Visual control method for transfer assembly process

Abstract

The present application relates to the technical field of turnaround assembly and discloses a method for visual control of the turnaround assembly process, comprising the following steps: collecting a parts list involved in the turnaround assembly process and assembly information of each part; generating an assembly tree for each turnaround based on the parts list and assembly information; optimizing the assembly tree of each turnaround using a genetic algorithm in combination with visual simulation assembly; allocating an assembly line to each turnaround based on the assembly tree, and time-aligning the assembly trees of any group of adjacent assembly lines; dividing the assembly trees into cross-tree assembly groups based on the time-aligned assembly trees; and performing synchronous assembly of turnarounds across assembly lines based on the assembly trees and the cross-tree assembly groups. The present application can improve the parallel assembly efficiency of differentiated turnarounds, reduce assembly time, and lower assembly costs and energy consumption.

Description

Technical Field

[0001] The present application relates to the technical field of turnkey assembly, and in particular to a method for visual control of a turnkey assembly process. Background Art

[0002] Turnkey assembly follows certain standard specifications and processes. This standardized approach ensures basic quality requirements, but it often lacks flexibility when faced with the diverse models and specifications of turnkey machines, as well as diverse customer customization requirements. Traditional static assembly processes lack effective mechanisms to address this variability.

[0003] Generally speaking, turntable equipment is often customized. When assembling turntables with different customization details, the assembly details differ substantially, leading to significant differences in the assembly process. Turntable components from different batches often vary in part types and sizes. For example, some turntables require the addition of special sound insulation and vibration damping components during assembly due to stringent customer requirements for operating noise. Other turntable applications require high heat dissipation performance, requiring the redesign of the air duct structure and optimization of the component layout during assembly. These special steps are not universally applicable to turntable assembly. Conventional technology requires the coordinated efforts of multiple experienced personnel, each responsible for several similar assembly steps on one or multiple adjacent production lines. Therefore, conventional technology cannot maximize the efficiency of parallel assembly across multiple production lines regardless of the different assembly requirements of turntables; it can only achieve parallel assembly of turntables of the same specifications.

[0004] For example, the patent application with publication number CN113759860A discloses a visual assembly system, assembly method, equipment, and medium. The visual assembly system includes an assembly data configuration module, an assembly process integrated control module, and a touch-screen display module. The assembly data configuration module is used to acquire and save assembly data. The assembly process integrated control module is used to provide an operation interface, allowing the user to associate the assembly data with the corresponding assembly process and display the corresponding assembly data according to the preset assembly process during the assembly process. The display module is used to display the corresponding assembly data according to the content display instructions of the assembly process integrated control module. This technical solution can achieve visual guidance of the assembly process during the trial production process, improve assembly efficiency, and reduce the error rate of assembly operations.

[0005] For example, the patent application with publication number CN111077867A discloses a dynamic simulation method for aircraft engine assembly quality based on digital twins, including: constructing a digital carrier of assembly information based on digital twin technology, obtaining assembly process information through multi-source sensing equipment, and dynamically simulating the status of the product assembly process in combination with the assembly process mechanism. The simulation results are efficiently fed back to the assembly personnel through visualization technology, and the assembly quality is dynamically adjusted, realizing a closed loop of the assembly process in assembly production, achieving the effect of improving the assembly quality of aircraft engines, and providing a new dynamic guidance method for product assembly process control.

[0006] The above patents all have the problem raised by this background technology: they are unable to adapt to the differentiated assembly requirements of turnaround machines to maximize the parallel assembly efficiency of multiple production lines.

[0007] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the application and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the Invention

[0008] Based on the above technical problems, the present application provides a method for visual control of the turntable assembly process to realize parallel assembly control of turntables with different assembly specifications to adapt to various differentiated assembly task requirements.

[0009] To solve the above technical problems, this application provides the following technical solutions:

[0010] A visual control method for a transfer assembly process includes the following steps:

[0011] Collect the parts list involved in the turnaround assembly process and the assembly information for each part;

[0012] generating an assembly tree for each transfer machine based on the parts list and assembly information;

[0013] Genetic algorithms are used in combination with visual assembly simulation to optimize the assembly tree of each transfer machine;

[0014] Assign an assembly line to each transfer machine based on the assembly tree, and perform time alignment on the assembly trees of any group of adjacent assembly lines;

[0015] Based on the time-aligned assembly tree, the cross-tree assembly groups are divided;

[0016] Based on the assembly tree and the cross-tree assembly group, synchronous assembly of transfer machines across assembly lines is performed.

[0017] As a preferred solution of the method for visual control of the transfer assembly process described in this application, the method for generating the assembly tree is as follows:

[0018] S201: Set the root node as the transfer machine and all parts as leaf nodes;

[0019] S202: searching for a parent node for each leaf node based on the assembly information; the parent node of any leaf node is the leaf node or the root node corresponding to the assembly position of the part corresponding to the leaf node;

[0020] S203: Connect all nodes according to the relationship between parent and child nodes to generate an assembly tree;

[0021] S204: Check the assembly logic relationship between all parts corresponding to all nodes in each layer of the assembly tree, and adjust the structure of the assembly tree; the details are as follows:

[0022] If there is an assembly sequence between the parts corresponding to any two nodes at the same level in the assembly tree, the two nodes corresponding to the parts with an assembly sequence will be adjusted to adjacent positions and connected with a dotted line, with the node corresponding to the part with the earlier assembly sequence on the left and the node corresponding to the part with the later assembly sequence on the right.

[0023] As a preferred solution of the method for visual control of the transfer machine assembly process described in this application, the method for optimizing the assembly tree of each transfer machine is as follows:

[0024] S301: Set the population size, crossover probability, and mutation probability, and generate N assembly trees for each transfer machine based on the method described in S201-S204, where N is the population size;

[0025] S302: Generate N individuals based on the N assembly trees to form an initial population;

[0026] S303: Calculate the fitness of each individual;

[0027] S304: performing selection operation, crossover operation, and mutation operation in sequence to generate the next generation population; wherein the object of the crossover operation is any pair of chromosomes with the same number in the two individuals participating in the crossover operation;

[0028] S305: Repeat S303-S304 until the highest fitness of all individuals in the population converges;

[0029] S306: Obtain the assembly tree corresponding to the individual with the highest fitness as the optimization result of the assembly tree corresponding to the transition.

[0030] As a preferred solution of the visualization control method of the transfer assembly process described in the present application, the method of generating the individuals based on the assembly tree is as follows: let the assembly tree have a total of n layers, then any of the individuals contains n-1 chromosomes, numbered as chromosome 1, chromosome 2,..., chromosome n-1 respectively; wherein chromosome i is a one-dimensional array composed of the nodes of the i+1th layer from top to bottom of the assembly tree, each element in the array corresponds to a node, and the arrangement order of the elements in the array is that the nodes of the i+1th layer are sorted from left to right.

[0031] As a preferred solution of the visualization control method for the transfer assembly process described in this application, the method for calculating the fitness of each individual is as follows:

[0032] Obtain the assembly sequence of each individual by recording the assembly order of each part in the assembly tree corresponding to each individual from top to bottom and from left to right, and forming an assembly sequence;

[0033] The assembly process of the transfer machine is simulated through the digital twin system to obtain the fitness parameters when the transfer machine is assembled according to the assembly sequence of each individual;

[0034] Calculating the assembly benefit score, process optimization score, and resource utilization rate of each individual based on the fitness parameters;

[0035] The fitness of each individual is calculated based on the assembly benefit score, process optimization score, and resource utilization.

[0036] As a preferred solution of the method for visual control of the transfer assembly process described in the present application, the time alignment of the assembly trees of any group of adjacent assembly lines specifically includes:

[0037] The assembly process of the transfer machine is simulated through the digital twin system to obtain the node features of each node in the assembly tree corresponding to each transfer machine; the node features include part type, assembly tool, and spatial coordinates;

[0038] Based on the node features, the similarity between nodes is calculated, and a cross-tree association matrix is ​​constructed to record the similarity;

[0039] Based on the cross-tree association matrix, the assembly trees of adjacent assembly lines are time-aligned, specifically including:

[0040] Setting a similarity threshold; traversing the cross-tree association matrix, marking two nodes corresponding to elements in the cross-tree association matrix that are greater than the similarity threshold as similar node pairs;

[0041] Mark the layers of the two nodes in each set of similar node pairs in the corresponding assembly tree as alignment layers;

[0042] The alignment layers between the assembly trees of adjacent assembly lines are subjected to conflict detection. If the conflict detection fails, the alignment layers are adjusted until the alignment layers between the assembly trees of adjacent assembly lines pass the conflict detection, thereby completing the temporal alignment of the assembly trees.

[0043] As a preferred embodiment of the method for visual control of the assembly process of a transfer machine described in the present application, wherein: any group of adjacent assembly lines is two spatially adjacent assembly lines; any assembly line is used to perform assembly of a corresponding transfer machine according to an assembly tree; any production line is only divided into a group of adjacent assembly lines; and the method of assigning an assembly line to each transfer machine based on the assembly tree specifically includes: if any two assembly trees have the same structure, assigning the two assembly trees with the same structure to a group of adjacent assembly lines;

[0044] Any element in the cross-tree association matrix represents the similarity between two nodes on different assembly trees in a group of adjacent assembly lines; let the two assembly trees of the adjacent assembly lines be C and D respectively, then the element in the p-th row and q-th column of the cross-tree association matrix represents the similarity between the p-th node in assembly tree C and the q-th node in assembly tree D.

[0045] As a preferred solution of the visualization control method for the turnaround assembly process described in the present application, wherein: based on the time-aligned assembly tree, the cross-tree assembly groups are divided; specifically including:

[0046] Divide any set of similar node pairs in any set of alignment layers in the time-aligned assembly tree into a cross-tree assembly group;

[0047] Based on any node in the spanning tree assembly group as a starting point, searching for nodes that can be added to the spanning tree assembly group on the corresponding assembly tree and expanding the spanning tree assembly group, specifically including:

[0048] Determine whether any associated node of the starting point satisfies a constraint condition for joining a corresponding assembly group; if any associated node satisfies the constraint condition, add the associated node to the corresponding assembly group; the associated nodes include the parent node and child nodes of the starting point, and nodes at the same level of the assembly tree as the starting point.

[0049] As a preferred solution of the method for visual control of the turnkey assembly process described in the present application, wherein: determining whether the associated node satisfies the constraint condition specifically includes: reading the part type, assembly tool, and spatial coordinates of each node in the spanning tree assembly group;

[0050] If the associated node has the same part type as any node in the spanning tree assembly group, the constraint condition is satisfied;

[0051] If the number of assembly tools introduced by the associated node is less than m, and the average distance between the associated node and all nodes in the spanning tree assembly group is less than a preset distance threshold, the constraint condition is satisfied.

[0052] As a preferred solution of the visualization control method of the transfer assembly process described in the present application, wherein: based on the assembly tree and the cross-tree assembly group, the synchronous assembly of the transfer across the assembly line is performed, specifically including: obtaining the assembly sequence of the corresponding transfer based on the assembly tree; performing the synchronous assembly of the transfer across the assembly line according to the parts assembly order recorded in the assembly sequence and the cross-tree assembly group; wherein any cross-tree assembly group shares assembly resources.

[0053] Compared with the prior art, the beneficial effects achieved by this application are as follows:

[0054] This application achieves dynamic optimization of the assembly process by adjusting the tree structure in real time through intelligent algorithms. The assembly sequence and method can be flexibly changed based on factors such as equipment differences and environmental changes, improving the adaptability and efficiency of the assembly process. Intelligent optimization techniques such as genetic algorithms are used to find the optimal assembly sequence, reducing assembly time, costs, and energy consumption.

[0055] This application solution can assist the same worker to operate synchronously on the production line of transfer machines of different specifications. By aligning and grouping cross-tree nodes, workers can handle similar process nodes of multiple production lines at the same time, breaking through the limitation of traditional solutions that can only assemble transfer machines of the same model in parallel, dynamically coordinating the assembly rhythm of key nodes of multiple transfer machines, avoiding global progress delays caused by delays in a single production line, and improving overall assembly coordination efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] To more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present application. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0057] Figure 1 A flow chart of the visual control method for the transfer assembly process provided in this application;

[0058] Figure 2 This is a flowchart of the assembly tree optimization method provided in this application. DETAILED DESCRIPTION

[0059] The technical solution of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. Unless there is a conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0060] This embodiment introduces a visual control method for the transfer assembly process. Figure 1 , the method comprises the following steps:

[0061] Collect the parts list involved in the turnaround assembly process and the assembly information for each part;

[0062] The assembly information includes basic part information, assembly simulation animation, three-dimensional model, assembly process operation content, parts requirements, and assembly resource requirement summary;

[0063] Each of the parts is a transfer component that needs to be assembled and cannot be disassembled; in other words, each of the parts is the smallest unit involved in the assembly process.

[0064] generating an assembly tree for each transfer machine based on the parts list and assembly information;

[0065] Reference Figure 2 , the generation method of the assembly tree is as follows:

[0066] S201: Set the root node as the transfer machine and all parts as leaf nodes;

[0067] S202: searching for a parent node for each leaf node based on the assembly information; the parent node of any leaf node is the leaf node or the root node corresponding to the assembly position of the part corresponding to the leaf node;

[0068] S203: Connect all nodes according to the relationship between parent and child nodes to generate an assembly tree;

[0069] S204: Check the assembly logic relationship between all parts corresponding to all nodes in each layer of the assembly tree, and adjust the structure of the assembly tree; the details are as follows:

[0070] If there is an assembly order between the parts corresponding to any two nodes at the same level in the assembly tree, the two nodes corresponding to the parts with an assembly order will be adjusted to adjacent positions and connected with a dotted line, with the node corresponding to the part with the earlier assembly order on the left and the node corresponding to the part with the later assembly order on the right.

[0071] In order to avoid assembly logic errors when optimizing the assembly sequence later, it is necessary to ensure that the nodes at each layer in the assembly tree are relatively independent assembly units, and the assembly work on the nodes at the same layer should not interfere with each other; for nodes in any layer that interfere with each other during the assembly process and have an assembly logic sequence, they should be arranged adjacent to each other in a logical order and connected with dotted lines. Then, when optimizing the assembly tree later, the assembly sequence connected by the dotted lines will be retained, thereby avoiding assembly logic errors.

[0072] Genetic algorithms are used in combination with visual assembly simulation to optimize the assembly tree of each transfer machine;

[0073] The method to optimize the assembly tree of each transfer machine is as follows:

[0074] S301: Set the population size, crossover probability, and mutation probability, and use the method described in S201-S204 as follows:

[0075] Each transfer machine generates N assembly trees, where N is the population size;

[0076] S302: Generate N individuals based on the N assembly trees to form an initial population;

[0077] Assuming that the assembly tree has a total of n layers, any of the individuals contains n-1 chromosomes, numbered chromosome 1, chromosome 2, ..., chromosome n-1; wherein chromosome i is a one-dimensional array consisting of the nodes of the i+1th layer from top to bottom of the assembly tree, each element in the array corresponds to a node, and the order of the elements in the array is the order from left to right of the nodes of the i+1th layer;

[0078] S303: Calculate the fitness of each individual; the method is as follows:

[0079] Obtain the assembly sequence of each individual by recording the assembly order of each part in the assembly tree corresponding to each individual from top to bottom and from left to right, and forming an assembly sequence;

[0080] The assembly process of the transfer machine is simulated through the digital twin system to obtain the fitness parameters when the transfer machine is assembled according to the assembly sequence of each individual;

[0081] Calculating the assembly benefit score, process optimization score, and resource utilization rate of each individual based on the fitness parameters;

[0082] The fitness of each individual is calculated based on the assembly benefit score, process optimization score, and resource utilization.

[0083] The digital twin model can simulate the real assembly process in real time, including the interaction between each link and resource consumption. The preferred fitness parameters in this embodiment include time benefit score, energy consumption coefficient, redirection ratio, tooling change ratio, number of interferences, assembly path length, and assembly line utilization rate;

[0084] Furthermore, the calculation formula for the preferred fitness in this embodiment is as follows:

[0085] ;

[0086] in, represents the fitness of the jth individual, and the value range of j is 1, 2, ..., N; represents the assembly benefit score of the jth individual; represents the process optimization score of the jth individual; represents the resource utilization rate of the jth individual; and is the weight coefficient, which is set according to experience;

[0087] The calculation formula of the assembly efficiency score is as follows:

[0088] ;

[0089] in, It represents the time efficiency score of turnkey assembly according to the assembly sequence represented by the jth individual, expressed by the number of assembled parts completed per unit time; The energy consumption coefficient of the assembly sequence represented by the jth individual is represented by the normalized value of energy consumption;

[0090] The calculation formula of the process optimization score is as follows:

[0091] ;

[0092] in, It represents the redirection ratio of the assembly sequence represented by the jth individual; that is, the ratio of the number of process reordering caused by objective reasons to the total number of assembly processes during the assembly process;

[0093] It represents the proportion of tooling changes for turnkey assembly according to the assembly sequence represented by the jth individual; that is, the ratio of the total tooling change time to the total assembly time within the assembly cycle; tooling change refers to the act of replacing or adjusting the original tooling equipment, tools or fixtures.

[0094] It represents the number of interferences during the assembly process of the assembly sequence represented by the jth individual, that is, the number of times two or more components cannot be assembled normally due to collision or obstruction between them due to design or positioning errors.

[0095] The resource utilization rate is calculated as follows:

[0096] ;

[0097] in, The jth individual represents the assembly line utilization rate for turnkey assembly of the assembly sequence represented by the jth individual. This refers to the ratio of the effective working time of the assembly line during the assembly process to the total available time. Effective working time refers to the time the assembly line actually performs assembly operations, i.e., the time from line start to line stop, excluding ineffective time caused by mold changes, adjustments, and material supply delays, and the time it actually spends in production. Total available time refers to the time period during which the assembly line can theoretically operate continuously each day. This refers to the total time the line can be used for production each day, excluding breaks and unplanned downtime.

[0098] It represents the assembly path length of the assembly sequence represented by the jth individual for turnkey assembly, and refers to the cumulative moving distance of all parts of the turnkey during the assembly process.

[0099] S304: performing selection operation, crossover operation, and mutation operation in sequence, and generating the next generation population;

[0100] The crossover operation includes single-point crossover and ordered crossover; the object of the crossover operation is any pair of chromosomes with the same number in the two individuals participating in the crossover operation;

[0101] The mutation operations include exchange mutation, reverse mutation, and adjacent exchange mutation;

[0102] Chromosomes with the same number represent nodes from the same layer on different assembly trees; crossover operations are only performed on nodes at the same layer, ensuring the structural stability of the assembly tree and avoiding logical errors in the assembly order.

[0103] Swap mutation involves randomly selecting two different positions on a chromosome and swapping the elements between them. This may improve the ordering of the sequence. Reverse mutation involves randomly selecting a continuous subsequence on a chromosome and reversing its order. Insertion mutation involves randomly removing an element from a chromosome and reinserting it at a random position afterward, similar to an element jumping within the sequence. Adjacent swap mutation involves randomly selecting two adjacent elements on a chromosome and swapping their positions.

[0104] The single-point crossover method works as follows: at one or more predetermined crossover points, portions of the parent chromosome's sequence are swapped to form a daughter chromosome. For example, if there are two chromosomes A and B, and a crossover point p is chosen, the daughter chromosome will consist of the first p elements of A plus the last Mp elements of B, and vice versa. M represents the number of elements on the chromosome involved in the crossover.

[0105] The ordered crossover method works as follows: Select a subset of subsequences from a pair of parent sequences, preserve the order of the elements within the subsequence, and insert it into the corresponding position of the other parent sequence, so that the resulting subsequences are as close to the required order as possible. This crossover method is particularly well-suited for sequence optimization problems, as it ensures that the resulting subsequences retain the relative order of the parent sequences.

[0106] The assembly tree itself covers the assembly constraint information between parts. Using a genetic algorithm to select the best assembly tree designed in this application is simpler and more efficient than directly performing genetic mutation on the assembly sequence and then constraining it with external conditions.

[0107] S305: Repeat S303-S304 until the highest fitness of all individuals in the population converges;

[0108] S306: Obtain the assembly tree corresponding to the individual with the highest fitness as the optimization result of the assembly tree corresponding to the transition.

[0109] Assign an assembly line to each transfer machine based on the assembly tree, and perform time alignment on the assembly trees of any group of adjacent assembly lines;

[0110] Any group of adjacent assembly lines is two spatially adjacent assembly lines; any assembly line is used to perform assembly of the corresponding transfer machine according to an assembly tree; any production line is only divided into a group of adjacent assembly lines; the assembly line is assigned to each transfer machine based on the assembly tree, specifically including: if any two assembly trees have the same structure, the two assembly trees with the same structure are assigned to a group of adjacent assembly lines.

[0111] Prioritize allocating two assembly trees with the same structure to a group of adjacent assembly lines, and then allocate the remaining assembly trees with different structures. This is conducive to the parallel assembly of the corresponding assembly trees with the same structure on adjacent assembly lines, thereby improving assembly efficiency.

[0112] The time alignment of the assembly trees of any group of adjacent assembly lines specifically includes:

[0113] The assembly process of the transfer machine is simulated through the digital twin system to obtain the node characteristics of each node in the assembly tree corresponding to each transfer machine;

[0114] Based on the node features, the similarity between nodes is calculated, and a cross-tree association matrix is ​​constructed to record the similarity;

[0115] The node features include part type, assembly tools, and spatial coordinates; for example, the part type is a classification label such as bolts and gears; the assembly tools are the wrench models and hydraulic equipment required to assemble the corresponding parts; the spatial coordinates are obtained based on the digital twin system, and the spatial coordinates are the three-dimensional spatial coordinates of the assembly position of the corresponding part in the assembly workshop.

[0116] The similarity between the calculated nodes is calculated based on the node features corresponding to each item of the two nodes; a preferred method for calculating the similarity between nodes in this embodiment is as follows: the basic similarity between any two nodes is 0; if the part types of the two nodes are the same, the similarity is increased by 0.6, otherwise the similarity is increased by 0; if there is an identical assembly tool between the two nodes, the similarity is increased by 0.1, and the upper limit of this similarity is 0.3; the normalized Euclidean distance of the spatial coordinates of the two nodes is calculated, and the similarity is adjusted based on the normalized Euclidean distance, and the larger the normalized Euclidean distance, the higher the similarity; if the similarity is adjusted based on the time coordinates of the two nodes, the larger the intersection between the time ranges of assembling the corresponding parts of the two nodes, the higher the similarity.

[0117] Any element in the cross-tree association matrix represents the similarity between two nodes on different assembly trees in a group of adjacent assembly lines; let the two assembly trees of the adjacent assembly lines be C and D respectively, then the element in the p-th row and q-th column of the cross-tree association matrix represents the similarity between the p-th node in assembly tree C and the q-th node in assembly tree D, where p and q are both positive integers.

[0118] Based on the cross-tree association matrix, the assembly trees of adjacent assembly lines are time-aligned, specifically including:

[0119] Setting a similarity threshold; traversing the cross-tree association matrix, marking two nodes corresponding to elements in the cross-tree association matrix that are greater than the similarity threshold as similar node pairs;

[0120] Mark the layers of the two nodes in each set of similar node pairs in the corresponding assembly tree as alignment layers;

[0121] The alignment layers between the assembly trees of adjacent assembly lines are subjected to conflict detection. If the conflict detection fails, the alignment layers are adjusted until the alignment layers between the assembly trees of adjacent assembly lines pass the conflict detection, thereby completing the temporal alignment of the assembly trees.

[0122] The preferred method of conflict detection and alignment layer adjustment in this embodiment is as follows:

[0123] Use straight lines to connect any set of alignment layers of two assembly trees; if there is any intersection or cross between the straight lines connecting the alignment layers, the conflict detection will not pass. Delete the alignment layers that cause the straight lines to intersect or cross until the alignment layers between the assembly trees pass the conflict detection. For example, the third layer in assembly tree C and the third and fourth layers of assembly tree D are all alignment layers, then the straight lines connecting the third layer in assembly tree C with the third and fourth layers of assembly tree D intersect. Delete the alignment layers corresponding to the third layer in assembly tree C and the fourth layer of assembly tree D, and only retain the alignment layers corresponding to the third layer in assembly tree C and the third layer of assembly tree D, so that the alignment layers between assembly trees C and D pass the conflict detection.

[0124] Based on the time-aligned assembly tree, the cross-tree assembly groups are divided; specifically, the following are included:

[0125] Divide any set of similar node pairs in any set of alignment layers in the time-aligned assembly tree into a cross-tree assembly group;

[0126] Based on any node in the spanning tree assembly group as a starting point, searching for nodes that can be added to the spanning tree assembly group on the corresponding assembly tree and expanding the spanning tree assembly group, specifically including:

[0127] Determining whether any associated node of the starting point satisfies a constraint condition for joining a corresponding assembly group; if any associated node satisfies the constraint condition, adding the associated node to the corresponding assembly group; the associated nodes include a parent node and child nodes of the starting point, and nodes at the same level of the assembly tree as the starting point;

[0128] Determining whether the associated nodes satisfy the constraint conditions specifically includes: reading the part type, assembly tool, and spatial coordinates of each node in the cross-tree assembly group;

[0129] If the associated node has the same part type as any node in the spanning tree assembly group, the constraint condition is satisfied;

[0130] If the number of assembly tools introduced by the associated node is less than m, and the average distance between the associated node and all nodes in the spanning tree assembly group is less than a preset distance threshold, then the constraint condition is satisfied; m is a positive integer.

[0131] The assembly tools introduced by the associated node are those required by the associated node but not required by any other node in the cross-tree assembly group. The Euclidean distance between the associated node and each node in the cross-tree assembly group is calculated based on spatial coordinates and averaged to obtain the average distance between the associated node and all nodes in the cross-tree assembly group. Based on these constraints, the nodes in the cross-tree assembly group can be guaranteed to have uniform assembly tools and uniform operating ranges.

[0132] Based on the assembly tree and the cross-tree assembly group, synchronous assembly of the cross-assembly line transfer is performed; specifically, the steps include:

[0133] Based on the assembly tree, the assembly sequence of the corresponding transfer is obtained; according to the assembly sequence of the parts and the cross-tree assembly group recorded in the assembly sequence, the cross-assembly line transfer is synchronously assembled; wherein, any cross-tree assembly group shares assembly resources, and when the assembly

[0134] Assembly resources shared by cross-tree assembly groups include assembly workers, assembly tools, and parts libraries. For fully or semi-automated assembly lines, these resources also include robotic arms and automated material carts. Each cross-tree assembly group performs continuous assembly as a unit, completing the assembly of all parts within the group before assembling parts outside the group. By dividing cross-tree assembly groups and sharing assembly resources, resource allocation costs between different assembly lines are optimized, improving assembly efficiency and reducing assembly costs.

[0135] Ideally, when performing simultaneous assembly of a transfer machine across assembly lines, a digital twin system can be used to simultaneously simulate the assembly visually, displaying the operational process of each assembly step in real time. This allows for remote monitoring and guidance of on-site workers on the correct installation steps, reducing operational errors and improving assembly accuracy. By comparing the simulated assembly with the actual assembly progress in real time, the assembly process can be monitored, delays or potential problems can be quickly identified, and appropriate measures can be taken to ensure that the transfer machine assembly work is completed on time.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0137] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are protected by the present application.

Claims

1. A visual control method for a transfer assembly process, characterized by: The following steps are involved: Collect the parts list involved in the turnaround assembly process and the assembly information for each part; generating an assembly tree for each transfer machine based on the parts list and assembly information; Genetic algorithms are used in combination with visual assembly simulation to optimize the assembly tree of each transfer machine; Assign an assembly line to each transfer machine based on the assembly tree, and perform time alignment on the assembly trees of any group of adjacent assembly lines, including: The assembly process of the transfer machine is simulated through the digital twin system to obtain the node features of each node in the assembly tree corresponding to each transfer machine; the node features include part type, assembly tool and spatial coordinates; Based on the node features, the similarity between nodes is calculated, and a cross-tree association matrix is ​​constructed to record the similarity; Based on the cross-tree association matrix, the assembly trees of adjacent assembly lines are time-aligned, specifically including: Setting a similarity threshold; traversing the cross-tree association matrix, marking two nodes corresponding to elements in the cross-tree association matrix that are greater than the similarity threshold as similar node pairs; Mark the layers of the two nodes in each set of similar node pairs in the corresponding assembly tree as alignment layers; Perform conflict detection on the alignment layers between the assembly trees of adjacent assembly lines. If the conflict detection fails, adjust the alignment layers until the alignment layers between the assembly trees of adjacent assembly lines pass the conflict detection, thus completing the temporal alignment of the assembly trees. Based on the time-aligned assembly tree, the cross-tree assembly groups are divided, including: Divide any set of similar node pairs in any set of alignment layers in the time-aligned assembly tree into a cross-tree assembly group; Based on any node in the spanning tree assembly group as a starting point, searching for nodes that can be added to the spanning tree assembly group on the corresponding assembly tree and expanding the spanning tree assembly group, specifically including: Determining whether any associated node of the starting point satisfies a constraint condition for joining a corresponding assembly group; if any associated node satisfies the constraint condition, adding the associated node to the corresponding assembly group; the associated nodes include a parent node and child nodes of the starting point, and nodes at the same level of the assembly tree as the starting point; Determining whether the associated nodes satisfy the constraint conditions specifically includes: reading the part type, assembly tool, and spatial coordinates of each node in the spanning tree assembly group; If the associated node has the same part type as any node in the spanning tree assembly group, the constraint condition is satisfied; If the number of assembly tools introduced by the associated node is less than m, and the average distance between the associated node and all nodes in the spanning tree assembly group is less than a preset distance threshold, then the constraint condition is satisfied; Based on the assembly tree and the cross-tree assembly group, synchronous assembly of transfer machines across assembly lines is performed.

2. The method for visual control of the transfer assembly process according to claim 1, characterized in that: The method for generating the assembly tree is as follows: S201: Set the root node as the transfer machine and all parts as leaf nodes; S202: searching for a parent node for each leaf node based on the assembly information; the parent node of any leaf node is the leaf node or the root node corresponding to the assembly position of the part corresponding to the leaf node; S203: Connect all nodes according to the relationship between parent and child nodes to generate an assembly tree; S204: Check the assembly logic relationship between all parts corresponding to all nodes in each layer of the assembly tree, and adjust the structure of the assembly tree; the details are as follows: If there is an assembly order between the parts corresponding to any two nodes at the same level in the assembly tree, the two corresponding nodes will be adjusted to adjacent positions and connected with a dotted line, with the node corresponding to the part with the earlier assembly order on the left and the node corresponding to the part with the later assembly order on the right.

3. The visual control method for the transfer assembly process according to claim 2, characterized in that: The method for optimizing the assembly tree of each transfer machine is as follows: S301: Set the population size, crossover probability, and mutation probability, and generate N assembly trees for each transfer machine based on the method described in S201-S204, where N is the population size; S302: Generate N individuals based on the N assembly trees to form an initial population; S303: Calculate the fitness of each individual; S304: performing selection operation, crossover operation, and mutation operation in sequence to generate the next generation population; wherein the object of the crossover operation is any pair of chromosomes with the same number in the two individuals participating in the crossover operation; S305: Repeat S303-S304 until the highest fitness of all individuals in the population converges; S306: Obtain the assembly tree corresponding to the individual with the highest fitness as the optimization result of the assembly tree corresponding to the transition.

4. The method for visual control of the transfer assembly process according to claim 3, wherein: The method for generating the individuals based on the assembly tree is as follows: suppose the assembly tree has a total of n layers, then any of the individuals contains n-1 chromosomes, numbered as chromosome 1, chromosome 2,..., chromosome n-1 respectively; wherein chromosome i is a one-dimensional array composed of nodes of the i+1th layer from top to bottom of the assembly tree, each element in the array corresponds to a node, and the arrangement order of the elements in the array is that the nodes of the i+1th layer are sorted from left to right.

5. The visual control method for the transfer assembly process according to claim 4, characterized in that: The method for calculating the fitness of each individual is as follows: Obtain the assembly sequence of each individual by recording the assembly order of each part in the assembly tree corresponding to each individual from top to bottom and from left to right, and forming an assembly sequence; The assembly process of the transfer machine is simulated through the digital twin system to obtain the fitness parameters when the transfer machine is assembled according to the assembly sequence of each individual; Calculating the assembly benefit score, process optimization score and resource utilization rate of each individual based on the fitness parameters; The fitness of each individual is calculated based on the assembly benefit score, the process optimization score and the resource utilization rate.

6. The visual control method for the transfer assembly process according to claim 5, characterized in that: Any group of adjacent assembly lines is two spatially adjacent assembly lines; any assembly line is used to perform the assembly of a corresponding transfer machine according to an assembly tree; Any production line is only divided into a group of adjacent assembly lines; the assembly line is assigned to each transfer machine based on the assembly tree, specifically including: if any two assembly trees have the same structure, the two assembly trees with the same structure are assigned to a group of adjacent assembly lines; Any element in the cross-tree association matrix represents the similarity between two nodes on different assembly trees in a group of adjacent assembly lines; let the two assembly trees of the adjacent assembly lines be C and D respectively, then the element in the p-th row and q-th column of the cross-tree association matrix represents the similarity between the p-th node in assembly tree C and the q-th node in assembly tree D.

7. The visual control method for the transfer assembly process according to claim 6, characterized in that: Based on the assembly tree and the cross-tree assembly group, synchronous assembly of the transfer across the assembly line is performed, specifically including: obtaining the assembly sequence of the corresponding transfer based on the assembly tree; performing synchronous assembly of the transfer across the assembly line according to the parts assembly order recorded in the assembly sequence and the cross-tree assembly group; wherein any cross-tree assembly group shares assembly resources.

Images