Welding sequence offline planning method and device, computer equipment and storage medium

Abstract

The invention discloses a welding sequence off-line planning method and device, computer equipment and a storage medium, and relates to the technical field of ship intelligent welding, welding is planned in an off-line mode, the reliability of a planning result is guaranteed, the planning result can be directly output to actual welding production as a result, and the production efficiency is improved. A double-target optimization algorithm of indexes of two dimensions of welding quality and welding no-load stroke is adopted for welding planning, the no-load stroke is effectively reduced on the premise that the welding quality requirement is met, and the ship production efficiency is improved. The current situation that a traditional welding planning process excessively depends on manual judgment is solved, the welding efficiency is improved on the premise that the welding quality is guaranteed, and therefore the ship production cost is reduced.

Description

Offline planning method, apparatus, computer equipment and storage medium for welding sequence Technical Field

[0001] This invention relates to the field of intelligent welding technology for ships, and in particular to an offline welding sequence planning method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Welding is a core process in shipbuilding, accounting for 70% to 90% of the total production cycle, and directly affects the integrity, safety, and efficiency of the structure. Traditional welding path planning methods typically rely on manual or semi-automated strategies. Some optimization algorithms are applied to welding planning scenarios, such as spot welding of body-in-white and welding of steel structures in buildings. As the basic unit of ship welding operations, welds on ship assembly parts are characterized by irregular distribution and diverse specifications. Some welds have a large spatial span and interfere with each other during the welding process. At the same time, welding quality determines the distribution of residual stress and mechanical properties, thus affecting the quality of the hull.

[0003] The results of offline welding path planning are primarily limited by welding process quality requirements and rely heavily on the engineer's experience. As two factors affecting shipyard production quality and efficiency, welding no-load travel and welding quality cannot be simultaneously achieved in many typical welding cases. Summary of the Invention

[0004] The purpose of this invention is to provide an offline welding sequence planning method, apparatus, computer equipment, and storage medium to solve the problem that current offline welding planning methods cannot simultaneously address the issues of welding idle stroke and welding quality. By setting reducing idle stroke and ensuring welding quality as dual optimization objectives, this invention shortens the welding idle stroke and improves operational efficiency while ensuring welding quality.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] On the one hand, an offline welding sequence planning method is provided, the steps of which include:

[0007] Obtain three-dimensional model data of the welding equipment and the workpiece to be welded;

[0008] The weld seams on the three-dimensional model data of the workpiece to be welded are traversed offline to obtain weld seam spatial information, and the weld seam spatial information is parsed to extract weld seam feature data.

[0009] The weld feature data is optimized using two dimensions: welding quality and welding no-load stroke, to obtain the weld planning result.

[0010] Welding paths are generated based on the weld planning results, and the welding paths are simulated and verified. Unreasonable welding paths are adjusted.

[0011] The weld planning results, the adjusted welding path, and the process parameter information are integrated into a robot operation instruction file, which is then sent to the welding robot to perform automated welding operations.

[0012] Furthermore, the step of traversing the weld seams on the 3D model data of the workpiece to be welded offline to obtain weld seam spatial information, and parsing the weld seam spatial information to extract weld seam feature data includes:

[0013] Identify each weld seam on the workpiece to be welded, and obtain weld seam spatial information for each weld seam separately;

[0014] Identify the weld type for each weld, which includes fillet welds and vertical fillet welds;

[0015] For each weld type, corresponding weld feature data is extracted from the weld space information. The weld feature data includes the three-dimensional coordinates of the weld start and end points, the normal vectors of the two side panels of the weld, the weld length, and the weld leg height.

[0016] Furthermore, the dual-objective optimization of the weld feature data based on two dimensions—welding quality and welding no-load stroke—includes:

[0017] Get the total number of welds n, and set w n Let W represent the set of n welds, where W = {w1, w2, ..., wn}. n};

[0018] Let V represent the start and end points of the weld, where V = {v1, v2, ... v}. 2n-1 ,v 2n}, the elements {v} in V 2i-1 ,v 2i} Depends on the i-th weld in W {w i The order and direction of};

[0019] Using D = {d ij |(i,j)∈V} represents the distance matrix of elements in V, where |D|=|V|×|V|, d ij It represents two arbitrary elements {v} in V. i ,v j The Euclidean distance between them;

[0020] The weld planning problem can be represented as: G = (W, V, D);

[0021] Using the welding direction and welding sequence as decision variables, and the welding quality and welding no-load stroke as optimization indicators, a multi-objective optimization problem expression for weld planning is established:

[0022] min S, S=[F1,F2]

[0023]

[0024] minF1=∑d ij x ij ;

[0025]

[0026] Where S represents a multi-objective optimization function; F1 represents the total welding operation path, which includes the welding stroke and the idle stroke; F2 represents the degree of conformity between the current welding sequence and welding direction and the preset welding process rules; and R... j (w i ) represents the i-th weld seam w i Does it meet the j-th welding process rule? α j F2 represents the weight of the j-th welding process rule; F2 = 0 indicates that each weld conforms to the welding process rule, and F2 = 1 indicates that each weld violates the welding process rule.

[0027] In constraint st, x ij ∈{0,1} represents x ij It is a binary quantity, and when x ij When d = 1, ij It belongs to the solution path, otherwise x ij When d = 0, ij It does not belong to the solution path; This indicates that the depth of each weld is 1. This indicates that the extension of each weld seam is 1. This indicates that the solution path is restricted to have one and only one solution.

[0028] Furthermore, the weld planning results obtained by the solution include:

[0029] The welding sequence and welding direction of the weld are encoded and decoded, and the welding sequence and welding direction information of the group of stand-up welds are mapped to a two-dimensional matrix;

[0030] The two-dimensional matrix uses a 2xn matrix to describe individuals. The first row uses natural number encoding to represent the order of the welds, and the second row uses binary encoding to represent the direction of the welds, where 0 represents the weld along the positive direction of the coordinate axis and 1 represents the weld along the negative direction of the coordinate axis.

[0031] The bi-objective optimization problem expression of weld planning is solved using the Non-Dominated Sorting Genetic Algorithm (NSGA-II). The objective function of the NSGA is the idle travel distance and the welding quality. The algorithm performs population initialization, crossover, mutation, non-dominated sorting, and selection processes in sequence. After the termination condition of the NSGA is met, the solution set of the Pareto front is output. The welding planning result is selected based on the solution set of the Pareto front.

[0032] Furthermore, the process of sequentially performing population initialization, crossover, mutation, non-dominated sorting, and selection includes:

[0033] The process of setting up population crossover involves: the offspring inheriting a gene segment from either of the two parents as a basis, and then selecting the node with the shortest spatial distance among the adjacent nodes of the parent chromosomes as the next gene segment of the offspring. The remaining genes are gradually constructed through the population crossover process.

[0034] The process of setting up population mutation includes: the individual to be mutated is referenced to the best individual in the current population, the corresponding gene fragment of the individual to be mutated is queried based on the gene fragment randomly selected from the best individual, and the mutated offspring are obtained by recombination based on the 3-opt method.

[0035] Furthermore, the selection of welding planning results based on the solution set of the Pareto front includes:

[0036] Obtain the solution set of the Pareto front, where the horizontal axis of the solution set represents the welding no-load stroke and the vertical axis of the solution set represents the welding quality, and perform normalization processing.

[0037] The welding no-load stroke and the welding quality are weighted according to production needs to select the corresponding welding planning result.

[0038] Furthermore, the step of generating a welding path based on the weld planning results, performing simulation verification on the welding path, and adjusting unreasonable welding paths includes:

[0039] The robot motion solver calculates the welding torch end posture based on the weld planning results, solves the values ​​of the robot body axis and external axis to obtain the solution results, and performs simulated welding based on the solution results to verify the feasibility of the operation path.

[0040] Based on the bounding box algorithm, collisions are detected in real time during the simulated welding process. Collision locations are highlighted to show the interference position, the operation paths that do not collide are retained, and the operation paths that collide are adjusted.

[0041] On the other hand, an offline welding sequence planning device is provided, the device comprising:

[0042] The model import module is used to acquire 3D model data of the welding equipment and the workpiece to be welded;

[0043] The weld seam analysis module is used to traverse the weld seams on the three-dimensional model data of the workpiece to be welded in an offline manner, obtain weld seam spatial information, and parse the weld seam spatial information to extract weld seam feature data.

[0044] The welding planning module is used to perform dual-objective optimization on the weld feature data based on two dimensions: welding quality and welding no-load stroke, and solve for the weld planning result.

[0045] The motion simulation module is used to generate a welding path based on the weld planning results, perform simulation verification on the welding path, and adjust unreasonable welding paths.

[0046] The post-processing module is used to integrate the weld planning results, the adjusted welding path, and the process parameter information into a robot operation instruction file, and to send the robot operation instruction file to the welding robot to perform automated welding operations.

[0047] In another aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of an offline welding sequence planning method.

[0048] In another aspect, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of an offline welding sequence planning method.

[0049] The beneficial effects of this invention are that it ensures the reliability of welding planning results through offline methods, and the results can be directly output to actual welding production. The welding planning employs a dual-objective optimization algorithm with two dimensions: welding quality and welding no-load travel. This effectively reduces no-load travel while meeting welding quality requirements, thereby improving ship production efficiency. Furthermore, it addresses the issue of traditional welding planning processes relying too heavily on manual judgment, improving welding efficiency while ensuring welding quality, thus reducing ship production costs. Attached Figure Description

[0050] Figure 1 is a flowchart of an offline welding sequence planning method in one embodiment of this application;

[0051] Figure 2 is a logic diagram of an offline welding sequence planning method in one embodiment of this application;

[0052] Figure 3 is a schematic diagram of the coding of welding sequence and welding direction in one embodiment of this application;

[0053] Figure 4 is a flowchart of the NSGA-II algorithm in one embodiment of this application;

[0054] Figure 5 is a schematic diagram of a heuristic crossover operator based on a greedy strategy in one embodiment of this application;

[0055] Figure 6 is a schematic diagram of a mutation operator based on the optimal individual in one embodiment of this application;

[0056] Figure 7 shows a typical ship assembly model in one embodiment of this application;

[0057] Figure 8 shows the Pareto front curve obtained by the NSGA-II algorithm in one embodiment of this application;

[0058] Figure 9 is a schematic diagram of the welding planning results output in one embodiment of this application;

[0059] Figure 10 is a structural block diagram of an offline welding sequence planning device in one embodiment of this application;

[0060] Figure 11 is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0061] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] As shown in Figure 1, this invention discloses an offline welding sequence planning method for offline planning of welding sequence of ship assembly components. The method includes the following steps:

[0063] Step S1: Obtain the three-dimensional model data of the welding equipment and the workpiece to be welded;

[0064] Step S2: Traverse the weld seams on the three-dimensional model data of the workpiece to be welded offline to obtain weld seam spatial information, and parse the weld seam spatial information to extract weld seam feature data;

[0065] Step S3: Based on the two dimensions of welding quality and welding no-load stroke, perform dual-objective optimization on the weld feature data to obtain the weld planning result;

[0066] Step S4: Generate a welding path based on the weld planning results, perform simulation verification on the welding path, and adjust any unreasonable welding paths.

[0067] Step S5: Integrate the weld planning results, the adjusted welding path, and the process parameter information into a robot operation instruction file, and send the robot operation instruction file to the welding robot to perform automated welding operations.

[0068] In other words, as shown in Figure 2, the 3D model data of the welding equipment and the workpiece to be welded are imported into the offline software through the model import module; the weld seam analysis module traverses the weld seams on the imported workpiece to be welded, analyzes and obtains the spatial position information of the weld seams; the welding planning module is applied to perform dual-objective optimization on the analyzed weld seam data based on two dimensions: welding quality and welding no-load stroke, to obtain the weld seam planning result; the welding path is generated according to the planning result, and the welding path is simulated in the motion simulation module to verify the rationality of the planning result; the post-processing module is applied to output the verified welding planning result, welding path and process parameters into robot operation instructions through scripting language, and the robot and welding torch execute automated welding operations according to the operation instructions.

[0069] The model import module is compatible with mainstream 3D model file formats, including .obj, .dxf, .ply, .stl, and .dae.

[0070] Furthermore, the step of traversing the weld seams on the 3D model data of the workpiece to be welded offline to obtain weld seam spatial information, and parsing the weld seam spatial information to extract weld seam feature data includes:

[0071] Identify each weld seam on the workpiece to be welded, and obtain weld seam spatial information for each weld seam separately;

[0072] Identify the weld type for each weld, which includes fillet welds and vertical fillet welds;

[0073] For each weld type, corresponding weld feature data is extracted from the weld space information. The weld feature data includes the three-dimensional coordinates of the weld start and end points, the normal vectors of the two side panels of the weld, the weld length, and the weld leg height.

[0074] Furthermore, the dual-objective optimization of the weld feature data based on two dimensions—welding quality and welding no-load stroke—includes:

[0075] Get the total number of welds n, and set w n Let W represent the set of n welds, where W = {w1, w2, ..., wn}. n};

[0076] Let V represent the start and end points of the weld, where V = {v1, v2, ... v}. 2n-1 ,v 2n}, the elements {v} in V 2i-1 ,v 2i} Depends on the i-th weld in W {w i The order and direction of};

[0077] Using D = {d ij |(i,j)∈V} represents the distance matrix of elements in V, where |D|=|V|×|V|, d ij It represents two arbitrary elements {v} in V. i ,v j The Euclidean distance between them;

[0078] The weld planning problem can be represented as: G = (W, V, D);

[0079] Using the welding direction and welding sequence as decision variables, and the welding quality and welding no-load stroke as optimization indicators, a multi-objective optimization problem expression for weld planning is established:

[0080] minS, S=[F1,F2]

[0081]

[0082] minF1=∑d ij x ij ;

[0083]

[0084] Where S represents a multi-objective optimization function; F1 represents the total welding operation path, which includes the welding stroke and the idle stroke; F2 represents the degree of conformity between the current welding sequence and welding direction and the preset welding process rules; and R... j (w i ) represents the i-th weld seam w i Does it meet the j-th welding process rule? α j F2 represents the weight of the j-th welding process rule; F2 = 0 indicates that each weld conforms to the welding process rule, and F2 = 1 indicates that each weld violates the welding process rule.

[0085] In constraint st, x ij ∈{0,1} represents x ij It is a binary quantity, and when x ij When d = 1, ij It belongs to the solution path, otherwise x ij When d = 0, ij It does not belong to the solution path; This indicates that the depth of each weld is 1. This indicates that the extension of each weld seam is 1. This indicates that the solution path is restricted to have one and only one solution.

[0086] This embodiment transforms the weld planning problem into a combinatorial optimization problem of the Travelling Salesman Problem (TSP) type, G = (W, V, D); thus, the planning problem can be specified and solved.

[0087] Figure 3 shows a schematic diagram of the encoding of welding sequence and welding direction. An improved NSGA-II algorithm is used, with the weld start point in the weld information as the algorithm input, and the welding sequence and welding direction as decision variables for encoding.

[0088] Furthermore, the weld planning results obtained by the solution include:

[0089] The welding sequence and welding direction of the weld are encoded and decoded, and the welding sequence and welding direction information of the group of stand-up welds are mapped to a two-dimensional matrix;

[0090] The two-dimensional matrix uses a 2xn matrix to describe individuals. The first row uses natural number encoding to represent the order of the welds, and the second row uses binary encoding to represent the direction of the welds, where 0 represents the weld along the positive direction of the coordinate axis and 1 represents the weld along the negative direction of the coordinate axis.

[0091] The bi-objective optimization problem expression of weld planning is solved using the Non-Dominated Sorting Genetic Algorithm (NSGA-II). The objective function of the NSGA is the idle travel distance and the welding quality. The algorithm performs population initialization, crossover, mutation, non-dominated sorting, and selection processes in sequence. After the termination condition of the NSGA is met, the solution set of the Pareto front is output. The welding planning result is selected based on the solution set of the Pareto front.

[0092] Figure 4 shows the flowchart of the NSGA-II algorithm. The Pareto front is a core concept in multi-objective optimization problems, also known as the Pareto optimal boundary. Essentially, it describes the solution set where, under multi-objective conditions, it is impossible to further optimize one objective without compromising the other objectives.

[0093] As shown in Figures 5 and 6, the process of sequentially performing population initialization, crossover, mutation, non-dominated sorting, and selection includes:

[0094] The process of setting up population crossover involves: the offspring inheriting a gene segment from either of the two parents as a basis, and then selecting the node with the shortest spatial distance among the adjacent nodes of the parent chromosomes as the next gene segment of the offspring. The remaining genes are gradually constructed through the population crossover process.

[0095] The process of setting up population mutation includes: the individual to be mutated is referenced to the best individual in the current population, the corresponding gene fragment of the individual to be mutated is queried based on the gene fragment randomly selected from the best individual, and the mutated offspring are obtained by recombination based on the 3-opt method.

[0096] As shown in Figure 5, the NSGA-II algorithm employs a heuristic crossover operator based on a greedy strategy instead of a random crossover process. Unlike typical single-point or multi-point crossover methods, this method avoids duplicate nodes in chromosome characters. Specifically, the offspring inherits a gene segment from either of the two parents as a starting point, and then selects the node with the shortest spatial distance among adjacent nodes of the parent chromosomes as the next gene segment for the offspring. The remaining genes are gradually constructed using this operator. The proposed crossover operator, by adopting a greedy strategy, avoids the generation of invalid individuals and expands the search space of the population.

[0097] As shown in Figure 6, the NSGA-II algorithm employs a mutation operator based on global optimal solution information. The individual to be mutated references the best individual in the current population, queries the corresponding gene segment of the individual to be mutated based on a randomly selected gene segment from the best individual, and then recombines it using the 3-opt method to obtain the mutated offspring. This method ensures that the new individual retains the gene segments of the current best individual, improving the convergence speed of the algorithm.

[0098] Furthermore, the selection of welding planning results based on the solution set of the Pareto front includes:

[0099] Obtain the solution set of the Pareto front, where the horizontal axis of the solution set represents the welding no-load stroke and the vertical axis of the solution set represents the welding quality, and perform normalization processing.

[0100] The welding no-load stroke and the welding quality are weighted according to production needs to select the corresponding welding planning result.

[0101] Furthermore, the step of generating a welding path based on the weld planning results, performing simulation verification on the welding path, and adjusting unreasonable welding paths includes:

[0102] The robot motion solver calculates the welding torch end posture based on the weld planning results, solves the values ​​of the robot body axis and external axis to obtain the solution results, and performs simulated welding based on the solution results to verify the feasibility of the operation path.

[0103] Based on the bounding box algorithm, collisions are detected in real time during the simulated welding process. Collision locations are highlighted to show the interference position, the operation paths that do not collide are retained, and the operation paths that collide are adjusted.

[0104] The method of the present invention includes the following steps in its specific implementation:

[0105] (1) Import the three-dimensional model of the workpiece to be welded through the model import module. Taking the typical ship group standing workpiece to be welded as shown in Figure 7 as an example, the length direction is 5702mm, the width direction is 3757mm, the thickness of the bottom plate is 12mm, and the thickness of the standing plate is 20mm.

[0106] (2) The weld seams on the imported workpiece are traversed through the weld seam analysis module to parse and obtain the weld seam spatial information, including the three-dimensional coordinates of the start and end points of the weld seam, the normal vectors of the two side panels of the weld seam, the weld seam length, and the weld leg height. The weld seam types on the workpiece to be welded include flat fillet welds and vertical fillet welds, with the longest weld seam being 2510mm.

[0107] (3) Applying the improved NSGA-II algorithm mentioned in this invention, the weld start point in the weld information is used as the algorithm input, and the welding sequence and welding direction of the weld are encoded as decision variables as shown in Figure 3. The algorithm goes through the sorting, selection, crossover, and mutation processes shown in Figure 4. The crossover process is shown in Figure 5. The offspring inherits a gene segment from any of the two parents as a basis, and then selects the node with the shortest spatial distance among the adjacent nodes of the parent chromosomes as the next gene segment of the offspring. The remaining genes are gradually constructed through this operator. The mutation process is shown in Figure 6. The individual to be mutated refers to the best individual in the current population. According to the gene segment randomly selected from the best individual, the corresponding gene segment of the individual to be mutated is queried, and the mutated offspring are recombined based on the 3-opt method. After the algorithm termination condition is met or the algorithm iteration limit is reached, the welding planning result is output, as shown in Figure 8, which is the solution set of the Pareto front. The horizontal axis represents the welding idle stroke, and the vertical axis represents the welding quality and is normalized. According to the actual production needs, point P in Figure 8 is selected as the planning result for output, as shown in Figure 9.

[0108] (4) Apply the welding planning results to the motion simulation module, obtain the real-time position and working posture of the welding equipment through kinematic calculation, and detect whether there is a collision between the parts. The results verified by motion simulation will be retained.

[0109] (5) The post-processing module is used to output the verified welding planning results, welding path and process parameters into robot operation instructions through scripting language. The robot and welding gun perform automated welding operations according to the operation instructions.

[0110] Experimental results show that, on selected experimental workpieces, this method effectively reduced the no-load travel by 6.95% compared to other methods, thereby improving ship production efficiency.

[0111] In one embodiment, as shown in FIG10, a welding sequence offline planning device 10 is provided, including: a model import module 1, a weld seam analysis module 2, a welding planning module 3, a motion simulation module 4, and a post-processing module 5.

[0112] The model import module 1 is used to acquire three-dimensional model data of the welding equipment and the workpiece to be welded.

[0113] The weld seam analysis module 2 is used to traverse the weld seams on the three-dimensional model data of the workpiece to be welded in an offline manner, obtain weld seam spatial information, and parse the weld seam spatial information to extract weld seam feature data.

[0114] The welding planning module 3 is used to perform dual-objective optimization on the weld feature data based on two dimensions: welding quality and welding no-load stroke, and solve to obtain the weld planning result.

[0115] The motion simulation module 4 is used to generate a welding path based on the weld planning results, perform simulation verification on the welding path, and adjust unreasonable welding paths.

[0116] The post-processing module 5 is used to integrate the weld planning results, the adjusted welding path, and the process parameter information into a robot operation instruction file, and to send the robot operation instruction file to the welding robot to perform automated welding operations.

[0117] Specific limitations regarding the offline welding sequence planning device can be found in the limitations of the offline welding sequence planning method described above, and will not be repeated here. Each module in the aforementioned offline welding sequence planning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0118] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram is shown in Figure 11. The computer device includes a processor, memory, a network interface, and a database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores offline welding sequence planning data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements an offline welding sequence planning method.

[0119] Those skilled in the art will understand that the structure shown in Figure 11 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0120] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0121] Obtain three-dimensional model data of the welding equipment and the workpiece to be welded;

[0122] The weld seams on the three-dimensional model data of the workpiece to be welded are traversed offline to obtain weld seam spatial information, and the weld seam spatial information is parsed to extract weld seam feature data.

[0123] The weld feature data is optimized using two dimensions: welding quality and welding no-load stroke, to obtain the weld planning result.

[0124] Welding paths are generated based on the weld planning results, and the welding paths are simulated and verified. Unreasonable welding paths are adjusted.

[0125] The weld planning results, the adjusted welding path, and the process parameter information are integrated into a robot operation instruction file, which is then sent to the welding robot to perform automated welding operations.

[0126] For specific limitations on the steps implemented by the processor when executing a computer program, please refer to the limitations on the offline planning method for soldering sequence mentioned above, which will not be repeated here.

[0127] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0128] Obtain three-dimensional model data of the welding equipment and the workpiece to be welded;

[0129] The weld seams on the three-dimensional model data of the workpiece to be welded are traversed offline to obtain weld seam spatial information, and the weld seam spatial information is parsed to extract weld seam feature data.

[0130] The weld feature data is optimized using two dimensions: welding quality and welding no-load stroke, to obtain the weld planning result.

[0131] Welding paths are generated based on the weld planning results, and the welding paths are simulated and verified. Unreasonable welding paths are adjusted.

[0132] The weld planning results, the adjusted welding path, and the process parameter information are integrated into a robot operation instruction file, which is then sent to the welding robot to perform automated welding operations.

[0133] For specific limitations on the steps implemented when a computer program is executed by a processor, please refer to the limitations on the offline planning method for welding sequence mentioned above, which will not be repeated here.

[0134] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An offline planning method for the welding sequence of ship assembly components, characterized in that, The process includes the following steps: acquiring three-dimensional model data of the welding equipment and the workpiece to be welded; traversing the weld seams on the three-dimensional model data of the workpiece to be welded offline to acquire weld seam spatial information; and parsing the weld seam spatial information to extract weld seam feature data. The weld feature data is optimized using two dimensions: welding quality and welding no-load stroke. The weld planning result is obtained by solving the problem. A welding path is generated based on the weld planning result. The welding path is then simulated and verified. Any unreasonable welding paths are adjusted. The weld planning results, the adjusted welding path, and the process parameter information are integrated into a robot operation instruction file, which is then sent to the welding robot to perform automated welding operations.

2. The offline planning method for welding sequence of ship assembly components according to claim 1, characterized in that, The process of traversing the weld seams on the 3D model data of the workpiece to be welded offline to obtain weld seam spatial information and extracting weld seam feature data by parsing the weld seam spatial information includes: identifying each weld seam on the workpiece to be welded and obtaining weld seam spatial information for each weld seam; identifying the weld seam type for each weld seam, which includes flat fillet welds and vertical fillet welds; and extracting corresponding weld seam feature data from the weld seam spatial information for each weld seam type, wherein the weld seam feature data includes the 3D coordinates of the start and end points of the weld seam, the normal vectors of the two side panels of the weld seam, the weld seam length, and the weld leg height.

3. The offline planning method for welding sequence of ship assembly components according to claim 1, characterized in that, The dual-objective optimization of the weld feature data based on two dimensions—welding quality and welding no-load stroke—includes: obtaining the total number of welds n, and setting w. n Let W represent the set of n welds, where W = {w1, w2, ..., wn}. n }; Let the start and end points of the weld be represented by V, where V = {v1, v2, ... v} 2n-1 ,v 2n }, the elements {v} in V 2i-1 ,v 2i } Depends on the i-th weld in W {w i The order and direction of}; using D = {d ij |(i,j)∈V} represents the distance matrix of elements in V, where |D|=|V|×|V|, d ij It represents two arbitrary elements {v} in V. i ,v j The Euclidean distance between the nodes is used; the weld planning problem is expressed as: G=(W,V,D); the welding direction and welding sequence are used as decision variables, and the welding quality and welding no-load stroke are used as optimization indicators for multi-objective optimization, establishing the expression for the bi-objective optimization problem of weld planning: min S,S=[F1,F2] my F1=∑d ij x ij ; Where S represents a multi-objective optimization function; F1 represents the total welding operation path, which includes the welding stroke and the idle stroke; F2 represents the degree of conformity between the current welding sequence and welding direction and the preset welding process rules; and R... j (w i ) represents the i-th weld seam w i Does it meet the j-th welding process rule? α j Let represent the weight of the j-th welding process rule; F2 = 0 indicates that every weld conforms to the welding process rule, and F2 = 1 indicates that every weld violates the welding process rule; in the constraint condition st, x ij ∈{0,1} represents x ij It is a binary quantity, and when x ij When d = 1, ij It belongs to the solution path, otherwise x ij When d = 0, ij It does not belong to the solution path; This indicates that the depth of each weld is 1. This indicates that the extension of each weld seam is 1. This indicates that the solution path is restricted to have one and only one solution.

4. The offline planning method for welding sequence of ship assembly components according to claim 3, characterized in that, The solution to obtain the weld planning result includes: encoding and decoding the welding sequence and welding direction of the welds, mapping the welding sequence and welding direction information of the group of welds to a two-dimensional matrix; using a 2xn matrix to describe individuals in the two-dimensional matrix, the first row uses natural number encoding to represent the weld sequence, and the second row uses binary encoding to represent the weld direction, where 0 represents the weld along the positive direction of the coordinate axis and 1 represents the weld along the negative direction of the coordinate axis; using a non-dominated sorting genetic algorithm to solve the bi-objective optimization problem expression of the weld planning, the objective function of the non-dominated sorting genetic algorithm is the idle travel and the welding quality, and performing population initialization, crossover, mutation, non-dominated sorting, and selection processes in sequence, outputting the solution set of the Pareto front after satisfying the termination condition of the non-dominated sorting genetic algorithm, and selecting the welding planning result according to the solution set of the Pareto front.

5. The offline planning method for welding sequence of ship assembly components according to claim 4, characterized in that, The process of initializing, crossing over, mutating, non-dominated sorting, and selecting the population includes: The crossover process involves the offspring inheriting a gene segment from either parent, then selecting the node with the shortest spatial distance among adjacent nodes on the parent chromosomes as the next gene segment for the offspring; the remaining genes are gradually constructed through the crossover process. The mutation process involves the individual to be mutated referencing the best individual in the current population, querying the corresponding gene segment of the individual to be mutated based on a randomly selected gene segment from the best individual, and recombining it using the 3-opt method to obtain the mutated offspring.

6. The offline planning method for welding sequence of ship assembly components according to claim 4, characterized in that, The step of selecting the welding planning result based on the solution set of the Pareto front includes: obtaining the solution set of the Pareto front, where the horizontal axis of the solution set represents the welding no-load stroke and the vertical axis of the solution set represents the welding quality, and performing normalization processing; allocating the weights of the welding no-load stroke and the welding quality according to production requirements to select the corresponding welding planning result.

7. The offline planning method for welding sequence of ship assembly components according to claim 1, characterized in that, The process of generating a welding path based on the weld planning results, simulating and verifying the welding path, and adjusting unreasonable welding paths includes: using a robot motion solver to calculate the welding torch end posture based on the weld planning results, solving for the values ​​of the robot body axis and external axis to obtain the solution results, performing simulated welding based on the solution results to verify the feasibility of the operation path; and using a bounding box algorithm to determine in real time whether a collision occurs during the simulated welding process, highlighting the collision position to show the interference position, retaining the operation path without collision, and adjusting the operation path with collision.

8. A welding sequence offline planning device, characterized in that, The device includes: a model import module for acquiring three-dimensional model data of welding equipment and workpiece to be welded; a weld seam analysis module for offline traversal of weld seams on the three-dimensional model data of the workpiece to be welded, acquiring weld seam spatial information, and parsing the weld seam spatial information to extract weld seam feature data; a welding planning module for performing dual-objective optimization on the weld seam feature data based on two dimensions: welding quality and welding no-load stroke, to obtain the weld seam planning result; a motion simulation module for generating a welding path based on the weld seam planning result, simulating and verifying the welding path, and adjusting unreasonable welding paths; and a post-processing module for integrating the weld seam planning result, the adjusted welding path, and process parameter information into a robot operation instruction file, and sending the robot operation instruction file to the welding robot to perform automated welding operations.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images