distance from the starting point to the ending point and the average length of each segment of the pipeline node; then, move towards the ending point dimension by dimension, adding random perturbations. The specific calculation formula is as follows:

[0105]

[0106] ;

[0107] In the formula, S and E are the start and end points of the pipeline, respectively, r is the random disturbance, a is the control parameter, j is the current dimension, and norm is the vector norm. Let be the j-th dimension of the i-th solution.

[0108] Randomly generating pipeline nodes is the initialization formula of the standard ALA algorithm.

[0109] The specific operation of orthogonality optimization is to find the dimension that differs most from the previous point point by point, and keep that dimension the same as the dimension of the previous point. Orthogonality is achieved by forcibly keeping the dimension with the largest difference between the previous point and the current point the same. A point in space can be represented by three-dimensional coordinates. As long as the two dimensions of two adjacent points in the three-dimensional coordinates of three points are consistent, the two line segments formed by the three points are orthogonal.

[0110] ALA faces different cabin environments. To ensure the robustness of the algorithm, pipeline nodes are generated along a straight line. The control parameters are affected by the obstacle density. If the obstacle density is greater than 0.1, then a equals 0.5; otherwise, a equals 0.2.

[0111] S3: Enter the iteration process. If the energy factor is greater than 1, there is a 30% probability of long-distance migration and a 70% probability of digging holes. The purpose of long-distance migration and digging holes is to explore a better solution.

[0112] If the energy factor is less than or equal to 1, then the search for food and the avoidance of predators will proceed randomly with equal probability, in order to explore a better solution.

[0113] The predator avoidance phase employs a pheromone-guided Levy flight search strategy.

[0114] The Levy flight search strategy aims to improve the randomness of the algorithm's search. To achieve a balance between randomness and exploration effectiveness, this application introduces adaptive incremental pheromone guidance, specifically including:

[0115] First, update the pheromones. The pheromone update formula is as follows:

[0116] ;

[0117] In the formula, p is the pheromone, fit is the fitness of the current solution, and e is the natural constant.

[0118] After the pheromone is updated, the average pheromone concentration in the pipeline is calculated to see if it is greater than the threshold. If it is, an offset operation is performed; otherwise, Levy flight continues.

[0119] After the algorithm completes the predator avoidance phase, it performs a pheromone-guided offset. This offset occurs when the updated average pheromone concentration in the pipeline is greater than a threshold (0.1 in this embodiment), as shown in the following formula:

[0120] ;

[0121] In the formula, p mean The average pheromone concentration in the pipeline is calculated by summing the pheromones at each node and dividing by the number of nodes.

[0122] S4: Periodically employ elite retention and neighborhood search strategies to continuously update the current global optimal solution, ensuring the continuity of development.

[0123] The optimal solution often contains more information and should be the focus of mining. To avoid over-mining and getting stuck in local optima, and to balance the time cost of global search and local optimization, periodic execution is adopted. Every 10 iterations, the optimal solution in the current iteration is selected for neighborhood search. After 5 neighborhood searches, the optimal solution in the current iteration and the global optimal solution are updated according to the greedy criterion.

[0124] The greedy criterion is: calculate the fitness values ​​of two solutions based on the fitness function, and select the one with the smaller fitness value for retention and replacement.

[0125] The neighborhood search equation is as follows:

[0126] ;

[0127] In the formula, X pbest X is the optimal solution in the current iteration. neighborThe solution obtained after performing a neighborhood search on the optimal solution in the current iteration.

[0128] S5: Adaptive orthogonal constraint relaxation strategy and boundary restrictions are adopted to ensure that the updated solution during the iteration process can meet the feasibility of pipeline layout.

[0129] Orthogonality is a primary requirement in pipeline design. As a swarm intelligence algorithm, ALA's random search process makes it difficult to guarantee the orthogonality of the pipeline results. Therefore, an adaptive orthogonal constraint relaxation strategy is designed. In the early stages of iteration, orthogonal constraints are relaxed to expand the search space, while in later stages, strict constraints are applied to meet engineering requirements. Specifically, the adaptive orthogonal constraint relaxation strategy involves randomly selecting a dimension every even-numbered iterations and restricting movement to that dimension. During movement, a relaxation factor is used to maintain a certain proportion of the dimension unchanged. The specific control process involves setting a random number between 0 and 1. For each dimension, a random number is generated and compared with the relaxation factor. If the relaxation factor is greater than the random number, the dimension remains unchanged.

[0130] Each iteration constraint reduces the exploratory nature of the algorithm, while an even number of iterations can guarantee orthogonality without losing the ability to freely explore the entire system.

[0131] The relaxation factor formula is as follows:

[0132] .

[0133] The boundary constraint operation process is as follows: compare the value of a certain dimension with the upper limit of that dimension and keep the minimum of the two; compare it with the lower limit of that dimension and keep the maximum of the two.

[0134] S6: Output Results. In this step, the final global optimal solution is calculated and determined. In the final global optimal solution, each three consecutive dimensions correspond to the coordinates of the pipeline nodes in three-dimensional space. The solution vector is reconstructed into a pipeline node coordinate matrix for visualization output.

[0135] The specific calculation process for the global optimal solution involved in the entire pipeline layout process is as follows:

[0136] The fitness value of each solution is calculated using a fitness function based on the pipe length, number of bends, obstacle collision penalty, orthogonality, and smoothness.

[0137] The fitness values ​​are compared, and the solution with the smallest fitness value is the global optimal solution.

[0138] The fitness function is expressed as follows:

[0139] ;

[0140] In the formula, fit is the fitness value of the current solution, and L, W, Obs, Ort and S are the pipe length, number of bends, obstacle collision penalty, orthogonality and smoothness, respectively.

[0141] In summary, the fitness value of each solution can be calculated using the fitness function based on the information of each solution. The smaller the fitness value, the better the solution. The solution with the smallest fitness value is the global optimum.

[0142] The main purpose of the fitness function is to ensure that the algorithm's iterative process is executed in accordance with the expected goals (including pipeline length, number of bends, obstacle collision penalty, orthogonality and smoothness).

[0143] Example 2

[0144] Reference Figures 2-13 To demonstrate the effectiveness of the improvements in this application, experimental simulations were conducted on grid maps with three different obstacles and different start and end points. Figure 4 , Figure 7 , Figure 10 and Figure 13 In the diagram, the red line represents the fitness curve of this invention, and the blue line represents the fitness curve of the traditional ALA.

[0145] The algorithm used in this invention is compiled using MATLAB in a Windows 11 environment. First, a simulation model space for obstacles is established. The equipment models in the layout space can be considered as obstacles where pipelines cannot be placed. The space is divided equally using a grid method, with a grid granularity of 1, a population size N of 30, a maximum number of iterations Tmax of 100, and 12 path points (including the start and end points). Compared with the traditional ALA algorithm, the experimental results are shown in Table 1. Environments 1-4 use different grid sizes for simulation, with the number of grids on the x, y, and z axes being 20×20×10, 20×20×10, 20×20×20, and 50×50×50, respectively. The algorithm proposed in this invention shows good performance in pipeline layout under different complex environments.

[0146] Table 1. Comparison of experimental results between this application and the traditional ALA algorithm.

[0147]

[0148] In summary, based on the artificial lemming algorithm, through improvements to multiple strategies and variable storage design, the standard algorithm is improved into a method that can solve the complex three-dimensional piping design problem of FLNG ship compartments, thereby improving the quality and search efficiency of piping layout in complex FLNG ship compartments.

[0149] The above description is merely an optional embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the content of the present invention under the concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for three-dimensional piping layout in complex FLNG ship compartments, characterized in that, Includes the following steps: S1. Initialize basic parameters: Initialize the basic parameters of the artificial lemming algorithm, construct a three-dimensional space based on the FLNG compartment deck area and equipment location, and set the corresponding parameters. S2. Calculate the initial solution using an intelligent initial strategy based on obstacle density, and take the optimal solution in the initial solution as the current global optimal solution; The intelligent initial strategy based on obstacle density in S2 includes: S2-1. An initial solution is generated by combining the generation of pipeline nodes along a straight line with the random generation of pipeline nodes. S2-2. Perform orthogonality optimization after generating the initial solution; S2-3. Dynamic adjustment based on obstacle density; S3. Enter the iteration process and calculate the energy factor. If the energy factor is greater than 1, there is a 30% probability of long-distance migration and a 70% probability of digging a hole; otherwise, there is an equal probability of randomly foraging for food or avoiding predators. S4. Periodically employ elite retention and neighborhood search strategies to continuously update the current global optimal solution; S5. Execute the adaptive orthogonal constraint relaxation strategy and boundary constraint operation; The adaptive orthogonal constraint relaxation strategy in S5 is as follows: every even-numbered iteration, a dimension is randomly selected and movement is restricted to that dimension. The boundary constraint operation process is as follows: compare the value of a certain dimension with the upper limit of that dimension, and keep the minimum of the two; compare the value of the certain dimension with the lower limit of that dimension, and keep the maximum of the two. S6. Calculate and determine the final global optimal solution. In the final global optimal solution, each three consecutive dimensions correspond to the coordinates of the pipeline node in three-dimensional space. Reconstruct the solution vector into a pipeline node coordinate matrix and output it.

2. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 1, characterized in that, The basic parameters for initializing the artificial lemming algorithm in S1 include setting the maximum number of iterations, population size, and number of waypoints. A three-dimensional space is constructed based on the FLNG cabin deck area and equipment location, and the map size, start point, end point, and obstacle parameters of the three-dimensional space are set.

3. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 1 or 2, characterized in that, The specific calculation process for the global optimal solution is as follows: The fitness value of each solution is calculated based on the pipe length, number of bends, obstacle collision penalty, orthogonality, and smoothness. The fitness values ​​are compared, and the solution with the smallest fitness value is the global optimal solution. The formula for calculating fitness value is as follows: ； In the formula, fit is the fitness value of the current solution, and L, W, Obs, Ort and S are the pipe length, number of bends, obstacle collision penalty, orthogonality and smoothness, respectively.

4. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 1, characterized in that, The specific process of generating pipeline nodes along a straight line in S2-1 is as follows: first, calculate the distance from the starting point to the ending point and the average length of each segment of the pipeline node; Then, it approaches the endpoint dimension by dimension, adding random perturbations; The specific calculation formula for generating pipe nodes along a straight line is as follows: ； ； In the formula, Let be the j-th dimension of the i-th solution, S and E be the starting and ending points of the pipeline, r be the random disturbance, a be the control parameter, j be the current dimension, norm be the vector norm, and dim be the upper limit of the dimension.

5. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 1, characterized in that, The specific operation of the orthogonality optimization process in S2-2 is as follows: find the dimension that differs most from the previous point point by point, and keep that dimension the same as the dimension of the previous point.

6. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 4, characterized in that, In S2-3, the obstacle density is used to dynamically adjust the control parameters. If the obstacle density is greater than 0.1, then a equals 0.5; otherwise, a equals 0.

2.

7. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 3, characterized in that, The predator avoidance phase in S3 employs a pheromone-guided Levy flight search strategy, specifically including: The pheromone update formula is as follows: In the formula, For the updated pheromones, The pheromone level before the update is given, and e is the natural constant. After the pheromone is updated, the average pheromone concentration in the pipeline is calculated to see if it is greater than the threshold. If it is, an offset operation is performed; otherwise, Levy flight continues.

8. The method for three-dimensional piping layout of complex FLNG ship compartments according to claim 3, characterized in that, In S4, every 10 iterations, the optimal solution in the current iteration is selected for neighborhood search. After 5 neighborhood searches, the optimal solution in the current iteration and the global optimal solution are updated according to the greedy criterion. The greedy criterion is: calculate the fitness values ​​of two solutions, and keep and replace the one with the smaller fitness value.

Images