Automatic container terminal stockpiling strategy dynamic energy efficiency simulation optimization method and system

Abstract

The invention relates to an automatic container terminal stockpiling strategy dynamic energy efficiency simulation optimization method and system. Obtaining the layout of the target wharf, the operation process of the multi-stage operation and a stockpiling strategy; according to the layout, the operation process, the stockpiling strategies and preset simulation parameters, energy consumption and operation time data under different stockpiling strategies are obtained by using a pre-constructed multi-agent-based multi-stage operation simulation model of the automatic container terminal, and dynamic energy efficiency indexes of the automatic container terminal under the different stockpiling strategies are calculated; according to the dynamic energy efficiency indexes, the energy efficiency imbalance of the multi-stage operation of the automatic container wharf under different stockpiling strategies is calculated; and comparing the dynamic energy efficiency index with the energy efficiency imbalance under each stockpiling strategy, and determining a final stockpiling strategy according to a comparison result. Compared with the prior art, the method has the advantages that the method is suitable for automatic container wharfs with different layouts, the operation efficiency is improved, and energy efficiency optimization of the wharf under the uncertainty of the operation sequence is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of container terminals, and in particular to a method and system for dynamic energy efficiency simulation optimization of storage strategies for automated container terminals. Background Art

[0002] Automated Container Terminal (ACT) is an important part of the modern logistics system, and the energy efficiency of the terminal has become the key to the competitiveness of the port.

[0003] The process of transporting containers from outside the yard to inside the yard or from inside the yard to outside the yard is usually called a multi-stage operation process. There are three main types of equipment involved in multi-stage operations: Automated Quay Crane (AQC), Automated Guided Vehicle (AGV) and Automated Rail-Mounted Gantry crane (ARMG). In this long-term operation process, the terminal faces the uncertainty of the container operation sequence. If the terminal operator does not respond in time, it will lead to a decrease in terminal operation efficiency and is more likely to cause the terminal to be paralyzed. The replacement of storage strategies can effectively deal with the uncertainty of container operation sequences. In other words, for terminals with different layouts, different storage strategies have different corresponding terminal energy efficiencies. By calculating and comparing the energy efficiency of the terminal, a more appropriate storage strategy can be selected.

[0004] At present, there are still some deficiencies in the research on the energy efficiency of multi-level operations of automated container terminals: (1) The existing energy efficiency calculation methods for automated container terminals are mainly applicable to the effect of replacing equipment under the same layout. There are few energy efficiency calculation methods for similar equipment under different layouts, which makes it difficult to compare the differences in terminal energy efficiency between similar equipment under different layouts; (2) The evaluation criteria for layout differences are relatively simple and cannot comprehensively reflect the energy efficiency of terminals under different layouts; (3) Mathematical modeling methods have difficulties in calculating the energy efficiency of automated container terminals. It is difficult to accurately calculate the energy consumption and operation time that conform to the actual situation at the same time, and it is impossible to refine the calculation results; (4) The randomness of ship arrivals and the variability of random tasks in the operation process of automated container terminals also affect the calculation of terminal energy efficiency. Therefore, for automated container terminals, how to select a more appropriate storage strategy under different terminal layouts to achieve energy efficiency optimization of terminals under uncertainty of operation sequence. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a dynamic energy efficiency simulation optimization method and system for the stacking strategy of an automated container terminal, which can select a more suitable stacking strategy while considering different terminal layouts, and realize the energy efficiency optimization of the terminal under the uncertainty of the operation sequence.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] According to a first aspect of the present invention, there is provided a method for dynamic energy efficiency simulation optimization of stacking strategies of automated container terminals, comprising the following steps: obtaining the layout of a target terminal, the operation process of multi-level operations and the stacking strategy; obtaining energy consumption and operation time data under different stacking strategies using a pre-built multi-agent-based automated container terminal multi-level operation simulation model according to the layout, the operation process, the stacking strategy and preset simulation parameters; calculating dynamic energy efficiency indicators of automated container terminals under different stacking strategies according to the energy consumption and operation time data; calculating the energy efficiency imbalance of multi-level operations of automated container terminals under different stacking strategies according to the dynamic energy efficiency indicators; comparing the dynamic energy efficiency indicators and energy efficiency imbalance under each stacking strategy, and determining a final stacking strategy according to the comparison result.

[0008] As an optimal technical solution, the process of constructing the multi-level operation simulation model of the multi-agent-based automated container terminal specifically includes: determining the time parameters of different equipment in the multi-level operation according to the operation process; determining the agent set and agent parameters; determining the multi-level operation constraints; determining the operation sequence uncertainty of the multi-level operation of the target terminal; based on the layout, the operation process, the stacking strategy, the time parameters, the agent set and agent parameters, the multi-level operation constraints and the operation sequence uncertainty, constructing the multi-level operation simulation model of the multi-agent-based automated container terminal under different stacking strategies.

[0009] As a preferred technical solution, the multi-level operation constraints include constraints on the operation process, constraints on continuous operations, and constraints on intelligent agents.

[0010] As a preferred technical solution, the dynamic energy efficiency indicators specifically include throughput, equipment energy efficiency, energy efficiency generated by container stays, and container area bay density.

[0011] As a preferred technical solution, the throughput is expressed as:

[0012]

[0013] In the formula, TEU IO AQC.ct represents the throughput of the target terminal during the time (t1, t2). iIndicates the number of containers loaded and unloaded by all automated quay cranes, ET.ct i represents the number of containers transported by all external container trucks, and T represents the time set when the automated quay crane agent and the external container truck agent receive container task i.

[0014] As a preferred technical solution, the equipment energy efficiency includes: when any automatic guided vehicle completes a preset number of container tasks within a unit time, the actual time, average time, maximum average time and minimum average time for the automatic guided vehicle to complete a single container task.

[0015] As a preferred technical solution, the energy efficiency generated by the container stay includes: for any ordinary container in any bay, the average time for the ordinary container to be taken out of or stored in the corresponding bay; for any dangerous goods container in any bay, the average time for the dangerous goods container to be taken out of or stored in the corresponding bay; and for any refrigerated goods container in any bay, the average time for the refrigerated goods container to be taken out of or stored in the corresponding bay.

[0016] As a preferred technical solution, the bay density in the container area includes: the average density of containers stored in the bays and the distribution of the number of containers stored in all bays in the yard.

[0017] As a preferred technical solution, the calculation expression of the energy efficiency imbalance is:

[0018]

[0019] and

[0020]

[0021] Where Z represents the Singh index, n i Indicates efficiency x i The number of devices corresponding to represents the Singh imbalance index, N represents the total number of devices, and L i Indicates that the efficiency is less than x i The proportion of the cumulative efficiency of the equipment to the total efficiency, p i Indicates that the efficiency is less than x i The cumulative number of devices accounts for the proportion of the total number of devices, and e represents the total number of efficiency values.

[0022] According to a second aspect of the present invention, there is provided a dynamic energy efficiency simulation optimization system for stacking strategies of automated container terminals, the system being used to implement the method described, comprising: a basic data acquisition module for acquiring the layout of a target terminal, the operation process of multi-level operations and the stacking strategy; an energy consumption and operation time data acquisition module for acquiring energy consumption and operation time data under different stacking strategies using a pre-built multi-agent-based automated container terminal multi-level operation simulation model according to the layout, the operation process, the stacking strategy and preset simulation parameters; a dynamic energy efficiency index calculation module for calculating the dynamic energy efficiency index of the automated container terminal under different stacking strategies according to the energy consumption and operation time data; an energy efficiency imbalance calculation module for calculating the energy efficiency imbalance of the automated container terminal multi-level operations under different stacking strategies according to the dynamic energy efficiency index; and a stacking strategy determination module for comparing the dynamic energy efficiency index and energy efficiency imbalance under each stacking strategy, and determining the final stacking strategy according to the comparison result.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. The method provided by the present invention considers the terminal layout, multi-level operation process, stacking strategy, operation sequence uncertainty and multiple parameters to construct a multi-agent-based automated container terminal multi-level operation simulation model. The multi-level operation process of the automated container terminal is described by constructing the simulation model. The model includes sub-tasks of multi-level operations, covers the operation processes of AQC, AGV and ARMG, can be applied to automated container terminals with different layouts, can accurately simulate and optimize the multi-level operation process of automated container terminals, improve operation efficiency and effect, can comprehensively reflect the energy efficiency of terminals under different layouts, refine the energy efficiency calculation results, and thus help to select a more appropriate stacking strategy, and realize the energy efficiency optimization of the terminal under the uncertainty of the operation sequence;

[0025] 2. The dynamic energy efficiency index designed by the present invention specifically includes throughput, operating equipment energy efficiency, energy efficiency generated by container stays, and container area bay density. On this basis, the Singh index is further used to calculate the imbalance of terminal equipment scheduling, which can evaluate the energy efficiency performance of different automated container terminal layouts under uncertainty in real time, and provide a reference for terminal operators to more reasonably deal with uncertainty and allocate resources;

[0026] 3. The present invention compares the energy efficiency of automated container terminals with different stacking strategies under uncertainty, and can select a suitable stacking strategy by calculating and comparing dynamic energy efficiency indicators, which helps terminal operators to reduce the negative impact of uncertainty on automated container terminal operations and improve the terminal's response capability and resilience to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of a method flow chart provided by an embodiment of the present invention;

[0028] Figure 2 Schematic diagram of three-dimensional scenes of four dock layouts provided in embodiments of the present invention;

[0029] Figure 3 The operation process and task parameters provided by the embodiment of the present invention;

[0030] Figure 4 It is a schematic diagram of the operation logic of a multi-agent-based multi-level operation simulation model of an automated container terminal provided by an embodiment of the present invention;

[0031] Figure 5 is the simulation calculation result of the annual throughput of the container terminal in the embodiment of the present invention;

[0032] Figure 6 is the simulation calculation result of the AGV operation time distribution in the embodiment of the present invention;

[0033] Figure 7 is the simulation calculation result of the energy consumption distribution of the AGV operation in the embodiment of the present invention;

[0034] Figure 8 It is the simulation calculation result of the maximum value of AGV operation time in the embodiment of the present invention;

[0035] Fig. 9 It is the simulation calculation result of the maximum energy consumption of AGV in the embodiment of the present invention;

[0036] Fig.10 is the simulation calculation result of the average time that the container stays in the bay in the embodiment of the present invention;

[0037] Fig.11 is the simulation calculation result of the average energy consumption of the container in the bay according to the embodiment of the present invention;

[0038] Fig.12 is the bay density distribution result of the vertical layout using the random stockpiling strategy in the embodiment of the present invention;

[0039] Fig.13 It is the bay density distribution result of the vertical layout using the partitioned stockpiling strategy in the embodiment of the present invention;

[0040] Fig.14 It is the bay density distribution result of the parallel cantilever-free layout adopting the random stacking strategy in the embodiment of the present invention;

[0041] Fig.15 It is the bay density distribution result of the parallel cantilever-free layout adopting the partitioned stacking strategy in the embodiment of the present invention;

[0042] Fig.16 is the bay density distribution result of the U-shaped layout using the random stockpiling strategy in the embodiment of the present invention;

[0043] Fig.17 is the bay density distribution result of the U-shaped layout adopting the partitioned stockpiling strategy in the embodiment of the present invention;

[0044] Fig.18 is the bay density distribution result of the parallel double cantilever layout adopting the random stacking strategy in the embodiment of the present invention;

[0045] Fig.19 It is the bay density distribution result of the parallel double cantilever layout adopting the partitioned stacking strategy in the embodiment of the present invention;

[0046] Fig. 20 is the Singh index of the average time for the AGV to complete a container task in the embodiment of the present invention;

[0047] Fig.21 is the Singh index of the average operation time of the container in the bay in the embodiment of the present invention;

[0048] Fig. 22 It is the Singh index of the bay density in the box area in the embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes, but the protection scope of the present invention is not limited to the following embodiments.

[0050] Example

[0051] like Figure 1 As shown, this embodiment provides a dynamic energy efficiency simulation optimization method for stacking strategies of automated container terminals under uncertain operation sequences, aiming to optimize the energy efficiency of automated container terminals when facing uncertain operation sequences. The optimization method is mainly through changing different stacking strategies. The energy efficiency of the same terminal under different stacking strategies is different. In order to reflect the dynamic energy efficiency of the multi-level operation process of the terminal, the dynamic energy efficiency indicators of the automated container terminal that need to be calculated are mainly considered from five aspects: throughput, equipment energy efficiency, energy efficiency generated by container stays, container area bay density and imbalance. This method mainly constructs a multi-level operation simulation model of an automated container terminal based on multiple agents, calculates the energy consumption and operation time data under different stacking strategies, and then calculates the dynamic energy efficiency indicators and energy efficiency imbalance of the automated container terminal under different stacking strategies. The specific steps are as follows:

[0052] Step 1: Determine the automated container terminal layout (i.e., ACT layout), the operation process of multi-level operations, and more than two storage strategies. Among them, the ACT layout is related to the equipment configuration. There are three main types of equipment involved in multi-level operations: AQC, AGV, and ARMG. The multi-level operation process refers to the order of loading and unloading containers. The terminal's storage strategy refers to a series of plans and methods used by the terminal to manage and arrange the location and method of container storage.

[0053] Step 2: Determine the time parameters of the multi-level operation subtasks of the automated container terminal. The subtasks are decomposed according to different equipment groups. They can be divided into AQC subtasks, AGV subtasks and ARMG subtasks. The time consumed by each subtask is recorded as the time parameter. The specific tasks and their time parameters are shown in Table 1.

[0054] Table 1 Time parameters of multi-level operation subtasks in automated container terminals

[0055]

[0056]

[0057] Step 3: Determine the set of agents and their parameters in the multi-agent based automated container terminal multi-level operation simulation model.

[0058] Agents are the basic units that make up a multi-agent system. Each agent has certain attributes. Several agents of similar types form an agent set, and several agent sets form a multi-agent system. If an agent has a certain attribute and the attribute has a specific value, it can be expressed as agent.x=1, which means that the value of the attribute x of the agent agent is 1.

[0059] (1) Agent Collection

[0060] The specific contents of each agent set are shown in Table 2.

[0061] Table 2 Specific contents of the agent set

[0062]

[0063]

[0064] (2) Agent parameters

[0065] The specific contents of each agent parameter are shown in Table 3.

[0066] Table 3 Specific content of agent parameters

[0067]

[0068] (3) Other collections

[0069] The specific contents of the remaining sets are shown in Table 4.

[0070] Table 4 Specific contents of the remaining sets

[0071] L A collection of container tasks; HP The set of loading and unloading points (1,2,3...hp)∈HP; SL The positions (1, 2, 3...sl)∈SL on a container ship for storing containers.

[0072] (4) Other parameters

[0073] The details of the remaining parameters are shown in Table 5.

[0074] Table 5 Details of other parameters

[0075]

[0076] (5) Decision variables

[0077] The specific content of the decision variables is shown in Table 6.

[0078] Table 6 Detailed content of decision variables

[0079]

[0080]

[0081] Step 4: Determine multi-level job constraints.

[0082] (1) Constraints of the work process

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091] Constraint (1) indicates that in the unloading process, AQC starts to unload container l after AGV arrives at the loading and unloading point. Constraint (2) indicates that in the loading process, AQC starts to load container l after AGV arrives at the loading and unloading point. Constraint (3) indicates that in the unloading process, ARMG starts to unload container l after AGV arrives at the loading and unloading point. Constraint (4) indicates that in the loading process, ARMG starts to load container l after AGV arrives at the loading and unloading point. Constraint (5) indicates that in the unloading process, AGV transports container l from the loading and unloading point at AQC to the loading and unloading point at ARMG. Constraint (6) indicates that in the loading process, AGV transports container l from the loading and unloading point at ARMG to the loading and unloading point at AQC. Constraint (7) indicates that the equipment in the multi-level operation chain in the unloading and loading process completes the corresponding container subtasks one by one, where k represents the kth step in the AQC, AGV, and ARMG subtask processes, and k' represents the next step after the kth step. In the loading process, the AQC subtask process is α1, α2, α3, and α4, and the operation is performed in the sequence of k = 1-2-3-4. The AGV subtask process is β1, β2, β3, β5, β7, β8, β9, β10, β11, β12, β13, β14, β15, β16, β17, β18, β19, β20, β21, β22, β23, β24, β25, β26, β27, β28, β29, β30, β31, β32, β33, β34, β35, β36, β37, β38, β39, β40, β41, β42, β43, β44, β45, β46, β47, β48, β49, β49, β50, β51 10 The tasks are performed in the order of k=1-2-3-5-7-8-10. The ARMG subtask process is γ1, γ2, γ3, γ4, and the operations are performed in the order of k=1-2-3-4. In the unloading process, the AQC subtask process is α1, α2, α3, α5, and the operations are performed in the order of k=1-2-3-5. The AGV subtask process is β1, β2, β3, β4, β6, β7, β9, β1 11 , and the operations are performed in the order of k = 1-2-3-4-6-7-9-11. The ARMG subtask process is γ1, γ2, γ3, γ5, and the operations are performed in the order of k = 1-2-3-5. Constraint (8) indicates that any AQC, AGV, and ARMG can only carry one container l at a time.

[0092] (2) Constraints of continuous operation

[0093]

[0094]

[0095] Constraint (9) indicates that for a continuous unloading process, after the AGV transports container l to the loading and unloading point at ARMG, it returns to AQC to transport container l+1. Constraint (10) indicates that for a continuous loading process, after the AGV transports container l to the loading and unloading point at AQC, it returns to ARMG to transport container l+1.

[0096] (3) Constraints on the Agent

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] Constraint (11) indicates that two AGVs cannot appear at the same loading and unloading point at the same time. Constraint (12) indicates that AGV and external container trucks cannot appear at the same loading and unloading point at the same time. Constraint (13) indicates that two external container trucks cannot appear at the same loading and unloading point at the same time. Constraint (14) indicates the relationship between the time when AQC completes the container task and the time when AQC puts the container on the AGV in the unloading process. Constraint (15) indicates the relationship between the time when AQC completes the container task and the time when AQC puts the container on the AGV in the loading process. Constraint (16) indicates the relationship between the time when AGV completes the container task and the time when ARMG takes the AGV away from the container in the unloading process. Constraint (17) indicates the relationship between the time when ARMG completes the container task and the time when ARMG puts the container on the AGV in the loading process. Constraint (18) indicates that for agv i, in the continuous unloading process, the AGV can only proceed to the next task after the unloading equipment takes the container from the AGV. Constraint (19) means that for the AGV i , in the continuous loading process, the AGV can only proceed to the next task after the unloading equipment takes the container from the AGV. Constraint (20) means that for armg i , in the continuous loading and unloading process, the ARMG will load the container onto the AGV only after the ARMG takes the container off the AGV. Constraint (21) indicates that for aqc i In the continuous loading and unloading process, the ARMG will load the container onto the AQC only after the AQC takes the container from the AGV. Constraint (22) indicates that only one box is loaded or unloaded at the container position with loading and unloading tasks on the ship. Constraint (23) indicates that two ARMGs cannot appear at the same loading and unloading point at the same time. Constraint (24) indicates that the two ARMGs cannot pass through each other to complete the loading and unloading of containers. Constraint (25) indicates that only one box is loaded or unloaded at the container position with loading and unloading tasks in the yard. Constraint (26) indicates that one box is stored in one container position in the yard. Constraint (27) indicates that the space below any container in the yard cannot be empty.

[0115] Step 5: Determine the uncertainty of the operation sequence of multi-level operations in multi-automated container terminals, and describe the uncertainty by assigning different attributes and generation probabilities of container agents in the multi-agent model. This uncertainty will cause changes in the actual operation sequence, and the terminal's ability to cope with the same uncertainty will also be different under different storage strategies.

[0116] Step 6: Select simulation software and build a multi-agent-based automated container terminal simulation model under different stacking strategies according to steps 1 to 5. In the simulation model, the number and running speed of AQC, AGV, and ARMG equipment should be set according to actual conditions, and the number and capacity of the yard should also be set.

[0117] Step 7: Run the model in step 6 to obtain the energy consumption and operation time data under different storage strategies. The data includes terminal throughput, equipment operating time and energy consumption, and yard storage quantity.

[0118] Step 8: Calculate the dynamic energy efficiency index of the automated container terminal under different stacking strategies based on the energy consumption and operation time data obtained in Step 7.

[0119] In order to more comprehensively analyze the energy efficiency characteristics of different automated container terminals in dealing with the uncertainty of container operation sequences during multi-level operations, the dynamic energy efficiency index is designed as follows:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129] ρ={ρ bay1 ,ρ bay2 ... bayn ... baymn} (37)

[0130] The throughput in the dynamic energy efficiency index is calculated by formula (28). Specifically, formula (28) means that within the time (t1, t2), the throughput of the terminal is recorded as TEU IO , which is equal to all containers loaded and unloaded by AQC and all containers transported by external container trucks. i Indicates the number of containers loaded and unloaded by all AQCs, ET.ct i Represents the number of containers transported by all external container trucks, T represents the time set when the agent receives the container task, and the agents here are AQC agents and external container truck agents.

[0131] The energy efficiency of the operating equipment in the dynamic energy efficiency index is calculated by formula (29) to formula (32). Where: Formula (29) means that for any AGV, if the AGV completes j container tasks in a unit time, the time it takes for the AGV to complete a container task is recorded as T agv , the maximum time for AGV to complete a container task is recorded as The minimum time for AGV to complete a container task is recorded as Formula (30) means that for any AGV, if the AGV completes j container tasks in a unit time, the average time for the AGV to complete a container task is recorded as Formula (31) means that for all AGVs, the maximum average time for AGV to complete a container task is recorded as Formula (32) means that the minimum average time for AGV to complete a container task is recorded as

[0132] The energy efficiency generated by container stay in the dynamic energy efficiency index is calculated by formula (33) to formula (35). Where: Formula (33) means that for any container in any bay, the average time for the container to be taken out / stored from the bay is recorded as N(bay ij ) represents the number of containers stored in the common container area bay ij. Formula (34) means that for any dangerous goods container in any bay, the average time for the container to be taken out / stored in the bay is recorded as N(bay ij ) represents the number of containers stored in the dangerous goods container area bay ij. Formula (35) means that for any refrigerated container in any bay, the average time for the container to be taken out / stored in the bay is recorded as N(bay ij ) represents the number of containers stored in the refrigerated container area at position ij.

[0133] The density of the container area in the dynamic energy efficiency index is calculated by formula (36) to formula (37). Where: Formula (36) means the average density of containers stored in the bay, denoted as d bay Formula (37) means the distribution of the number of containers stored in all bays in the yard.

[0134] Step 9: Based on step 8, calculate the imbalance of energy efficiency of multi-level operations in automated container terminals under different stacking strategies.

[0135] Singh index is used as an evaluation index for the imbalance of energy efficiency of multi-level operations in automated container terminals:

[0136] Assume that X represents a certain operating equipment resource, {0≤x1 <x2<…<x i <… <x e} is a set of efficiency values ​​of resource X in ascending order. Each element in the set is different and non-negative. e means that there are e elements in the efficiency set, that is, there are e efficiency values. The number of devices corresponding to each efficiency value is (n1, n2, ..., n i ,...,n e ), also contains e elements. The total efficiency value distribution is defined as {(x i ,n i ), i = 1, 2, ..., e}, indicating that the efficiency of the equipment resource X is x i The equipment has n i indivual.

[0137]

[0138]

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] Formula (38) represents the total number of equipment resources. Formula (39) represents the efficiency less than x i The cumulative number of devices. Formula (40) indicates that the efficiency is less than x i The ratio of the cumulative number of devices to the total number of devices. Formula (41) indicates that the efficiency is less than x i The cumulative efficiency of the equipment. Formula (42) represents the total efficiency of the equipment. Formula (43) represents the proportion of the cumulative efficiency of the equipment with efficiency less than xi to the total efficiency. Formula (44) represents the total splicing efficiency.

[0145] Since each efficiency x i The equipment resources can be divided into low-efficiency equipment groups {(x1,n1),…,(x i ,n i )} and high-efficiency equipment group Therefore use and represents the average efficiency of the low-efficiency equipment group and the high-efficiency equipment group, see equation (45) and equation (46):

[0146]

[0147]

[0148] From this, we can get the Singh point imbalance index, which measures the relative deviation of different efficiency equipment groups, as shown in formula (47):

[0149]

[0150] in It is called the equilibrium index, which represents the ratio of the inefficient group to the efficient group. Therefore When the Singh imbalance index is equal to 0, This indicates that inefficient devices and efficient devices have the same average efficiency, indicating that the system task allocation is completely balanced; when the Singh imbalance index is equal to 1, This means that all tasks are completed by efficient devices, indicating that the system task allocation is completely unbalanced. Usually, the Singh imbalance index is between 0 and 1. The closer the Singh imbalance index is to 1, the higher the degree of imbalance.

[0151] Another expression of the Singh imbalance index can be obtained through inverse operation, see formula (48):

[0152]

[0153] The Singh index is the weighted sum of all Singh point imbalance indices, with the weights being all points with efficiencies greater than x. i The proportion of devices in all devices, that is, Therefore, the final expression of the Singh index is shown in formula (49):

[0154]

[0155] The Singh index has the same properties as the Singh point imbalance index, Z∈[0,1]. When Z is 0, it indicates that the system task allocation is completely balanced. When Z is 1, it indicates that the system task allocation is completely unbalanced. The closer the Singh index is to 1, the higher the degree of imbalance.

[0156] Step 10: Select appropriate stockpiling strategies based on the calculated energy efficiency indicators and imbalances to optimize the energy efficiency of the terminal under uncertainty.

[0157] Next, this embodiment establishes a three-dimensional model in Flexsim, and adopts different stockpiling strategies for four common dock layouts to perform multi-level operation dynamic energy efficiency calculation optimization.

[0158] The three-dimensional scenes of the four dock layouts are as follows: Figure 2 As shown. The gray area represents the container area for ordinary containers, the blue area represents the container area for refrigerated containers, and the red area represents the container area for dangerous goods containers. The black rectangle on the edge of the yard represents the AGV handover area. The conveyor belt above is used to simulate the appearance and disappearance of container tasks, and the black area below is used to simulate the appearance area of ​​external container trucks, where the role of external container trucks is to transport containers to and from the automated terminal. The yellow I-shaped object is used to simulate the ARMG in the container area, and the blue line represents the lane. Layout (a) represents a vertical layout with no cantilever ARMG, layout (b) represents a parallel layout with no cantilever ARMG, and layout (c) represents a U-shaped layout with double cantilever ARMG. The yellow lane in layout (c) represents the external container truck lane, and layout (d) represents a parallel layout with double cantilever ARMG.

[0159] Two different stockpiling strategies were used for each layout. This approach allowed us to compare the terminal’s ability to cope with uncertainty:

[0160] (1) Consider the random stacking strategy of different container types: First, determine the cargo attributes of the container. If the cargo in the container is ordinary cargo, dangerous goods cargo, and refrigerated cargo, respectively, the container is transported to the corresponding ordinary cargo area, dangerous goods area, and refrigerated cargo area. When the container is transported to the yard area corresponding to its cargo attributes, the system will screen the density of the container areas in the yard area and select the container area with the smallest density to store the container, that is, the container area with the least number of containers stored in all slots of each container area. When the system selects the corresponding container area, it randomly selects the slot with the smallest density in the container area to store the container, that is, the number of containers stored in all slots of the selected slot is the least. When the container is placed in the corresponding slot, the task of the container ends, and the next container repeats this operation.

[0161] (2) Considering the zoning storage strategy for different container types: First, divide each container area into areas. For the layout of vertically placed containers (Layout (a), Layout (c)), each container area is divided into an import sub-box area and an export sub-box area. The import sub-box area is close to the shore side, and the export sub-box area is close to the sea side. The import and export sub-box areas are used to store import and export containers. For the layout of horizontally placed containers (Layout (b), Layout (d)), each container area is also divided into an import sub-box area and an export sub-box area. The import and export sub-box areas are close to the left and right sides, respectively, and the import and export sub-box areas are used to store import and export containers. Then, the cargo attributes of the container are determined. If the cargo in the container is ordinary cargo, dangerous goods cargo, and refrigerated cargo, the container is transported to the corresponding ordinary cargo area, dangerous goods area, and refrigerated cargo area. When the container is transported to the yard area corresponding to its cargo attributes, the system will screen the density of the container areas in the yard area and select the container area with the smallest density to store the container, that is, the container area with the least number of containers stored in all bays of each container area. When the system selects the corresponding container area, it randomly selects the bay with the lowest bay density in the area to store the container, that is, the number of containers stored in all slots on the selected bay is the least. When the container is placed in the corresponding slot, the task of the container ends, and the next container repeats this operation.

[0162] The specific implementation steps are as follows:

[0163] Step 1: Determine the multi-stage operation mode as loading and unloading mode. The operation process is as follows: Figure 3 As shown in the figure, the target wharf layouts are vertical layout, parallel non-cantilever layout, U-shaped layout and parallel double-cantilever layout, and two strategies, namely partitioned stockpiling and random stockpiling, are adopted.

[0164] Step 2: The subtasks and their parameters of the multi-level job are as follows Figure 3 As shown, the parameters are set in the simulation environment.

[0165] Step 3-4: Determine the set of intelligent agents and their parameters, as well as multi-level task constraints.

[0166] Step 5: Uncertainty arises from the uncertainty of import and export box types.

[0167] For the container terminal in the case, there is a 30% probability of generating imported containers and a 70% probability of generating imported containers; among these imported and exported containers, there is an 80% probability of generating ordinary containers, a 10% probability of generating dangerous containers, and a 10% probability of generating refrigerated containers.

[0168] That is, for ct i , with a 30% probability of generating ct i .A1=in, generates ct with a probability of 70% i .A1=out, generating ct with 80% probability i .B1=ord, generates ct with a probability of 10% i .B1=dan, generates ct with a 10% probability i .B1=ref.

[0169] Step 6: Based on steps 1-5, a simulation model is constructed using Flexsim software, and the operation logic of the model is set according to the operation mode of step 1. The operation logic is as follows: Figure 4 shown.

[0170] Set the parameters of the terminal equipment agent and set the uncertainty in the import and export agent according to step 5.

[0171] In the simulation experiment, the parameters of AGV are shown in Table 7.

[0172] Table 7 AGV parameter settings

[0173]

[0174]

[0175] In the simulation experiment, the parameters of ARMG are shown in Table 8.

[0176] Table 8 ARMG parameter settings

[0177]

[0178] In the simulation experiment, the parameters of the yard are shown in Table 9.

[0179] Table 9 Yard parameter settings

[0180]

[0181] Step 7: Run the model to obtain the required data.

[0182] Step 8: Calculate the energy efficiency of the terminal based on the data. The simulation results are as follows: Figures 5 to 19 shown.

[0183] Step 9: Further calculate the energy efficiency imbalance. The simulation results are as follows: Figure 20 to Figure 22 shown.

[0184] Step 10: Analyze and select a suitable storage strategy for each layout based on the dynamic energy efficiency and its imbalance.

[0185] The simulation results show that different automated terminals can cope with the uncertainty in the multi-level operation process by changing the stacking strategy. At the same time, each automated terminal has its own stacking strategy, and the effect of each stacking strategy on the same automated terminal is also different.

[0186] For automated terminals under a vertical layout, from a macro perspective, the use of a random stacking strategy can achieve higher container throughput, lower equipment operating time and equipment energy consumption. From a micro perspective, the imbalance of the random stacking strategy is stronger. Therefore, it can be considered that automated terminals under a vertical layout can achieve efficient terminal operations by maintaining appropriate imbalance.

[0187] For automated terminals with parallel cantilever-free layouts, from a macro perspective, the use of partitioned stacking strategies can achieve higher container throughput, lower equipment operating time and equipment energy consumption; from a micro perspective, the imbalance of the partitioned stacking strategy is stronger. It should be noted that the imbalance under the partitioned stacking strategy is very different from the imbalance under the random stacking strategy, but the annual throughput is not much different. Therefore, it can be considered that automated terminals with parallel cantilever-free layouts can sacrifice a small amount of throughput to achieve a more stable operation process.

[0188] For automated terminals under the U-shaped layout, from a macro perspective, the random stacking strategy can achieve higher container throughput, and the partitioned stacking strategy can achieve lower equipment operating time and equipment energy consumption; from a micro perspective, the random stacking strategy has a stronger imbalance. It should be noted that the imbalance under the partitioned stacking strategy is slightly different from the imbalance under the random stacking strategy, but the annual throughput is not much different. Therefore, it can be considered that the automated terminal under the U-shaped layout is more suitable for the partitioned stacking strategy.

[0189] For the automated terminal under the parallel double-cantilever layout, from a macro perspective, the random stacking strategy can achieve higher container throughput, lower equipment operation time and equipment energy consumption. From a micro perspective, the imbalance of the random stacking strategy is weaker. Therefore, it can be considered that the automated terminal under the parallel double-cantilever layout is more suitable for the random stacking strategy.

[0190] In summary, automated terminals under vertical layout are suitable for random stockpiling strategy and partitioned stockpiling strategy, automated terminals under parallel cantilever-free layout are suitable for partitioned stockpiling strategy, automated terminals under U-shaped layout are suitable for partitioned stockpiling strategy, and automated terminals under parallel double cantilever layout are suitable for random stockpiling strategy.

[0191] Furthermore, this embodiment also provides a dynamic energy efficiency simulation optimization system for stacking strategy of automated container terminals, which is used to implement the aforementioned method, including a basic data acquisition module, an energy consumption and operation time data acquisition module, a dynamic energy efficiency index calculation module, an energy efficiency imbalance calculation module, and a stacking strategy determination module. The specific implementation process of each module is consistent with the aforementioned method and will not be repeated here.

[0192] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A dynamic energy efficiency simulation optimization method for automated container terminal storage strategy, characterized in that: The following steps are involved: Obtain the layout of the target terminal, the operation flow and stockpiling strategy of multi-level operations; According to the layout, the operation process, the stacking strategy and the preset simulation parameters, the energy consumption and operation time data under different stacking strategies are obtained by using a pre-built multi-agent-based automated container terminal multi-level operation simulation model; Calculate the dynamic energy efficiency index of the automated container terminal under different stacking strategies based on the energy consumption and operation time data; According to the dynamic energy efficiency index, the energy efficiency imbalance of multi-level operations in automated container terminals under different stacking strategies is calculated; Compare the dynamic energy efficiency indicators and energy efficiency imbalance under each stockpile strategy, and determine the final stockpile strategy based on the comparison results.

2. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 1 is characterized in that: The construction process of the multi-agent-based automated container terminal multi-level operation simulation model specifically includes: According to the operation process, determine the time parameters of different equipment in the multi-level operation; Determine the agent set and agent parameters; Determine multi-level job constraints; Determine the uncertainty of the operation sequence of multi-level operations at the target terminal; Based on the layout, the operation process, the stacking strategy, the time parameters, the agent set and agent parameters, the multi-level operation constraints and the operation sequence uncertainty, a multi-agent-based automated container terminal multi-level operation simulation model under different stacking strategies is constructed.

3. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 2 is characterized in that: The multi-level operation constraints include constraints on the operation flow, constraints on continuous operations, and constraints on intelligent agents.

4. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 1, characterized in that: The dynamic energy efficiency indicators specifically include throughput, equipment energy efficiency, energy efficiency generated by container stays, and container area bay density.

5. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 4 is characterized in that: The throughput is expressed as: In the formula, TEU IO AQC.ct represents the throughput of the target terminal during the time (t1, t2). i Indicates the number of containers loaded and unloaded by all automated quay cranes, ET.ct i represents the number of containers transported by all external container trucks, and T represents the time set when the automated quay crane agent and the external container truck agent receive container task i.

6. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 4 is characterized in that: The equipment energy efficiency includes: when any automatic guided vehicle completes a preset number of container tasks within a unit time, the actual time, average time, maximum average time and minimum average time for the automatic guided vehicle to complete a single container task.

7. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 4, characterized in that: The energy efficiency of container stay includes: For any common container in any bay, the average time it takes for the common container to be taken out of or stored in the corresponding bay; For a dangerous goods container in any bay, the average time it takes for the dangerous goods container to be taken out of or stored in the corresponding bay; and For a refrigerated container in any bay, the average time it takes for the refrigerated container to be taken out of or stored in the corresponding bay.

8. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 4, characterized in that: The bay density of the container area includes: the average density of containers stored in the bays and the distribution of the number of containers stored in all bays in the yard.

9. The method for dynamic energy efficiency simulation optimization of automated container terminal storage strategy according to claim 1, characterized in that: The calculation expression of the energy efficiency imbalance is: and Where Z represents the Singh index, n i Indicates efficiency x i The corresponding number of devices, Z (pi) represents the Singh imbalance index, N represents the total number of devices, and L i Indicates that the efficiency is less than x i The proportion of the cumulative efficiency of the equipment to the total efficiency, p i Indicates that the efficiency is less than x i The cumulative number of devices accounts for the proportion of the total number of devices, and e represents the total number of efficiency values.

10. A dynamic energy efficiency simulation optimization system for automated container terminal storage strategy, characterized in that: The system is used to implement the method described in any one of claims 1 to 9, including: Basic data acquisition module, used to obtain the layout of the target terminal, the operation process of multi-level operations and the stockpiling strategy; An energy consumption and operation time data acquisition module is used to acquire energy consumption and operation time data under different stockpiling strategies using a pre-built multi-agent-based automated container terminal multi-level operation simulation model according to the layout, the operation process, the stockpiling strategy and preset simulation parameters; A dynamic energy efficiency index calculation module, used to calculate the dynamic energy efficiency index of the automated container terminal under different storage strategies according to the energy consumption and operation time data; An energy efficiency imbalance calculation module, used to calculate the energy efficiency imbalance of multi-stage operations of an automated container terminal under different stacking strategies according to the dynamic energy efficiency index; and The stockpile strategy determination module is used to compare the dynamic energy efficiency indicators and energy efficiency imbalance under each stockpile strategy, and determine the final stockpile strategy based on the comparison results.

Images