Material distribution regulation and control method of robot laser welding work center and related device

Abstract

The invention relates to the technical field of robots, and discloses a robot laser welding work center material distribution regulation and control method and related devices.The method comprises the steps that a welding production plan and a real-time material inventory condition of a robot laser welding work center are obtained; performing feature extraction on the welding production plan and the material inventory condition to obtain target features of the welding production plan and the material inventory condition; the target features of the two are input into a biologically heuristic material delivery intelligent regulation and control model based on fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm, and a material delivery regulation and control strategy of the robot laser welding work center is obtained; and carrying out corresponding material distribution regulation and control on the robot laser welding work center according to the material distribution regulation and control strategy. According to the invention, timely supply of the materials can be guaranteed, and meanwhile, reduction of material distribution cost and reduction of stock overstock can be effectively realized.

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and in particular to a material distribution control method and related devices for a robot laser welding work center. Background Art

[0002] In the material distribution scenario of the robot laser welding work center, the common practice is to determine the approximate time of material distribution based on empirical rules, such as arranging distribution according to a fixed production rhythm or a preset time interval, without fully considering the actual working condition changes in the welding production process and the real-time consumption rate of materials. In terms of distribution path planning, most of them are based on the static layout diagram of the workshop, and some conventional and relatively short paths are selected. Once these paths are determined, they are rarely adjusted according to dynamic factors such as the operating status of the equipment in the workshop and traffic flow. As a result, problems such as path congestion and equipment failure may occur during the actual distribution process, affecting the distribution efficiency. For the distribution quantity of materials, the distribution quantity for each time is usually determined based on the average value of historical production data or fixed production process standards. There is a lack of accurate dynamic adjustment mechanism, and it is difficult to adapt to the fluctuations in material demand caused by factors such as product model changes and process improvements during the production process.

[0003] In summary, the existing material distribution technology solutions for robot laser welding work centers have obvious limitations. Due to the failure to effectively integrate welding production plans and real-time material inventory, and the lack of accurate and dynamic optimization capabilities for distribution time, path and quantity, it is difficult to ensure the timeliness of material distribution. It is impossible to effectively reduce material distribution costs and inventory backlogs while ensuring timely material supply. Summary of the invention The embodiment of the present invention provides a material distribution control method and related devices for a robot laser welding work center, which can effectively reduce material distribution costs and inventory backlogs while ensuring timely supply of materials.

[0004] An embodiment of the present invention provides a material distribution control method for a robot laser welding work center, comprising: Obtain welding production plans and real-time material inventory status of the robotic laser welding work center; Extracting features of both the welding production plan and the material inventory status to obtain target features of both; The target features of the two are input into a biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm to obtain the material distribution control strategy of the robot laser welding work center; the material distribution control strategy includes the material distribution time, distribution path and distribution quantity; According to the material distribution control strategy, corresponding material distribution control is performed on the robot laser welding work center.

[0005] Another embodiment of the present invention provides a material distribution control device for a robot laser welding work center, including: The acquisition module is used to obtain the welding production plan and real-time material inventory of the robot laser welding work center; A feature extraction module is used to extract features from both the welding production plan and the material inventory status to obtain target features of the two; A control strategy generation module is used to input the target features of the two into a biologically inspired material distribution intelligent control model based on the fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm to obtain a material distribution control strategy for the robot laser welding work center; the material distribution control strategy includes the material distribution time, distribution path and distribution quantity; The control module is used to perform corresponding material distribution control on the robot laser welding work center according to the material distribution control strategy.

[0006] Another embodiment of the present invention provides a material distribution and control device for a robot laser welding work center, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the material distribution and control method for the robot laser welding work center described in the above-mentioned embodiment of the invention is implemented.

[0007] Another embodiment of the present invention provides a storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the material distribution and control method of the robot laser welding work center described in the above-mentioned embodiment of the invention.

[0008] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: First, the welding production plan and real-time material inventory of the robot laser welding work center are obtained, and then the target features of the two are extracted by using a specific feature extraction method. These target features are then input into the biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm. In this model, the particle swarm optimization algorithm updates the speed and position of particles to explore the optimal combination of material distribution time, path and quantity according to its own and global optimal position; the ant colony optimization algorithm uses ants to select paths according to pheromone concentration and heuristic information, as well as the pheromone update mechanism, to continuously optimize the distribution path. The two algorithms work together, and after multiple iterations, they comprehensively evaluate the scheduling schemes represented by each particle and ant, and finally derive a material distribution control strategy that includes accurate distribution time, efficient distribution path and reasonable distribution quantity, and implement material distribution control for the robot laser welding work center accordingly. It can be seen that the embodiment of the present invention realizes the precise optimization of material delivery time, path and quantity through real-time acquisition and feature extraction of welding production plan and material inventory status, and utilizes the intelligent control model that integrates the advantages of two advanced algorithms, thereby effectively solving the problems of the existing technology and ensuring the timely supply of materials to the welding work center during the production process, while significantly reducing material distribution costs and inventory backlogs, and improving production efficiency and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flow chart of a material distribution control method for a robot laser welding work center provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a material distribution and control device for a robot laser welding work center provided by an embodiment of the present invention; Figure 3 It is a structural schematic diagram of a material distribution and control device for a robot laser welding work center provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] See also Figure 1 , is a flow chart of a material distribution control method for a robot laser welding work center provided by an embodiment of the present invention. The method comprises steps S10 to S13 S10, obtaining the welding production plan and real-time material inventory status of the robot laser welding work center; S11, extracting features of both the welding production plan and the material inventory status to obtain target features of both; S12, inputting the target features of the two into a biologically inspired material distribution intelligent control model based on the fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm, to obtain a material distribution control strategy for the robot laser welding work center; the material distribution control strategy includes a material distribution time, a distribution path, and a distribution quantity; S13, performing corresponding material distribution control on the robot laser welding work center according to the material distribution control strategy.

[0012] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: First, the welding production plan and real-time material inventory of the robot laser welding work center are obtained, and then the target features of the two are extracted by using a specific feature extraction method. These target features are then input into the biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm. In this model, the particle swarm optimization algorithm updates the speed and position of particles to explore the optimal combination of material distribution time, path and quantity according to its own and global optimal position; the ant colony optimization algorithm uses ants to select paths according to pheromone concentration and heuristic information, as well as the pheromone update mechanism, to continuously optimize the distribution path. The two algorithms work together, and after multiple iterations, they comprehensively evaluate the scheduling schemes represented by each particle and ant, and finally derive a material distribution control strategy that includes accurate distribution time, efficient distribution path and reasonable distribution quantity, and implement material distribution control for the robot laser welding work center accordingly. It can be seen that the embodiment of the present invention realizes the precise optimization of material delivery time, path and quantity through real-time acquisition and feature extraction of welding production plan and material inventory status, and utilizes the intelligent control model that integrates the advantages of two advanced algorithms, thereby effectively solving the problems of the existing technology and ensuring the timely supply of materials to the welding work center during the production process, while significantly reducing material distribution costs and inventory backlogs, and improving production efficiency and resource utilization.

[0013] As an example, in step S10, a two-way data interface is established with the enterprise production management system (such as ERP, MES, etc.), and through a standardized data transmission protocol (such as XML or JSON), detailed information of the welding production plan is obtained in real time, including product model, welding process requirements, production quantity of each product and estimated completion time, etc. For the material inventory, a high-precision sensor network is deployed in the material storage area, including laser ranging sensors, weight sensors and RFID readers. The laser ranging sensor calculates the approximate quantity of the material by measuring the height and volume of the material pile and combining the known material shape and density information; the weight sensor accurately measures the weight change of each shelf or storage box to obtain the real-time weight data of the material; the RFID reader scans the electronic tag on the material to identify the type, batch and unique identifier of the material, so as to accurately grasp the inventory quantity and location information of the material. At the same time, through data verification and redundant backup mechanisms, the acquired data is ensured to be accurate, and the data is stored in a local real-time database for use in subsequent steps.

[0014] As an improvement of the above embodiment, the feature extraction of both the welding production plan and the material inventory status to obtain the target features of the two includes: The text information including product description and process requirements in the welding production plan is segmented and POS tagged, and the Stanford NLP library, a natural language processing tool, is used to decompose the text into a sequence of words and POS tags; the Word2Vec word vector model is used to convert each word into a low-dimensional vector representation, so that words with similar semantics are close in distance in the vector space; the production plan features of the welding production plan are extracted by combining and calculating the vectors of words with similar semantics; the production plan features include: the earliest start time and the latest end time of the production plan, the set of robot laser welding workshop nodes, all material distribution paths of the robot laser welding workshop, the distribution starting position of the robot laser welding workshop, the unit distance transportation cost and the time when the materials arrive according to the production plan requirements; The principal component analysis algorithm (PCA algorithm) is used to reduce the dimension of the material inventory situation and extract the inventory characteristics of the material inventory situation; the inventory characteristics include: the total number of different materials involved in the production process of the robot laser welding work center, the minimum delivery quantity and inventory upper limit of the materials, the total demand for materials during the entire production plan period, and the inventory holding cost of the materials.

[0015] In this embodiment, the text information covering product description and process requirements of the welding production plan is processed. With the help of Stanford NLP library, word segmentation and part-of-speech tagging are performed to convert the text into a sequence of words and part-of-speech tags. Then, the Word2Vec word vector model is used to map the words into low-dimensional vectors. By combining similar semantic word vectors, production plan features such as the time range of the production plan, workshop nodes and paths, distribution starting positions, transportation costs, and material planned arrival times are accurately extracted. At the same time, the principal component analysis algorithm is used to reduce the dimension of the material inventory, and inventory features including the total number of materials, distribution volume and inventory upper and lower limits, total demand, and inventory holding costs are extracted. Through this series of processes, the welding production plan and material inventory are converted into target features with clear physical meaning and decision-making value, providing a key data foundation and decision-making basis for the subsequent intelligent control of material distribution.

[0016] As an improvement of the above embodiment, the target features of the two are input into a biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm to obtain the material distribution control strategy of the robot laser welding work center, including: Parameter setting: Parameter setting of the intelligent control model of material distribution based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm, including determining the particle swarm scale, number of ants, particle position and velocity vector dimension and value range, setting the maximum number of iterations, inertia weight, learning factor, pheromone volatility coefficient and enhancement coefficient, and obtaining the intelligent control model of material distribution with set parameters; Initialization elements: Based on the material distribution intelligent control model with set parameters, the particle position and velocity vector are initialized, the distribution time is set to an integer value within the production plan time, a non-repetitive workshop node path sequence that meets the material conditions is generated for the particle, and the ant is placed at the warehouse node to obtain the initialized particle and ant element; Update particle state: Update the particle speed according to the initialized particles and target parameters, and update it according to the particle's own historical optimal position and the global optimal position to obtain the particle with updated speed; Generate a scheduling plan: Use the particles with updated speed to update the particle positions, and generate a preliminary material distribution control strategy based on the updated positions; Ant operation: Based on the initialized ant elements, let the ants select the path according to the probability of pheromone and heuristic information. After completing the path selection, the pheromone matrix is ​​updated according to the pheromone volatility coefficient and enhancement coefficient to obtain the ant state after the pheromone matrix is ​​updated; Evaluate and update the optimal information: through the material distribution control strategy and the ant state after updating the pheromone matrix, and evaluate according to the objective function including delivery time, path and inventory cost, pheromone volatility coefficient and enhancement coefficient, update the historical optimal position and global optimal position in the particle swarm, update the optimal path and information of the ant colony, and obtain the updated optimal information; Iteration decision: Check whether the current number of iterations reaches the set maximum number of iterations. If so, output the final global optimal material distribution control strategy; if not, continue the iteration process with the current state.

[0017] More specifically, the above process is described as follows: (1) Initialization phase: After receiving the target features of the two, the biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm is initialized. The initialization process includes: Parameter setting: Determine the size of the particle swarm and the number of ants , defines the position vector of each particle and the velocity vector Dimensions , and preset the value range of each dimension; Initialization of particles and ants: Randomly generate the initial position and speed of particles: use a uniformly distributed random number generator to assign values ​​to the position vector of particles within the range of each dimension; for the delivery time dimension, randomly select integers between the earliest start time and the latest end time of the production plan; for the delivery path dimension, randomly generate a non-repeating node sequence from the node set of the robot laser welding workshop; for the delivery quantity dimension, randomly determine the value under the condition of satisfying the minimum delivery quantity of materials and the inventory upper limit; the initial value of the speed vector is also randomly generated within a pre-set reasonable range, such as the speed of the time dimension, the value range can be set to [-1, 1] time units, the speed of the path dimension is [-1, 1] node movement, and the speed of the quantity dimension is [-10, 10] material units, so that particles can explore extensively in the solution space in the initial stage; randomly determine the starting position of ant delivery: initially place all ants at the warehouse node and prepare to start path search based on pheromone and heuristic information; Key algorithm parameter settings: Set the maximum number of iterations , inertia weight , learning factor and , pheromone volatility coefficient and pheromone enhancement factor ; Among them, the inertia weight Initialized to 0.6, using a linear decreasing strategy ,in is the current iteration number; (2) Iteration phase, including: Particle velocity update: In each iteration In the process, particles are positioned according to their historical best position and the global optimal position , the particle velocity is updated through the velocity update formula: , Indicates At the iteration, the position vector of a particle has The velocity components in dimensions, where each particle corresponds to a material distribution control strategy, and the dimension corresponds to one of the decision variables of distribution time, path and quantity; The particle is At the iteration, the position vector The velocity component in the dimension, that is, the value after the velocity update of this iteration, will affect the change of the particle's position in the next iteration; is the inertia weight, which is used to balance the influence of the particle's current velocity on the next velocity update; and is the learning factor, Controls the degree to which particles approach their historical optimal positions. Control the degree to which particles approach the global optimal position; and They are all random numbers between 0 and 1, which are used to introduce randomness into the process of particles approaching their own historical optimal positions and global optimal positions respectively; It is The historical optimal position vector of the particle Middle The value of the dimension represents the particle in the past iteration process. The optimal value found in the dimensions of the decision variables; It is The particle in At the iteration, the position vector The current value of a dimension, that is, the value of the corresponding decision variable in the current material distribution control strategy; is the global optimal position vector Middle The value of the dimension, that is, the entire particle swarm is At the iteration, among the scheduling schemes represented by all particles, The optimal value found in the dimensions of decision variables; Particle position update: The particle obtains a new position according to the updated speed through the position update formula, and generates a new material distribution control strategy. The position update formula is: , is the particle's velocity after this velocity update. At the iteration, the position vector The new value in the dimension is obtained by adding the current position value to the updated speed value, which means that the particle moves to a new state in the dimension of the decision variable, thus generating a new material distribution control strategy attempt; Ant path selection and pheromone update: Path selection: The ant starts from the distribution starting point, the warehouse node, and selects the next moving location using a probabilistic formula based on pheromone concentration and heuristic information: ;in, Indicated in In the first iteration, the ants From Location Select Move to Location The ant decides the next node to go to in its feasible neighborhood based on the probability value, and then gradually builds a complete material distribution path; Represents the elements in the pheromone matrix, representing the In the first iteration, from the delivery location To delivery location pheromone concentration; is the weight coefficient of the pheromone, which determines the importance of pheromone concentration in the ant path selection process; is an element in the heuristic information matrix, according to the position To location The distance or cost is calculated as the inverse of ; is the weight coefficient of the heuristic information, and They work together to weigh the relative importance of pheromone concentration and heuristic information in ants’ path selection; It's an ant In Location The feasible neighborhood at the time of , which is a set determined by the actual topological structure and logistics rules of the robot laser welding workshop, contains the ant’s current location All possible options for the next position to move to when Pheromone update: When the ant completes a path traversal from the warehouse to the welding work center, it will update the ant according to the path length or delivery cost. Calculate the pheromone increment and update the pheromone matrix: , Indicated in At the iteration, from position To location pheromone concentration; is the pheromone volatility coefficient, which ranges from 0 to 1 and is used to simulate the natural volatility of pheromones as the number of iterations increases; It's an ant In the At the iteration, on the path The increment of pheromone released on the path is calculated differently depending on whether the ant passes through the path. In the The path of the iteration Contains the location To location When the ant passes this path, there will be a corresponding pheromone increment, and the increment size is related to the length or cost of the path the ant has walked. about, through Otherwise, the increment is 0, indicating that the path has not been occupied by ants. After, no pheromone is added; is the pheromone enhancement coefficient, which is a constant used to determine the intensity of the pheromone concentration on a path after the ants have walked through it; It's an ant In the The length or cost of the path traveled in the iteration; It's an ant In the The complete path taken in the first iteration is a sequence of workshop nodes, which represents the specific route of the ant starting from the warehouse node, the distribution starting point of the robot laser welding workshop, passing through multiple transfer nodes, and finally arriving at the welding work center node; (3) Evaluation and update phase, including: Scheme evaluation: The scheduling schemes represented by particles and ants are evaluated according to the preset objective function. The objective function is: ,in, and are 3 preset weight coefficients that satisfy ; Delivery time cost : , It is the total number of different materials involved in the production process of the entire robotic laser welding work center; It is the number of discrete time periods into which the time range involved in the entire production plan is divided according to certain time intervals; It is material In time slice The penalty cost per unit for late arrival reflects the delay caused by the material In a specific time slice The cost per unit of time lost in production due to failure to arrive at the welding work center on time; It is material The actual arrival time slice of the material The discrete time slice number where the welding work center is actually reached; It is material The time slice required to arrive according to the production plan is the material clearly specified in the pre-made welding production plan. The ideal time slice number that should arrive at the welding work center is used as a standard to measure the punctuality of the actual arrival time of the material; is the penalty index, which is a constant greater than 0 and is used to increase the penalty for delayed arrival of materials; the delivery path cost : , is the distribution path cost, which represents the sum of the path transportation related costs incurred during the distribution of all materials; It is a slave node To Node The unit distance transportation cost; is the indicator function, when and When it represents an adjacent node, it represents a material There is indeed a slave node on the delivery path To Node This section of road, ; Otherwise, if the two nodes are not adjacent in the actual path, ; is the path cost adjustment coefficient, which is a parameter greater than 0 and is used to adjust the impact of path length or cost in the objective function; inventory cost : , is the inventory cost; It is material The unit inventory holding cost covers all expenses incurred during the storage period of the materials; It is material In time slice The inventory level is obtained through real-time monitoring of material inventory or record data from the inventory management system, indicating the inventory level in a specific time slice. When the material The quantity remaining in the warehouse is used to calculate the material in this time slice The inventory holding cost of the material in the entire production cycle can be obtained by accumulating the inventory holding costs of all time slices. It is material The unit out-of-stock cost reflects the Loss costs in production caused by shortages due to insufficient inventory; It is material Total demand during the entire production planning period, extracting relevant information from the welding production plan; It is material In time slice The actual delivery volume, that is, the materials actually delivered to the welding work center The quantity is obtained through logistics distribution records or actual material receiving data statistics; is the inventory cost adjustment coefficient, which is a parameter greater than 0 and is used to adjust the overall impact of inventory cost in the objective function; Optimal solution update: For particle swarms, compare the objective function value of each particle's current position with the objective function value of the historical optimal position. If the current value is smaller, update the historical optimal position. Then find the global optimal position among the historical optimal positions of all particles, that is, find the position with the smallest objective function value among the historical optimal positions of all particles as the global optimal position, and record its corresponding material distribution control strategy. For ant colonies, record the optimal path and its objective function value found by the ants in each iteration, and compare them with the current global optimal path. If it is better, update the global optimal path and record the material distribution information corresponding to the path. Iteration termination judgment: judge whether the preset number of iterations has been reached. If so, stop the iteration and output the globally optimal material distribution scheduling strategy, that is, the distribution time, path and quantity represented by the position vector of the globally optimal particle; otherwise, return to the iterative process and continue to the next iteration, repeating the particle update, ant path selection and pheromone update, as well as the evaluation and update operations until the preset number of iterations is reached.

[0018] In this embodiment, after obtaining the target characteristics of the welding production plan and the material inventory situation, the operation process of the intelligent control model based on the fusion of the particle swarm optimization algorithm (PSO) and the ant colony optimization algorithm (ACO) is started. In the initialization stage, the various parameters of the particle swarm and the ant swarm are first determined, including the scale, vector dimension and value range, and the particles and ants are initialized to set the key parameters of the algorithm. In the iteration stage, the particles adjust the speed through a specific speed update formula based on their own and global optimal position information, and then use the position update formula to generate a new material distribution control strategy; the ants start from the warehouse node, select the path according to the probability formula of the pheromone concentration and heuristic information, and update the pheromone matrix according to the path situation after the path traversal. In the evaluation and update stage, the scheduling schemes represented by particles and ants are evaluated according to the objective function including delivery time, path, and inventory cost. The updated individual and global optimal positions are compared in the particle swarm, and the comparative optimal path is recorded in the ant colony. Through continuous iteration until the preset number of iterations is reached, the global optimal material delivery scheduling strategy is finally output, covering accurate delivery time, efficient delivery path, and reasonable delivery quantity, so as to realize the intelligent regulation of material delivery in the robot laser welding work center. Specifically, in the particle swarm optimization part, through the unique speed and position update mechanism, particles can efficiently explore the optimal combination of time, path, and quantity of material delivery under the guidance of global and local optimal solutions, effectively balance the global search and local development capabilities, avoid falling into the local optimal solution, and enhance the robustness of the algorithm. In the ant colony optimization part, the dynamic interaction of pheromone concentration and heuristic information is used, combined with the actual workshop topology and logistics rules, so that ants can continuously accumulate experience in path search and quickly find a better delivery path. As the iteration proceeds, high-quality paths are strengthened and low-quality paths are eliminated, which improves the efficiency and adaptability of path planning. The technical effect of this fusion algorithm is significant. Compared with traditional single algorithms or simple empirical scheduling methods, it can more accurately arrange material distribution according to production plans and inventory conditions, greatly reducing material distribution costs and reducing inventory backlog risks. At the same time, it ensures the timely supply of materials to the welding work center and improves the operational efficiency and resource utilization of the entire production system.

[0019] For ease of understanding, the following example is given here: Suppose a robot laser welding work center is producing a complex metal frame product that requires multiple types of steel and welding accessories as materials. The production plan requires a certain number of product welding tasks to be completed within the next week (divided into multiple time slices), and different processes require different times and quantities of materials. Material inventory is distributed in different warehouse locations in the workshop, and the inventory level of each material changes in real time.

[0020] In the initialization phase, the intelligent control model sets the value range of the particle delivery time dimension according to the earliest start time and the latest end time in the production plan; determines the search space of the particle delivery path according to the workshop node set; and limits the generation range of the particle delivery quantity according to the minimum material delivery quantity and the inventory upper limit. At the same time, ants are placed in the warehouse nodes to prepare for path search, and the key parameters of the algorithm are set, such as inertia weight, learning factor, pheromone volatilization and enhancement coefficient.

[0021] During the iteration process, particles continuously adjust the delivery plans they represent. For example, the delivery time of a steel material initially set by a particle may be too early, resulting in increased inventory costs. Through the speed update formula, referring to its own historical optimal position and the global optimal position, the speed of its delivery time dimension changes. After the position is updated, the delivery time is adjusted to a more appropriate time slice, and the delivery path and quantity are also optimized accordingly. Ants start from the warehouse and choose a path based on pheromone and heuristic information. For example, there are two paths leading to the welding work center. Although one is slightly longer, the current traffic conditions are good (heuristic information is better), and the other is shorter but passes through an area where equipment is being repaired (the pheromone concentration is lower). Ants will be more inclined to choose the former according to the probability formula, and after walking this path, they will update the pheromone matrix according to the path cost, so that subsequent ants are more likely to choose this high-quality path.

[0022] After multiple iterations and program evaluations, the intelligent control model finally outputs the globally optimal material distribution scheduling strategy. For example, it was determined that a batch of key steel materials would be delivered on Monday morning, and the distribution route avoided the area in the workshop where large equipment was being moved. The distribution quantity just met the production demand of the morning of that day and would not cause excessive inventory backlogs; for welding auxiliary materials, according to their consumption rate and inventory situation, appropriate distribution was arranged on Wednesday afternoon, and quickly delivered to the welding work center through the optimal distribution route. Through such intelligent scheduling, the timely supply of materials in the welding work center is ensured throughout the production process, and production is not stagnant due to material shortages. At the same time, the material distribution cost is effectively reduced, unnecessary inventory backlogs are avoided, and production efficiency and economic benefits are improved.

[0023] As an improvement of the above embodiment, the material distribution control of the robot laser welding work center according to the material distribution control strategy includes: According to the obtained material distribution scheduling strategy, the corresponding distribution task instructions are generated and sent to the corresponding distribution equipment of the robot laser welding work center through the wireless communication network, so that the distribution equipment can distribute materials according to the predetermined distribution path after receiving the corresponding distribution task instructions.

[0024] In this embodiment, based on the material distribution control strategy generated by integrating the particle swarm optimization algorithm and the ant colony optimization algorithm, it is further converted into executable distribution task instructions. With the help of the wireless communication network as a medium, these instructions are accurately sent to the corresponding distribution equipment of the robot laser welding work center, so that the distribution equipment can carry out material distribution work according to the key information such as the distribution path predetermined in the instructions. Its technical effect is very significant, realizing the seamless connection from the strategy generated by the intelligent control model to the actual material distribution operation, breaking the barrier between theoretical optimization and actual execution, ensuring the efficiency, accuracy and timeliness of material distribution, ensuring that the welding work center can smoothly carry out production as planned, avoiding the impact of problems such as untimely material supply or unreasonable distribution path on production progress, and at the same time, with the help of precise control strategies, effectively reducing the material distribution cost and inventory backlog, and improving the operating efficiency of the entire robot laser welding work center.

[0025] For example, in a large robot laser welding workshop, a batch of welding components for automobile parts manufacturing is being produced. Its production plan covers material requirements of various specifications of steel, welding wire, etc., and the quantity and urgency of the demand for each material in different time periods are different. Through the calculation of the previous intelligent control model, a material distribution control strategy is obtained, which clearly states that at around 10 o'clock in the morning, the steel of model A needs to be delivered from the warehouse located in the southwest corner of the workshop along the path passing through transfer points 1 and transfer points 2 to the welding work center in the northeast corner, with a delivery quantity of 50 bundles; and the welding wire is scheduled to depart from another warehouse at 2 o'clock in the afternoon and be sent to the welding work center through different transfer nodes, with a quantity of 10 boxes. Subsequently, based on this strategy, the system automatically generates corresponding distribution task instructions, which record in detail key information such as the name of the distribution material, the distribution time, the specific distribution path, and the distribution quantity. Then, through the wireless communication network covered in the workshop, such as Wi-Fi or industrial-specific wireless communication protocols, these instructions are sent to the distribution equipment such as the automatic guided vehicle (AGV) or forklift responsible for transportation. After receiving the command, the AGV started on time at 10 a.m., loaded 50 bundles of type A steel from the southwest corner warehouse according to the predetermined route, passed through transfer points 1 and 2 in turn, and successfully delivered the steel to the welding work center in the northeast corner, meeting the demand for the material at that time. Similarly, the forklift responsible for transporting welding wire at 2 p.m. also completed the distribution task accurately according to the command. Throughout the process, the material distribution was carried out in an orderly manner according to the plan, which not only ensured the timely supply of materials to the welding work center and enabled continuous and uninterrupted production, but also reduced unnecessary transportation costs and inventory backlogs due to the reasonable distribution route and time arrangement, effectively improving the overall production efficiency and economic benefits of the workshop.

[0026] See also Figure 2, is a schematic diagram of a material distribution control device for a robot laser welding work center provided by an embodiment of the present invention. The material distribution control device for a robot laser welding work center includes: An acquisition module 10 is used to acquire the welding production plan and real-time material inventory status of the robot laser welding work center; A feature extraction module 11 is used to extract features of both the welding production plan and the material inventory status to obtain target features of the two; A control strategy generation module 12 is used to input the target features of the two into a biologically inspired material distribution intelligent control model based on the fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm to obtain a material distribution control strategy for the robot laser welding work center; the material distribution control strategy includes the material distribution time, distribution path and distribution quantity; The control module 13 is used to perform corresponding material distribution control on the robot laser welding work center according to the material distribution control strategy.

[0027] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: First, the welding production plan and real-time material inventory of the robot laser welding work center are obtained, and then the target features of the two are extracted by using a specific feature extraction method. These target features are then input into the biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm. In this model, the particle swarm optimization algorithm updates the speed and position of particles to explore the optimal combination of material distribution time, path and quantity according to its own and global optimal position; the ant colony optimization algorithm uses ants to select paths according to pheromone concentration and heuristic information, as well as the pheromone update mechanism, to continuously optimize the distribution path. The two algorithms work together, and after multiple iterations, they comprehensively evaluate the scheduling schemes represented by each particle and ant, and finally derive a material distribution control strategy that includes accurate distribution time, efficient distribution path and reasonable distribution quantity, and implement material distribution control for the robot laser welding work center accordingly. It can be seen that the embodiment of the present invention realizes the precise optimization of material delivery time, path and quantity through real-time acquisition and feature extraction of welding production plan and material inventory status, and utilizes the intelligent control model that integrates the advantages of two advanced algorithms, thereby effectively solving the problems of the existing technology and ensuring the timely supply of materials to the welding work center during the production process, while significantly reducing material distribution costs and inventory backlogs, and improving production efficiency and resource utilization.

[0028] As an improvement of the above solution, the feature extraction module is specifically used for: The text information including product description and process requirements in the welding production plan is segmented and POS tagged, and the Stanford NLP library, a natural language processing tool, is used to decompose the text into a sequence of words and POS tags; the Word2Vec word vector model is used to convert each word into a low-dimensional vector representation, so that words with similar semantics are close in distance in the vector space; the production plan features of the welding production plan are extracted by combining and calculating the vectors of words with similar semantics; the production plan features include: the earliest start time and the latest end time of the production plan, the set of robot laser welding workshop nodes, all material distribution paths of the robot laser welding workshop, the distribution starting position of the robot laser welding workshop, the unit distance transportation cost and the time when the materials arrive according to the production plan requirements; The principal component analysis algorithm is used to perform dimensionality reduction processing on the material inventory situation and extract the inventory characteristics of the material inventory situation; the inventory characteristics include: the total number of different materials involved in the production process of the robot laser welding work center, the minimum delivery quantity and inventory upper limit of materials, the total demand for materials during the entire production plan period, and the inventory holding cost of materials.

[0029] As an improvement of the above solution, the control strategy generation module is specifically used for: Parameter setting: Parameter setting of the intelligent control model of material distribution based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm, including determining the particle swarm scale, number of ants, particle position and velocity vector dimension and value range, setting the maximum number of iterations, inertia weight, learning factor, pheromone volatility coefficient and enhancement coefficient, and obtaining the intelligent control model of material distribution with set parameters; Initialization elements: Based on the material distribution intelligent control model with set parameters, the particle position and velocity vector are initialized, the distribution time is set to an integer value within the production plan time, a non-repetitive workshop node path sequence that meets the material conditions is generated for the particle, and the ant is placed at the warehouse node to obtain the initialized particle and ant element; Update particle state: Update the particle speed according to the initialized particles and target parameters, and update it according to the particle's own historical optimal position and the global optimal position to obtain the particle with updated speed; Generate a scheduling plan: Use the particles with updated speed to update the particle positions, and generate a preliminary material distribution control strategy based on the updated positions; Ant operation: Based on the initialized ant elements, let the ants select the path according to the probability of pheromone and heuristic information. After completing the path selection, the pheromone matrix is ​​updated according to the pheromone volatility coefficient and enhancement coefficient to obtain the ant state after the pheromone matrix is ​​updated; Evaluate and update the optimal information: through the material distribution control strategy and the ant state after updating the pheromone matrix, and evaluate according to the objective function including delivery time, path and inventory cost, pheromone volatility coefficient and enhancement coefficient, update the historical optimal position and global optimal position in the particle swarm, update the optimal path and information of the ant colony, and obtain the updated optimal information; Iteration decision: Check whether the current number of iterations reaches the set maximum number of iterations. If so, output the final global optimal material distribution control strategy; if not, continue the iteration process with the current state.

[0030] As an improvement of the above solution, the control module is specifically used for: According to the obtained material distribution scheduling strategy, the corresponding distribution task instructions are generated and sent to the corresponding distribution equipment of the robot laser welding work center through the wireless communication network, so that the distribution equipment can distribute materials according to the predetermined distribution path after receiving the corresponding distribution task instructions.

[0031] See also Figure 3 , is a schematic diagram of a material distribution control device for a robot laser welding work center provided by an embodiment of the present invention. The material distribution control device for a robot laser welding work center of this embodiment includes: a processor 100, a memory 101, and a computer program stored in the memory 101 and executable on the processor 100, such as a device program. When the processor 100 executes the computer program, the steps in the above-mentioned device method embodiments are implemented. Alternatively, when the processor 100 executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0032] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the material distribution control device of the robot laser welding work center.

[0033] The material distribution and control equipment of the robot laser welding work center can be a main control device such as a desktop computer, a notebook and a cloud server. The material distribution and control equipment of the robot laser welding work center may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the schematic diagram is only an example of the material distribution and control equipment of the robot laser welding work center, and does not constitute a limitation on the material distribution and control equipment of the robot laser welding work center. It may include more or less components than shown in the figure, or combine certain components, or different components. For example, the material distribution and control equipment of the robot laser welding work center may also include input and output devices, network access devices, buses, etc.

[0034] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the material distribution control equipment of the robot laser welding work center, and uses various interfaces and lines to connect various parts of the material distribution control equipment of the entire robot laser welding work center.

[0035] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the material distribution control device of the robot laser welding work center by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0036] Wherein, if the module / unit integrated in the material distribution control equipment of the robot laser welding work center is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0037] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0038] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A material distribution control method for a robot laser welding work center, characterized in that: include: Obtain welding production plans and real-time material inventory status of the robotic laser welding work center; Extracting features of both the welding production plan and the material inventory status to obtain target features of both; The target features of the two are input into a biologically inspired material distribution intelligent control model based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm to obtain the material distribution control strategy of the robot laser welding work center; the material distribution control strategy includes the material distribution time, distribution path and distribution quantity; According to the material distribution control strategy, corresponding material distribution control is performed on the robot laser welding work center.

2. The material distribution control method of the robot laser welding work center according to claim 1 is characterized in that: The feature extraction of both the welding production plan and the material inventory status to obtain target features of both includes: The text information including product description and process requirements in the welding production plan is segmented and POS tagged, and the Stanford NLP library, a natural language processing tool, is used to decompose the text into a sequence of words and POS tags; the Word2Vec word vector model is used to convert each word into a low-dimensional vector representation, so that words with similar semantics are close in distance in the vector space; the production plan features of the welding production plan are extracted by combining and calculating the vectors of words with similar semantics; the production plan features include: the earliest start time and the latest end time of the production plan, the set of robot laser welding workshop nodes, all material distribution paths of the robot laser welding workshop, the distribution starting position of the robot laser welding workshop, the unit distance transportation cost and the time when the materials arrive according to the production plan requirements; The principal component analysis algorithm is used to perform dimensionality reduction processing on the material inventory situation and extract the inventory characteristics of the material inventory situation; the inventory characteristics include: the total number of different materials involved in the production process of the robot laser welding work center, the minimum delivery quantity and inventory upper limit of materials, the total demand for materials during the entire production plan period, and the inventory holding cost of materials.

3. The material distribution control method of the robot laser welding work center as claimed in claim 2 is characterized in that: The target features of the two are input into a biologically inspired material distribution intelligent control model based on the fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm to obtain a material distribution control strategy for the robot laser welding work center, including: Parameter setting: Parameter setting of the intelligent control model of material distribution based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm, including determining the particle swarm scale, number of ants, particle position and velocity vector dimension and value range, setting the maximum number of iterations, inertia weight, learning factor, pheromone volatility coefficient and enhancement coefficient, and obtaining the intelligent control model of material distribution with set parameters; Initialization elements: Based on the material distribution intelligent control model with set parameters, the particle position and velocity vector are initialized, the distribution time is set to an integer value within the production plan time, a non-repetitive workshop node path sequence that meets the material conditions is generated for the particle, and the ant is placed at the warehouse node to obtain the initialized particle and ant element; Update particle state: Update the particle speed according to the initialized particles and target parameters, and update it according to the particle's own historical optimal position and the global optimal position to obtain the particle with updated speed; Generate a scheduling plan: Use the particles with updated speed to update the particle positions, and generate a preliminary material distribution control strategy based on the updated positions; Ant operation: Based on the initialized ant elements, let the ants select the path according to the probability of pheromone and heuristic information. After completing the path selection, the pheromone matrix is ​​updated according to the pheromone volatility coefficient and enhancement coefficient to obtain the ant state after the pheromone matrix is ​​updated; Evaluate and update the optimal information: through the material distribution control strategy and the ant state after updating the pheromone matrix, and evaluate according to the objective function including delivery time, path and inventory cost, pheromone volatility coefficient and enhancement coefficient, update the historical optimal position and global optimal position in the particle swarm, update the optimal path and information of the ant colony, and obtain the updated optimal information; Iteration decision: Check whether the current number of iterations reaches the set maximum number of iterations. If so, output the final global optimal material distribution control strategy; if not, continue the iteration process with the current state.

4. The material distribution control method of the robot laser welding work center according to claim 1, characterized in that: The material distribution control strategy is used to control the robot laser welding work center accordingly, including: According to the obtained material distribution scheduling strategy, the corresponding distribution task instructions are generated and sent to the corresponding distribution equipment of the robot laser welding work center through the wireless communication network, so that the distribution equipment can distribute materials according to the predetermined distribution path after receiving the corresponding distribution task instructions.

5. A material distribution and control device for a robot laser welding work center, characterized in that: include: The acquisition module is used to obtain the welding production plan and real-time material inventory of the robot laser welding work center; A feature extraction module is used to extract features from both the welding production plan and the material inventory status to obtain target features of the two; A control strategy generation module is used to input the target features of the two into a biologically inspired material distribution intelligent control model based on the fusion of a particle swarm optimization algorithm and an ant colony optimization algorithm to obtain a material distribution control strategy for the robot laser welding work center; the material distribution control strategy includes the material distribution time, distribution path and distribution quantity; The control module is used to perform corresponding material distribution control on the robot laser welding work center according to the material distribution control strategy.

6. The material distribution control device for the robot laser welding work center as claimed in claim 5, characterized in that: The feature extraction module is specifically used for: The text information in the welding production plan, including product descriptions and process requirements, is segmented and POS tagged. The Stanford NLP library, a natural language processing tool, is used to decompose the text into a sequence of words and POS tags. The Word2Vec word vector model is used to convert each word into a low-dimensional vector representation, so that words with similar semantics are close in distance in the vector space. Extracting production plan features of the welding production plan by combining and operating vectors of words with similar semantics; The production plan characteristics include: the earliest start time and the latest end time of the production plan, the set of robot laser welding workshop nodes, all material distribution paths of the robot laser welding workshop, the distribution starting position of the robot laser welding workshop, the unit distance transportation cost, and the time when the materials arrive according to the production plan requirements; The principal component analysis algorithm is used to perform dimensionality reduction processing on the material inventory situation and extract the inventory characteristics of the material inventory situation; the inventory characteristics include: the total number of different materials involved in the production process of the robot laser welding work center, the minimum delivery quantity and inventory upper limit of materials, the total demand for materials during the entire production plan period, and the inventory holding cost of materials.

7. The material distribution control device for the robot laser welding work center as claimed in claim 5, characterized in that: The control strategy generation module is specifically used for: Parameter setting: Parameter setting of the intelligent control model of material distribution based on the fusion of particle swarm optimization algorithm and ant colony optimization algorithm, including determining the particle swarm scale, number of ants, particle position and velocity vector dimension and value range, setting the maximum number of iterations, inertia weight, learning factor, pheromone volatility coefficient and enhancement coefficient, and obtaining the intelligent control model of material distribution with set parameters; Initialization elements: Based on the material distribution intelligent control model with set parameters, the particle position and velocity vector are initialized, the distribution time is set to an integer value within the production plan time, a non-repetitive workshop node path sequence that meets the material conditions is generated for the particle, and the ant is placed at the warehouse node to obtain the initialized particle and ant element; Update particle state: Update the particle speed according to the initialized particles and target parameters, and update it according to the particle's own historical optimal position and the global optimal position to obtain the particle with updated speed; Generate a scheduling plan: Use the particles with updated speed to update the particle positions, and generate a preliminary material distribution control strategy based on the updated positions; Ant operation: Based on the initialized ant elements, let the ants select the path according to the probability of pheromone and heuristic information. After completing the path selection, the pheromone matrix is ​​updated according to the pheromone volatility coefficient and enhancement coefficient to obtain the ant state after the pheromone matrix is ​​updated; Evaluate and update the optimal information: through the material distribution control strategy and the ant state after updating the pheromone matrix, and evaluate according to the objective function including delivery time, path and inventory cost, pheromone volatility coefficient and enhancement coefficient, update the historical optimal position and global optimal position in the particle swarm, update the optimal path and information of the ant colony, and obtain the updated optimal information; Iteration decision: Check whether the current number of iterations reaches the set maximum number of iterations. If so, output the final global optimal material distribution control strategy; If not reached, continue the iteration process with the current state.

8. The material distribution control device for the robot laser welding work center as claimed in claim 5, characterized in that: The control module is specifically used for: According to the obtained material distribution scheduling strategy, the corresponding distribution task instructions are generated and sent to the corresponding distribution equipment of the robot laser welding work center through the wireless communication network, so that the distribution equipment can distribute materials according to the predetermined distribution path after receiving the corresponding distribution task instructions.

9. A material distribution and control device for a robot laser welding work center, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the material distribution and control method of the robot laser welding work center as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the material distribution and control method for the robot laser welding work center as described in any one of claims 1 to 4.

Images