Optimization method and system for layout design of complex equipment piping system

Abstract

The present invention provides a method and system for optimizing the layout design of a complex equipment pipeline system, including: step 1: establishing a layout space; step 2: meshing the layout space according to a preset granularity, simplifying the equipment according to the axis-parallel bounding box method, and setting the start and end points of the pipeline; step 3: setting the parameters of the ant colony algorithm; step 4: solving the optimal path for pipeline layout and obtaining the optimal pipeline layout result; step 5: generating a three-dimensional solid model of the pipeline. The present invention proposes an improved scheme for heuristic information, which increases the tendency of ants to move to the end point and the tendency to move by inertia, improves the convergence speed, reduces the formation of elbows, and improves the stability of the pipeline system.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline layout design, and in particular to a method and system for optimizing the layout design of a pipeline system of complex equipment. Background Art

[0002] Pipeline layout design is a key link in the design of complex equipment, which directly affects the various costs and performance of the equipment. At present, the piping design of complex equipment is heavily dependent on the experience of designers, and there are problems such as low design efficiency and unstable quality, which seriously restricts the development of equipment reliability, safety, economy and maintainability. The piping system of complex equipment still faces the key problem of three-dimensional automated layout. The development of optimization algorithms and artificial intelligence technology has made it possible to realize automatic piping layout design, which can improve the efficiency of automatic piping layout design, reduce the burden on designers, and provide a theoretical basis and technical basis for reducing costs, improving quality and increasing efficiency in the construction of complex equipment piping systems.

[0003] The current optimization algorithms for automatic pipeline layout include deterministic algorithms (maze algorithm, Dijkstra algorithm, A* algorithm) and non-deterministic algorithms (heuristic algorithms such as ant colony algorithm, genetic algorithm, particle swarm algorithm, etc.), which solve the given physical, economic, reliability, convenience, appearance and other constraints by converting them into objective functions and constraints. However, in existing studies, the positive feedback mechanism of the ant colony algorithm causes its final convergence result to be limited by the initial solution, resulting in low path search efficiency and easy to fall into local optimality. At the same time, the greed and short-sightedness of the algorithm make it easy to generate redundant "concave pocket" structures when encountering obstacles, resulting in low layout quality.

[0004] Patent document CN115563719A discloses a method for optimizing the layout of a double-layer single pipeline in an aircraft fuel tank. The method generates a point cloud set based on a three-dimensional model of the aircraft fuel tank, extracts the pipeline laying space from the point cloud set based on the three-dimensional coordinates of the two end points of the pipeline and converts it into a three-dimensional grid map, generates an initial solution set of paths that meet the constraints based on the three-dimensional grid map using an ant colony algorithm, generates an initial arm layout plan for each initial path and calculates the length of each arm, uses a genetic algorithm to iteratively optimize to obtain the optimal solution for the arm layout of each path, and uses the plan with the smallest sum of the total arm layout length and the pipeline path length to guide the next round of search, and finally obtains a layout plan with the optimal sum of the pipeline length and the arm length. However, this patent cannot completely solve the existing technical problems, nor can it meet the needs of the present invention. Summary of the invention

[0005] In view of the defects in the prior art, the purpose of the present invention is to provide a method and system for optimizing the layout design of a complex equipment piping system.

[0006] The method for optimizing the layout design of a complex equipment piping system provided by the present invention comprises:

[0007] Step 1: Create layout space;

[0008] Step 2: Decompose the layout space into grids according to the preset granularity, simplify the equipment according to the axis-parallel bounding box method, and set the start and end points of the pipeline;

[0009] Step 3: Set the parameters of the ant colony algorithm;

[0010] Step 4: Solve the optimal path of pipeline layout and obtain the optimal pipeline layout result;

[0011] Step 5: Generate a 3D solid model of the pipeline.

[0012] Preferably, the parameters of the ant colony algorithm include: the population size is 10, the number of iterations is 50, the pheromone constant is 5, the initial values ​​of the upper and lower limits of the pheromone concentration are 10 and 0.1 respectively, and the pheromone volatility coefficient is 0.05.

[0013] Preferably, step 4 comprises:

[0014] Starting from the starting point, the adjacent grids of the current grid in the six directions of up, down, left, right, front, and back are used as the candidate positions for the next move, and the grids inside the obstacles, the grids beyond the layout space, and the grids occupied by other pipelines are removed to construct a feasible point set U;

[0015] According to the set of feasible points, the heuristic information of each feasible point is calculated in turn, considering the distance and position relationship between the end point and the current point, as well as the relationship between the current moving direction and the previous moving direction. The expression is:

[0016]

[0017] Where η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the next node that the ant will reach, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively;

[0018] direction(p i ,p i+1 ,G E ) is the end - point direction coefficient. When the direction of the ant's current movement is towards the end - point, its value is 1; otherwise, its value is 0. The definition of the end - point direction coefficient is as follows:

[0019]

[0020] bend(p i-1 ,p i ,p i+1 ) is the inertia - movement coefficient. When the ant moves straight in the direction of the previous step, its value is 1; otherwise, its value is 0. The definition of the inertia - movement coefficient is as follows:

[0021]

[0022] According to the set of feasible points, calculate the product of the pheromone concentration and the heuristic information of each feasible point in turn, and perform a normalization operation to obtain the probability that the ant transfers from the current position to each feasible point. The expression is:

[0023]

[0024] In the formula, (i, j, k) are the rectangular coordinates of the grid node, τijk is the pheromone concentration of the grid node (i, j, k), ηijk is the heuristic information of the grid node (i, j, k), and α and β are the relative importance degrees of the residual pheromone concentration and the heuristic information on the path respectively;

[0025] According to the calculated transfer probability, adopt the pseudo - random proportional selection strategy to determine the target point of the ant's next movement. The specific process is as follows: Generate a random number q. If q < q0, select the point with the maximum transfer probability as the target point of the next movement; otherwise, perform roulette selection according to the size of the transfer probability to determine the target point of the next movement:

[0026]

[0027] In the formula, q is a random number between 0 and 1, n represents the serial number of the feasible point, N is the number of feasible points in the set of feasible points U, max() returns the maximum value of the input values, roulette() is the roulette function, and q0 is the threshold of the pseudo - random proportional selection strategy. The expression is:

[0028]

[0029] In the formula, iterMax is the set maximum number of iterations;

[0030] The ant moves from the starting point to the next point, repeating the above process until it reaches the end point.

[0031] Preferably, all ants complete the path search, traverse all paths, and optimize the concave pocket structure in the path. The specific process is as follows: first, the path needs to be scanned to detect whether there are redundant concave pockets in the path and their positions. For the detected concave pockets, it is further determined whether these concave pockets can be removed. For the removable concave pockets, the removal operation is performed to obtain the optimized path;

[0032] For the optimized paths, their fitness values ​​are calculated in turn, and the calculation formula is:

[0033] fitness=ω1·f1+ω2·f2+ω3·f3+ω4·f4

[0034] Where f1, f2, f3 and f4 are the pipeline length, number of elbows, energy value and installation distance respectively, and ω1, ω2, ω3 and ω4 are the weights of these four optimization objectives respectively.

[0035] Preferably, the path search and optimization of this iteration is completed, and the pheromone concentration is updated, including pheromone volatilization and pheromone deposition;

[0036] The pheromone is volatilized according to the following formula:

[0037] τ ijk (t+1)=(1-ρ)·τ ijk (t)

[0038] Where, ρ is the pheromone volatility coefficient;

[0039] Pheromone deposition is carried out according to the following formula:

[0040] τ ijk (t+1)=τ ijk (t)+γ·Δτ ijk

[0041] γ=0.99 N-step

[0042]

[0043] Where γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration; N is the total number of steps on a path, and step is the current number of steps;

[0044] The pheromone concentration is dynamically limited so that the pheromone concentration is maintained within the preset range. The expression is:

[0045]

[0046] In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively;

[0047] The lower and upper limits of pheromone concentration vary as follows:

[0048] τ min =τ min0 ×0.9 mod(iter，10)

[0049] τ max =τ max0 ×1.1 mod(iter，10)

[0050] In the formula, iter is the current iteration number, τ min0 and τ max0 are the initial values ​​of the lower and upper limits of pheromone concentration, respectively, and mod(iter, 10) is the remainder of iter divided by 10;

[0051] After the pheromone update of all grids in the layout space is completed, this iteration ends and enters the next iteration; iterates continuously until the number of iterations reaches the set maximum number of iterations; according to the fitness function, the optimal solution generated during the iteration process is determined as the pipeline layout result.

[0052] The complex equipment piping system layout design optimization system provided by the present invention includes:

[0053] Module M1: Establish layout space;

[0054] Module M2: Decompose the layout space into grids according to the preset granularity, simplify the equipment according to the axis-parallel bounding box method, and set the start and end points of the pipeline;

[0055] Module M3: Setting the parameters of the ant colony algorithm;

[0056] Module M4: solve the optimal path of pipeline layout and obtain the optimal pipeline layout result;

[0057] Module M5: Generate pipeline 3D solid model.

[0058] Preferably, the parameters of the ant colony algorithm include: the population size is 10, the number of iterations is 50, the pheromone constant is 5, the initial values ​​of the upper and lower limits of the pheromone concentration are 10 and 0.1 respectively, and the pheromone volatility coefficient is 0.05.

[0059] Preferably, the module M4 comprises:

[0060] Starting from the starting point, the adjacent grids of the current grid in the six directions of up, down, left, right, front, and back are used as the candidate positions for the next move, and the grids inside the obstacles, the grids beyond the layout space, and the grids occupied by other pipelines are removed to construct a feasible point set U;

[0061] According to the set of feasible points, the heuristic information of each feasible point is calculated in turn, considering the distance and position relationship between the end point and the current point, as well as the relationship between the current moving direction and the previous moving direction. The expression is:

[0062]

[0063] Where η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the next node that the ant will reach, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively;

[0064] direction(p i ,p i+1 ,G E ) is the end point direction coefficient. When the ant is currently moving in the direction close to the end point, its value is 1, and in other cases its value is 0. The definition of the end point direction coefficient is as follows:

[0065]

[0066] bend(p i-1 ,p i ,p i+1 ) is the inertial travel coefficient. When the ant keeps moving in a straight line in the direction of the previous step, its value is 1, and in other cases its value is 0. The definition of the inertial travel coefficient is as follows:

[0067]

[0068] According to the set of feasible points, the product of the pheromone concentration and the heuristic information of each feasible point is calculated in turn, and normalized to obtain the probability of the ant transferring from the current position to each feasible point. The expression is:

[0069]

[0070] Where (i, j, k) is the rectangular coordinate of the grid node, τ ijk is the pheromone concentration of the grid node (i, j, k), η ijk is the heuristic information of the grid node (i, j, k), α and β are the relative importance of the residual pheromone concentration and the heuristic information on the path respectively;

[0071] According to the calculated transfer probability, a pseudo-random proportional selection strategy is used to determine the target point for the ant's next move. The specific process is as follows: a random number q is generated. If q < q0, the point with the maximum transfer probability is selected as the target point for the next move; otherwise, a roulette wheel selection is performed according to the size of the transfer probability to determine the target point for the next move:

[0072]

[0073] In the formula, q is a random number between 0 and 1, n represents the serial number of the feasible point, N is the number of feasible points in the feasible point set U, max() returns the maximum value of the input value, roulette() is the roulette function, q0 is the threshold of the pseudo-random proportional selection strategy, and the expression is:

[0074]

[0075] In the formula, iterMax is the maximum number of iterations set;

[0076] The ant moves from the starting point to the next point, repeating the above process until it reaches the end point.

[0077] Preferably, all ants complete the path search, traverse all paths, and optimize the concave pocket structure in the path. The specific process is as follows: first, the path needs to be scanned to detect whether there are redundant concave pockets in the path and their positions. For the detected concave pockets, it is further determined whether these concave pockets can be removed. For the removable concave pockets, the removal operation is performed to obtain the optimized path;

[0078] For the optimized paths, their fitness values ​​are calculated in turn, and the calculation formula is:

[0079] fitness=ω1·f1+ω2·f2+ω3·f3+ω4·f4

[0080] Where f1, f2, f3 and f4 are the pipeline length, number of elbows, energy value and installation distance respectively, and ω1, ω2, ω3 and ω4 are the weights of these four optimization objectives respectively.

[0081] Preferably, the path search and optimization of this iteration is completed, and the pheromone concentration is updated, including pheromone volatilization and pheromone deposition;

[0082] The pheromone is volatilized according to the following formula:

[0083] τ ijk (t+1)=(1-ρ)·τ ijk (t)

[0084] Where, ρ is the pheromone volatility coefficient;

[0085] Pheromone deposition is carried out according to the following formula:

[0086] τ ijk (t+1)=τ ijk (t)+γ·Δτ ijk

[0087] γ=0.99 N-step

[0088]

[0089] Where γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration; N is the total number of steps on a path, and step is the current number of steps;

[0090] The pheromone concentration is dynamically limited so that the pheromone concentration is maintained within the preset range. The expression is:

[0091]

[0092] In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively;

[0093] The lower and upper limits of pheromone concentration vary as follows:

[0094] τ min =τ min0 ×0.9 mod(iter，10)

[0095] τ max =τ max0 ×1.1 mod(iter，10)

[0096] In the formula, iter is the current iteration number, τ min0 and τ max0 are the initial values ​​of the lower and upper limits of pheromone concentration, respectively, and mod(iter, 10) is the remainder of iter divided by 10;

[0097] After the pheromone update of all grids in the layout space is completed, this iteration ends and enters the next iteration; iterates continuously until the number of iterations reaches the set maximum number of iterations; according to the fitness function, the optimal solution generated during the iteration process is determined as the pipeline layout result.

[0098] Compared with the prior art, the present invention has the following beneficial effects:

[0099] (1) The present invention proposes an improved scheme for heuristic information, which increases the tendency of ants to move toward the end point and the tendency to move by inertia, improves the convergence speed, reduces the formation of elbows, and improves the stability of the pipeline system;

[0100] (2) The present invention improves the pheromone volatilization and deposition rules, expands the gap between the better path and the worse path in the fitness function, and accelerates the convergence speed of the algorithm; at the same time, by dynamically processing the pheromone limit, the pheromone can play a good guiding role for the ant colony while reducing the possibility of falling into the local optimal solution;

[0101] (3) The present invention increases the inspection and processing of the path pocket structure, reduces the path length, reduces the number of elbows, and achieves further optimization of manufacturing costs and pipeline reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0103] Figure 1 Flowchart of optimization method for layout of piping system of complex equipment;

[0104] Figure 2 Flowchart for improving the ant colony algorithm;

[0105] Figure 3 This is a schematic diagram of the layout environment after modeling using the mesh decomposition method and the axis-parallel bounding box method;

[0106] Figure 4 Schematic diagram optimized for concave pockets;

[0107] Figure 5 This is a schematic diagram of the layout scene;

[0108] Figure 6 In order to use the method of the present invention Figure 5 Results of pipeline layout for a given layout scenario. DETAILED DESCRIPTION

[0109] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0110] Example 1

[0111] like Figure 1 The present invention provides an intelligent layout method for complex equipment piping under the influence of multiple factors, including: step 1: establishing a layout space; step 2: meshing the layout space according to a preset granularity, simplifying the equipment according to the axis-parallel bounding box method, and setting the starting point and end point of the pipeline; step 3: setting the parameters of the ant colony algorithm; step 4: solving the optimal path for pipeline layout and obtaining the optimal pipeline layout result; step 5: generating a three-dimensional solid model of the pipeline. The present invention realizes rapid design of piping layout while considering factors such as material cost, manufacturing cost, operation convenience, installation cost, functional constraints, physical constraints, assembly constraints, and reliability constraints, and provides a solution for the automatic generation of complex equipment piping schemes in industrial production.

[0112] To achieve the above objectives, the technical solution of the present invention is: to establish a mathematical description of the pipeline layout environment, including free space modeling, obstacle feature extraction and simplified processing, and mathematical expression of objective functions and constraints. These mathematical descriptions will provide a data basis for the pipeline layout algorithm and provide engineering designers with a clear problem definition and solution path; improve the heuristic information of the ant colony algorithm so that it can give full play to the guiding role of the heuristic information on the behavior of the ant colony, improve the efficiency of the algorithm and the quality of the layout; update the pheromone volatilization rules, deposition rules and pheromone limits, and balance the global search ability and rapid convergence ability of the algorithm; through the inspection and processing of the concave pockets, reduce the pipeline manufacturing cost and provide the reliability of the pipeline system. Compared with the existing technology, the effectiveness and efficiency of the present invention have been verified, providing an effective solution for realizing efficient and high-quality automated design of pipeline layout.

[0113] The specific process is:

[0114] S1, based on the traditional ant colony algorithm, redesigned the heuristic information, improved the state transfer strategy, formulated new pheromone update rules, introduced the path optimization mechanism, and proposed an improved ant colony algorithm. Figure 2 Flowchart for improving the ant colony algorithm;

[0115] S2, mesh the layout space according to a certain granularity, simplify various equipment according to the axis-parallel bounding box method, and set the starting point and end point of the pipeline. The layout scene modeled by the mesh decomposition method and the axis-parallel bounding box method is as follows Figure 3 As shown;

[0116] S3, set the parameters of the ant colony algorithm, the population size is 10, the number of iterations is 50, the pheromone constant is 5, the initial values ​​of the upper and lower limits of the pheromone concentration are 10 and 0.1 respectively, and the pheromone volatility coefficient is 0.05;

[0117] S4, at a certain iteration, an ant in the ant colony starts path search;

[0118] S5, starting from the starting point, the adjacent grids of the current grid in the six directions of up, down, left, right, front, and back are used as the candidate positions for the next move, and the grids inside the obstacles, the grids beyond the layout space, and the grids occupied by other pipelines are removed, so as to construct a set of feasible points;

[0119] S6, based on the set of feasible points, calculate the heuristic information of each feasible point in turn, considering the distance and position relationship between the end point and the current point, as well as the relationship between the current moving direction and the previous moving direction:

[0120]

[0121] Where η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the possible next node, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively.

[0122] direction(p i ,p i+1 ,G E ) is the end point direction coefficient. When the ant is currently moving in the direction close to the end point, its value is 1, and in other cases its value is 0. In this way, the ant colony will be more inclined to move towards the end point, shortening the length of the pipeline and reducing the material cost of the pipeline system. The definition of the end point direction coefficient is as follows:

[0123]

[0124] bend(pi-1 ,p i ,p i+1 ) is the inertial travel coefficient. When the ant keeps moving in a straight line in the previous step, its value is 1, and in other cases its value is 0. In this way, the ant colony tends to move by inertia during the movement, which is conducive to reducing the formation of elbows and reducing the manufacturing cost of the pipeline system. The definition of the inertial travel coefficient is as follows:

[0125]

[0126] S7, according to the feasible point set, calculate the product of the pheromone concentration and the heuristic information of each feasible point in turn, and perform normalization operation to obtain the probability of the ant transferring from the current position to each feasible point:

[0127]

[0128] Where, τ is the pheromone concentration of the grid node, η is the heuristic information of the grid node, α and β are the relative importance of the residual pheromone concentration and heuristic information on the path respectively;

[0129] S8, according to the calculated transfer probability, a pseudo-random proportional selection strategy is used to determine the target point for the ant's next move. The specific steps are: generate a random number q, if q < q0, then select the point with the maximum transfer probability as the next move target point. Otherwise, a roulette wheel selection is performed according to the size of the transfer probability to determine the next move target point:

[0130]

[0131] In the formula, q is a random number between 0 and 1, and roulette() is the roulette function. q0 is the threshold of the pseudo-random proportion selection strategy, which is adaptively adjusted according to the following change rules to make the algorithm transition from the exploration stage to the exploitation stage:

[0132]

[0133] S9, the ant moves from the starting point to the next point and repeats the process until it moves to the end point;

[0134] S10, all ants complete path search;

[0135] S11, traverse all paths and optimize the "concave pocket" structure in the path, such as Figure 4As shown. The specific steps are as follows: First, the path needs to be scanned to detect whether there are redundant "pockets" in the path and their locations. For the detected "pockets", it is necessary to further determine whether these "pockets" can be removed. For the removable "pockets", the removal operation is performed to obtain the optimized path;

[0136] S12, for the optimized paths, calculate their fitness values ​​in turn, and the calculation formula is:

[0137] fitness=ω1·f1+ω2·f2+ω3·f3+ω4·f4

[0138] In the formula, f1, f2, f3 and f4 are the pipeline length, number of elbows, energy value and installation distance respectively, and ω1, ω2, ω3 and ω4 are the weights of these four optimization objectives respectively;

[0139] S13, the path search and optimization of this iteration is completed, and the pheromone concentration is updated, including pheromone volatilization and pheromone deposition;

[0140] S14, volatilize the pheromone according to the following formula:

[0141] τ ijk (t+1)=(1-ρ)·τ ijk (t)

[0142] Where, ρ is the pheromone volatility coefficient;

[0143] S15, pheromone deposition is performed according to the following formula:

[0144] τ ijk (t+1)=τ ijk (t)+γ·Δτ ijk

[0145] Where ρ is the pheromone volatility coefficient, γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration. In order to enable the ants to find the end point quickly, the grid nodes closer to the end point on the path will receive more pheromone deposition, that is, they will change according to the following rules:

[0146] γ=0.99 N-step

[0147] Among them, N is the total number of steps of a path, and step is the current step number;

[0148] In order to widen the gap between the better path and the worse path in the fitness function, so that the pheromone concentration on the better path can be significantly increased, thereby accelerating the convergence of the algorithm, the following formula is used to calculate the pheromone concentration increment:

[0149]

[0150] S16, dynamically limiting the pheromone concentration so that the pheromone concentration is maintained within an appropriate range.

[0151]

[0152] In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively;

[0153] S17, the lower and upper limits of pheromone concentration change as follows:

[0154] τ min =τ min0 ×0.9 mod(iter，10)

[0155] τ max =τ max0 ×1.1 mod(oter，10)

[0156] In the formula, iter is the current iteration number, mod(iter, 10) is the remainder of iter divided by 10;

[0157] S18, after the pheromone update of all grids in the layout space is completed, this iteration ends and enters the next iteration;

[0158] S19, iterate continuously until the number of iterations reaches the set maximum number of iterations;

[0159] S20, determining the optimal solution generated in the iteration process as the pipeline layout result according to the fitness function;

[0160] S21, based on the piping system in the ship (for layout scenarios, see Figure 5 ) as an example, the method of the present invention is used for automatic layout, and the result is as follows Figure 6 shown.

[0161] Example 2

[0162] The present invention provides a complex equipment pipeline system layout design optimization system, comprising: module M1: establishing a layout space; module M2: gridding the layout space according to a preset granularity, simplifying the equipment according to an axis-parallel bounding box method, and setting the start and end points of the pipeline; module M3: setting parameters of an ant colony algorithm; module M4: solving the optimal path of the pipeline layout to obtain the optimal pipeline layout result; module M5: generating a three-dimensional solid model of the pipeline.

[0163] The parameters of the ant colony algorithm include: population size is 10, number of iterations is 50, pheromone constant is 5, initial values ​​of upper and lower limits of pheromone concentration are 10 and 0.1 respectively, and pheromone volatility coefficient is 0.05.

[0164] The module M4 comprises:

[0165] Starting from the starting point, the adjacent grids of the current grid in the six directions of up, down, left, right, front, and back are used as the candidate positions for the next move, and the grids inside the obstacles, the grids beyond the layout space, and the grids occupied by other pipelines are removed to construct a feasible point set U;

[0166] According to the set of feasible points, the heuristic information of each feasible point is calculated in turn, considering the distance and position relationship between the end point and the current point, as well as the relationship between the current moving direction and the previous moving direction. The expression is:

[0167]

[0168] Where η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the next node that the ant will reach, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively;

[0169] direction(p i ,p i+1 ,G E ) is the end point direction coefficient. When the ant is currently moving in the direction close to the end point, its value is 1, and in other cases its value is 0. The definition of the end point direction coefficient is as follows:

[0170]

[0171] bend(p i-1 ,p i ,p i+1 ) is the inertial travel coefficient. When the ant keeps moving in a straight line in the direction of the previous step, its value is 1, and in other cases its value is 0. The definition of the inertial travel coefficient is as follows:

[0172]

[0173] According to the set of feasible points, the product of the pheromone concentration and the heuristic information of each feasible point is calculated in turn, and normalized to obtain the probability of the ant transferring from the current position to each feasible point. The expression is:

[0174]

[0175] Where (i, j, k) is the rectangular coordinate of the grid node, τ ijk is the pheromone concentration of the grid node (i, j, k), η ijk is the heuristic information of the grid node (i, j, k), α and β are the relative importance of the residual pheromone concentration and the heuristic information on the path respectively;

[0176] According to the calculated transfer probability, a pseudo-random proportional selection strategy is used to determine the target point for the ant's next move. The specific process is as follows: a random number q is generated. If q < q0, the point with the maximum transfer probability is selected as the target point for the next move; otherwise, a roulette wheel selection is performed according to the size of the transfer probability to determine the target point for the next move:

[0177]

[0178] In the formula, q is a random number between 0 and 1, n represents the serial number of the feasible point, N is the number of feasible points in the feasible point set U, max() returns the maximum value of the input value, roulette() is the roulette function, q0 is the threshold of the pseudo-random proportional selection strategy, and the expression is:

[0179]

[0180] In the formula, iterMax is the maximum number of iterations set;

[0181] The ant moves from the starting point to the next point, repeating the above process until it reaches the end point.

[0182] All ants complete the path search, traverse all paths, and optimize the concave pocket structure in the path. The specific process is as follows: first, the path needs to be scanned to detect whether there are redundant concave pockets in the path and their locations. For the detected concave pockets, it is further determined whether these concave pockets can be removed. For the removable concave pockets, the removal operation is performed to obtain the optimized path;

[0183] For the optimized paths, their fitness values ​​are calculated in turn, and the calculation formula is:

[0184] fitness=ω1·f1+ω2·f2+ω3·f3+ω4·f4

[0185] Where f1, f2, f3 and f4 are the pipeline length, number of elbows, energy value and installation distance respectively, and ω1, ω2, ω3 and ω4 are the weights of these four optimization objectives respectively.

[0186] The path search and optimization of this iteration is completed, and the pheromone concentration is updated, including pheromone volatilization and pheromone deposition;

[0187] The pheromone is volatilized according to the following formula:

[0188] τ ijk (t+1)=(1-ρ)·τ ijk (t)

[0189] Where, ρ is the pheromone volatility coefficient;

[0190] Pheromone deposition is carried out according to the following formula:

[0191] τ ijk (t+1)=τ ijk (t)+γ·Δτ ijk

[0192] γ=0.99 N-step

[0193] In the formula, N is the total number of steps of a path, and step is the current number of steps;

[0194]

[0195] Where γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration;

[0196] The pheromone concentration is dynamically limited so that the pheromone concentration is maintained within the preset range. The expression is:

[0197]

[0198] In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively;

[0199] The lower and upper limits of pheromone concentration vary as follows:

[0200] τ min =τ min0 ×0.9 mod(iter，10)

[0201] τ max =τ max0×1.1 mod(iter，10)

[0202] In the formula, iter is the current iteration number, τ min0 and τ max0 are the initial values ​​of the lower and upper limits of pheromone concentration, respectively, and mod(iter, 10) is the remainder of iter divided by 10;

[0203] After the pheromone update of all grids in the layout space is completed, this iteration ends and enters the next iteration; iterates continuously until the number of iterations reaches the set maximum number of iterations; according to the fitness function, the optimal solution generated during the iteration process is determined as the pipeline layout result.

[0204] Those skilled in the art know that, in addition to implementing the system, device and its various modules provided by the present invention in a purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded microcontrollers by logically programming the method steps. Therefore, the system, device and its various modules provided by the present invention can be considered as a hardware component, and the modules included therein for implementing various programs can also be considered as structures within the hardware component; the modules for implementing various functions can also be considered as both software programs for implementing the method and structures within the hardware component.

[0205] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for optimizing the layout design of a complex equipment piping system, characterized in that: Including: Step 1: Establish a layout space; Step 2: Perform grid decomposition on the layout space according to a preset granularity, simplify the equipment according to the axis-aligned bounding box method, and set the starting point and ending point of the pipeline; Step 3: Set the parameters of the ant colony algorithm; Step 4: Solve the optimal path of the pipeline layout to obtain the optimal pipeline layout result; Step 5: Generate a 3D solid model of the pipeline; The said Step 4 includes: Starting from the starting point, take the adjacent grids in the six directions of up, down, left, right, front, and back of the current grid as the alternative next moving positions, remove the grids inside the obstacles, the grids outside the layout space, and the grids already occupied by other pipelines, so as to construct a set of feasible points U; According to the set of feasible points, calculate the heuristic information of each feasible point in turn, considering the distance and position relationship between the end point and the current point, and the relationship between the current moving direction and the previous moving direction. The expression is: In the formula, η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the next node that the ant will reach, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively; direction(p i ,p i+1 ,G E ) is the end point direction coefficient. When the ant is currently moving in the direction close to the end point, its value is 1, and in other cases its value is 0. The definition of the end point direction coefficient is as follows: bend(p i-1 ,p i ,p i+1 ) is the inertial travel coefficient. When the ant keeps moving in a straight line in the direction of the previous step, its value is 1, and in other cases its value is 0. The definition of the inertial travel coefficient is as follows: According to the set of feasible points, calculate the product of the pheromone concentration and the heuristic information of each feasible point in turn, and perform a normalization operation to obtain the probability that the ant transfers from the current position to each feasible point. The expression is: Where (i, j, k) is the rectangular coordinate of the grid node, τ ijk is the pheromone concentration of the grid node (i, j, k), η ijk is the heuristic information of the grid node (i, j, k), α and β are the relative importance of the residual pheromone concentration and the heuristic information on the path respectively; According to the calculated transfer probability, adopt the pseudo-random proportional selection strategy to determine the target point for the next move of the ant. The specific process is: generate a random number q. If q < q0, select the point with the maximum transfer probability as the target point for the next move; otherwise, perform roulette selection according to the size of the transfer probability to determine the target point for the next move: In the formula, q is a random number between 0 and 1, n represents the serial number of the feasible point, N is the number of feasible points in the set of feasible points U, max() returns the maximum value in the input values, roulette() is the roulette function, and q0 is the threshold of the pseudo-random proportional selection strategy. The expression is: In the formula, iterMax is the set maximum number of iterations; The ant moves from the starting point to the next point, and continuously repeats the above process until it moves to the end point.

2. The complex equipment piping system layout design optimization method according to claim 1 is characterized in that: The parameters of the ant colony algorithm include: the population size is 10, the number of iterations is 50, the pheromone constant is 5, the initial values of the upper and lower limits of the pheromone concentration are 10 and 0.1 respectively, and the pheromone evaporation coefficient is 0.

05.

3. The complex equipment piping system layout design optimization method according to claim 1 is characterized in that: All ants have completed the path search and traversed all paths. Optimize the concave pocket structure in the path. The specific process is: First, scan the path to detect whether there are redundant concave pockets in the path and their positions. For the detected concave pockets, further judge whether these concave pockets can be removed. For the removable concave pockets, perform the removal operation to obtain the optimized path; For the optimized path, calculate its fitness value in turn. The calculation formula is: fitness = ω1·f1 + ω2·f2 + ω3·f3 + ω4·f4 In the formula, f1, f2, f3, and f4 are the pipeline length, the number of elbows, the energy value, and the installation distance respectively, and ω1, ω2, ω3, and ω4 are the weights of these four optimization objectives.

4. The complex equipment piping system layout design optimization method according to claim 3 is characterized in that: The path search and optimization of this iteration are completed, and the pheromone concentration is updated, including two parts: pheromone evaporation and pheromone deposition; Perform pheromone evaporation according to the following formula: t ijk (t+1)=(1ρ)·τ ijk (t) In the formula, ρ is the pheromone evaporation coefficient; Pheromone deposition is carried out according to the following formula: t ijk (t+1)=τ ijk (t)+γ·Δτ ijk γ=0.99 N-step Where γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration; N is the total number of steps on a path, and step is the current number of steps; The pheromone concentration is dynamically limited so that the pheromone concentration is maintained within a preset range. The expression is: In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively; The variation rules of the lower and upper limits of the pheromone concentration are as follows: t min =t min0 ×0.9 mod(iter,10) t max =t max0 ×1.1 mod(iter,10) In the formula, iter is the current iteration number, τ min0 and τ max0 are the initial values ​​of the lower and upper limits of pheromone concentration, respectively, and mod(iter, 10) is the remainder of iter divided by 10; After the pheromone of all grids in the layout space is updated, this iteration ends and enters the next iteration; continuously iterate until the number of iterations reaches the set maximum number of iterations; According to the fitness function, determine the optimal solution generated during the iteration as the pipeline layout result.

5. A complex equipment piping system layout design optimization system, characterized in that: Including: Module M1: Establish a layout space; Module M2: Decompose the layout space into grids according to the preset granularity, simplify the equipment according to the axis-aligned bounding box method, and set the starting and ending points of the pipeline; Module M3: Set the parameters of the ant colony algorithm; Module M4: Solve the optimal path of the pipeline layout to obtain the optimal pipeline layout result; Module M5: Generate a 3D solid model of the pipeline; The said Module M4 includes: Starting from the starting point, take the adjacent grids of the current grid in the six directions of up, down, left, right, front, and back as the alternative next moving positions, remove the grids inside the obstacles, the grids outside the layout space, and the grids already occupied by other pipelines, so as to construct a set of feasible points U; According to the set of feasible points, calculate the heuristic information of each feasible point in turn, considering the distance and position relationship between the end point and the current point, and the relationship between the current moving direction and the previous moving direction. The expression is: In the formula, η is the heuristic information value, p i is the grid node where the ant is currently located, p i-1 is the last node the ant passed, p i+1 is the next node that the ant will reach, G E is the grid corresponding to the end point, dis(p i ,G E ) represents the grid node p i With G E The Manhattan distance between the direction (p i ,p i+1 ,G E ) and bend(p i-1 ,p i ,p i+1 ) are the terminal direction coefficient and the inertial travel coefficient respectively; direction(p i ,p i+1 ,G E ) is the end point direction coefficient. When the ant is currently moving in the direction close to the end point, its value is 1, and in other cases its value is 0. The definition of the end point direction coefficient is as follows: bend(p i -1,p i ,p i+1 ) is the inertial travel coefficient. When the ant keeps moving in a straight line in the direction of the previous step, its value is 1, and in other cases its value is 0. The definition of the inertial travel coefficient is as follows: According to the set of feasible points, calculate the product of the pheromone concentration and the heuristic information of each feasible point in turn, and perform a normalization operation to obtain the probability that the ant transfers from the current position to each feasible point. The expression is: Where (i, j, k) is the rectangular coordinate of the grid node, τ ijk is the pheromone concentration of the grid node (i, j, k), η ijk is the heuristic information of the grid node (i, j, k), α and β are the relative importance of the residual pheromone concentration and the heuristic information on the path respectively; According to the calculated transfer probability, adopt the pseudo-random proportional selection strategy to determine the target point for the ant's next move. The specific process is: generate a random number q. If q < q0, select the point with the maximum transfer probability as the next moving target point; otherwise, perform roulette selection according to the size of the transfer probability to determine the next moving target point: In the formula, q is a random number between 0 and 1, n represents the serial number of the feasible point, N is the number of feasible points in the set of feasible points U, max() returns the maximum value in the input value, roulette() is the roulette function, and q0 is the threshold of the pseudo-random proportional selection strategy. The expression is: In the formula, iterMax is the set maximum number of iterations; The ant moves from the starting point to the next point, and continuously repeats the above process until it moves to the end point.

6. The complex equipment piping system layout design optimization system according to claim 5 is characterized in that: The parameters of the ant colony algorithm include: the population size is 10, the number of iterations is 50, the pheromone constant is 5, the initial values of the upper and lower limits of the pheromone concentration are 10 and 0.1 respectively, and the pheromone evaporation coefficient is 0.

05.

7. The complex equipment piping system layout design optimization system according to claim 5 is characterized in that: All ants have completed the path search and traversed all paths. Optimize the concave pocket structure in the path. The specific process is: First, scan the path to detect whether there are redundant concave pockets in the path and their positions. For the detected concave pockets, further judge whether these concave pockets can be removed. For the removable concave pockets, perform the removal operation to obtain the optimized path; For the optimized path, calculate its fitness value in turn. The calculation formula is: fitness=ω1·f1+ω2·f2+ω3·f3+ω4·f4 Where f1, f2, f3 and f4 are the pipeline length, number of elbows, energy value and installation distance respectively, and ω1, ω2, ω3 and ω4 are the weights of these four optimization objectives respectively.

8. The complex equipment piping system layout design optimization system according to claim 7 is characterized in that: The path search and optimization of this iteration is completed, and the pheromone concentration is updated, including pheromone volatilization and pheromone deposition; The pheromone is volatilized according to the following formula: t ijk (t+1)=(1-ρ)·τ ijk (t) Where, ρ is the pheromone volatility coefficient; Pheromone deposition is performed according to the following formula: t ijk (t+1)=τ ijk (t)+γ·Δτ ijk γ=0.99 N-step Where γ is the pheromone reward coefficient, Δτ ijk =1 / fitness is the increment of pheromone concentration; N is the total number of steps on a path, and step is the current number of steps; The pheromone concentration is dynamically limited so that the pheromone concentration is maintained within the preset range. The expression is: In the formula, τ min and τ max are the lower and upper limits of pheromone concentration, respectively; The lower and upper limits of pheromone concentration vary as follows: t min =t min 0×0.9 mod(iter,10) t max =t max0 ×1.1 mod(iter,10) In the formula, iter is the current iteration number, τ min0 and τ max0 are the initial values ​​of the lower and upper limits of pheromone concentration, respectively, and mod(iter, 10) is the remainder of iter divided by 10; After the pheromone of all grids in the layout space is updated, this iteration ends and enters the next iteration; iterates continuously until the number of iterations reaches the set maximum number of iterations; According to the fitness function, the optimal solution generated in the iterative process is determined as the pipeline layout result.

Images