Integrated method for multi-AGV path planning and multi-row equipment layout optimization

Abstract

This invention provides an integrated method and system for multi-AGV path planning and multi-row equipment layout. The method includes: establishing a multi-objective optimization model for multi-row equipment layout; solving for the total processing time in the iteration: for a given layout, a multi-AGV path planning method based on the time window A* algorithm is used, combined with information on the product's process route and processing time, to solve for the shortest time required for the workshop to complete the order; an ant colony algorithm combined with a genetic algorithm is used to solve the model; and dynamic simulation based on discrete events is used to verify the accuracy of the solution. This invention solves the technical problems of poor equipment layout and production evaluation effects in workshops, especially in mixed-flow processing production lines.

Description

Technical Field

[0001] This invention belongs to the field of factory virtual simulation and optimization, specifically involving a method and system for integrating multi-AGV path planning and multi-row equipment layout. Background Technology

[0002] With the rapid development of technology, manufacturing, as a fundamental industry, is increasingly demonstrating its importance in the national economy. In the design process of a production system, determining the equipment layout is an indispensable and even crucial aspect. For every production system, a reasonable equipment layout can improve its production efficiency. AGVs play an indispensable role in intelligent logistics systems. They are not only a primary means for enterprises to achieve digital and automated production models, but also a major indicator of the level of intelligence in manufacturing. Therefore, the quality of their path planning directly determines the production efficiency of an intelligent manufacturing system. For example, the existing invention patent application document CN110442105A, entitled "A Digital Workshop for Lithium Battery Materials," includes: a digital production equipment layer, comprising intelligent production equipment, intelligent transmission equipment, intelligent inspection equipment, and intelligent public infrastructure equipment; a process control layer: based on an intelligent simulation system and a PLC-based underlying process control system, a lithium battery material production workshop network system is established using industrial Ethernet to construct a lithium battery material production process data acquisition and monitoring system, enabling interconnection and information exchange among various intelligent devices; and a production execution layer: an intelligent manufacturing execution system for the lithium battery material production workshop is established with the goal of intelligent optimization scheduling and quality control, realizing production progress management, material management, equipment management, quality management, and intelligent production scheduling. Furthermore, the method described in CN105354648A, entitled "Modeling and Optimization Method for AGV Scheduling Management," includes the following steps: Step 1: AGV running path network modeling; Step 2: Path network map storage; Step 3: Binary modeling of multi-AGV systems; and Step 4: Dynamic path optimization method using local search.

[0003] Comprehensive optimization of equipment layout and AGV paths can significantly reduce logistics costs in the production and processing process. These two issues are also two very typical problems in the field of industrial engineering, and there is a close connection between them. In order to further optimize the factory logistics system, it is necessary to explore the coupling relationship between AGV path planning and equipment layout to achieve a match with productivity levels.

[0004] As part of a smart workshop, the efficiency of AGV operation is closely related to the workshop's production efficiency. Regarding facility layout, the scheduling of AGVs, when orders are confirmed, greatly influences the evaluation index of whether such a layout is good or bad. Therefore, we need to optimize the scheduling of AGVs so that it can be used as an evaluation index for equipment layout.

[0005] Equipment layout is the process of determining the spatial location of production units, such as processing equipment, within a given production workshop. For multi-row equipment layouts, compared to single-row and circular equipment layouts, it is also necessary to consider the issue of row crossing during the equipment layout process.

[0006] In summary, existing technologies have technical problems such as poor equipment layout in workshops, especially in mixed-flow processing production lines, and poor production evaluation results. Summary of the Invention

[0007] The technical problem to be solved by this invention is how to solve the technical problem of poor production evaluation results in the reconstruction of equipment layout in workshops, especially mixed-flow processing production lines.

[0008] This invention solves the above-mentioned technical problems by employing the following technical solution: An integrated method for multi-AGV path planning and multi-row equipment layout includes:

[0009] S1. Collect workshop operation data, solve for no less than two clustering indices, perform linear weighted processing of the clustering indices to obtain the model objective function, and establish a multi-objective optimization model for multi-row equipment layout based on the model objective function and preset model constraints.

[0010] S2. For the pre-set layout, the A* algorithm based on the time window is used for multi-AGV path planning. Combined with workshop operation data, the total processing time in the iteration is obtained.

[0011] S3. Using an ant colony algorithm combined with a genetic algorithm, the final layout is determined based on the total processing time during the iteration.

[0012] S4. Perform dynamic simulation based on preset discrete events to verify the accuracy of the multi-objective optimization model for multi-row device layout according to the final layout.

[0013] This invention implements an integrated multi-AGV path planning and multi-row equipment layout method. It constructs a model using clustering indices and model constraints, calculates the total processing time using the A* algorithm, and obtains the final layout using an ant colony algorithm combined with a genetic algorithm. The constructed model is then dynamically simulated and verified. This method optimizes equipment layout in the workshop and evaluates production, enabling more efficient completion of corresponding production tasks and guiding workshop production optimization. This method is applicable to equipment layout reconstruction in mixed-flow machining production lines and also provides valuable insights for similar workshop layouts.

[0014] In a more specific technical solution, step S1 also includes the following specific steps:

[0015] Step S11: Determine the objective function of the model using the following logic:

[0016] min Z=μ1T+μ2C+μ3K (1)

[0017] Where T is the order completion time, C is the transportation cost, K is the clustering index between devices, and μ1, μ2, and μ3 represent the weights of order completion time, transportation cost, and clustering index between devices, respectively.

[0018] Step S12: Define the clustering index K between devices using the following logic:

[0019]

[0020] Where, d ij S is the distance between devices i and j. t This is a set of equipment that performs the same type of processing operation, where m is the number of processing operations.

[0021] S13. Determine the weights among the three objectives, and determine the weights of the clustering index using the entropy weight method;

[0022] S14. Determine the preset model constraints based on the spatial constraints between devices.

[0023] In a more specific technical solution, step S13 includes:

[0024] Step S13 also includes:

[0025] S131. Using the following logic, obtain the preprocessed data using the min-max normalization method:

[0026]

[0027] Among them, X ij For sample i, X is the data before standardization of the j-th indicator. i For the data in sample i, Y ijThe data is the standardized data of the j-th indicator for sample i.

[0028] S132. Calculate the preprocessed data to obtain the weight P of the clustering index. ij :

[0029]

[0030] S133. Based on the weight of the clustering index, calculate the information entropy E of the index using the following logic. j :

[0031]

[0032] S134. Use the following logic to process information entropy to determine the weight of the clustering index:

[0033]

[0034] Where E j The information entropy of indicator j, k is the number of indicators, and w j The weights of the calculated clustering index.

[0035] In a more specific technical solution, step S2 also includes:

[0036] S21. Construct a grid map so that each device is located at a point in the grid;

[0037] S22. Minimize the workshop processing time according to the preset AGV scheduling rules;

[0038] S23. For AGVs that have been dispatched for transportation tasks, use the A* algorithm based on time windows to plan the path of the dispatched AGVs in the grid map.

[0039] S24. To address node conflicts and opposing conflicts, adjust the AGV path by constructing AGV occupancy time windows for different nodes.

[0040] S25. Repeat steps S23 to S24 until all products are processed to calculate the total processing time in the iteration.

[0041] This invention utilizes the A* algorithm based on time windows to obtain time windows for passing through nodes during AGV path planning. By using time windows to detect whether a conflict will occur, a waiting measure is adopted to wait at the node before the conflict node, thereby reducing the collision risk during the operation of each AGV.

[0042] In a more specific technical solution, step S23 of the A* algorithm based on time windows further includes:

[0043] S231. Obtain the previous task path for all AGV paths except the current AGV path.

[0044] S232. Set the non-loading / unloading points on the previous task path as obstacles;

[0045] S233. Combining the start and end points and obstacles of the previous task, call the A* algorithm to obtain the applicable path;

[0046] S234. Obtain the time window of the nodes traversed by the current task based on the applicable path;

[0047] S235. Compare the current time window with the total time window to determine whether there is a node conflict;

[0048] S236. If so, wait for time t and shift the time window.

[0049] S237. Obtain the new AGV path;

[0050] S238. If not, add the current time window that is the AGV path to the total time window.

[0051] In a more specific technical solution, in step S23, a turning penalty term P is introduced into the evaluation function of the time-window-based A* algorithm:

[0052]

[0053] In the formula, (x f y f Let be the coordinates of the parent node of point S, and C be a constant. s ,y s ) represents the coordinates of the current node S.

[0054] This invention takes into account that AGVs need to change speed when changing direction in actual work. It introduces a turning penalty term P into the evaluation function of the A* algorithm, which reduces energy consumption and improves transportation efficiency.

[0055] In a more specific technical solution, step S3 includes:

[0056] S31. Initialize the population and update the pheromone concentration. Execute step S2 to solve for the total processing time and processing and transportation cost in the iteration. Based on the spatial characteristics of the initial layout, calculate the clustering index between the devices and update the pheromone concentration accordingly.

[0057] S32. Calculate the ant state transition probability based on the pheromone concentration, and process it according to the model objective function to obtain the intermediate population;

[0058] S33. Using a genetic algorithm, crossover and mutation are performed on the intermediate population to obtain the final offspring based on the preset fitness parameters.

[0059] S34. Update the pheromone concentration on the path through the offspring, and iterate through steps S32 to S34 to obtain the final layout.

[0060] This invention employs an ant colony algorithm combined with a genetic algorithm to solve the problem. This method updates the pheromone concentration based on the clustering index, obtains the intermediate population by processing with the objective function, and achieves iterative updates of the offspring through differential mutation. This method can overcome the shortcomings of the ant colony algorithm in the convergence process, which is not fast enough, and the genetic algorithm in the convergence process, which is prone to getting trapped in local optima.

[0061] In a more specific technical solution, in step S4, the path of the dispatched AGV is planned in the grid map using the A* algorithm based on time windows.

[0062] In a more specific technical solution, in step S4, a simulation model is run to obtain the processing and transportation equipment status of the production line, and compared with the final layout to obtain the accuracy verification results of the multi-objective optimization model for the multi-row equipment layout.

[0063] In a more specific technical solution, the integrated multi-AGV path planning and multi-row equipment layout system includes:

[0064] The model building module is used to collect workshop operation data, solve for no less than two clustering indices, linearly weight the clustering indices to obtain the model objective function, and establish a multi-objective optimization model for multi-row equipment layout based on the model objective function and preset model constraints.

[0065] The total time processing module is used to perform multi-AGV path planning based on the A* algorithm with a time window for the preset layout, and to calculate the total processing time in the iteration by combining workshop operation data.

[0066] The layout solving module uses an ant colony algorithm combined with a genetic algorithm to solve for the final layout based on the total processing time during the iteration. The layout solving module is connected to the total time processing module.

[0067] The dynamic simulation verification module is used to perform dynamic simulation based on preset discrete events. According to the final layout, it verifies the accuracy of the multi-objective optimization model for multi-row device layout. The dynamic simulation verification module is connected to the layout solution module and the model building module.

[0068] Compared with existing technologies, this invention has the following advantages: The integrated multi-AGV path planning and multi-row equipment layout method implemented in this invention constructs a model using clustering indices and model constraints, calculates the total processing time using the A* algorithm, and obtains the final layout using an ant colony algorithm combined with a genetic algorithm. This allows for dynamic simulation verification of the constructed model, optimizing equipment layout in the workshop and evaluating production to complete corresponding production tasks more efficiently, thus guiding workshop production optimization. This method is applicable to equipment layout reconstruction of mixed-flow processing production lines and also provides valuable insights for similar workshop layouts.

[0069] This invention utilizes the A* algorithm based on time windows to obtain time windows for passing through nodes during AGV path planning. By using time windows to detect whether a conflict will occur, a waiting measure is adopted to wait at the node before the conflict node, thereby reducing the collision risk during the operation of each AGV.

[0070] This invention takes into account that AGVs need to change speed when changing direction in actual work. It introduces a turning penalty term P into the evaluation function of the A* algorithm, which reduces energy consumption and improves transportation efficiency.

[0071] This invention employs an ant colony algorithm combined with a genetic algorithm to solve the problem. This method updates the pheromone concentration based on the clustering index, obtains the intermediate population by processing with the objective function, and achieves iterative updates of the offspring through differential mutation. This method can overcome the shortcomings of the ant colony algorithm in the convergence process, which is not fast enough, and the genetic algorithm in the convergence process, which is prone to getting trapped in local optima.

[0072] This invention solves the technical problems of poor production evaluation results in the prior art, particularly in the reconstruction of equipment layout in workshops, especially in mixed-flow processing production lines. Attached Figure Description

[0073] Figure 1 This is a schematic diagram of the basic steps of the integrated multi-AGV path planning and multi-row equipment layout method in Embodiment 1 of the present invention;

[0074] Figure 2 This is a flowchart of the A* algorithm based on time windows in Embodiment 1 of the present invention;

[0075] Figure 3 This is a schematic diagram illustrating the specific construction steps of the multi-row device layout multi-objective optimization model in Embodiment 1 of the present invention;

[0076] Figure 4 This is a schematic diagram illustrating the specific steps for calculating the total processing time in Embodiment 1 of the present invention;

[0077] Figure 5 This is a schematic diagram illustrating the specific steps of model solving in Embodiment 1 of the present invention;

[0078] Figure 6 This is a schematic diagram of the ant colony algorithm process that incorporates a genetic algorithm in Embodiment 1 of the present invention;

[0079] Figure 7 This is a schematic diagram illustrating the specific verification steps for the accuracy of the solution scheme in Embodiment 1 of the present invention;

[0080] Figure 8 This is a schematic diagram of the data flow processing of the integrated multi-AGV path planning and multi-row equipment layout system in Embodiment 2 of the present invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. 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.

[0082] Example 1

[0083] like Figure 1 As shown, the integrated multi-AGV path planning and multi-row equipment layout method provided by the present invention includes the following basic steps:

[0084] Step S1: Establish a multi-objective optimization model for the layout of multiple rows of devices;

[0085] In this embodiment, given the workpiece processing route and the processing time on each device, for the given order, the three objectives of total processing time, total transportation cost, and clustering index between devices under a certain layout are solved. The linear weighted result of the three is used as the objective function of the model, and the model constraint is the spatial condition constraint between each device.

[0086] Step S2: Solve for the total processing time in the iteration;

[0087] In this embodiment, given a layout, a multi-AGV path planning method based on the time window A* algorithm is used to solve for the shortest time required for the workshop to complete the order, combined with information on the product's process route and processing time.

[0088] like Figure 2 As shown, in this embodiment, the A* algorithm based on time windows further includes the following specific steps:

[0089] S21. Obtain the previous task path of other AGVs;

[0090] S22. Set non-loading / unloading points on the path as obstacles;

[0091] S23. Combine the task start and end points with obstacles to call the A* algorithm to obtain the path;

[0092] S24. Obtain the time window of the nodes traversed by the current task based on the path;

[0093] S25. Compare the current time window with the total time window to determine if there are any node conflicts;

[0094] S26. If so, calculate the waiting time t and shift the time window.

[0095] S27. Obtain a new path;

[0096] S28. If not, add the current time window to the total time window.

[0097] Step S3: Solve the model using an ant colony algorithm that combines genetic algorithms;

[0098] Step S4: Verify the accuracy of the solution scheme using dynamic simulation based on discrete events.

[0099] like Figure 3 As shown, in this embodiment, step S1 further includes the following specific steps:

[0100] Step S11: Determine the expression of the objective function, defined as follows:

[0101] min Z=μ1T+μ2C+μ3K (1)

[0102] Where T is the order completion time, C is the transportation cost, K is the clustering index between devices, and μ1, μ2, and μ3 represent the weights of order completion time, transportation cost, and clustering index between devices, respectively.

[0103] Step S12: Define the clustering index between devices;

[0104] In this embodiment, the clustering index characterizes equipment that processes the same procedure and needs to be spatially concentrated to meet the needs of production personnel for observation, maintenance, and centralized management. Its value is defined as follows:

[0105]

[0106] Where, d ij S is the distance between devices i and j. t Let m be the set of equipment that performs the same process, and m be the number of processes.

[0107] Step S13: Determine the weights among the three objectives. Use the entropy weighting method to determine the weights of the three indicators. Preprocess the data using the min-max standardization method, then calculate the magnitude of the variation of the indicators, i.e., the proportion of each indicator. Finally, calculate the information entropy of the indicators and determine the weights based on the information entropy.

[0108]

[0109]

[0110]

[0111]

[0112] Among them, X ij For sample i, X is the data before standardization of the j-th indicator. i For the data in sample i, Y ij For the standardized data of the j-th indicator of sample i, E j The information entropy of indicator j, k is the number of indicators, and w j The weights of the calculated clustering index.

[0113] Step S14: Determine model constraints. For the equipment layout model, the constraints are spatial constraints.

[0114] In this embodiment, spatial constraints include, but are not limited to: equipment shape constraints (aspect ratio), safety distance constraints between equipment, non-overlapping constraints of equipment in space, and constraints that the equipment layout does not exceed the workshop area.

[0115] like Figure 4 As shown, in this embodiment, step S2 further includes the following specific steps:

[0116] Step S21: Construct a grid map, with each device located at a point in the grid;

[0117] Step S22: Determine the AGV scheduling rule as FIFS to minimize the workshop processing time;

[0118] Step S23: For the AGVs dispatched with transportation tasks, plan their paths in the grid map using the A* algorithm.

[0119] In this embodiment, considering that AGVs need to change speeds when changing direction in actual operation, a turning penalty term P is introduced into the evaluation function of the A* algorithm to minimize energy consumption and improve transportation efficiency:

[0120]

[0121] Among them, (xf y f Let be the coordinates of the parent node of point S, and C be a constant. s ,y s () represents the coordinates of the current node S. This is done to reduce the number of turns the AGV makes.

[0122] Step S24: There is a risk of collision during the operation of multiple AGVs. To address node conflicts and head-on conflicts, the AGV paths are adjusted by constructing AGV occupancy time windows for different nodes.

[0123] In this embodiment, the path adjustment of the AGV can be implemented by, for example, obtaining the time window of the node to be passed when the AGV is planning the path, detecting whether a conflict will occur through the time window, and adopting a waiting measure to wait at the node before the conflict node to avoid collision.

[0124] Step S25: Repeat steps S23 to S24 until all products are processed, and calculate the final processing time.

[0125] like Figure 5 As shown, in this embodiment, step S3 includes the following specific steps:

[0126] Step S31: Initialize the population and update the pheromone concentration, i.e. the initial layout of the equipment, which is generated randomly. Then, the total processing time and transportation cost during the processing are calculated using the method in step S2. The clustering index between the equipment is calculated based on the spatial characteristics of the initial layout, which serves as the basis for updating the pheromone concentration.

[0127] Step S32: Calculate the ant state transition probability based on the pheromone concentration and select a path to calculate the circumvention cost, which is the objective function in step S1 to obtain the intermediate population.

[0128] Step S33: Input the intermediate population from step S32 into the genetic algorithm and perform crossover and mutation. Here, the crossover operator is PMX crossover operator, and the mutation is two-point exchange mutation. The final offspring are obtained according to the fitness.

[0129] Step S34: Update the pheromone concentration on the path through the offspring, and repeat steps S32 to S34 until the iteration terminates and the final layout is obtained.

[0130] like Figure 6 As shown, in this embodiment, the ant colony algorithm combined with the genetic algorithm further includes the following specific steps:

[0131] S31', Initialization;

[0132] S32', Update the pheromone concentration table;

[0133] S33' Determine if the loop has ended; if so, end the loop.

[0134] S34' If not, then set the current ant k=1;

[0135] S35' Calculate the state transition probability and select a path, then calculate the roaming cost;

[0136] S36'、k=k+1;

[0137] S37' Determine if k is greater than m; if not, proceed to step S35'.

[0138] S38' If so, then use the solution of the ant colony algorithm as the initial population input;

[0139] S39', crossover mutation;

[0140] S310' Calculate the new offspring fitness;

[0141] S311', Update pheromones.

[0142] like Figure 7 As shown, in this embodiment, step S4 includes the following specific steps:

[0143] Step S41: Establish a digital twin workshop model;

[0144] In this embodiment, the process route of the products processed in the workshop, the processing time data of the products on each machine, and the equipment layout obtained in step 3 are obtained to perform parametric modeling of the AGV system, processing equipment, material storage area and the overall workshop.

[0145] Step S42: Run the simulation model;

[0146] In this embodiment, the model is run to obtain statistics on the total processing time of the production line, logistics and transportation efficiency, and equipment utilization, which are then compared and analyzed with the results obtained from the algorithm solution.

[0147] Example 2

[0148] like Figure 8 As shown in this embodiment, the production line of a certain agricultural machinery production workshop is a mixed-flow processing production line with three products, denoted as A, B, and C. Each of the three products has up to 14 processes, and the processing equipment type and processing time for each process are given. The processes for different products are different, and the required processing time is also different. Materials between different equipment are transported by AGV. It is known that the distance between the equipment is equal and the moving speed of the AGV between the equipment is constant. For a certain type of order, the layout optimization of a total of 28 equipment, including the raw material storage location, is required.

[0149] The integrated multi-AGV path planning and multi-row equipment layout optimization method provided by this invention is used to optimize the layout of workshop equipment.

[0150] In this embodiment, a multi-objective optimization model for multi-row equipment layout is established. 500 layouts are randomly generated, and their corresponding processing time, logistics and transportation costs, and clustering indices between equipment are solved. These are used as initial data input into the entropy weight method to determine the weights of the objective function as follows: 0.3219, 0.3370, and 0.3411.

[0151] In this embodiment, the relevant parameters of the ant colony algorithm and the genetic algorithm are set as follows: the number of ants is set to 27 (the number of devices), the pheromone heuristic factor is 0.8, the expected heuristic factor is 1.1, the number of iterations is 100 generations, the pheromone increase constant term is 1, the pheromone evaporation factor is 0.5, the crossover rate is 0.8, and the mutation rate is 0.2.

[0152] In this embodiment, the optimal layout is obtained by solving the model, and the results are verified using simulation software.

[0153] In summary, the integrated multi-AGV path planning and multi-row equipment layout method implemented in this invention constructs a model using clustering indices and model constraints, calculates the total processing time using the A* algorithm, and obtains the final layout using an ant colony algorithm combined with a genetic algorithm. This allows for dynamic simulation verification of the constructed model, optimizing equipment layout in the workshop and evaluating production to more efficiently complete corresponding production tasks and guide workshop production optimization. This method is applicable to equipment layout reconstruction of mixed-flow processing production lines and also provides valuable insights for similar workshop layouts.

[0154] This invention utilizes the A* algorithm based on time windows to obtain time windows for passing through nodes during AGV path planning. By using time windows to detect whether a conflict will occur, a waiting measure is adopted to wait at the node before the conflict node, thereby reducing the collision risk during the operation of each AGV.

[0155] This invention takes into account that AGVs need to change speed when changing direction in actual work. It introduces a turning penalty term P into the evaluation function of the A* algorithm, which reduces energy consumption and improves transportation efficiency.

[0156] This invention employs an ant colony algorithm combined with a genetic algorithm to solve the problem. This method updates the pheromone concentration based on the clustering index, obtains the intermediate population by processing with the objective function, and achieves iterative updates of the offspring through differential mutation. This method can overcome the shortcomings of the ant colony algorithm in the convergence process, which is not fast enough, and the genetic algorithm in the convergence process, which is prone to getting trapped in local optima.

[0157] This invention solves the technical problems of poor production evaluation results in the prior art, particularly in the reconstruction of equipment layout in workshops, especially in mixed-flow processing production lines.

[0158] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An integrated multi-AGV path planning and multi-row equipment layout method, characterized in that, The method comprises: S1, collecting workshop operation data, solving not less than two group indicators, linearly weighting processing the group indicators to obtain a model objective function, and establishing a multi-row equipment layout multi-objective optimization model according to the model objective function and a preset model constraint; wherein the group indicator represents that the equipment processing the same process needs to be concentrated in space to meet the needs of production personnel for observation, maintenance and centralized management; S2, for a preset layout, performing multi-AGV path planning based on a time window A* algorithm, and combining the workshop operation data to solve total processing time in iteration; S3, using an ant colony algorithm combined with a genetic algorithm to solve a final layout according to the total processing time in iteration; S4, performing dynamic simulation based on a preset discrete event to verify the accuracy of the multi-row equipment layout multi-objective optimization model according to the final layout; The step S1 further comprises the following specific steps: Step S11, determining the model objective function by using the following logic: min Z = μ1T + μ2C + μ3K (1) Wherein T is the time of completing an order, C is the transportation cost, K is the group indicator between the equipment, and μ1, μ2 and μ3 represent the weights of the order completion time, the transportation cost and the group indicator between the equipment respectively; Step S12, defining the group indicator between the equipment by using the following logic: where d ij is the distance between devices i, j, S t is the set of devices processing the same process, and m is the number of processes. S13, determining the weights among the three objectives, and determining the weight of the group indicator according to an entropy weight method; S14, determining the preset model constraint according to the space condition constraint between the equipment.

2. The integrated multi-AGV path planning and multi-line equipment layout method of claim 1, wherein, The step S13 comprises: The step S13 further comprises: S131, using the following logic to obtain preprocessed data by using a minimum maximum standardization method: wherein X ij is the data before standardization of the jth indicator of sample i, X i is the data in sample i, Y ij is the data after standardization of the jth indicator of sample i; S132, calculating the preprocessed data to obtain the proportion of the group indicator: S133, according to the proportion of the group indicator, solving the information entropy of the indicator by using the following logic: S134, processing the information entropy by using the following logic to determine the weight of the group indicator: where E j Information entropy of index j, k is the number of indexes, w j is the weight of the calculated clustering index.

3. The integrated multi-AGV path planning and multi-line equipment layout method of claim 1, wherein, The step S2 further comprises: S21, constructing a grid map, so that each equipment is respectively located at a point in the grid; S22, minimizing the processing time of the workshop according to a preset AGV scheduling rule; S23, for the dispatched transportation task AGV, planning the path of the dispatched transportation task AGV in the grid map by using the time window based A* algorithm; S24, for node conflict and opposite conflict, adjusting the path of the AGV by constructing the AGV occupation time window condition of different nodes; S25, repeatedly executing the steps S23 to S24 until all products are processed to calculate the total processing time in iteration.

4. The integrated multi-AGV path planning and multi-line equipment layout method of claim 3, wherein, In the step S23, the time window based A* algorithm further comprises: S231, obtaining the current AGV path and the last task path of the remaining AGV paths; S232, setting the non-loading and unloading points on the last task path as obstacles; S233, combine the starting point and end point of the last task and the obstacle, call A* algorithm to obtain the applicable path; S234, obtain the time window of the current task passing node according to the applicable path; S235, compare the time window of the current task with the total time window to determine whether there is a node conflict; S236, if yes, wait for time t and translate the time window; S237, obtain the new AGV path; S238, if no, add the time window of the current AGV path to the total time window.

5. The integrated multi-AGV path planning and multi-line equipment layout method of claim 3, wherein, In the evaluation function of the A* algorithm based on the time window in step S23, a turning penalty term P is introduced: where (x f , y f ) are the coordinates of the parent node of point S, C is a constant, and (x s , y s ) are the coordinates of the current node S.

6. The integrated multi-AGV path planning and multi-line equipment layout method of claim 1, wherein, Step S3 includes: S31, initialize the population and update the pheromone concentration, execute step S2 to obtain the total processing time and processing and transportation cost in the iteration, according to the spatial characteristics of the initial layout, obtain the clustering index between the devices to update the pheromone concentration; S32, calculate the state transition probability of the ant according to the pheromone concentration to obtain the intermediate population according to the model objective function; S33, use genetic algorithm to perform crossover and mutation according to the intermediate population to obtain the final offspring according to the pre-set fitness parameter; S34, update the pheromone concentration on the path through the offspring, and iteratively execute steps S32 to S34 to obtain the final layout.

7. The integrated multi-AGV path planning and multi-line equipment layout method of claim 1, wherein, In step S4, for the dispatched transportation task AGV, the A* algorithm based on the time window is used to plan the path of the dispatched transportation task AGV in the grid map.

8. The integrated multi-AGV path planning and multi-line equipment layout method of claim 1, wherein, In step S4, run the simulation model to obtain the processing and transportation equipment situation of the production line, compare it with the final layout to obtain the accuracy verification result of the multi-row device layout multi-objective optimization model.

9. The system of integrated multi-AGV path planning and multi-row equipment layout, applied to the method of integrated multi-AGV path planning and multi-row equipment layout as claimed in claim 1, characterized in that, The system includes: A model construction module is used to collect workshop operation data, solve at least two clustering indexes, linearly weight the clustering indexes to obtain a model objective function, and establish a multi-row device layout multi-objective optimization model according to the model objective function and pre-set model constraints; wherein the clustering index represents that the devices processing the same process need to be concentrated in space to meet the needs of production personnel for observation, maintenance and centralized management; A total time processing module is used to plan the multi-AGV path based on the A* algorithm based on the time window for the pre-set layout, and solve the total processing time in the iteration based on the workshop operation data; A layout solving module is used to solve the final layout according to the total processing time in the iteration by using the ant colony algorithm combined with the genetic algorithm, and the layout solving module is connected with the total time processing module; A dynamic simulation verification module is used to perform dynamic simulation based on pre-set discrete events to verify the accuracy of the multi-row device layout multi-objective optimization model according to the final layout, and the dynamic simulation verification module is connected with the layout solving module and the model construction module.

Images