Multi-rotor linear motor scheduling method based on improved genetic algorithm

Abstract

The invention relates to a multi-rotor linear motor scheduling method based on an improved genetic algorithm. The multi-rotor linear motor scheduling method comprises six steps of population initialization, fitness evaluation, selection operation, crossover operation, mutation operation, iteration and termination. According to the method, the scheduling problem of the multi-mover linear motor is solved by utilizing the global search capability and the complex problem processing capability of the improved genetic algorithm, the problems of multi-target performance, multi-constraint performance and multi-dynamic performance in the multi-mover scheduling process are solved, and the optimal scheduling performance and production efficiency are realized.

Description

Technical Field

[0001] The present invention relates to a multi-rotor scheduling method, and more particularly to a multi-rotor linear motor scheduling method based on an improved genetic algorithm, belonging to the field of motion control. Background Art

[0002] In the fields of modern industrial production, logistics transportation, automation control, etc., scheduling problems are ubiquitous and becoming increasingly complex. As an important driving device, linear motors have been widely used in many fields such as high-speed maglev trains, precision machining, and logistics conveyor lines. In a multi-rotor linear motor system, since multiple rotors (i.e., moving parts) can move independently or cooperatively on a linear track at the same time, the flexibility and efficiency of the system are further improved. However, how to efficiently schedule these rotors to achieve optimal performance and efficiency has become an urgent problem to be solved. Facing the multi-rotor scheduling problem with multi-objectives, multi-constraints, and multi-dynamics, traditional linear programming and dynamic programming methods have limitations such as high computational complexity, difficulty in dealing with non-linear constraints, and lack of flexibility.

[0003] In view of the limitations of traditional scheduling methods and the advantages of genetic algorithms, a multi-rotor linear motor scheduling method based on genetic algorithms is proposed according to the characteristics of the multi-rotor linear motor scheduling problem. This method uses the global search ability of genetic algorithms and the ability to handle complex problems to solve the multi-rotor linear motor scheduling problem in order to achieve optimal scheduling performance and production efficiency. Summary of the Invention

[0004] Aiming at the above deficiencies existing in the prior art, the technical problem to be solved by the present invention is to overcome the problems of multi-objectives, multi-constraints, and multi-dynamics in the multi-rotor scheduling process, and provide a multi-rotor linear motor scheduling method based on an improved genetic algorithm. By establishing a motion planning mathematical model for multi-rotor linear motor scheduling and combining specific constraints (such as the number of rotors, buffer size, task dependency, etc.) in the multi-rotor linear motor scheduling process, the scheduling result is optimized, and finally the optimal solution is output through specific termination conditions.

[0005] The technical solution adopted by the invention to achieve the above object is: a multi-rotor linear motor scheduling method based on an improved genetic algorithm, including: 6 steps of initializing the population, fitness evaluation, selection operation, crossover operation, mutation operation, iteration and termination.

[0006] For the initialization of the population, first, a coding method based on the task sequence is adopted. A task processing sequence is represented as a chromosome, where each gene in the chromosome represents a task number, and each task number appears only once in the chromosome. Second, a certain number (set as N, i.e., the population size) of chromosomes are randomly generated to form the initial population. Each chromosome is a random permutation of task numbers.

[0007] For the fitness evaluation, first, a fitness function is established. The fitness function is used to evaluate the quality of each individual (scheduling scheme). The fitness value is usually the reciprocal of the task completion time C max , that is, the reciprocal of the time when all tasks are completed on the machine. The fitness function is as follows

[0008]

[0009] where Cmax i is the total completion time of individual i. The larger the fitness value, the better the scheduling scheme. In the fitness function, in addition to evaluating the task completion time, the actual production constraints in the scheduling are also considered, such as the number of movers, buffer size, and task dependencies. If the scheduling scheme of a certain generation violates these constraint conditions, the fitness value is adjusted by adding a penalty term, so as to guide the algorithm to generate solutions that meet the constraints.

[0010] For the selection operation, binary tournament selection is adopted, that is, two individuals are randomly selected from the population, their fitness values are compared, and the individual with a better fitness value is selected as the parent individual. This process will continue until enough parent individuals are generated.

[0011]

[0012] where P i is the probability that the i-th individual is selected, and fitness i is the fitness value of this individual. Through binary tournament selection, individuals with higher fitness values will have a greater chance of becoming parents.

[0013] The crossover operation is used to generate offspring individuals and adopts partially mapped crossover (PMX). The purpose of the crossover operation is to generate new individuals by exchanging gene segments of two parent individuals and maintain the legality of the task sequence.

[0014] For the mutation operation, with a certain mutation probability P m , the mutation operation is performed on each offspring chromosome. By randomly exchanging the task sequence, the genes of the individual are changed, thereby increasing the diversity of the population and helping the algorithm to jump out of the local optimal solution.

[0015] Regarding the iteration and termination, the maximum number of iterations is preset. When the algorithm runs to the maximum number of iterations or the optimal fitness value does not change significantly for several consecutive generations, it can be considered that the algorithm has converged and the iteration stops.

[0016] The present invention has the following advantages and beneficial effects:

[0017] 1. The present invention uses an improved genetic algorithm to optimize the multi-mover scheduling task. By simulating the natural evolution process, the genetic algorithm can perform a global search in the solution space, effectively avoiding falling into local optimal solutions. The genetic algorithm has strong robustness and is insensitive to initial conditions, and can stably converge to the optimal solution or an approximate optimal solution under different initial conditions. The genetic algorithm searches the solution space through the evolution of the population, has implicit parallelism, and can significantly improve the search efficiency.

[0018] 2. The present invention uses the partially mapped crossover method for the crossover operation. When solving the multi-mover task scheduling problem, the partially mapped crossover can well ensure the uniqueness of each gene (representing a process) in the offspring chromosome through the mechanism of exchanging the matching segments and resolving conflicts, avoiding the situation of repeated scheduling or missed scheduling, and thus ensuring that the generated solution is a legal scheduling scheme.

[0019] 3. The present invention uses the partially mapped crossover method for the crossover operation. In the multi-mover task scheduling problem, there is a certain order relationship between processes, and these order relationships are crucial for generating an effective scheduling scheme. When the partially mapped crossover method exchanges the matching segments, it can retain the order relationship of the processes within the matching segment in the parent chromosome, and at the same time, it will also try to maintain the relative order of other parts during the conflict resolution process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of the system execution process of the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following further describes the present invention in detail with reference to the drawings and embodiments.

[0022] Embodiment 1

[0023] A multi-mover linear motor scheduling method based on an improved genetic algorithm includes six steps: initializing the population, fitness evaluation, selection operation, crossover operation, mutation operation, and iteration and termination.

[0024] Regarding the initialization of the population, first, a coding method based on the task order is adopted. A task processing order is represented as a chromosome, and each gene in the chromosome represents a task number, and each task number appears only once in the chromosome;

[0025] Secondly, a certain number (set as N, i.e., the population size) of chromosomes are randomly generated to form an initial population;

[0026] Each chromosome is a random permutation of task numbers.

[0027] For the fitness evaluation, first, a fitness function is established. The fitness function is used to evaluate the quality of each individual (scheduling scheme). The fitness value is usually the reciprocal of the task completion time C max That is, the reciprocal of the time when all tasks are completed on the machine. The fitness function is as follows

[0028]

[0029] where, Cmax i is the total completion time of individual i. The larger the fitness value, the better the scheduling scheme. In the fitness function, in addition to evaluating the task completion time, the actual production constraints in the scheduling are also considered, such as the number of movers, buffer size, and task dependencies. If the scheduling scheme of a certain generation violates these constraint conditions, the fitness value is adjusted by adding a penalty term, so as to guide the algorithm to generate solutions that meet the constraints.

[0030] For the selection operation, binary tournament selection is adopted, that is, two individuals are randomly selected from the population, their fitness is compared, and the individual with better fitness is selected as the parental individual. This process will continue until enough parental individuals are generated.

[0031]

[0032] where, P i is the probability that the i-th individual is selected, and fitness i is the fitness value of this individual. Through binary tournament selection, individuals with higher fitness will have a greater chance of becoming parents.

[0033] For the crossover operation, which is used to generate offspring individuals, partially matched crossover (PMX) is adopted. The purpose of the crossover operation is to generate new individuals by exchanging gene segments of two parental individuals and maintain the legality of the task order.

[0034] For the mutation operation, with a certain mutation probability P m , the mutation operation is performed on each offspring chromosome. By randomly exchanging the task order, the genes of the individual are changed, so as to increase the diversity of the population and help the algorithm jump out of the local optimal solution.

[0035] Regarding the iteration and termination, the maximum number of iterations is preset. When the algorithm runs to the maximum number of iterations or the optimal fitness value does not change significantly for several consecutive generations, it can be considered that the algorithm has converged and the iteration stops.

[0036] In the mutation operation mentioned above, the selection of the mutation probability Pm has an important impact on the performance of the algorithm. An overly high mutation probability may lead to the instability of the algorithm, while an overly low mutation probability may cause the algorithm to fall into a local optimal solution. Therefore, in practical applications, it is necessary to determine the appropriate mutation probability Pm according to the complexity and scale of the problem through experiments or experience.

[0037] In addition, in order to further improve the search efficiency and performance of the algorithm, some other strategies can also be introduced, such as the elitist retention strategy, adaptive mutation probability, and crossover probability, etc. The elitist retention strategy can retain the optimal individuals in each generation and prevent them from being destroyed in the subsequent evolution process; the adaptive mutation probability and crossover probability can be dynamically adjusted according to the evolutionary state of the population and the fitness values of individuals, thereby improving the convergence speed and search ability of the algorithm.

[0038] In summary, a multi-mover linear motor scheduling method based on an improved genetic algorithm proposed by the present invention can effectively solve the multi-mover linear motor scheduling problem and achieve optimal scheduling performance and production efficiency by adopting strategies such as encoding based on task order, establishing a fitness function, adopting binary tournament selection, partially matched crossover, and mutation operation. At the same time, by introducing other strategies, the search efficiency and performance of the algorithm can be further improved, providing an effective solution for the multi-mover scheduling problem in actual industrial production.

[0039] Embodiment 2

[0040] Figure 1 This is the execution flowchart of the method of the present invention. The present invention will be further described in detail below with reference to the accompanying drawings and embodiments:

[0041] A multi-mover linear motor scheduling method based on an improved genetic algorithm includes:

[0042] Step S1: Initialize and set the relevant parameters of the algorithm. The specific parameters include: the maximum number of iterations, the individual mutation probability P m , the crossover probability P c ;

[0043] Step S2: According to the size of the buffer capacity and the total number of movers k, initialize the population size N. The initial population size is randomly generated N non-repeating task execution permutations.

[0044] Step S3: Calculate the fitness value and determine whether the fitness converges. If the optimal fitness value of the population does not change significantly for several consecutive generations, it is considered that the algorithm has converged and the iteration is stopped.

[0045] Step S4: Determine whether the maximum number of iterations is reached. If so, stop the iteration;

[0046] Step S5: Binary tournament selection. Randomly select two chromosomes from the population, compare their fitness values, and select the chromosome with the higher fitness value as the parent. Repeat this process until a set of parent chromosomes with a quantity equal to the population size N is selected.

[0047] Step S6: Partially matched crossover operation (PMX). Randomly select two chromosomes from the set of parent chromosomes, determine a matching segment by randomly selecting two crossover points, exchange the matching segments of the two parent chromosomes to obtain two intermediate offspring. When gene duplicates appear in the intermediate offspring, according to the gene correspondence relationship within the matching segment, replace the duplicate genes in the non-matching segment to ensure that each task number appears only once in each offspring chromosome.

[0048] Step S7: Random swap mutation. With an individual mutation probability P m Perform mutation operations on each offspring chromosome. The specific method is: randomly select two gene positions in the chromosome and exchange their values.

[0049] Step S8: Population update. Replace the chromosomes in the original population with the new chromosomes generated after selection, crossover, and mutation operations to form a new generation of population.

[0050] Step S9: Output the result. When the maximum number of iterations is reached or the fitness value converges, stop the iteration. Select the chromosome with the highest fitness value from the final population, decode it into a specific task processing order, and this order is the approximate optimal solution to the multi-mover linear motor scheduling problem. At the same time, output the corresponding makespan Cmax.

[0051] In the process of implementing the present invention, it is also possible to combine specific industrial application scenarios to perform targeted optimization and improvement on the algorithm. For example, for certain specific task dependency relationships and buffer capacity limitations, more refined fitness functions and constraint conditions can be designed to improve the accuracy and feasibility of the scheduling scheme.

[0052] In addition, the present invention can also be combined with other intelligent optimization algorithms, such as particle swarm optimization algorithm, ant colony algorithm, etc., to form a hybrid scheduling strategy to further improve the search efficiency and performance of the algorithm. Through continuous exploration and practice, the present invention is expected to provide a more efficient and accurate solution to the multi-mover linear motor scheduling problem.

[0053] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0054] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A scheduling method for a multi-moving-sub linear motor based on an improved genetic algorithm, comprising: Six steps: initializing the population, fitness evaluation, selection operation, crossover operation, mutation operation, iteration, and termination.

2. The improved genetic optimization algorithm and linear motor control method according to claim 1, wherein For the described population initialization, first, a coding method based on the task sequence is adopted. A task processing sequence is represented as a chromosome, where each gene in the chromosome represents a task number, and each task number appears only once in the chromosome. Secondly, a certain number (set as N, i.e., the population size) of chromosomes are randomly generated to form the initial population. Each chromosome is a random permutation of task numbers.

3. The method for controlling a linear motor based on an improved genetic optimization algorithm according to claim 2, wherein, For the fitness evaluation described above, first, a fitness function is established. The fitness function is used to evaluate the quality of each individual (scheduling plan), and the fitness value is usually the reciprocal of the task completion time C max , that is, the reciprocal of the time taken for all tasks to be completed on the machine. The fitness function is as follows Among them, Cmax i is the total completion time of individual i. The larger the fitness value, the better the scheduling scheme. In the fitness function, in addition to evaluating the completion time of tasks, the actual production constraints in the scheduling are also considered, such as the number of movers, buffer size, and task dependencies. If the scheduling scheme of a certain generation violates these constraint conditions, the fitness value is adjusted by adding a penalty term, so as to guide the algorithm to generate solutions that meet the constraints.

4. The method for scheduling a multi-rotor linear motor based on an improved genetic algorithm according to claim 3, characterized in that, For the described selection operation, binary tournament selection is used, that is, two individuals are randomly selected from the population, their fitness values are compared, and the individual with better fitness is selected as the parental individual. This process continues until enough parental individuals are generated. where P i is the probability that the i-th individual is selected, and fitness i is the fitness value of this individual. Through binary tournament selection, individuals with higher fitness values will have a greater chance of becoming parents.

5. The improved genetic optimization algorithm-based linear motor control method according to claim 4, wherein The described crossover operation is used to generate offspring individuals and adopts partially matched crossover (PMX). The purpose of the crossover operation is to generate new individuals by exchanging gene segments of two parental individuals and maintain the legality of the task sequence.

6. The improved genetic optimization algorithm and linear motor control method according to claim 5, characterized in that The mutation operation is performed on each offspring chromosome with a certain mutation probability P m , by randomly swapping the task order to change the genes of individuals, thereby increasing the diversity of the population and helping the algorithm to jump out of local optima.

7. The improved genetic optimization algorithm and linear motor control method according to claim 6, characterized in that, For the described iteration and termination, the maximum number of iterations is preset in advance. When the algorithm runs to the maximum number of iterations or the optimal fitness value no longer changes significantly in several consecutive generations, it can be considered that the algorithm has converged and the iteration stops.

Images