Annular rail RGV scheduling method

Abstract

The invention provides a circular orbit RGV scheduling method, and the method comprises the steps: generating an initial chromosome population based on the information of goods in a warehouse and the initial position of an RGV; constructing a fitness function based on the efficiency and energy consumption of the RGV scheduling process, and calculating the fitness value of each chromosome; selecting high-quality chromosomes for breeding according to the fitness value, and reserving the chromosomes with high fitness; performing crossover operation on the selected chromosomes to generate new chromosomes; performing mutation operation on part of chromosomes to obtain a newly generated chromosome population; and performing a next round of iteration based on the newly generated chromosome population until a preset iteration termination condition is reached. The method effectively improves the efficiency and energy consumption management of RGV scheduling, and is suitable for optimal scheduling of the warehouse circular track system.

Description

Technical Field

[0001] The present invention relates to the field of automated equipment dispatching and control, and in particular to a circular track RGV dispatching method. Background Art

[0002] In recent years, with the rapid development of emerging information technologies such as artificial intelligence, the Internet of Things, and intelligent control, the warehousing and logistics industry has gradually moved towards a new era of flexibility, automation, and intelligence. As an important part of the modern warehousing and logistics system, the automated storage / retrieval system (AS / RS) has gradually replaced traditional warehouses and played a key role in improving inventory control, increasing space and time utilization, and increasing cost-effectiveness. With its efficient storage and retrieval capabilities, the automated warehouse is widely used in industries such as pharmaceutical production, tobacco distribution, automobile manufacturing, and military logistics, bringing significant economic benefits to various production companies.

[0003] In automated warehouses, Rail Guided Vehicle (RGV), as an automatic handling device integrating multiple high-tech technologies, is widely used in cargo handling due to its advantages of fast transportation speed, strong flexibility and high intelligence. However, with the expansion of the scale of storage systems and the increase in transportation demand, the RGV system faces challenges such as track congestion and reduced handling efficiency when scheduling on circular tracks. Most existing studies focus on the optimization of the minimum completion time of RGV or the improvement of algorithms, but rarely consider energy consumption issues in actual production environments. In addition, the scheduling optimization problem of RGV on circular tracks often encounters problems such as slow algorithm convergence and easy to fall into local optimal solutions due to the large search space. Therefore, it is particularly important to further study and optimize the scheduling algorithm of the system. Summary of the invention

[0004] The purpose of the present invention is to provide a circular track RGV scheduling method to solve the technical problems in the prior art such as the easy occurrence of local optimal solutions, low RGV scheduling efficiency and high energy consumption cost.

[0005] In a first aspect, the present invention provides a circular track RGV scheduling method, the method comprising:

[0006] Based on the cargo information in the warehouse and the initial position of RGV, an initial chromosome population is generated;

[0007] Based on the efficiency and energy consumption of the RGV scheduling process, a fitness function is constructed to calculate the fitness value of each chromosome;

[0008] Selecting high-quality chromosomes for reproduction according to the fitness value, and retaining chromosomes with high fitness;

[0009] Perform crossover operation on the selected chromosome to generate new chromosome;

[0010] Perform mutation operations on some chromosomes to obtain newly generated chromosome populations;

[0011] Based on the newly generated chromosome population, the next round of iteration is performed until a preset iteration termination condition is reached.

[0012] Optionally, the step of constructing a fitness function based on the efficiency and energy consumption of the RGV scheduling process and calculating the fitness value of each chromosome includes:

[0013] The energy consumption optimization target f1 and time optimization target f2 are calculated according to the following formulas:

[0014]

[0015]

[0016] The fitness function is constructed based on the energy consumption optimization target and the time optimization target. The formula of the fitness function is:

[0017]

[0018] Among them, E i represents the energy consumption in each time period t; Represents the total running time of the i-th time period; a and b are positive numbers.

[0019] Optionally, the energy consumption optimization target f1 in E i for:

[0020]

[0021] The time optimization objective f2 is t i go for:

[0022]

[0023] in, Indicates the roller speed of RGV under load, Indicates the roller speed of RGV in no-load state. num_jam indicates the number of jams that occurred in the i-th time period; num_mission: the number of missions completed in the i-th time period; The total time that the RGV runs without any congestion; The total time wasted by the i-th RGV in traffic jam; The time for loading and unloading the i-th RGV; t up : Elevator working time.

[0024] Optionally, selecting high-quality chromosomes for reproduction according to the fitness value and retaining chromosomes with high fitness includes:

[0025] The roulette wheel method is used to set the selection probability according to the proportion of individual fitness in the total fitness, and the roulette wheel is drawn;

[0026] Based on the result of the roulette wheel extraction, a secondary selection is performed using a selection function to obtain a chromosome with high fitness. The formula of the selection function is:

[0027]

[0028] Among them, x represents the relative difference between the individual fitness and the average fitness of the population.

[0029] Optionally, the method of performing secondary selection based on the result of the roulette wheel extraction using a selection function to obtain a chromosome with high fitness specifically includes:

[0030] Calculate the average fitness of the population drawn by roulette;

[0031] Calculating the difference between the fitness of each chromosome and the average fitness to obtain the relative difference between each chromosome and the average fitness of the population;

[0032] The probability of each chromosome being selected is calculated based on the relative difference, and a chromosome with high fitness is obtained.

[0033] Optionally, performing a crossover operation on the selected chromosome to generate a new chromosome includes:

[0034] Four gene nodes are randomly selected in the gene sequence of the chromosome, and two gene fragments are correspondingly crossed to generate a new chromosome.

[0035] Optionally, performing a mutation operation on a portion of the chromosomes to obtain a newly generated chromosome population includes:

[0036] A semi-random mutation method was used for region-specific mutation;

[0037] A plurality of mutation positions are randomly generated on the chromosome, and a mutation operation is performed on the mixed-coding chromosome.

[0038] Optionally, the method further comprises encoding the RGV transport task by combining decimal coding and binary coding to form a chromosome.

[0039] Optionally, the newly generated chromosome population includes elite individuals, large-scale crossover individuals and new generations.

[0040] Optionally, performing the next round of iteration based on the newly generated chromosome population until a preset iteration termination condition is reached includes:

[0041] When the predetermined number of iterations or running time is reached, the iteration is stopped; or,

[0042] When the fitness value of the optimal solution at this moment meets the preset threshold, the iteration stops.

[0043] According to the solution of the present invention, the fitness function is constructed by comprehensively considering efficiency and energy consumption, and the chromosomes are selected, crossed and mutated to optimize the scheduling process of RGV. This method not only improves the scheduling efficiency, but also effectively reduces energy consumption, avoiding the inefficiency and high energy consumption problems of traditional algorithms in scheduling. At the same time, through continuous iterative optimization, the circular track RGV can achieve more efficient resource scheduling in large-scale intensive transportation scenarios with multiple types of in-and-out storage, improve overall operating efficiency, and optimize resource allocation.

[0044] Furthermore, the technical solution of the present invention effectively avoids the common local optimal solution problem of traditional genetic algorithms in circular track RGV scheduling by introducing a new selection function and an optimized offspring population structure, thereby significantly improving the global optimality of the solution. This method effectively improves the handling efficiency during the scheduling process and is suitable for large-scale intensive transportation scenarios of multiple types of inbound and outbound storage. Therefore, the overall operating efficiency of the circular track RGV system in the automated warehouse has been significantly improved.

[0045] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic flow chart of a circular track RGV scheduling method according to an embodiment of the present invention is shown;

[0047] Figure 2 A schematic diagram of the arrangement of a ring-shaped RGV system according to an embodiment of the present invention is shown;

[0048] Figure 3 It shows a flow chart of FOSB finished product storage in a ring RGV system according to one embodiment of the present invention;

[0049] Figure 4 A schematic diagram of a chromosome encoding structure according to an embodiment of the present invention is shown;

[0050] Figure 5 A graph showing a secondary selection probability function according to one embodiment of the invention;

[0051] Figure 6 A schematic diagram of the structure of chromosome variation according to an embodiment of the present invention is shown;

[0052] Figure 7 A schematic diagram of an initialization interface of a circular RGV warehousing operation system according to an embodiment of the present invention is shown;

[0053] Figure 8 A schematic diagram of a parameter input interface of a circular RGV storage operation system according to an embodiment of the present invention is shown;

[0054] Fig. 9 A schematic diagram of a ring RGV warehousing operation system solution generation interface according to an embodiment of the present invention is shown;

[0055] Fig.10 A schematic diagram of fitness iteration curves of an improved algorithm and an original algorithm according to an embodiment of the present invention is shown;

[0056] Fig.11 A schematic diagram of iterative curves of traffic jam times of an improved algorithm and an original algorithm according to an embodiment of the present invention is shown;

[0057] Fig.12 A schematic diagram of iterative curves of dynamic speed and static speed fitness according to an embodiment of the present invention is shown;

[0058] Fig.13 A schematic diagram of iterative curves of dynamic speed and static speed traffic jam times according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0060] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0061] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] Figure 1 FIG. 2 shows a schematic flow chart of a method for scheduling a circular track RGV according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0063] Step S100, generating an initial chromosome population based on the cargo information in the warehouse and the initial position of the RGV.

[0064] The initial chromosome population is generated based on the cargo information in the warehouse and the initial position of the RGV. Each chromosome represents a possible RGV scheduling solution or path, which includes different scheduling strategies and path selection. This step provides a diverse solution space for the genetic algorithm. By generating multiple initial solutions, it ensures the comprehensiveness and diversity of the subsequent optimization process, helps to explore more possible solutions, and increases the chance of finding the global optimal solution.

[0065] Step S200, based on the efficiency and energy consumption of the RGV scheduling process, construct a fitness function and calculate the fitness value of each chromosome.

[0066] A fitness function is constructed based on the efficiency and energy consumption of the RGV scheduling process, and the fitness value of each chromosome is calculated. The fitness function evaluates the quality of each scheduling solution, usually taking into account scheduling efficiency and energy consumption. By quantifying the performance of chromosomes, it ensures that the genetic algorithm can effectively evaluate based on scheduling efficiency and energy consumption. Chromosomes with high fitness values ​​represent better scheduling solutions, which help to screen out solutions with better performance and guide the subsequent selection and optimization process.

[0067] Step S300, selecting high-quality chromosomes for reproduction according to the fitness value, and retaining chromosomes with high fitness.

[0068] Select high-quality chromosomes for reproduction based on fitness values. By retaining chromosomes with high fitness, it is ensured that excellent scheduling solutions can be retained and passed on during the genetic process, improving the overall quality and optimization effect of subsequent populations. By selecting high-quality chromosomes, high-quality solutions will be given priority in entering the next generation, which can accelerate the process of finding a better solution.

[0069] Step S400, performing a crossover operation on the selected chromosome to generate a new chromosome.

[0070] The crossover operation can generate potential new solutions by combining the characteristics of different chromosomes, thereby increasing the diversity of the population and the search space. This helps to discover new and potentially better scheduling solutions and improve global optimization capabilities.

[0071] Step S500, performing mutation operation on some chromosomes to obtain a newly generated chromosome population.

[0072] Perform mutation operations on some chromosomes, randomly change part of the chromosome's genetic information, and generate new chromosomes. The mutation operation introduces new gene variants, which can break the local optimal solution of the current population, increase the diversity of the population, help avoid falling into the local optimal solution, and promote a wider search and discovery of potentially better solutions.

[0073] Step S600, performing the next round of iteration based on the newly generated chromosome population until a preset iteration termination condition is reached.

[0074] The iterative process continuously optimizes the population and gradually improves the fitness value of the chromosome. Through multiple rounds of iterations, the scheduling plan can be continuously improved, gradually approaching the global optimal solution, ensuring that the final solution has high optimization performance.

[0075] The technical solution of the present invention constructs a fitness function by comprehensively considering efficiency and energy consumption, and performs selection, crossover and mutation operations on chromosomes, thereby optimizing the scheduling process of RGV. This method not only improves scheduling efficiency, but also effectively reduces energy consumption, avoiding the inefficiency and high energy consumption problems of traditional algorithms in scheduling. At the same time, through continuous iterative optimization, the circular track RGV can achieve more efficient resource scheduling in large-scale intensive transportation scenarios with multiple types of in-and-out storage, improve overall operating efficiency, and optimize resource allocation.

[0076] In the present invention, step S200 constructs a fitness function based on the efficiency and energy consumption of the RGV scheduling process, and calculates the fitness value of each chromosome, specifically including:

[0077] Step S210: Calculate the energy consumption optimization target f1 and the time optimization target f2 according to the following formulas:

[0078]

[0079]

[0080] According to the duality principle, Can be converted to Therefore, the ratio of the two can be used to represent the total fitness taking into account the two objectives.

[0081] Step S220, construct the fitness function based on the energy consumption optimization target and the time optimization target, the formula is:

[0082] Among them, E i represents the energy consumption in each time period t; Represents the total running time of the i-th time period; a and b are positive numbers.

[0083] In some embodiments, since the general scenario pays more attention to the transport efficiency target, the influence of the efficiency part is increased in the fitness calculation, and the influence of the energy consumption part is reduced. By setting the values ​​of a and b, the constants affecting the weights of each part of the fitness function are adjusted. The values ​​of a and b can be between 0 and 10, and the specific values ​​are determined according to the weight requirements for energy consumption and time optimization.

[0084] Specifically, the E in the above energy consumption optimization target f1 is i for:

[0085]

[0086] t in the time optimization objective f2 i go for:

[0087]

[0088] in, Indicates the roller speed of RGV under load, Indicates the roller speed of RGV in no-load state. num_jam indicates the number of jams that occurred in the i-th time period; num_mission: the number of missions completed in the i-th time period; The total time that the RGV runs without any congestion; The total time wasted by the i-th RGV in traffic jam; The time for loading and unloading the i-th RGV; t up : Elevator working time.

[0089] The method of selecting high-quality chromosomes for reproduction according to the fitness value and retaining chromosomes with high fitness in step S300 of the present invention specifically includes:

[0090] Step S310, using the roulette method, setting the selection probability according to the proportion of the individual fitness in the total fitness, and performing roulette drawing.

[0091] The roulette wheel method is used to set the selection probability according to the proportion of each chromosome's fitness in the total fitness. Through roulette wheel extraction, the selection probability of each chromosome is proportional to its fitness, and chromosomes with high fitness have a greater chance of being selected. The roulette wheel method is a simple and effective selection strategy that ensures that chromosomes with high fitness have a greater chance of entering the next generation through probabilistic selection. This method naturally retains the better solutions and helps improve the efficiency and performance of the genetic algorithm. The randomness of the roulette wheel can also prevent premature convergence to the local optimal solution, allowing the algorithm to keep exploring in the global solution space.

[0092] Step S320: Based on the result of the roulette wheel extraction, a selection function is used to perform secondary selection to obtain a chromosome with high fitness. The formula of the selection function is:

[0093]

[0094] Among them, x represents the relative difference between the individual fitness and the average fitness of the population.

[0095] On the basis of roulette extraction, a secondary selection is performed using a selection function to further screen chromosomes with high fitness. Secondary selection further optimizes the fitness selection process through the selection function. The selection function can effectively amplify the fitness difference, further increasing the probability of chromosomes with high fitness being selected, thereby enhancing the genetic algorithm's retention of excellent solutions. This method allows high-quality solutions to be more strongly propagated in the population, which helps to accelerate convergence and improve overall optimization performance.

[0096] The above step S320 uses a selection function to perform secondary selection based on the result of the roulette wheel extraction to obtain a chromosome with high fitness. The specific method includes the following steps:

[0097] Step S321, calculating the average fitness of the population obtained by roulette wheel extraction.

[0098] Calculating the average fitness of a population helps determine the gap between the fitness levels of individual chromosomes and the overall population, provides a reference value for relative difference calculation and selection probability determination, and helps to more accurately evaluate and screen chromosomes with high fitness.

[0099] Step S322, calculating the difference between the fitness of each chromosome and the average fitness, and obtaining the relative difference between each chromosome and the average fitness of the population.

[0100] The fitness value of each chromosome is compared with the average fitness of the colony calculated in step S321, and the difference between the fitness of each chromosome and the average fitness is calculated. This difference reflects the performance of the individual chromosome relative to the colony. By calculating the difference between each chromosome and the average fitness of the colony, the performance of each chromosome in the colony can be intuitively understood. The calculation of this difference helps to distinguish the pros and cons of chromosomes, so that the chromosomes with higher fitness have a greater probability of being retained in the next selection step, thereby improving the optimization effect of the algorithm.

[0101] Step S323, calculating the probability of each chromosome being selected based on the relative difference, and obtaining chromosomes with high fitness.

[0102] According to the relative difference calculated in step S322, the selection function is used to calculate the probability of each chromosome being selected. The larger the relative difference, the higher the probability of the chromosome being selected. Calculating the selection probability based on the relative difference can effectively amplify the fitness difference, so that better quality chromosomes have a greater chance of being selected into the next generation. This step ensures that the genetic algorithm can retain more individuals with good performance, thereby improving the overall optimization ability and convergence speed of the algorithm. At the same time, through accurate probability calculation, the random retention of inferior chromosomes is also avoided, making the entire algorithm more efficient during the evolution process.

[0103] In some embodiments, the crossover operation on the selected chromosome is specifically to randomly select four gene nodes in the gene sequence of the chromosome, cross the two gene segments accordingly, and generate a new chromosome.

[0104] Four gene nodes are randomly selected from the gene sequence of each selected chromosome. These gene nodes divide the gene sequence of the chromosome to form two gene fragments. These two gene fragments are crossed with the corresponding gene fragments of another chromosome, that is, the positions of the two gene fragments are exchanged to generate a new chromosome. The randomness in this process ensures that the crossover operation is not limited to certain specific gene regions, and genes are recombined in a wider range. New gene combinations can be generated by randomly selecting gene nodes and performing crossover operations on gene fragments. This randomness greatly enriches the diversity of the population, prevents the gene sequences of chromosomes from being too similar, and thus reduces the risk of the algorithm falling into a local optimal solution. The crossover operation introduces new gene combinations, allowing the algorithm to explore more possible solution spaces. This improvement in global search capabilities helps the algorithm more effectively approach the global optimal solution of the problem.

[0105] In some embodiments, the method of performing mutation operation on part of chromosomes to obtain a newly generated chromosome population in step S500 specifically includes the following steps:

[0106] S510, using a semi-random mutation method to perform region-specific mutation.

[0107] The mutation operation is performed in a semi-random manner, that is, the location and range of the mutation are not completely random, but are performed in a specified chromosome region. By performing mutations in specified regions, the possibility of making adjustments to genes that may contain key information regions is enhanced, thereby improving the optimization effect of the mutation operation on the solution. Compared with completely random mutations, semi-random mutations reduce the risk of destroying excellent gene combinations and help retain high-quality individuals in the population.

[0108] S520, randomly generate multiple mutation positions on the chromosome, and perform mutation operations on the mixed-coded chromosome.

[0109] Multiple mutation positions are randomly generated on the chromosome, and the genes at these positions are mutated. The mixed-coded chromosome contains decimal and binary codes. Through random mutation, the gene positions of different codes can be modified. The mutation may include flipping binary bits, changing decimal values, etc., to introduce new genetic diversity. By performing mutations at multiple positions, new gene combinations can be effectively introduced to avoid the population being too single, which helps to explore a wider solution space. The mutation operation can break the original gene combination structure and may produce better solutions, thereby improving the overall fitness of the population.

[0110] In some embodiments, the newly generated chromosome population includes elite individuals, large-scale crossover individuals, and new generations.

[0111] Elite individuals refer to individuals retained by secondary selection during the parent selection process. Large-scale crossover individuals refer to individuals obtained after crossover and mutation of individuals obtained in the first selection process. The new generation refers to individuals obtained after crossover and mutation between elite individuals. By retaining elite individuals, the continuation of high-quality genes is ensured, large-scale crossover introduces new gene combinations, and the new generation further enriches the population diversity and ensures the global search capability of the algorithm. This population structure design helps to quickly find individuals close to the optimal solution and accelerate the convergence of the algorithm.

[0112] In some embodiments, the RGV transport task is encoded by combining decimal coding and binary coding to form a chromosome.

[0113] Combining the two encoding methods can more accurately represent complex task scheduling requirements and improve the expressiveness of the encoding. This encoding method enables the fitness function to more comprehensively evaluate the pros and cons of individuals, thereby improving the optimization effect.

[0114] In some embodiments, the iteration termination condition of the above step S600 includes stopping the iteration when a predetermined number of iterations or running time is reached, or stopping the iteration when the fitness value of the optimal solution at this moment meets a preset threshold.

[0115] By setting the maximum number of iterations and running time, the algorithm is prevented from continuing to consume computing resources after the effect has been significantly improved, ensuring efficiency. When the preset fitness threshold is met, the iteration is stopped to ensure the quality of the solution obtained and avoid falling into the situation of local optimal or suboptimal solution.

[0116] According to the above embodiment, by introducing a new selection function and an optimized offspring population structure, the common local optimal solution problem of the traditional genetic algorithm in the circular track RGV scheduling is effectively avoided, thereby significantly improving the global optimality of the solution. This method effectively improves the handling efficiency during the scheduling process and is suitable for large-scale intensive transportation scenarios of multiple types of inbound and outbound storage. Therefore, the overall operating efficiency of the circular track RGV system in the automated warehouse has been significantly improved.

[0117] The reliability analysis and optimization method for the edge recognition accuracy of the inner diameter of the heat exchanger tube hole of the present invention will be further explained below with a specific implementation process.

[0118] Figure 2 A schematic diagram of the arrangement of a ring-shaped RGV system according to an embodiment of the present invention is shown. Figure 3 The figure shows a flow chart of FOSB finished product storage in a ring RGV system according to an embodiment of the present invention. Figure 2 and Figure 3 As shown in the figure, in the initial state of the circular RGV system, the RGV is in an empty state. After receiving the operation instruction, it starts from the origin to perform the task. During the entire process of finished product warehousing, the warehouse control system (WCS) dispatches the stacker to take out the empty turnover box specified by the warehouse management system (WMS) from the specific cargo position of the shelf, and transports it to the roller conveyor at the entrance of the lane, waiting for the empty RGV on the circular track to ship. The RGV then travels counterclockwise to the workstation, waiting for manual loading of the fully transparent wafer packaging box (Front Opening Shipping Box, FOSB) into the empty turnover box. After the boxing is completed, the RGV transports the finished product to the designated lane entrance platform, and the stacker then returns the finished product to the shelf, thus completing the entire finished product warehousing task.

[0119] Figure 4 FIG. 2 shows a schematic diagram of a chromosome encoding structure according to an embodiment of the present invention. Figure 4 As shown, the RGV handling task is encoded by combining decimal coding and binary coding to form a chromosome. The chromosome coding of the RGV task point adopts integer coding. The task lanes and elevators are numbered in sequence. The task lanes are numbered 0-5, and the elevators are numbered 6 and 7. For the chromosome coding of the RGV speed part, binary coding is adopted. When the maximum speed of the RGV is 3m / s, a five-bit binary number can be used to represent the running speed and loading and unloading speed of the RGV respectively. The speed accuracy after decoding can reach 0.1m / s. The result of the operation sequence generated by the coding is as follows:

[0120]

[0121] In the generated operation sequence, the target lanes for the i-th RGV to receive the empty box outbound operation are numbered from 0 to 5 in a counterclockwise direction; the target stations for the FOSB manual packing operation at the workstation are numbered 6 and 7 respectively; the target lanes for the finished product warehousing operation are numbered the same as the empty box outbound operation, from 0 to 5. Each RGV and each operation task lane are represented by decimal coding, while the speed variable is encoded in binary. The RGV speed and roller speed are each represented by five binary digits to ensure high-precision control of speed changes. For example, in the randomly generated chromosome sequence "10651000010000, 21631010100101, 33711101001100, 45741011001001, 52621000100011, 65701111101011", the sequence "10651000001000" means that the first RGV first goes to aisle 0 to perform the empty turnover box outbound operation, then runs to workstation 6 to wait for workers to load FOSB into empty turnover boxes, and finally transports to aisle 5 to complete the finished product warehousing operation. "10000" in the sequence is converted to decimal 16, and "01000" is converted to decimal 8. Since the maximum representation value of a five-bit binary number is 32, and the operating speed ranges of the RGV and roller conveyor have been defined, the operating speed of the RGV and the operating speeds of the roller conveyor and RGV rollers at this time can be calculated.

[0122] For the ring track RGV system, it is crucial to choose an appropriate population size. Controlling the population size between 50 and 200 can avoid excessive consumption of computing resources while ensuring the efficiency and diversity of the algorithm search. After analysis, 100 chromosomes were randomly generated in this embodiment, and the population size of 100 was finally determined. The encoding of each chromosome consists of five parts, namely the RGV number, the empty turnover box outbound lane number, the workstation packing position number, the RGV running speed and the roller speed.

[0123] The fitness function is constructed based on the transport efficiency and operating energy consumption of the circular RGV, and the fitness value of each chromosome is calculated. The high-quality chromosomes are selected for reproduction according to the fitness value, and the double selection process is used to enhance the accuracy and diversity of population selection. In the first selection, the roulette method is used to set the selection probability of each individual according to the proportion of its fitness in the total fitness, and a group of individuals are initially selected through roulette extraction. The second selection introduces a new selection function: Figure 5 FIG. 4 shows a graph of a secondary selection probability function according to an embodiment of the invention. Figure 5 As shown in the figure, this function cleverly combines the activation function Sigmoid in deep learning, and extends its domain to the entire interval (-1,1) through the inverse hyperbolic tangent function, thereby ensuring that the probability distribution covers the complete curve during the population selection process. In the secondary selection, for individuals with lower fitness, this function will significantly reduce their selection probability weight, making it more difficult to be retained in subsequent selections, while for individuals with higher fitness, their selection weight will be increased accordingly to improve their chances of survival. Specifically, the average fitness of the population after one selection is first calculated, and the difference is made with the fitness of each individual to obtain the deviation relative to the average value. Then, the minimum and maximum values ​​of the deviation are used as the interval endpoints of the domain, and the entire interval is mapped to (-1, 1). After being mapped by this function, the final selection probability of each individual is further calculated by the Sigmoid function. Finally, individuals are randomly selected from the population selected once based on these probabilities to ensure that the retained individuals are more adaptable.

[0124] After the selection operation is completed, this embodiment adopts the method of two-segment gene crossover. Since the individual gene sequence in the model is long, the single-point or two-point crossover has limited effect on the individual's gene structure. Therefore, four nodes are randomly selected in the crossover operation, and the two gene segments between the parent nodes are correspondingly crossed to generate new chromosomes. This crossover method ensures that the diversity of individual gene sequences is retained and recombined during the gene recombination process by operating on multiple points of the parent gene.

[0125] Figure 6 FIG. 2 shows a schematic diagram of a chromosome variation structure according to an embodiment of the present invention. Figure 6 As shown, a semi-random mutation method is used to mutate the selected chromosome, and a specific area is specified for mutation. In addition, multiple mutation positions are randomly generated on the chromosome, and the mixed-coded chromosome is mutated to ensure the diversity of gene positions during the mutation process. After the mutation operation is completed, the generated offspring population is divided into three parts: elite individuals, large-scale crossover individuals, and new generations. Elite individuals come from individuals retained by secondary selection during the parent selection process; large-scale crossover individuals are individuals generated by crossover and mutation operations after a single selection; and the new generation is generated by crossover and mutation between elite individuals. The iterative process of the entire genetic algorithm sets two main termination conditions. In this embodiment, the first termination condition is selected, that is, the algorithm is stopped when the number of iterations reaches a preset 1000 generations.

[0126] In order to realize the visualization of the scheduling problem solving process of the circular RGV system in the stereoscopic warehouse, this embodiment uses Python's third-party library PyQt5 to develop and demonstrate a circular RGV warehousing operation system, aiming to display the key data and operation steps in the scheduling process through a graphical interface, thereby providing an intuitive analysis tool and user interaction interface. Figure 7 A schematic diagram of an initialization interface of a circular RGV warehousing operation system according to an embodiment of the present invention is shown. Figure 8 A schematic diagram of a parameter input interface for a circular RGV warehousing operation system according to an embodiment of the present invention is shown. Fig. 9 FIG. 1 is a schematic diagram of a ring RGV storage operation system solution generation interface according to an embodiment of the present invention. Figure 7 and Figure 8 As shown, the user first accesses the initialization interface of the system. Clicking the "Start" button will jump to the parameter input interface, and clicking "Exit" will close the initialization interface. In the parameter input interface, the user can enter relevant data parameters related to the RGV scheduling problem, such as "number of RGVs", "number of tasks", "maximum speed", "efficiency weight" and "energy consumption weight". In some embodiments, the user sets the number of RGVs to 3, the number of tasks to 3, the maximum speed to 3, the efficiency weight to 0.8, and the energy consumption weight to 0.2. Fig. 9 As shown in the figure, after the user enters the relevant parameters of the scheduling problem to be solved and clicks the "Confirm" button, the system will generate the optimal solution to the scheduling problem. The solution includes the task sequence of each RGV, the operating speed of each RGV, and the roller loading and unloading transportation speed.

[0127] Fig.10 A schematic diagram of fitness iteration curves of an improved algorithm and an original algorithm according to an embodiment of the present invention is shown. Fig.11The diagram shows the iterative curve of the number of traffic jams of the improved algorithm and the original algorithm according to an embodiment of the present invention. Based on the improved genetic algorithm, the scheduling problem of the circular RGV system under both constant speed and variable speed conditions is solved. For the variable speed condition, the aforementioned encoding method is used to represent the task scheduling and speed configuration. To ensure the rationality and effectiveness of the comparison, under the constant speed condition, the running speed of the RGV and the roller speed are set to the average of all RGV and roller speeds under the variable speed condition. Fig.10 and Fig.11 As shown in the figure, the unimproved algorithm converged at the 230th iteration, and the final fitness converged to 1.7. The improved algorithm began to converge at the 250th iteration, and broke away from the local optimal solution at the 750th iteration, and finally converged to 2.3. The improved algorithm finally improved the fitness value by 30%. Although its convergence speed was slightly reduced, it had better search performance and could effectively break the local optimal solution. Since the increase in congestion will lead to an increase in transportation time and energy consumption, the change trend of the number of congestion is opposite to the change trend of the fitness value.

[0128] Fig.12 A schematic diagram of iterative curves of dynamic speed and static speed fitness according to an embodiment of the present invention is shown. Fig.13 FIG. 1 shows a schematic diagram of an iterative curve of dynamic speed and static speed traffic jam times according to an embodiment of the present invention. Fig.12 and Fig.13 As shown, the red curve represents the dynamic speed iteration process, and the black curve represents the fixed speed iteration process. When the number of RGVs is set to 6, the dynamic speed scheduling scheme begins to converge at the 100th iteration and eventually reaches a fitness value of 2.3. After the 200th iteration, the fitness value tends to stabilize, showing good convergence and faster convergence speed. In contrast, in the dual-objective optimization, the fitness improvement of the fixed speed scheduling scheme is relatively limited. Since congestion will increase the handling time and energy consumption, Fig.13 The trend of the number of blockages and the fitness value is opposite. Fig.12 and Fig.13 ,It can be clearly seen that in the circular track RGV scheduling, the dynamic speed scheduling scheme shows a better scheduling effect after ,iterative optimization compared with the fixed speed scheduling scheme.

[0129] According to the embodiment, the scheduling problem of the circular track RGV system is optimized by improving the genetic algorithm, which effectively improves the scheduling efficiency and energy consumption management. By combining the roulette method and the two-selection strategy of the new selection function, and adopting two-segment gene crossover and semi-random mutation methods, the fitness of the population is optimized and the local optimal solution is avoided. In addition, the composition of the offspring population is clarified, so that the convergence of the algorithm is guaranteed. The visualization system developed based on PyQt5 enables users to intuitively input parameters and view optimization results, which enhances the practicality and user experience of the system. Overall, these optimization measures provide an efficient and reliable solution to practical scheduling problems.

[0130] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0131] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A circular track RGV scheduling method, It is characterized in that The method comprises: Based on the cargo information in the warehouse and the initial position of RGV, an initial chromosome population is generated; Based on the efficiency and energy consumption of the RGV scheduling process, a fitness function is constructed to calculate the fitness value of each chromosome; Selecting high-quality chromosomes for reproduction according to the fitness value, and retaining chromosomes with high fitness; Perform crossover operation on the selected chromosome to generate new chromosome; Perform mutation operations on some chromosomes to obtain newly generated chromosome populations; Based on the newly generated chromosome population, the next round of iteration is performed until a preset iteration termination condition is reached.

2. The circular track RGV scheduling method according to claim 1, It is characterized in that The efficiency and energy consumption of the RGV scheduling process are used to construct a fitness function, and the fitness value of each chromosome is calculated, including: The energy consumption optimization target f is calculated according to the following formulas 1 and time optimization objective f 2 : The fitness function is constructed based on the energy consumption optimization target and the time optimization target. The formula of the fitness function is: Among them, E i represents the energy consumption in each time period t; Represents the total running time of the i-th time period; a and b are positive numbers.

3. The circular track RGV scheduling method according to claim 2, It is characterized in that The energy consumption optimization target f 1 E i for: The time optimization objective f 2 t in i go for: in, Indicates the roller speed of RGV under load. Indicates the roller speed of RGV in no-load state. num_jam indicates the number of jams that occurred in the i-th time period; num_mission: the number of missions completed in the i-th time period; The total time that the RGV runs without any congestion; The total time wasted by the i-th RGV in traffic jam; The time for loading and unloading the i-th RGV; t up : Elevator working time.

4. The circular track RGV scheduling method according to claim 1, It is characterized in that The step of selecting high-quality chromosomes for reproduction according to the fitness value and retaining chromosomes with high fitness includes: The roulette wheel method is used to set the selection probability according to the proportion of individual fitness in the total fitness, and the roulette wheel is drawn; Based on the result of the roulette wheel extraction, a secondary selection is performed using a selection function to obtain a chromosome with high fitness. The formula of the selection function is: Among them, x represents the relative difference between the individual fitness and the average fitness of the population.

5. The circular track RGV scheduling method according to claim 4, It is characterized in that The method of performing secondary selection based on the result of the roulette wheel extraction using a selection function to obtain a chromosome with high fitness specifically includes: Calculate the average fitness of the population drawn by roulette; Calculating the difference between the fitness of each chromosome and the average fitness to obtain the relative difference between each chromosome and the average fitness of the population; The probability of each chromosome being selected is calculated based on the relative difference, and a chromosome with high fitness is obtained.

6. The circular track RGV scheduling method according to claim 1, It is characterized in that The performing a crossover operation on the selected chromosome to generate a new chromosome comprises: Four gene nodes are randomly selected in the gene sequence of the chromosome, and two gene fragments are correspondingly crossed to generate a new chromosome.

7. The circular track RGV scheduling method according to claim 1, It is characterized in that The performing mutation operation on some chromosomes to obtain a newly generated chromosome population comprises: A semi-random mutation method was used for region-specific mutation; A plurality of mutation positions are randomly generated on the chromosome, and a mutation operation is performed on the mixed-coding chromosome.

8. The circular track RGV scheduling method according to claim 1, It is characterized in that The newly generated chromosome population includes elite individuals, large-scale crossover individuals and new generations.

9. The circular track RGV scheduling method according to claim 1, It is characterized in that The method also includes encoding the RGV transport task by combining decimal coding and binary coding to form a chromosome.

10. The circular track RGV scheduling method according to claim 1, It is characterized in that The step of performing the next round of iteration based on the newly generated chromosome population until a preset iteration termination condition is reached includes: When the predetermined number of iterations or running time is reached, the iteration is stopped; or, When the fitness value of the optimal solution at this moment meets the preset threshold, the iteration stops.

Images