Cross-layer cross-roadway shuttle vehicle warehousing system configuration optimization method

Abstract

The invention discloses a cross-layer cross-roadway shuttle vehicle warehousing system configuration optimization method. The method comprises the following steps: determining a warehouse-in and warehouse-out operation process of a cross-layer cross-roadway shuttle vehicle warehousing system; establishing a mixed integer optimization mathematical model according to the operation process, and defining a model hypothesis; determining a target and a constraint condition of a mixed integer optimization model; designing an improved non-dominated sorting genetic algorithm II fused with a clone selection algorithm to solve the mathematical model; and the optimal layer-column ratio configuration of the cross-layer and cross-roadway shuttle vehicle warehousing system is determined. According to the method, a mixed integer optimization model is established by taking a cross-layer cross-roadway shuttle vehicle storage system as a research object and taking travel time and energy consumption required for completing a task as targets, and an improved non-dominated sorting genetic algorithm II is adopted to solve a problem to determine the optimal configuration of the system. According to the method, a design scheme can be provided for designers in the warehouse design stage, and the optimal configuration of the warehousing system and the model selection of the equipment speed acceleration are determined.

Description

Technical Field

[0001] The present invention relates to the technical field of configuration optimization in the field of intelligent warehousing, and specifically to an optimization method for the configuration of a cross-layer and cross-aisle shuttle car warehousing system. Background Art

[0002] With the development of intelligent manufacturing technology, the demand for automated warehousing systems has increased significantly. The development of the shuttle car warehousing system improves the performance of the automated storage system by gradually updating equipment and enhancing equipment performance. As an extended system of the shuttle car storage and retrieval system, the cross-layer and cross-aisle shuttle car warehousing system has characteristics such as high flexibility and low system cost. However, in the early stage of system construction, how to reasonably plan the number of floors, columns, and aisles of the warehouse as well as the optimal equipment parameters, and design a system architecture that can not only meet functional requirements but also have scalability is a key issue in the initial stage of warehousing system construction.

[0003] In the past, the configuration of warehouses mostly relied on experience to determine, which often led to the inability of the system storage to meet actual needs. With the continuous advancement of enterprise development, the storage requirements of the warehousing system have been continuously improved, mainly reflected in the following aspects: one is to increase storage capacity, the second is to improve storage efficiency, and the third is to save time costs. In addition, reducing the energy consumption of the warehousing system has also become a more important consideration factor.

[0004] On the other hand, using only travel time or energy consumption as the evaluation index for optimal configuration is one-sided and cannot obtain the optimal configuration of the system. In fact, it may not improve the system operation efficiency and may even reduce the system operation efficiency. Only by comprehensively considering the system can the cross-layer and cross-aisle shuttle car warehousing system exhibit optimal performance in future actual operations. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method for the configuration of a cross-layer and cross-aisle shuttle car warehousing system, to solve technical problems such as insufficient storage capacity and high system cost caused by relying on empirical construction of warehouses in the early stage of warehouse construction, so as to achieve the purpose of saving system costs and sustainable development.

[0006] According to an embodiment of the present application, an optimization method for the configuration of a cross-layer and cross-aisle shuttle car warehousing system is provided, including: S1: According to the operation process of the cross-layer and cross-aisle shuttle car warehousing system, establish a bi-objective mathematical model with travel time and energy consumption as the objectives. The bi-objective mathematical model includes model assumptions, a travel time objective function, an energy consumption objective function, and model constraint conditions; S2: According to the bi-objective mathematical model, by expanding the search solution range and strengthening local search, fuse the non-dominated sorting genetic algorithm II with the clonal selection algorithm to obtain an improved non-dominated sorting genetic algorithm II; S3: Solve the bi-objective mathematical model using the improved non-dominated sorting genetic algorithm II to obtain the travel time and energy consumption required to complete the task. S4: Determine the system comprehensive cost based on the travel time and energy consumption required to complete the task, and then determine the optimal configuration of the cross-layer and cross-aisle shuttle car storage system.

[0007] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects: For the optimal configuration of the cross-layer and cross-aisle shuttle car storage system, a bi-objective mathematical model considering travel time and energy consumption is constructed, and the improved non-dominated sorting genetic algorithm II is used to solve the mathematical model to obtain the optimal layer-column ratio configuration and equipment parameters of the cross-layer and cross-aisle shuttle car storage system, solving the technical problems such as insufficient storage capacity and high system cost caused by relying on empirical construction of the warehouse in the initial stage of warehouse construction, thereby achieving the purpose of saving system cost and sustainable development. Description of the Drawings

[0008] Figure 1 is a method for optimizing the configuration of a cross-layer and cross-aisle shuttle car storage system shown according to an exemplary embodiment.

[0009] Figure 2 is a schematic layout diagram of a cross-layer and cross-aisle shuttle car storage system shown according to an exemplary embodiment.

[0010] Figure 3 is a flowchart of the operation of a cross-layer and cross-aisle shuttle car storage system shown according to an exemplary embodiment.

[0011] Figure 4 is a speed model diagram of a transport device shown according to an exemplary embodiment.

[0012] Figure 5 is an algorithm coding diagram of the improved non-dominated sorting genetic algorithm II shown according to an exemplary embodiment.

[0013] Figure 6 is a schematic diagram of the chromosome mutation operation of the improved non-dominated sorting genetic algorithm II shown according to an exemplary embodiment.

[0014] Figure 7 is a schematic diagram of the chromosome crossover operation of the improved non-dominated sorting genetic algorithm II shown according to an exemplary embodiment.

[0015] Figure 8 is a flowchart of the improved non-dominated sorting genetic algorithm II shown according to an exemplary embodiment.

[0016] Figure 9 is a comparison of the comprehensive costs of cross-layer and cross-aisle shuttle car storage systems under different algorithms and configurations shown according to an exemplary embodiment. Detailed implementation manners

[0017] The following further describes in detail the detailed implementation manners of the present invention with reference to the accompanying drawings. It should be understood that the detailed implementation manners described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0018] Figure 1 It is an optimization method for the configuration of a cross-layer and cross-aisle shuttle car storage system shown according to an exemplary embodiment. Refer to Figure 1 , an optimization method for the configuration of a cross-layer and cross-aisle shuttle car storage system provided by an embodiment of the present invention may include: S1: According to the operation process of the cross-layer and cross-aisle shuttle car storage system, establish a bi-objective mathematical model with travel time and energy consumption as objectives. The bi-objective mathematical model includes model assumptions, a travel time objective function, an energy consumption objective function, and model constraint conditions; Specifically, as Figure 2 , is a schematic diagram of a cross-layer and cross-aisle shuttle car storage system. The shuttle car moves along the y direction in the aisle to perform storage and retrieval tasks. The transfer vehicle moves along the x direction at the aisle entrance for cross-aisle movement of tasks. The elevator moves along the z axis to perform cross-layer movement of tasks. Determine the inbound and outbound operation process of the cross-layer and cross-aisle shuttle car storage system (refer to Figure 3 ), as follows: System operation process: The system receives a task sequence, the number of tasks is n, the retrieval task is set to 0, and the storage task is set to 1. The system identifies the type of each task and performs the corresponding storage or retrieval operation accordingly. After each task is executed, the system checks whether the currently executed tasks have reached the maximum number of tasks n. If it reaches "n", the task is terminated. Otherwise, continue to execute the next task; Retrieval operation process: The system receives a task sequence . If represents a retrieval task, then . Subsequently, perform the retrieval operation. First, call the shuttle car, which reaches the target position to grab the bin and brings it to buffer 1; then call the transfer vehicle, which travels to the aisle entrance of the layer, completes the picking, transports the bin to buffer 2, and finally calls the elevator. The elevator transports the bin to the I / O point to complete the retrieval task.

[0019] Storage operation process: The system receives a storage task instruction, that is, . Determine the three-dimensional coordinates of the storage task. First, call the shuttle car and move it to the I / O port to pick up the bin. Second, call the elevator, which brings the bin to buffer 2 on the z i layer. Subsequently, call the transfer vehicle, and the transfer vehicle transports the bin to buffer 1 in the aisle x i ; call the shuttle car, and the shuttle car transports the bin to the storage bin to complete a storage task.

[0020] After determining the operation process of the cross - layer and cross - roadway shuttle car storage system, it is necessary to establish a bi - objective mathematical model with travel time and energy consumption as the objectives, specifically as follows: (1) First, determine the model assumptions: Specifically: Assumption 1: The shuttle car can only carry one bin at a time; Assumption 2: The loading and unloading time of the shuttle car is equal; Assumption 3: Adopt the random storage strategy, and the position of the transport equipment is the position for the final task processing; Assumption 4: Each transport equipment in the system operates independently; Assumption 5: During the operation of the shuttle car, adopt the first - in - last - out strategy, that is: if the first shuttle car enters the roadway before the second shuttle car, the first shuttle car needs to leave the roadway after the second shuttle car completes the task; Assumption 6: The system executes composite command operations. However, when the task only involves retrieval / storage tasks, single - command operations are executed; Assumption 7: Call the transport equipment according to the principle of the shortest time.

[0021] (2) Determine the objective function of the bi - objective mathematical model: Objective 1: The objective function of travel time: Specifically: As Figure 4 shown, according to whether the transport equipment can run at the maximum speed, the travel time of the system can be divided into Mode 1: The transport equipment does not reach the maximum speed; Mode 2: The transport equipment reaches the maximum speed.

[0022] Mode 1: The transport equipment does not reach the maximum speed. When the transport equipment runs in the limited space of the warehouse and cannot reach the maximum driving speed, its movement is limited to the acceleration and deceleration processes, and the corresponding travel - time function is expressed as follows: ; ; ; ; In the formula, D is the driving distance of the transport equipment, a is the acceleration of the transport equipment, is the highest speed reached by the transport equipment, t 1 is the running time at the highest speed, and T is the total running time of the transport equipment accelerating first and then decelerating.

[0023] Mode 2: The transport equipment can reach its maximum driving speed. The operation process of the equipment can be divided into three stages: first is the constant acceleration stage, then it maintains the maximum speed for a period of time, and finally is the constant deceleration stage until it reaches the destination. The corresponding travel time function for this process, the time required for the transport equipment to reach the maximum speed, and the total travel time are as follows: ; ; where D is the driving distance of the transport equipment, V max is the maximum speed reached by the transport equipment, is the uniform running time of the transport equipment.

[0024] Objective 2: Energy consumption objective function: According to the double-objective mathematical model, the energy consumption objective function is calculated as follows: The traction force required when the shuttle car accelerates and the energy consumption to overcome the traction force are: ; ; where, F T is the traction force during acceleration, a s is the acceleration of the shuttle car, m s is the weight of the shuttle car, g is the acceleration due to gravity, C r is the friction coefficient, is the resistance coefficient of the variable-speed rotating mass (1.15), is the efficiency of the transport equipment, W S_A is the energy consumption consumed when the shuttle car accelerates; The braking force of the shuttle car and the energy consumption required to overcome the braking force are expressed as follows: ; ; where, F B is the traction force when the shuttle car brakes, W S_D is the energy consumption consumed when the shuttle car decelerates; The traction force F C of the shuttle car during uniform motion and the energy W S_C required to overcome the traction force are expressed as follows: ; ; where, F C is the traction force of the shuttle car during uniform motion, W S_C is the energy consumption consumed when the shuttle car is moving at a uniform speed; The total energy consumption required for the shuttle car to complete the inbound and outbound tasks is expressed as follows: ; The transfer vehicle moves horizontally, and its motion mode is similar to that of the shuttle vehicle, so the power calculation of the transfer vehicle is simplified. The total energy consumption required for the transfer vehicle to accelerate, decelerate, and move at a constant speed is expressed as follows:

[0025] The lifting force F of the elevator L and the energy consumption W required when the elevator accelerates L_A , are expressed as follows:

[0026]

[0027] In the formula, F L is the traction force when the elevator accelerates, m L is the weight of the elevator, a L is the acceleration of the elevator, W L_A is the energy consumption consumed by the elevator during acceleration; The traction force F when the elevator brakes B and the energy consumption W required by the elevator during deceleration L_D , are expressed as follows:

[0028]

[0029] In the formula, F B is the braking force when the elevator decelerates, W L_D is the energy consumption required by the elevator.

[0030] The traction force F when the elevator runs at a constant speed at the maximum speed C and the energy consumption W required to maintain the constant speed operation L_C , are expressed as follows:

[0031]

[0032] The total energy consumption required for the elevator to complete the inbound and outbound tasks is expressed as follows:

[0033] (3) Model constraints of the double-objective mathematical model: Constraint 1: Only one carrier device can carry the bin at each execution stage of the current task, and the expression is as follows:

[0034] In the formula, N sThe number of transportation devices in stage s, s is the stage of task execution, K i The i-th task currently being executed, k represents the transportation device number; Constraint 2: Ensure that the start time of the task is non - negative, and the expression is as follows:

[0035] In the formula, represents the start time of the K i -th task on device k in stage s; Constraint 3: Ensure that under the same transportation device, the current task starts later than the previous task, and the expression is as follows:

[0036] Constraint 4: Ensure that the completion time of the i - th task is earlier than the start time of the (i + 1)-th task, and the expression is as follows:

[0037] In the formula, is the operation time of the K i -th task in stage s; Constraint 5: Ensure that task K i has at most one subsequent task under the same device, and the expression is as follows:

[0038] Constraint 6: Ensure that task K i has at most one preceding task under the same device, and the expression is as follows:

[0039] Constraint 7: Limit the total travel time of the task to be not less than the travel time of any single task, and the expression is as follows:

[0040]

[0041]

[0042] In the formula, T max is the time required to complete the task; In the multi - objective optimization problem, each objective restricts each other, so there is no unique optimal solution. In this context, many solutions are non - dominated with each other, forming a Pareto solution set. And there are problems such as the inability to obtain the optimal solution and the difference in the objective scale of the solutions in the Pareto solution set, which can be handled by weighted summation and normalization of the solutions. The travel time and energy consumption of the cross - layer and cross - roadway shuttle car storage system are normalized, and a weight coefficient is introduced to establish a two - objective mathematical model of the system comprehensive cost, and the expression is as follows:

[0043]

[0044]

[0045]

[0046] wherein, is the normalized value of the travel time of the i-th experimental run under the j-th configuration, is the normalized value of the energy consumption of the i-th experiment under the j-th configuration, and n e is the weighted value of the comprehensive cost of the system after n experiments, m is the total number of warehouse configurations, and n is the total number of experiments conducted under each configuration. and are the weighting coefficients of travel time and energy consumption; C j is the comprehensive cost of the system.

[0047] S2: According to the double-objective mathematical model, by expanding the search solution range and strengthening local search, a non-dominated sorting genetic algorithm II is obtained by integrating the non-dominated sorting genetic algorithm and the clonal selection algorithm; this step includes the following sub-steps: S21: A three-layer hybrid integer coding method is used for chromosome representation, where each chromosome contains three-dimensional coordinate information for expressing the coordinates of each batch of storage and retrieval tasks; Specifically, the coding method determines the form of the solution to the problem, and the coding method adopted for chromosomes largely determines the performance of the algorithm. Hybrid integer coding has characteristics such as being concise and easy for crossover and mutation operations, which can improve the algorithm's performance and optimization ability. In this application, a three-layer hybrid integer coding method is used for chromosome representation, where each chromosome contains three-dimensional coordinate information for a series of storage and retrieval tasks. As Figure 5 shown, the three-dimensional coordinates of the first task are (1, 6, 8), indicating that the task is located at row 6, column 1, on the 8th floor.

[0048] S22: Tournament selection is adopted, and the non-dominated sorting levels of chromosomes are compared pairwise. Chromosomes with lower levels are preferentially retained and passed on to the next generation; if the non-dominated sorting levels of two chromosomes are the same, then selection is made according to the niche technology, and individuals with larger crowding distances are preferentially selected and retained for the next generation; Specifically, tournament selection is adopted. Two chromosomes are selected for non-dominated sorting, and those with lower levels are preferentially retained and passed on to the next generation. If the non-dominated sorting levels of two chromosomes are the same, then selection can be made according to the niche technology. Individuals with larger crowding distances are preferentially selected and retained for the next generation. The niche selection method is as follows:

[0049] where f lt is the function value corresponding to the t-th objective in the l-th iteration, Pop is the population size, and M is the number of objectives.

[0050] S23: Calculate the fitness value of each chromosome, and calculate the individual cloning scale according to the fitness value. The individual cloning scale is proportional to the chromosome fitness value. The higher the fitness value, the more the individual cloning quantity; S24: Classify the population into excellent, medium, and inferior populations, retain the excellent and medium populations, and perform adaptive mutation and crossover recombination on the individuals in the excellent and medium populations; determine the mutation probability of the excellent and medium populations , by comparing the size of the random number rand() to determine the mutation mechanism of the chromosome, with two modes respectively. Mode 1: When , adopt adaptive mutation; Mode 2: , adopt crossover recombination; the mutation point needs to be generated within the coordinate range.

[0051] Specifically, the original non-dominated sorting genetic algorithm II has the disadvantage of falling into local optimum. The clonal selection algorithm simulates the clonal selection mechanism in the immune system. The core is to strengthen local search by expanding the search solution range. Combining the non-dominated sorting genetic algorithm II with the clonal selection algorithm can effectively avoid the algorithm from falling into local optimum. The individual cloning scale can be calculated according to the fitness value of the chromosome, thereby enhancing the search depth and breadth of the improved non-dominated sorting genetic algorithm II.

[0052]

[0053]

[0054]

[0055] where is the artificially set maximum fitness value, is the system comprehensive cost. The smaller the comprehensive cost, the higher the fitness value and the larger the cloning scale; is the individual cloning scale, n j is the population size of the j-th generation, N c is the population cloning scale, is the calculation intermediate value, is the i-th individual in the j-th generation, D i is the individual concentration.

[0056] The cloned individuals have the characteristic of high similarity and are prone to falling into local optima. Therefore, a dynamic adaptive mutation probability is adopted to improve the population diversity. Compared with the traditional fixed mutation probability, the dynamic adaptive mutation algorithm selects the mutation probability according to the chromosome similarity situation and can efficiently approach a better solution. The mutation probability expressions for the excellent and medium populations are as follows:

[0057]

[0058]

[0059] In the formula, is the mutation probability of the excellent population, P med is the mutation probability of the medium population, c t is the number of current iterations, m t is the maximum number of iterations, n is the number of tasks, and len() is the length function.

[0060] By comparing with the random number to determine the mutation mechanism of the chromosome. When , an adaptive dynamic mutation mechanism is adopted, cross recombination is adopted. To avoid generating invalid solutions due to the mutation coordinates exceeding the TTAASBS / RS three-dimensional coordinate range, the mutation points need to be generated within the coordinate range.

[0061] The said adaptive mutation is as follows: Through adaptive mutation, mutation sites are generated, and all gene sites of the chromosome mutate; when the x coordinate of the chromosome changes, the corresponding y coordinate and z coordinate also change; Specifically, through adaptive dynamic mutation, mutation sites are generated, and all three gene sites of the chromosome mutate. When the x coordinate of the chromosome changes, the corresponding y coordinate and z coordinate also change, as Figure 6 shown. First, randomly generate the initial mutation sites 3 and 6; subsequently, the x, y, and z genes of points 3 and 6 mutate along the coordinate axes to ensure that the mutated genes are within the coordinate value range. Finally, new offspring individuals are generated.

[0062] The said cross recombination is as follows: Chromosome - to - chromosome cross recombination is adopted to change the order of gene positions without changing the gene content, and cross recombination is performed once per iteration; the cross operation fully considers the coordinate range of the task and limits the operation within the coordinate range of the cross - layer and cross - roadway shuttle car storage system.

[0063] Specifically, the traditional Non-dominated Sorting Genetic Algorithm II (NSGA-II) generally uses a simulated binary crossover operator to generate offspring. After crossover, the chromosome segments have a high degree of repetition and can lead to the problem of narrowing the search range. In crossover recombination, inter-chromosome crossover recombination is adopted to change the order of gene positions without changing the gene content, and a crossover recombination method is performed once per iteration. The crossover operation fully considers the coordinate range of the tasks and limits the operation within the coordinate range of the cross-level and cross-aisle shuttle car storage system. As Figure 7 shown, first, randomly select 2 parent chromosomes, and randomly select a point position in the gene positions of the chromosomes as a fixed point position to inherit to the offspring chromosomes (such as Figure 7 the dark-colored point positions). Secondly, the remaining unfixed parent genes are inserted into the offspring chromosomes in a specific order. The genes of Parent I are inserted into the chromosome of Offspring II to form crossover recombination between chromosomes.

[0064] By integrating the advantages of the Non-dominated Sorting Genetic Algorithm II and the Clonal Selection Algorithm, and combining improved selection, crossover, and mutation, an improved Non-dominated Sorting Genetic Algorithm II is obtained, which reduces the probability of the local convergence problem that occurs in the original algorithm when solving multi-objective optimization problems. Figure 8 This is the entire process of using the improved Non-dominated Sorting Genetic Algorithm II to solve the multi-objective optimization problem of the configuration optimization of the cross-level and cross-aisle shuttle car storage system.

[0065] S3: Solve the dual-objective mathematical model to obtain the travel time and energy consumption required to complete the tasks; Specifically, taking the cross-level and cross-aisle shuttle car storage system that a certain company actually needs to build in its future factory as an example, the future factory plans to build a small material warehouse to meet the in-line material storage needs. The enterprise has reserved a space with a length of 15 m, a width of 12 m, and a height of 12 m for this warehouse. According to the warehouse location requirements, a total of 30 groups of warehouse layer-column ratio schemes can be generated. Select 8 groups of configurations that can meet the reserved space size requirements of the warehouse from the 30 groups of schemes, and calculate the steel cost of the shelves (see Table 1). The unit price of steel is yuan / m, and the steel cost can be calculated by the following formula, Cost m represents the total steel cost. As can be seen from Table 1, the configurations of groups 6, 15, and 18 have the lowest cost.

[0066]

[0067] In the formula, z is the number of shelves, h is the height of each layer, x is the number of rows of shelves, w is the width of the storage grid, y is the number of columns of shelves, l is the length of the storage grid, and α is the unit price of steel.

[0068] Table 1 is the warehouse configuration plan;

[0069] To determine the optimal configuration of the cross - layer and cross - aisle shuttle car storage system in a three - group configuration, the following will conduct configuration optimization research on the cross - layer and cross - aisle shuttle car storage system under three groups of configurations: 6, 15, and 18. A total of 54 inbound and outbound tasks are generated, with 27 storage tasks and 27 retrieval tasks respectively. Simulation experiments are carried out in MATLAB software. At the same time, the influence of the inconsistent positions of storage and retrieval tasks in the three configurations on the experimental results is considered. 30 repeated experiments are carried out for each configuration, and the mean value is calculated to reduce the contingency of the results.

[0070] To further determine the optimal speed and acceleration of the equipment, research is carried out on the comprehensive system cost of the carrying equipment at different speeds and accelerations. By adjusting the speed and acceleration of the carrying equipment, 48 different experiments are designed. At the same time, combined with the actual needs of the factory, the weight coefficients are set to 0.6 and 0.4 respectively, and the normalized comprehensive system cost is calculated, as shown in Table 2.

[0071] Table 2 is a comparison table of the configurations of the cross - layer and cross - aisle shuttle car storage system;

[0072] S4: Determine the comprehensive system cost based on the travel time and energy consumption required to complete the tasks, and then determine the optimal configuration of the cross - layer and cross - aisle shuttle car storage system. This step includes the following sub - steps: S41: Assign weight coefficients to the travel time and energy consumption required to complete the tasks; Specifically, first, the weights of the travel time and energy consumption are determined by comparing through the analytic hierarchy process. The sum of the weighted weights is 1, that is . The weights of the travel time and energy consumption for a certain company's future factory project are mainly discussed and compared by the company's logistics engineers. Finally, the weight of the travel time is determined to be 0.6, and the weight of the energy consumption is 0.4. The final comprehensive system cost is calculated through linear weighting.

[0073] Normalization can convert all objectives into dimensionless values, enabling them to be compared and optimized on the same scale, thus avoiding a certain objective dominating the optimization process due to a large dimension. Through reasonable normalization and weighting processing, it is possible to better guide the optimization algorithm to find a suitable compromise solution and ensure the best balance among different objectives.

[0074] S42: Input the travel time and energy consumption with assigned weight coefficients into the double - objective mathematical model to obtain the comprehensive system cost; The various objectives restrict each other, so there is no unique optimal solution. In this scenario, many solutions are non-dominated with each other, forming a Pareto solution set. However, there are problems with the Pareto solution set such as being unable to obtain the optimal solution and the objective scale difference of the solutions, which can be handled through weighted summation and normalization of the solutions.

[0075] Specifically: Taking the configuration of 6 groups as an example for the calculation of the system comprehensive cost. The non-dominated solution set obtained by solving 54 tasks using the improved non-dominated sorting genetic algorithm II is used. Each group of configurations is run 30 times, and the Pareto solution set with the lowest average travel time in the first rank of the non-dominated solution set is selected as the candidate solution. The lower the rank, the better the quality of the solution. The maximum and minimum values of the travel time in this group of solution sets are determined. Specifically: The maximum travel time is 575.92, the minimum travel time is 496.91 s, and the average travel time is 562.9 s. The normalization formula is used:

[0076] The maximum energy consumption required to solve 54 tasks is 3.67 kWh, the minimum is 2.52 kWh, and the average energy consumption is 2.84 kWh. The normalization formula is used:

[0077] The weighted summation is used to obtain the system comprehensive cost of the configuration of 6 groups, and finally it is: 。

[0078] The remaining system comprehensive costs can be obtained by the same calculation.

[0079] S43: Determine the optimal configuration of the cross-layer and cross-aisle shuttle car storage system according to the system comprehensive cost. The optimal configuration is the optimal layer-column ratio configuration and the speed and acceleration of the equipment.

[0080] Figure 9 It is a comparison chart of the system comprehensive costs of the cross-layer and cross-aisle shuttle car storage system under different algorithms and configurations. Figure 9 In (a), (b), and (c) respectively represent the comprehensive costs solved by the improved non-dominated sorting genetic algorithm II, the basic non-dominated sorting genetic algorithm II, and the genetic algorithm under three configurations, proving that the improved non-dominated sorting genetic algorithm II has advantages in solving the performance of the cross-layer and cross-aisle shuttle car storage system compared with the genetic algorithm and the basic non-dominated sorting genetic algorithm II.

[0081] Figure 9 In (d) gives the comparison of the comprehensive costs when solving three configurations using the improved non-dominated sorting genetic algorithm II. From Figure 9From (d) in it, among the 16 groups of comparative experiments, the experimental results show that the comprehensive cost of system configuration 6 is better than that of configurations 15 and 18. It is only slightly higher than that of configuration 15 in the 10th group of experiments, which indicates that configuration 6 can obtain better system performance compared with configurations 15 and 18. Combining configuration 6 numbers 3 and 5, by increasing the speed and acceleration of the shuttle car and the transfer vehicle respectively, the system comprehensive cost is reduced by 7.52% and 6.70% respectively. While simultaneously increasing the speed and acceleration of the elevator, as in configuration 6 number 7 in Table 2, the system comprehensive cost can be reduced by 13.73%. Therefore, compared with the impact of the change in the speed and acceleration of the elevator on the system comprehensive cost, the impact of the change in the speed and acceleration of the shuttle car and the transfer vehicle on the system comprehensive cost is smaller.

[0082] Through experimental comparison, the final configuration of the cross-layer and cross-aisle shuttle car storage system is finally determined to be 9 floors, 28 rows and 20 columns. The speed and acceleration of the shuttle car are 2.5 m / s and 1.5 m / s respectively 2 , and the speed and acceleration of the transfer vehicle are 2.5 m / s and 1.5 m / s 2 , and the speed and acceleration of the elevator are 4.0 m / s and 2.0 m / s respectively 2 .

[0083] This application proposes an improved non-dominated sorting genetic algorithm II integrated with a clonal selection algorithm to solve the configuration optimization problem of the cross-layer and cross-aisle shuttle car storage system, analyzes the performance of warehouses with different layer-column ratios, obtains the optimal warehouse configuration plan, and provides theoretical guidance for the configuration selection in the initial stage of warehouse construction. Further, the influence of the power parameters of the transport equipment on the performance of the cross-layer and cross-aisle shuttle car storage system is discussed, and the key factors affecting the system configuration are obtained.

[0084] After considering the specification and the content disclosed herein in practice, those skilled in the art will easily think of other implementation schemes of this application. This application aims to cover any variations, uses or adaptive changes of this application. These variations, uses or adaptive changes follow the general principles of this application and include the common general knowledge or conventional technical means in the technical field not disclosed in this application. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of this application are pointed out by the claims.

[0085] It should be understood that this application is not limited to the precise structure already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is only limited by the appended claims.

Claims

1. A method for optimizing the configuration of a cross-layer and cross-aisle shuttle storage system, characterized in that: include: S1: According to the operation process of the cross-layer and cross-aisle shuttle storage system, a dual-objective mathematical model with travel time and energy consumption as objectives is established, wherein the dual-objective mathematical model includes model assumptions, travel time objective function, energy consumption objective function, and model constraints; S2: According to the dual-objective mathematical model, the local search is strengthened by expanding the search solution range, and the non-dominated sorting genetic algorithm II is integrated with the clonal selection algorithm to obtain an improved non-dominated sorting genetic algorithm II; S3: using the improved non-dominated sorting genetic algorithm II to solve the dual-objective mathematical model to obtain the travel time and energy consumption required to complete the task; S4: Determine the comprehensive cost of the system based on the travel time and energy consumption required to complete the task, and then determine the optimal configuration of the cross-level and cross-aisle shuttle storage system.

2. The method according to claim 1, characterized in that The model assumptions include: Assumption 1: The shuttle can only carry one container at a time; Assumption 2: The time for the shuttle to load and unload the bin is equal; Assumption 3: A random storage strategy is adopted, and the location of the carrier is the location of the final processing task; Assumption 4: Each carrier in the system operates independently; Assumption 5: During the operation of the shuttles, the first-in-last-out strategy is adopted, that is, if the first shuttle enters the lane before the second shuttle, the first shuttle must leave the lane after the second shuttle completes its task; Assumption 6: The system performs compound command operations, however, when the task involves only retrieval / storage tasks, it performs single command operations; Assumption 7: The transport equipment is deployed in the shortest possible time.

3. The method according to claim 1, characterized in that The travel time objective function is: Depending on whether the vehicle can run at the maximum speed, the system's travel time is divided into Mode 1: the vehicle does not reach the maximum speed; Mode 2: the vehicle reaches the maximum speed.

4. The method according to claim 1, characterized in that: The energy consumption objective function includes the energy consumption of the shuttle vehicle, the energy consumption of the transfer vehicle, and the energy consumption of the hoist; wherein: The energy consumption of the shuttle W s The calculation formula is as follows: ； In the formula, is the energy consumed when the shuttle accelerates. is the energy consumption required for the shuttle to decelerate. is the energy consumed by the shuttle at a constant speed; Energy consumption of the transfer vehicle The calculation formula is as follows: ； In the formula, Indicates that the transfer vehicle accelerates energy consumption, Indicates the energy consumption of the transfer vehicle when it slows down. It indicates the uniform speed energy consumption of the transfer vehicle; The energy consumption of the hoist The calculation formula is as follows: ； In the formula, Indicates the energy consumption of the elevator acceleration. Indicates the energy consumption of elevator deceleration. Indicates the uniform speed energy consumption of the elevator.

5. The method according to claim 1, characterized in that: The model constraints include: Constraint 1: Only one carrier can carry a container at each execution stage of the current task; Constraint 2: Ensure that the start time of the task is non-negative; Constraint 3: Ensure that the current task starts later than the previous task under the same carrier; Constraint 4: Ensure that the completion time of the i-th task is earlier than the start time of the i+1-th task; Constraint 5: Ensure that task K i There is at most one subsequent task under the same device; Constraint 6: Ensure that task K i There is at most one predecessor task under the same device; Constraint 7: The total travel time of the tasks is constrained to be no less than the travel time of any single task.

6. The method according to claim 1, characterized in that The improved non-dominated sorting genetic algorithm II is obtained by integrating the non-dominated sorting genetic algorithm with the clonal selection algorithm, including: S21: A three-layer mixed integer coding method is used to represent chromosomes, where each chromosome contains three-dimensional coordinate information to express the coordinates of each batch of storage and retrieval tasks; S22: Tournament selection is used to compare the non-dominated order of chromosomes in pairs, and the chromosomes with lower order are preferentially retained for inheritance to the next generation; if the non-dominated order of two chromosomes is the same, selection is performed according to the niche technique, and individuals with larger crowding distance are preferentially selected and retained for the next generation; S23: Calculate the fitness value of each chromosome, and calculate the individual clone size according to the fitness value. The individual clone size is proportional to the chromosome fitness value. The higher the fitness value, the more individual clones there are. S24: Classify the population into superior, medium, and inferior populations, retain the superior and medium populations, perform adaptive mutation and crossover recombination on the superior and medium population individuals; determine the mutation probability of the superior and medium populations , by comparison The size of the random number rand() determines the chromosome mutation mechanism. There are two modes: Mode 1: When , using adaptive mutation; Mode 2: , using crossover recombination; mutation points need to be generated within the coordinate range.

7. The method according to claim 6, characterized in that The calculation of chromosome fitness value and individual clone size is as follows: ; ; ; In the formula is the maximum fitness value set artificially, is the comprehensive cost of the system. The smaller the comprehensive cost, the higher the fitness value. The larger it is, the larger the clone size will be; is the size of individual clones, n j is the population size of the jth generation, N c is the population clone size, To calculate the median value, For individual i in generation j, D i is the individual concentration.

8. The method according to claim 6, characterized in that The calculation formula for the mutation probability of superior and medium populations is as follows: ； ； ； In the formula, is the mutation probability of the superior population, P med is the medium population mutation probability, is the number of current iterations, is the maximum number of iterations, n is the number of tasks, and len() is the length function.

9. The method according to claim 6, characterized in that The adaptive mutation is as follows: through adaptive mutation, mutation sites are generated, and all gene sites of the chromosome are mutated; when the x coordinate of the chromosome changes, the corresponding y coordinate and z coordinate also change; The crossover recombination is: using crossover recombination between chromosomes to change the order of gene positions without changing the gene content, and performing a crossover recombination method once per iteration; the crossover operation fully considers the coordinate range of the task and limits the operation to the coordinate range of the cross-layer and cross-aisle shuttle storage system.

10. The method according to claim 1, characterized in that According to the travel time and energy consumption required to complete the task, the comprehensive cost of the system is determined, and then the optimal configuration of the cross-layer and cross-aisle shuttle storage system is determined, including: S41: assigning a weight coefficient to the travel time and energy consumption required to complete the task; S42: inputting the travel time and energy consumption after assigning weight coefficients into the dual-objective mathematical model to obtain the system comprehensive cost; S43: Determine the optimal configuration of the cross-layer and cross-aisle shuttle storage system based on the comprehensive cost of the system, wherein the optimal configuration is the optimal layer ratio configuration and the speed and acceleration of the equipment.

Images