Automatic Layout Calculation Method for Ship Pipelines

Abstract

The present invention provides a method for calculating the automatic layout of ship pipelines, comprising the steps of: S1: analyzing and simplifying a ship cabin, extracting the spatial geometric outline of the ship cabin; S2: using cells to represent the spatial geometric outline; S3: using a grid diagram to record the information of the cells; S4: optimizing the pipe layout path based on the A* algorithm; S5: performing multi-pipeline pipe layout. The method for calculating the automatic layout of ship pipelines of the present invention realizes the solution of the single-pipe pipe layout path based on the A* algorithm according to the cabin structure and pipe layout rules, and realizes the solution of the multi-pipe pipe layout path using the pipeline priority sorting and cost value correction method; the design quality is stable and the design efficiency is high.

Description

Technical Field

[0001] The invention relates to the technical field of automatic pipeline layout, and in particular to a calculation method for automatic layout of ship pipelines. Background Art

[0002] Ship piping system (hereinafter referred to as "piping system") is a general term for pipelines, mechanical equipment and detection instruments used to transport and remove liquids or gases in ships to complete set tasks. Piping systems are divided into power piping and ship systems, among which power piping includes: fuel piping, lubrication piping, cooling piping, compressed air piping, exhaust piping; ship systems include: bilge water system, ballast water system, fire protection system, ventilation system, water supply system, refrigeration and air conditioning system, cargo oil system, etc. The above piping systems are very complex and large. Studies have shown that in the drawings of the engine professional, piping system drawings account for more than 70%. Piping system design is one of the keys to ship design, and piping system is the key part of the ship power system. High-quality, high-efficiency and low-cost piping system design plays an irreplaceable role in the safety, economy, rationality of the overall layout, operability, maintenance and ensuring the normal operation and safe navigation of various ship machinery.

[0003] In the research of pipeline automatic layout algorithm, Fu Yili and others from Harbin Institute of Technology have successively carried out research on pipeline automatic layout methods based on particle swarm algorithm and chaos algorithm, and have achieved certain results. However, both the pipeline automatic layout methods based on particle swarm algorithm and chaos algorithm only focus on the problem of automatic optimization of pipeline paths, and still lack consideration of engineering constraints. Christian Vander Velden and others proposed a pipeline intelligent layout method integrating knowledge engineering technology (KBE). This method uses the improved A* algorithm to optimize the path in a three-dimensional space grid and embeds some engineering constraints in the algorithm, which greatly improves the quality of automatically generated pipelines. However, since this method considers fewer engineering constraints and the engineering constraint knowledge is written into the program code, it cannot realize real-time editing and management of knowledge, resulting in great limitations in its practicality. In addition, this method does not pay attention to the mutual influence between multiple pipelines. Fan Xiaoning studied the pipeline laying system based on ant colony algorithm and optimized the algorithm parameters through a large number of simulation experiments. The collaborative evolution method was introduced, a multi-ant colony collaboration mechanism was established, and the parallel laying of multiple pipelines was realized. An overall layout with good synergy effect was obtained. The model used in computational simulation is a virtual model that is not artificially constructed, and there is a gap between it and the physical model. In addition, the entire calculation process is completely closed-loop, and human intellectual resources cannot play a role. Lu Huibiao used the path template method in the study of ship engine room piping layout, and implemented the actual ship layout using VBA language on the AutoCAD platform. However, AutoCAD is rarely used as a ship 3D modeling software by the mainstream, and the input data in the article comes from the database, which is still different from the actual production. TERUAKI ITO uses genetic algorithms to solve pipeline path planning, and based on the prototype system developed by this method, designers can interactively and collaboratively design pipeline paths in the system. However, this algorithm can only be used for two-dimensional plane pipe layout, and it has not been integrated with three-dimensional CAD. Dong Zongran combined taboo search and co-evolution methods to solve the important order problem in pipeline laying, and optimized the individual generation efficiency with the help of A* algorithm.

[0004] In general, many scholars have conducted research on ship piping layout, but most of the current methods remain at the algorithm level and do not involve the development of practical tools for applying the algorithm. The efficiency of the algorithms used is often not efficient enough and cannot be compared with the speed of manual pipe laying. Most of its experimental verifications are based on simple mathematical models, and there are few tests based on actual ship models. In short, current research is still a long way from being implemented. Moreover, for a long time, piping design has relied on the experience of designers, with large fluctuations in design quality and low design efficiency, and it is impossible to effectively consider issues such as material cost, manufacturability, and maintainability. Summary of the invention

[0005] In view of the deficiencies in the above-mentioned prior art, the present invention provides a method for automatic layout calculation of ship pipelines. According to the cabin structure and piping rules, the single-pipe piping path solution is realized based on the A* algorithm, and the multi-pipe piping path solution is realized by using pipeline priority sorting and cost value correction method; the design quality is stable and the design efficiency is high.

[0006] In order to achieve the above object, the present invention provides a method for calculating the automatic layout of ship pipelines, comprising the steps of:

[0007] S1: Analyze and simplify a ship cabin to extract the spatial geometric outline of the ship cabin;

[0008] S2: using cells to represent the geometric outline of the space;

[0009] S3: using a grid diagram to record the information of the cell;

[0010] S4: Optimize the pipe routing path based on the A* algorithm;

[0011] S5: Perform multi-pipe routing.

[0012] Preferably, in the step S1: using an AABB bounding box algorithm, the spatial geometric outline of the input ship cabin in three-dimensional space is identified.

[0013] Preferably, in the step S2:

[0014] A cube is selected as the shape of the cell, and the scale of the cell is determined by the pipe diameter; based on the bounding box information of the spatial geometric outline, the actual spatial position corresponding to the cell, whether the space of the cell can be piped, the area where the cell is located, and the entity type to which the cell belongs are determined; all the cells are saved in a large array, and all information is recorded on the cell.

[0015] Preferably, the information recorded on the cell includes the three-dimensional position information of the cell center and the penalty value of the cell space layout.

[0016] Preferably, in the step S3: the overall size of the grid map is the length of the entire space in all directions, and the length is the number of cells; the number of cells = actual length / cell scale; when the unit grid size is set to be relatively large, the number of cells obtained by space division is reduced accordingly; the grids in the grid map can only be interconnected with adjacent grids; the positions of the cells in the grid map are recorded using a three-dimensional array; the coordinates of the three-dimensional array start from 0, and the conversion of the three-dimensional coordinates of the cells to array indexes is completed.

[0017] Preferably, in step S3: use formula (1) to complete the conversion of the three-dimensional coordinates of the cell into an array index:

[0018]

[0019] Among them, A′ represents the point index in the grid map, A represents the three-dimensional original coordinates of the center point of the cell, pMin represents the minimum coordinates in the entire calculation space, cellLength represents the cell scale, and round() is the rounding method; since scaling requires removing the cell scale and taking an integer value, the result must be adjusted when it cannot be divided evenly.

[0020] Preferably, in step S4, the requirement for optimizing the pipe routing path is to start from the starting point in a specified direction, pass through favorable areas as much as possible in a given space, avoid unfavorable areas, and finally reach the target point from the specified direction; a mathematical model is used to characterize the requirement for optimizing the pipe routing path:

[0021]

[0022] Among them, f(pipe) represents the evaluation index of the quality of pipeline design; Length(pipe) represents the total length of the pipeline; Bend(pipe) represents the number of bends in the pipeline; Punishment(pipe) represents the penalty value of the pipe layout in the cell space; α, β and γ are coefficients, and the coefficients have no fixed values ​​and need to be determined through experiments and experience; A and B are constraints; node represents a node, pipeSet represents a pipeline; totalArea represents the overall space; constraint A means that all nodes passed by the pipeline are within the scope of the overall space; obstacleSet represents the points occupied by known physical obstacles; constraint B means that the pipeline does not overlap with the points occupied by known physical obstacles.

[0023] Preferably, in step S4, using the A* algorithm to perform pathfinding includes the following steps:

[0024] S41: adding the information of the cell at the starting point S to an open table;

[0025] S42: Determine whether the open table is empty. If so, the problem is unsolvable and the algorithm ends. Otherwise, proceed to step S43.

[0026] S43: taking out the best node bestnode from the open table, and adding the best node bestnode to a close table;

[0027] S44: Determine whether the best node bestnode is the end point T, if so, complete the path finding and end the program, otherwise proceed to step S45;

[0028] S45: Obtain an adjacent point of the best node bestnode, calculate the priority value h(S,v) of the adjacent point, and add it to the open table;

[0029] S46: Go to step S42;

[0030] The priority value h(S,v) satisfies formula (3):

[0031] h(S,v)=f(S,v)+g(v,T) (3);

[0032] Among them, f(S,v) represents the cost from the starting point S to the current point v, and g(v,T) represents the cost from the current point v to the end point T. Since the path from v to the end point T has not been determined, g(v,T) is a prediction expression;

[0033] f(S,v) is calculated according to formulas (4) to (7):

[0034] Length(S,v i )=Length(S,v i-1 )+1 (4);

[0035]

[0036] Punishment(S,v i )=Punishment(S,v i-1 )+Punishment(v i ) (6);

[0037] f(S,v i )=α×Length(S,v i )+β×Bend(S,v i )+γ×Punishment(S,v i ) (7);

[0038] Among them, Length(S,v i ) represents the starting point S to the current point v i The length of Bend(S,v i ) represents the starting point S to the current point v i The number of pipe bends; Represents a symbolic function, a vector With vector The function value is 1 when the inner product is 0, otherwise it is 0; Punishment(S,vi ) represents the starting point S to the current point v i The penalty value of

[0039] g(v,T) is calculated according to formula (8):

[0040] g(v,T)=|vx-Tx|+|vy-Ty|+|vz-Tz| (8);

[0041] Wherein, x, y, and z represent the x-axis coordinate value, y-axis coordinate value, and z-axis coordinate value of the spatial coordinate, respectively.

[0042] Preferably, in the step S5: first, the priority of the pipeline is set, and the pipes are laid out from high to low according to the priority. When laying new pipes, the penalty value of the cell is updated according to the adsorption relationship or repulsion relationship between the laid pipes and the current pipes, and the current pipe path is obtained using a single-pipeline laying method; this is performed in sequence until all pipe laying tasks are completed.

[0043] The present invention adopts the above technical solution, so it has the following beneficial effects:

[0044] According to the cabin structure and piping rules, the AABB (Axis-aligned bounding box) algorithm is used to extract the spatial contour of the ship cabin, the cell is used to represent the spatial contour, and the grid map is used to store all the cell information. The analog-to-digital conversion method is designed to solve the conversion of the actual physical coordinates of the cell to the grid map index. The mathematical model is used to characterize the piping requirements, and the node priority value evaluation method is designed. The single-pipe piping path solution is realized based on the A* algorithm. The multi-pipe piping path solution is realized using the pipeline priority sorting and cost value correction method. It has the advantages of stable design quality and high design efficiency, and the pipeline spacing of the pipeline is uniform, neat and beautiful. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a method for calculating the automatic layout of ship pipelines according to an embodiment of the present invention;

[0046] Figure 2 A flow chart of optimizing the pipe routing using the A* algorithm of an embodiment of the present invention;

[0047] Figure 3 This is a pipe arrangement effect diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0048] According to the attached figure Figure 1 to Figure 3 , give the preferred embodiments of the present invention, and describe them in detail so that the functions and features of the present invention can be better understood.

[0049] See also Figure 1 to Figure 3 A method for calculating the automatic layout of ship pipelines in an embodiment of the present invention comprises the following steps:

[0050] S1: Analyze and simplify a ship cabin and extract the spatial geometric outline of the ship cabin;

[0051] The AABB bounding box algorithm is used to identify the spatial geometric outline of the input ship cabin in three-dimensional space.

[0052] The specific method is to calculate the coordinate boundaries of all points in the ship cabin, x1, y1, z1, x2, y2, z2. For any point P(x, y, z) on a spatial geometric body, there is

[0053]

[0054] Then obtain the deterministic representation of the bounding box:

[0055] BoundingBox B=[pMin(x1,y1,z1),pMax(x2,y2,z2)].

[0056] S2: Use cells to represent spatial geometric outlines;

[0057] A cube is selected as the shape of the cell, and the scale of the cell is determined by the pipe diameter. Based on the bounding box information of the spatial geometric contour, the actual spatial position corresponding to the cell, whether the cell space can be piped, the area where the cell is located, and the entity type to which the cell belongs (if an entity exists) are determined. All cells are saved in a large array, and all information is recorded on the cell.

[0058] The information recorded on the cell includes the three-dimensional position information of the cell center and the penalty value of the cell space pipe layout. The penalty value is smaller for areas suitable for pipe layout, and larger for unsuitable locations.

[0059] S3: Use grid diagram to record cell information;

[0060] The overall size of the grid map is the length of the entire space in all directions, and the length is the number of cells; the number of cells = actual length / cell scale; when the unit grid size is set to be relatively large, the number of cells obtained by space division is reduced accordingly; the grids in the grid map can only be interconnected with adjacent grids; the position of the cells in the grid map is recorded using a three-dimensional array; the coordinates of the three-dimensional array start from 0, and the conversion of the three-dimensional coordinates of the cell to the array index is completed.

[0061] Use formula (1) to convert the cell's three-dimensional coordinates to array indexes:

[0062]

[0063] Among them, A′ represents the point index in the grid map, A represents the original three-dimensional coordinates of the cell center point, pMin represents the minimum coordinates in the entire calculation space, cellLength represents the cell scale, and round() is the rounding method; since scaling requires removing the cell scale and taking an integer value, the result must be adjusted when it cannot be divided evenly.

[0064] S4: Optimize the pipe routing path based on the A* algorithm;

[0065] The requirement for optimizing the pipe routing path is to start from the starting point in the specified direction, pass through favorable areas as much as possible in the given space, avoid unfavorable areas, and finally reach the target point from the specified direction. A mathematical model is used to describe the requirements for optimizing the pipe routing path:

[0066]

[0067] Among them, f(pipe) is an evaluation index for the quality of pipeline design; Length(pipe) is the total length of the pipeline; Bend(pipe) is the number of bends in the pipeline; Punishment(pipe) is the penalty value for pipe layout in the cell space; α, β and γ are coefficients. The coefficients have no fixed values ​​and need to be determined through experiments and experience; A and B are constraints; node is a node, pipeSet is a pipeline; totalArea is the overall space; constraint A means that all nodes passed by the pipeline are within the scope of the overall space; obstacleSet is the points occupied by known physical obstacles; constraint B means that the pipeline does not overlap with the points occupied by known physical obstacles.

[0068] Pathfinding using the A* algorithm includes the following steps:

[0069] S41: adding the information of the cell of the starting point S to an open table;

[0070] S42: Determine whether the open list is empty. If so, the problem is unsolvable and the algorithm ends. Otherwise, proceed to step S43.

[0071] S43: Take out the best node bestnode from the open table, and add the best node bestnode to a close table;

[0072] S44: Determine whether the best node bestnode is the end point T. If so, complete the path finding and end the program. Otherwise, proceed to step S45.

[0073] S45: Get an adjacent point of the best node bestnode, calculate the priority value h(S,v) of the adjacent point, and add it to the open table;

[0074] S46: Go to step S42;

[0075] The smaller the priority value h(S,v), the better. The priority value h(S,v) satisfies formula (3):

[0076] h(S,v)=f(S,v)+g(v,T) (3);

[0077] Among them, f(S,v) represents the cost from the starting point S to the current point v, and g(v,T) represents the cost from the current point v to the end point T. Since the path from v to the end point T has not been determined, g(v,T) is a prediction expression;

[0078] f(S,v) is calculated according to formulas (4) to (7):

[0079] Length(S,v i )=Length(S,v i-1 )+1 (4);

[0080]

[0081] Punishment(S,v i )=Punishment(S,v i-1 )+Punishment(v i ) (6);

[0082] f(S,v i )=α×Length(S,v i )+β×Bend(S,v i )+γ×Punishment(S,v i ) (7);

[0083] Among them, Length(S,v i ) represents the starting point S to the current point v i The length of Bend(S,v i ) represents the starting point S to the current point v i The number of pipe bends; Represents a symbolic function, a vector With vector The function value is 1 when the inner product is 0, otherwise it is 0; Punishment(S,v i ) represents the starting point S to the current point v i The penalty value of

[0084] g(v,T) is calculated according to formula (8):

[0085] g(v,T)=|vx-Tx|+|vy-Ty|+|vz-Tz| (8);

[0086] Wherein, x, y, and z represent the x-axis coordinate value, y-axis coordinate value, and z-axis coordinate value of the spatial coordinate, respectively.

[0087] S5: Perform multi-pipe routing.

[0088] First, set the pipeline priority and lay out the pipes from high to low priority. When laying new pipes, update the penalty value of the cell based on the adsorption or repulsion relationship between the laid pipes and the current pipe, and use the single-pipeline laying method to obtain the current pipe path. Continue in sequence until all pipe laying tasks are completed.

[0089] In the embodiment of the present invention, a method for calculating the automatic layout of ship pipelines is used to test a model of the second platform on the port side of a certain type of ship. The machine configuration used in the test is: Intel(R)Core(TM)i7-6700 CPU@3.40GHZ3.41GHZ, RAM 16GB, GPU AMD Radeon(TM)R7 350. The starting point and end point information of the pipe group are shown in Table 1:

[0090] The average time to calculate and generate the overall pipeline solution is about 13ms. Figure 3 As shown, it can be found that the pipeline spacing in the obtained solution is uniform, neat and beautiful.

[0091] Table 1. Starting point and ending point information table

[0092]

[0093] The present invention is described in detail above in conjunction with the embodiments of the accompanying drawings. A person skilled in the art can make various variations of the present invention according to the above description. Therefore, certain details in the embodiments should not constitute a limitation of the present invention, and the scope of protection of the present invention shall be defined by the scope of the attached claims.

Claims

1. A method for calculating the automatic layout of ship pipelines, comprising the steps of: S1: Analyze and simplify a ship cabin to extract the spatial geometric outline of the ship cabin; S2: using cells to represent the geometric outline of the space; S3: using a grid diagram to record the information of the cell; S4: Optimize the pipe routing path based on the A* algorithm; S5: perform multi-pipeline routing; In S1: using the AABB bounding box algorithm, identifying the spatial geometric outline of the input ship cabin in the three-dimensional space; In S2: a cube is selected as the shape of the cell, and the scale of the cell is determined by the pipe diameter; according to the bounding box information of the spatial geometric outline, the actual spatial position corresponding to the cell, whether the space of the cell can be piped, the area where the cell is located, and the entity type to which the cell belongs are determined; All the cells are stored in a large array, and all information is recorded on the cells; The information recorded on the cell includes the three-dimensional position information of the center of the cell and the penalty value of the spatial pipe layout of the cell; In S3: the overall size of the grid map is the length of the entire space in all directions, and the length is the number of cells; the number of cells = actual length / cell scale; the grids in the grid map can only be interconnected with adjacent grids; the positions of the cells in the grid map are recorded using a three-dimensional array; the coordinates of the three-dimensional array start from 0, and the conversion of the three-dimensional coordinates of the cells to array indexes is completed; In S3: Formula (1) is used to complete the conversion of the three-dimensional coordinates of the cell into array indexes: Among them, A′ represents the point index in the grid map, A represents the three-dimensional original coordinates of the center point of the cell, pMin represents the minimum coordinates in the entire calculation space, cellLength represents the cell scale, and round() is a rounding method.

2. The method for calculating the automatic layout of ship pipelines according to claim 1 is characterized in that: In S4, the requirement for optimizing the pipe routing path is to start from the starting point in a specified direction, pass through favorable areas in a given space, avoid unfavorable areas, and finally reach the target point from the specified direction; a mathematical model is used to describe the requirement for optimizing the pipe routing path: Among them, f(pipe) represents the evaluation index of the quality of pipeline design; Length(pipe) represents the total length of the pipeline; Bend(pipe) represents the number of bends in the pipeline; Punishment(pipe) represents the penalty value of pipe layout in the cell space; α, β and γ are coefficients; A and B are constraints; node represents the node, pipeSet represents the pipeline; totalArea represents the overall space; constraint A means that all nodes passed by the pipeline are within the scope of the overall space; obstacleSet represents the points occupied by known physical obstacles; constraint B means that the pipeline does not overlap with the points occupied by known physical obstacles.

3. The method for calculating the automatic layout of ship pipelines according to claim 2 is characterized in that: In S4, the path finding using the A* algorithm includes the following steps: S41: adding the information of the cell at the starting point S to an open table; S42: Determine whether the open table is empty. If so, the problem is unsolvable and the algorithm ends. Otherwise, proceed to step S43. S43: taking out the best node bestnode from the open table, and adding the best node bestnode to a close table; S44: Determine whether the best node bestnode is the end point T, if so, complete the path finding and end the program, otherwise proceed to step S45; S45: Obtain an adjacent point of the best node bestnode, calculate the priority value h(S,v) of the adjacent point, and add it to the open table; S46: Go to step S42; The priority value h(S,v) satisfies formula (3): h(S,v)=f(S,v)+g(v,T) (3); Among them, f(S,v) represents the cost from the starting point S to the current point v, and g(v,T) represents the cost from the current point v to the end point T; f(S,v) is calculated according to formulas (4) to (7): Length(S,v i )=Length(S,v i-1 )+1 (4); Punishment(S,v i )=Punishment(S,v i-1 )+Punishment(v i ) (6); f(S,v i )=α×Length(S,v i )+β×Bend(S,v i )+γ×Punishment(S,v i ) (7); Among them, Length(S,v i ) represents the starting point S to the current point v i The length of Bend(S,v i ) represents the starting point S to the current point v i The number of pipe bends; Represents a symbolic function, a vector With vector The function value is 1 when the inner product is 0, otherwise it is 0; Punishment(S,v i ) represents the starting point S to the current point v i The penalty value of g(v,T) is calculated according to formula (8): g(v,T)=|vx-Tx|+|vy-Ty|+|vz-Tz| (8); Wherein, x, y, and z represent the x-axis coordinate value, y-axis coordinate value, and z-axis coordinate value of the spatial coordinate, respectively.

4. The method for calculating the automatic layout of ship pipelines according to claim 3 is characterized in that: In S5: first, the priority of the pipeline is set, and the pipeline is laid out from high to low according to the priority. When laying a new pipe, the penalty value of the cell is updated according to the adsorption relationship or repulsion relationship between the laid pipe and the current pipe, and the current pipeline path is obtained using the single-pipeline laying method; this is carried out in sequence until all pipe laying tasks are completed.

Images