Automatic wharf cold dangerous cargo storage yard layout simulation method and system considering manual and unmanned mixed operation

Abstract

The invention discloses an automatic wharf cold dangerous cargo storage yard layout simulation method and system considering manual and unmanned mixed operation, and belongs to the field of automatic wharf cold dangerous cargo storage yard layout. Firstly, data are collected to generate an overall layout set A; secondly, randomly selecting one overall layout from A to obtain a first-stage optimization model, and solving the first-stage optimization model by using a genetic algorithm; then, constructing a second-stage physical model, and recording the second-stage physical model as a simulation model set; and finally, evaluating, comparing and selecting different simulation models to obtain an optimal layout and scale combination. The automatic wharf cold dangerous cargo storage yard layout simulation system comprises a data acquisition module, a layout mode generation module, a scale optimization module, a simulation model construction module, a comprehensive evaluation module and a judgment module. According to the method, layout and scale collaborative optimization is achieved through two-stage optimization and simulation, the port operation efficiency, the resource utilization rate and safety can be improved, meanwhile, energy consumption and carbon emission are reduced, and scientific decision support is provided for port design.

Description

Technical Field

[0001] The present invention belongs to the field of the layout of refrigerated and dangerous goods yards in automated terminals, and relates to a simulation method and system for the layout of refrigerated and dangerous goods yards in automated terminals considering mixed manual and unmanned operations. Background Art

[0002] At present, the layout design of refrigerated container yards in automated terminals often ignores the safety issues of port workers working in the port area. In fact, it is difficult to fully realize unmanned operation for refrigerated and dangerous goods at present. Therefore, it is necessary to find an efficient and safe simulation method for the yard layout in the case of unmanned machinery and manual operation in the same space. In addition, in the past, when carrying out yard layout, due to the imperfect evaluation system and excessive subjective assumptions in the early stage of modeling, it is very likely to lead to problems such as an incomplete design logic chain of the yard, low operation efficiency, potential safety hazards, and resource waste.

[0003] To solve the above problems, Chinese invention patent CN119417340A proposes a method for transporting and operating special containers in an automated terminal. This method sends a ship unloading operation plan through the front end of the TOS system, and the scheduling system obtains the operation plan for special container types and matches the IGV equipped with a cargo rack to meet the requirements of special container types, so as to achieve efficient transportation and operation of special containers. However, this patent mainly focuses on the optimization of the transportation and operation process of special containers and does not fully consider the safety and efficiency of the yard layout. Especially for refrigerated and dangerous goods, this patent lacks safety guarantee measures when technical workers cooperate with automated facilities and equipment, and does not establish a perfect evaluation system to optimize the yard layout to ensure operation efficiency and rational utilization of resources. In terms of yard operation, Chinese invention patent CN119579038A proposes a method and storage medium for yard operation in an automated terminal, which improves the yard operation efficiency and the utilization rate of gantry cranes by optimizing the sorting and scheduling of operation tasks. However, this patent mainly focuses on the scheduling optimization of yard operation tasks and does not involve the specific methods for transporting and operating special containers. Especially, the safety and efficiency of the layout of refrigerated and dangerous goods yards still need to be improved.

[0004] In summary, how to provide a simulation system for the layout of refrigerated and dangerous goods yards in an automated terminal considering mixed manual and unmanned operations, which can comprehensively and accurately conduct a comprehensive analysis of the layout of refrigerated and dangerous goods yards in an automated terminal with mixed manual and unmanned operations, so as to achieve layout optimization, has also become an urgent problem to be solved. Summary of the Invention

[0005] In order to solve the problem of cold and dangerous goods storage, which has been long neglected when discussing the yard layout in the prior art, especially the safety, resource utilization efficiency and operation efficiency problems in the scenario of mixed manual and unmanned operations. The present invention provides a simulation method and system for the layout of cold and dangerous goods yards in automated terminals considering mixed manual and unmanned operations, which can comprehensively and accurately conduct a comprehensive analysis of the layout of cold and dangerous goods yards in automated terminals with mixed manual and unmanned operations, realize the optimized layout, thereby improving the port benefits and resource utilization rate, while ensuring the safety of cold and dangerous goods and meeting the green energy-saving requirements of modern ports.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A simulation method for the layout of cold and dangerous goods yards in automated terminals considering mixed manual and unmanned operations, comprising the following steps:

[0008] The first step is to collect the characteristic data of the yard to be built in the automated terminal and the parameters of the facilities and equipment in the automated terminal to be built, specifically as follows:

[0009] The characteristic data of the yard to be built in the automated terminal mainly includes: geographical location, available land depth, terrain conditions, natural conditions, main goods, number of days of terminal operation, estimated annual throughput, predicted ship type, estimated number of technical workers and total port construction budget.

[0010] The parameters of the facilities and equipment in the automated terminal to be built respectively consider the parameters of several facilities and equipment such as horizontal transportation equipment, yard cranes and quay cranes. The parameters considered for each facility and equipment specifically include: unit price of the facility and equipment, operation efficiency (including loading and unloading efficiency and transportation efficiency), energy consumption per unit hour, endurance capacity, maximum acceleration.

[0011] The second step is to generate an overall layout set. Based on the characteristic data of the yard to be built in the automated terminal and the parameters of the facilities and equipment in the automated terminal obtained in the first step, n overall layouts are generated to form an overall layout set A. The generation of the overall layout is specifically as follows:

[0012] Step 2.1, based on the characteristic data of the yard to be built in the automated terminal obtained in the first step and the determination rules of the yard layout method, judge whether the yard to be built in the automated terminal can be built into a vertical, horizontal or U-shaped yard layout method, and determine the available yard layout methods:

[0013] If the estimated annual throughput is extremely small, the available land depth is small, the terrain conditions are restricted, and the total port construction budget is low, then the vertical layout is not available;

[0014] If the estimated annual throughput is extremely large, the terrain conditions are complex or the development is restricted, then the horizontal layout is not available;

[0015] If the expected annual throughput is large, but the terrain conditions are complex, the U-shaped layout is not available;

[0016] Determine one or more available yard layout methods.

[0017] Step 2.2: Based on the parameters of the automated terminal facilities and equipment to be built in the first step and the selection and layout specifications of each facility and equipment, judge the availability of various facilities and equipment. The specific selection and layout specifications of each facility and equipment include the industry specifications JTS165-2013 General Design Code for Seaports and JTS144-1-2010 Code for Port Engineering Loads.

[0018] Step 2.3: Combine the facilities and equipment with the yard layout methods to form an overall layout set A = {S1, S2, S3... S i ...S n}, where i ∈ 1......n.

[0019] The third step: Construct and solve the first-stage optimization model. Arbitrarily select an overall layout S from the overall layout set A generated in the second step i , construct the first-stage mathematical model, obtain the optimization model of the scale of the cold and dangerous goods yard facilities and equipment of the automated terminal, and use the genetic algorithm to solve the first-stage optimization model; specifically as follows:

[0020] Step 3.1: Determine the decision variables and key parameters of the optimization model.

[0021] The decision variables include: the number of quay cranes ; the number of yard cranes ; the number of horizontal transportation equipment ;

[0022] The key parameters include: the handling efficiency of a single quay crane ; the handling efficiency of a single yard crane ; the transportation efficiency of a single horizontal transportation equipment ; the unit price of the equipment is divided into the unit price of the yard crane , the unit price of the quay crane , the unit price of the horizontal transportation equipment , the total construction budget of the automated terminal to be built is G, means a positive integer, the number of days of terminal operation , the expected annual throughput Q.

[0023] Step 3.2: Construct the objective function of the optimization model for the scale of the cold and dangerous goods yard facilities and equipment of the automated terminal. Ensure the highest operating efficiency U, maximize the throughput of the system bottleneck link, and ensure the balance of the collaborative operation ability of the quay crane, yard crane and horizontal transportation tool:

[0024] (1)

[0025] Step 3.2, construct the constraint conditions of the optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal, as follows;

[0026] (2)

[0027] Equation (2) is that the total purchase cost of quay cranes, yard cranes and horizontal transport equipment cannot exceed the total construction budget G of the terminal. Ensure equipment configuration within the scope of economic feasibility.

[0028] (3)

[0029] (4)

[0030] (5)

[0031] Equations (3)-(5) ensure that the combined handling capacities of quay cranes, yard cranes and horizontal transport equipment in the terminal meet or exceed the requirements of the annual throughput Q respectively. That is, the handling capacity of each piece of equipment (quantity multiplied by the efficiency of a single unit) should at least be able to handle the average daily cargo volume , so as to ensure the operation efficiency and service level of the terminal.

[0032]

[0033] Equation (6) is a non - negative integer constraint.

[0034] Step 3.3, use the genetic algorithm to solve the optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal, and obtain the optimized results of the scale of each facility and equipment.

[0035] Fourth step, construct the second - stage physical model, that is, the simulation model of the cold and dangerous goods yard of the automated terminal, denoted as the simulation model set B = {M1, M2, M3…M i …M n}, where i ∈ 1......n, as follows:

[0036] Step 4.1, in the simulation software AnyLogic, model all the overall layouts in A respectively. The overall layout S i is modeled as the corresponding simulation model M i . S i The facilities and equipment in are determined as the facility and equipment agents in M i . In particular, in order to reflect the influence of human - machine interaction in the model, the present invention also adds technical worker agents to the model. Traverse the overall layout set A = {S1, S2, S3...S i ...S n}, and correspondingly construct a set of simulation models B = {M1, M2, M3…M i …M n}.

[0037] Step 4.2: Determine the scale of agents. Each simulation model uses the same scale of facilities and equipment, and its scale is the optimized result of the scale of each facility and equipment obtained in the third step. In particular, the scale of agents such as technical workers is determined according to the expected number of technical workers in the yard feature data of the to-be-built automated terminal in the first step.

[0038] Step 4.3: Construct the behavior patterns of agents. In the simulation model, the facility and equipment agents use the random functions built into the simulation software to generate random behavior patterns of the agents. For the technical worker agents, the random movement behavior of humans is described separately by Lévy Flight, and the description formula is as follows:

[0039] The position of the technical worker is described by two-dimensional coordinates (xp, yp), and its position change is expressed as:

[0040] (7)

[0041] (8)

[0042] where μxp and μyp are the drift coefficients in the xp and yp directions; σxp and σyp are the diffusion coefficients in the xp and yp directions; 、 are the increments of independent Wiener processes.

[0043] For the convenience of calculation, a method based on the Mantegna algorithm is specifically used to generate the step size of Lévy Flight , and the probability density function of its step size s is:

[0044] (9)

[0045] where β is the distribution index, ranging from (0, 2], which controls the heavy-tailed characteristics of the step size, and D is the normalization constant.

[0046] Step 4.4, the model is set to simulate the loading and unloading operations of the cold and dangerous goods yard in an automated terminal for one year, and the average operating rate of horizontal transportation equipment, the average operating rate of yard cranes, the average rate of quay cranes, labor costs, construction costs, the utilization rate of horizontal transportation equipment, the utilization rate of storage space, the utilization rate of yard cranes, the utilization rate of quay cranes, the total loading and unloading time of horizontal transportation equipment, the total loading and unloading time of yard cranes, the total loading and unloading time of quay cranes, the average charging time of horizontal transportation equipment, the cargo throughput, the proportion of automated facilities, the total energy consumption of refrigerated containers, and when the horizontal transportation equipment or yard crane identifies that there are technical workers on the front path, the position coordinates and driving speeds of both are output.

[0047] Step 5, evaluate each simulation model. First, construct an evaluation index system for the layout of the cold and dangerous goods yard in an automated port area. Select the required port operation data according to the evaluation index system, and use the entropy weight method and TOPSIS method for processing to obtain the comprehensive scores of each simulation model respectively. Note that the i-th simulation model is the i-th evaluation object. Specifically as follows:

[0048] Step 5.1, construct an evaluation index system for the layout of the cold and dangerous goods yard in an automated port area. The evaluation system consists of evaluation indexes in six dimensions, and the evaluation indexes include automation level, cost investment, resource utilization, operation efficiency, human-machine interaction safety, and green energy conservation. The evaluation indexes are divided into benefit-type indexes and cost-type indexes. The indexes for processing the cost investment part are regarded as cost-type indexes, and the rest of the indexes are considered benefit-type indexes.

[0049] Step 5.2, calculate the comprehensive collision time T. The comprehensive collision time T in the evaluation index system is not the data that can be directly output by the simulation model. It is an innovative human-machine interaction safety evaluation index designed by the present invention and needs to be calculated separately. The specific calculation is given later.

[0050] Step 5.3, select the required port operation data according to the evaluation index system, and then perform dimensionless processing on the required operation data and the comprehensive collision time T obtained in Step 5.2. The specific process is as follows:

[0051] (10)

[0052] Among them, X max is the maximum value of the j-th index of the evaluation system; X min is the minimum value of the j-th index of the evaluation system; x ij is the preprocessed data; X ij is the original value in the evaluation system.

[0053] Step 5.4, calculate the information entropy and weight of the required operation data and the comprehensive collision time T obtained in Step 5.2:

[0054]

[0055] Among them, is the information entropy of the j-th index, is the weight of the j-th index; is for the i-th simulation model M i , that is, the probability of the j-th index of the i-th evaluation object occurring; n represents n evaluation objects, that is, all the simulation models {M1, M2, M3... M i ... M n} in set B; m represents there are m evaluation indexes.

[0056] Step 5.5 Calculate the positive ideal solution and the negative ideal solution . Among them represents the optimal value combination of each index. represents the worst value combination of each index:

[0057] (14)

[0058] (15)

[0059] } (16)

[0060] (17)

[0061] Among them, represents the positive ideal solution value of the first index; represents the positive ideal solution value of the n-th index; represents the negative ideal solution value of the first index; represents the negative ideal solution value of the n-th index; represents the j-th index; J1 represents the set of benefit-type indexes; For the benefit-type indexes, take the maximum value of this index in all the schemes; For the benefit-type indexes, take the minimum value of this index in all the schemes; represents the set of cost-type indexes; For the cost-type indexes, take the minimum value of this index in all the schemes; For the cost-type indexes, take the maximum value of this index in all the schemes.

[0062] Step 5.6, use to represent the distance of the i-th simulation model M i to the positive ideal solution; , represents the distance of the i-th simulation model M i to Distance to the negative ideal solution, calculate the relative closeness degree of the i-th evaluation object to the positive and negative ideal solutions , the formula is as follows:

[0063] (18)

[0064] Step 6, determine the optimal layout. According to the results obtained in Step 5 evaluate and compare different simulation models, where 0 ≤ δ i ≤ 1, according to the numerical size for sorting, the larger the value, the closer it is to the optimal level. Specifically as follows:

[0065] If the overall layout S i corresponding to M i is the simulation model of the cold and dangerous goods yard in the automated terminal with the largest closeness degree , then it is considered that the scale of facilities and equipment adopted by the M i model and the overall layout are the optimal;

[0066] If the overall layout S i corresponding to M i is not the simulation model of the cold and dangerous goods yard in the automated terminal with the largest closeness degree, and the model with the largest closeness degree is M , then take the S k corresponding to M k as the overall layout selected in Step 3, and then complete Steps 3 to 6 in sequence until the scale of facilities and equipment and the overall layout are both optimal. k

[0067]

[0067] Furthermore, the evaluation system in Step 5.2 is composed of evaluation indicators in six dimensions, and the evaluation indicators include automation level, cost input, resource utilization, operation efficiency, human-machine interaction safety, and green energy conservation. Specifically:

[0068] For the aspect of automation level, the quantitative index of the proportion of automated facilities is adopted.

[0069] For the aspect of cost input, labor cost and construction cost are adopted.

[0070] For the aspect of resource utilization, four evaluation indicators with high usage frequency in the evaluation system are used, namely the utilization rate of horizontal transportation equipment, the utilization rate of stacking space, the utilization rate of yard cranes, and the utilization rate of quay cranes.

[0071] For the aspect of operation efficiency, it is described by the total loading and unloading time of transportation equipment, the average charging time of horizontal transportation equipment, and the cargo throughput, where the total loading and unloading time of transportation equipment includes: the total loading and unloading time of horizontal transportation equipment, the total loading and unloading time of yard cranes, and the total loading and unloading time of quay cranes.

[0072] In terms of the green energy conservation, aiming at the characteristics of the refrigerated container yard, the total energy consumption of refrigerated containers is especially introduced to describe the energy consumption problem caused by the continuous power supply of refrigerated containers.

[0073] In terms of the human-machine interaction safety, the average operation speed of the equipment can reflect the safety of human-machine interaction to a certain extent. Aiming at the human-machine interaction situation formed within a small range, the present invention introduces the comprehensive collision time T. Specifically as follows:

[0074] The parameter setting is as follows: when the horizontal transportation equipment or the yard crane recognizes that there is a technical worker on the front path, the projection coordinates of the position of the horizontal transportation equipment or the yard crane on the equipment movement track are: , and the projection coordinates of the two-dimensional coordinates (xp, yp) of the position of the technical worker on the equipment movement track are: . The traveling speed of the horizontal transportation equipment or the yard crane is v (vector), and this speed direction is taken as the positive direction. The traveling speed of the technical worker is v r , the maximum acceleration of the facility equipment is a, the lane width is taken as L. When the equipment recognizes a person, the relative distance between the person and the equipment is expressed as . The calculation method of the comprehensive collision time T is as follows:

[0075] (19)

[0076] (20)

[0077] (21)

[0078] (22)

[0079] (23)

[0080] (24)

[0081] (25)

[0082] (26)

[0083] (27)

[0084] Among them, the coefficient μ is a conversion coefficient, and its purpose is to unify the two states of human-machine interaction. describes the relative position between the person and the equipment. The distances between the device and the person in two critical states are described. For these two critical states, one is that the distance between the device and the person is equal to the braking distance of the device, and the other is that the distance between the device and the person is equal to the distance that the vehicle travels at the original speed without braking when the person finishes crossing the road. y describes the distance that the vehicle decelerates during the time when the person crosses the road.

[0085] The comprehensive meaning of the collision time T is: when on the same straight line at this time = 1, actually the formula after the TTC transformation is used to describe the relative safety of human - machine interaction; when That is, when the person and the device are moving in the same direction, if 0 > That is, the person is in front of the vehicle, and the distance between the person and the vehicle is less than the braking distance of the vehicle, then it is considered that there is a potential danger; when When, that is, the person and the device are moving towards each other or in opposite directions. If moving towards each other , there is danger; if moving in opposite directions , it is safe;

[0086] When When, that is, when their running directions are perpendicular to each other = 0, actually the formula after the PET transformation is used to describe the relative safety of human - machine interaction: when When, it is considered dangerous; when Or When, it is considered very safe;

[0087] To sum up, the value range of the comprehensive collision time T shown is: When, it is considered that the human - machine interaction is in a safe state, and the larger the value of T, the safer it is; , then it is considered that there is a potential danger in the human - machine interaction.

[0088] An automated terminal cold - dangerous goods yard layout simulation system considering mixed manual and unmanned operations processes each step in the automated terminal cold - dangerous goods yard layout simulation method in a modular way. The automated terminal cold - dangerous goods yard layout simulation system includes a data acquisition module, a layout mode generation module, a scale optimization module, a simulation model construction module, a comprehensive evaluation module, and a judgment module; specifically:

[0089] The data acquisition module: This data acquisition module is mainly used to obtain the characteristic data of the to - be - built automated terminal yard and the parameters of the to - be - built automated terminal facilities and equipment from the port environment. It covers three functional units: data cleaning, formatting, and classified storage, pre - processes the data to ensure the integrity and accuracy of the data, and provides high - quality input data for the subsequent modules.

[0090] The layout mode generation module: This layout mode generation module is connected to the data collection module and inputs the yard feature data of the port and the performance data of the port facilities and equipment. The layout mode generation module is based on the yard layout determination rules and equipment layout specifications, covering three functional units: layout method determination, multi - scheme generation, and index establishment, and supports the automatic generation and management of multiple feasible overall yard layouts. Finally, it outputs structured data containing the overall layout set A = {S1, S2, S3,..., S i ,..., S n}, and stores it for future reference.

[0091] The scale optimization module: This scale optimization module is connected to the layout mode generation module and the data collection module, covering two functional units: parameter update and model solution. Randomly select an overall layout mode S i (if it is not the first iteration, it is to obtain the overall layout mode S k returned from the judgment module) from the layout mode generation module, input the yard feature data of the port and the performance data of the port facilities and equipment into the optimization model of the scale of the facilities and equipment in the cold and dangerous goods yard of the automated terminal, update the model parameters, and then solve the optimization model. Finally, it outputs the scale optimization results of each facility and equipment and feeds them back to the simulation model construction module.

[0092] The simulation model construction module: This simulation model construction module is connected to the data collection module, the layout mode generation module, and the scale optimization module, covering three functional units: parameter update, model construction, and dynamic simulation. Input the yard feature data, the parameters of the facilities and equipment, the overall layout set, and the scale optimization results of each facility and equipment, update the relevant parameters of the simulation model, and construct different simulation models M1, M2, M3…M i …M n of the cold and dangerous goods yard of the automated terminal (where i ∈ 1......n). It supports simulating the dynamic process of mixed operation of manual and unmanned equipment and can reflect the random drift and diffusion characteristics of technical workers, and finally outputs various port operation data.

[0093] The comprehensive evaluation module: This comprehensive evaluation module is connected to the simulation model construction module and obtains various port operation data. The module is based on the evaluation index system, covering three functional units: data pre - processing, weight assignment (entropy weight method), and multi - index comprehensive ranking (using the TOPSIS method), supports the parallel evaluation of multiple simulation models, and finally outputs the closeness index (0 ≤ δ i ≤ 1) of each simulation model.

[0094] The judgment module: This judgment module is connected to the comprehensive evaluation module and includes two functional units: comparison and decision-making, and feedback control. It inputs the closeness indexes of each simulation model, ranks and optimizes different overall layout and facility scale plans. This module automatically identifies the optimal model with the highest closeness. If the overall layout corresponding to the optimal model is the overall layout S adopted by the scale optimization module i (if it is not the first iteration, it is S k ), it is considered that the facility scale and the overall layout are optimal, the system terminates the iteration, and outputs the final plan; otherwise, it feeds back the overall layout corresponding to the optimal model to the scale optimization module to start a new round of iteration. Until the facility scale and the overall layout are optimal, the dynamic collaborative optimization of the layout and facility scale is finally realized.

[0095] The beneficial effects of the present invention are as follows:

[0096] The present invention processes the overall layout and facility scale through a two-stage method, comprehensively uses the optimization model of the cold and dangerous goods yard facilities and equipment scale of the automated terminal driven by data, the simulation model of the cold and dangerous goods yard of the automated terminal, and the evaluation system, improving the scientific nature of the entire process of port design. At the same time, the comprehensive collision time T index is innovatively introduced to improve the safety of human-machine interaction, and the green energy-saving index is incorporated to reduce energy consumption and carbon emissions. Description of the Drawings

[0097] Figure 1 are the implementation steps of the layout method of the port cold and dangerous goods yard considering the interaction between manual and unmanned operations.

[0098] Figure 2 is the layout evaluation index system of the cold and dangerous goods yard in the automated port area.

[0099] Figure 3 are the agents composition of the simulation model construction module and the mutual relationship between the agents.

[0100] Figure 4 is the schematic diagram of the structure of the simulation system for the layout of the cold and dangerous goods yard in the automated terminal considering the mixed operation of manual and unmanned. Detailed Embodiments

[0101] The following further illustrates the present invention in combination with specific implementation cases.

[0102] A simulation method for the layout of the cold and dangerous goods yard in the automated terminal considering the mixed operation of manual and unmanned includes the following steps:

[0103] The first step is data collection. Collect the yard characteristic data of the to-be-built automated terminal and the parameters of the facilities and equipment of the to-be-built automated terminal, specifically as follows:

[0104] The data of the to-be-built automated terminal mainly includes: Taking a certain actual port as an example, its geographical location is in a southern bay of China. The available land depth is about 1300m, the terrain is flat and open, without special restrictions, and the natural conditions are relatively stable. The main cargo is refrigerated containers. The number of days of terminal operation is expected to be 300 days / year, the annual throughput is expected to be 1 million TEU, the predicted ship types are mainly container ships, the number of expected technical workers is 100, and the total budget for port construction is 1 billion yuan;

[0105] The parameters of the facilities and equipment of the to-be-built automated terminal. For horizontal transportation equipment, two schemes of AGV and L-AGV are considered. The acquisition cost of AGV is 1 million yuan per unit, the operation efficiency is 40 TEU / h, and the operation speed is 6 m / s; The acquisition cost of L-AGV is the same, and the operation efficiency and speed are the same as those of AGV, but the energy consumption is slightly higher than that of AGV. Both transportation tools use the method of replacing batteries to ensure power, with a battery life of 6 hours and a battery replacement time of 6 minutes. There is no power loss when stationary. The yard cranes involve cantilever ARMG and non-cantilever ARMG, with an operation efficiency of 25 TEU / h and an acquisition cost of 3 million yuan per unit. The quay crane operation efficiency is 30 TEU / h, and the acquisition cost is 5 million yuan per unit. Parameters such as the maximum acceleration also need to be collected in detail.

[0106] Step 2: Generate the overall layout set. Based on the yard feature data of the to-be-built automated terminal obtained in the first step and the parameters of the facilities and equipment of the to-be-built automated terminal, n overall layouts are generated to form the overall layout set A, which is specifically as follows:

[0107] Step 2.1, based on the yard feature data of the to-be-built automated terminal obtained in the first step, judge whether the to-be-built automated terminal can be built into a vertical, horizontal or U-shaped layout of the yard layout, and determine the available yard layout methods:

[0108] After evaluation, the annual throughput of this port is expected to be large, the available land depth and terrain conditions are both good, and the total budget for port construction is relatively sufficient. Therefore, both vertical layout and horizontal layout are feasible. The U-shaped layout can also be considered because the terrain conditions are not complex. However, considering various factors, finally it is determined that mainly two methods of horizontal layout and vertical layout are considered.

[0109] Step 2.2, based on the parameters of the facilities and equipment of the to-be-built automated terminal in the first step, judge the availability of various facilities and equipment. The specific selection and layout refer to the industry specifications JTS165-2013 General Design Code for Seaports and JTS144-1-2010 Code for Loads in Port Engineering. In this specific embodiment, it is also considered that under the horizontal layout, the cooperation between AGV and cantilever ARMG is more tacit, and they can efficiently complete the tasks of cargo transportation and loading and unloading; under the vertical layout, the combination of L-AGV and non-cantilever ARMG has more advantages.

[0110] Step 2.3. Combine the facility equipment with the yard layout mode, and combine the feasible yard layout modes with the transportation route forms of the horizontal transportation equipment to generate two overall layout modes: S1 (horizontal layout with AGV and cantilevered ARMG) and S2 (vertical layout with L-AGV and non-cantilevered ARMG), denoted as forming the overall layout mode set A = {S1, S2};

[0111] Step 3. Construct and solve the first-stage optimization model. Randomly select an overall layout S from the overall layout set A generated in the second step i , construct the first-stage mathematical model to obtain the optimization model for the scale of the cold and dangerous goods yard facilities and equipment in the automated terminal, and use the genetic algorithm to solve the first-stage optimization model; specifically as follows:

[0112] Step 3.1, based on this specific implementation case, randomly select the overall layout S1, and determine the decision variables and key parameters of the optimization model.

[0113] Step 3.2 Construct the objective function of the optimization model for the scale of the cold and dangerous goods yard facilities and equipment in the automated terminal. Ensure the highest operating efficiency U, maximize the throughput of the system bottleneck link, and ensure the balance of the collaborative operation capabilities of the quay crane, cantilevered ARMG, and AGV as shown in formula (1).

[0114] Step 3.2, construct the constraint conditions of the optimization model for the scale of the cold and dangerous goods yard facilities and equipment in the automated terminal as shown in formula (2). Specifically:

[0115] Formula (2) is that the total purchase cost of the quay crane, yard crane, and horizontal transportation equipment cannot exceed the total construction budget G of the terminal. Ensure equipment configuration within the economically feasible range.

[0116] Formulas (3) to (5) ensure that the combined handling capacities of the quay crane, yard crane, and horizontal transportation equipment in the terminal respectively meet or exceed the requirements of the annual throughput Q. That is, the handling capacity of each piece of equipment (quantity multiplied by the single-unit efficiency) should at least be able to handle the average daily cargo volume , so as to ensure the operation efficiency and service level of the terminal.

[0117] Formula (6) is a non-negative integer constraint.

[0118] Step 3.3, use the genetic algorithm to solve the optimization model for the scale of the cold and dangerous goods yard facilities and equipment in the automated terminal, and obtain the optimal scale of each facility and equipment corresponding to the overall layout S1: build 12 quay cranes, configure 20 cantilevered ARMGs, and 40 AGVs.

[0119] Step 4. Construct the second-stage physical model, that is, the simulation model of the cold and dangerous goods yard in the automated terminal, denoted as the simulation model set B = {M1, M2,} specifically as follows:

[0120] Step 4.1, in the simulation software AnyLogic, model all the overall layouts in A respectively. In the simulation software AnyLogic, model S1 and S2 in the overall layout set A respectively to form the corresponding simulation models M1 and M2. The facilities and equipment in S i are determined as the facility and equipment agents in M i . As shown in Figure 3 , it describes the agent composition of the simulation model construction module and the mutual relationship between agents.

[0121] Step 4.2, determine the agent scale. Each simulation model uses the same scale of facilities and equipment, and its scale is the optimal scale of facilities and equipment obtained by solving in the third step. Specifically, the scale of agents such as technical workers is determined according to the expected number of 100 technical workers in the yard characteristic data of the automated terminal to be built in the first step.

[0122] Step 4.3, construct the agent behavior mode. In the simulation model, the facility and equipment agents use the built-in random function of the simulation software to generate the random behavior mode of the agents. For the technical worker agents, the Lévy Flight is used alone to describe the random movement behavior of humans, and the description formula is as follows:

[0123] The position of the technical worker is described by the two-dimensional coordinates (xp, yp), and its position change is shown in Formulas (7) and (8).

[0124] For the convenience of calculation, the method based on the Mantegna algorithm is specifically used to generate the step size of the Lévy Flight, and its step size density function is shown in Formula (9).

[0125] Step 4.4, set the model to simulate the loading and unloading operations of the cold and dangerous goods yard of the automated terminal for one year, and output the average operating rate of the horizontal transport equipment, the average operating rate of the yard crane, the average rate of the quay crane, the labor cost, the construction cost, the utilization rate of the horizontal transport equipment, the utilization rate of the storage space, the utilization rate of the yard crane, the utilization rate of the quay crane, the total loading and unloading time of the horizontal transport equipment, the total loading and unloading time of the yard crane, the total loading and unloading time of the quay crane, the average charging time of the horizontal transport equipment, the cargo throughput, the proportion of automated facilities, the total energy consumption of the refrigerated containers, and when the horizontal transport equipment or the yard crane identifies that there are technical workers on the front path, the position coordinates and driving speeds of the two.

[0126] Fifth step, evaluate each simulation model. First, construct an evaluation index system for the layout of the cold and dangerous goods yard in the automated port area, select the required port operation data according to the evaluation index system, and use the entropy weight method and the TOPSIS method for processing to obtain the comprehensive scores of each simulation model respectively. Specifically as follows:

[0127] Step 5.1, construct an evaluation index system for the layout of the cold and dangerous goods yard in the automated port area, as Figure 2 shown. The evaluation system consists of evaluation indexes in six dimensions, and the evaluation indexes include automation level, cost input, resource utilization, operation efficiency, human-machine interaction safety, and green energy conservation. The evaluation indexes are divided into benefit-type indexes and cost-type indexes. The indexes related to the cost input part are regarded as cost-type indexes, and the rest of the indexes are regarded as benefit-type indexes.

[0128] Step 5.2, calculate the comprehensive collision time T. Combine formulas (19) to (27) to obtain the comprehensive collision time T in the evaluation index system.

[0129] Step 5.3, select the required port operation data according to the evaluation index system, and then perform dimensionless processing on the required operation data and the comprehensive collision time T obtained in Step 5.2, and the processing is as shown in formula (10).

[0130] Step 5.4, calculate the information entropy and weight of the required operation data and the comprehensive collision time T obtained in Step 5.2, such as formulas (11) to formula (13);

[0131] Step 5.5, calculate the positive ideal solution and the negative ideal solution , such as formulas (14) to (17). Among them represents the optimal value combination of each index. represents the worst value combination of each index.

[0132] Step 5.6, use to represent the distance of the i-th simulation model M i to the positive ideal solution; , represents the distance of the i-th simulation model M i to the negative ideal solution, and calculate to obtain the value of is 0.755, the value of is 0.245.

[0133] Sixth step, judge the optimal layout, and evaluate and compare different simulation models according to the obtained in the fifth step. In this implementation case, as shown in Table 1, the value of M1 corresponding to the layout mode S1 is 0.755, and the value of M2 corresponding to the layout mode S2 is 0.245. The M1 corresponding to the layout mode S1 is the simulation model with the highest score for the cold and dangerous goods yard of the automated terminal, and it is considered that the optimal facility and equipment scale and the overall layout mode are S1, that is, the horizontal layout with AGV and cantilever type ARMG.

[0134] Table 1. Evaluation Results of Specific Implementation Cases

[0135]

[0136] This embodiment also provides an automated terminal cold and dangerous goods yard layout simulation system considering manual and unmanned mixed operations, including a data acquisition module, a layout mode generation module, a scale optimization module, a simulation model construction module, a comprehensive evaluation module, and a judgment module, as follows:

[0137] Data acquisition module: As Figure 4 , the system first obtains the characteristic data of the yard to be built for the automated terminal and the parameters of the facilities and equipment of the automated terminal to be built from the port environment through the data acquisition module, covering three functional units of data cleaning, formatting, and classified storage. Based on this specific implementation case, the data cleaning unit is responsible for removing missing, duplicate, and abnormal data from the characteristic data of the yard to be built for the automated terminal and the parameters of the facilities and equipment of the automated terminal to be built; the formatting unit uniformly converts the collected multi-dimensional heterogeneous data including numerical data, categorical data, and time series data into a standard format; the classified storage unit then archives the yard characteristics and equipment parameters in an orderly manner according to the data type and usage, ensuring the integrity and accuracy of the data and supporting fast retrieval, providing high-quality input for the layout generation and scale optimization modules.

[0138] Layout mode generation module: As Figure 4 , the layout mode generation module accesses the data acquisition module and inputs the yard characteristic data of the port and the performance data of the port facilities and equipment. The layout mode generation module is based on the yard layout determination rules and equipment layout specifications, covering three functional units of layout method determination, multi-scheme generation, and index establishment. Based on the specific implementation case, the layout method determination unit screens out the feasible overall yard layout types as horizontal layout and vertical layout by analyzing the port geographical environment, cargo attributes, and equipment characteristics. The multi-scheme generation unit combines different layout forms with equipment to form multiple overall layout schemes, forming a horizontal layout with AGV and cantilever ARMG (Scheme S1), and a vertical layout with L-AGV and non-cantilever ARMG (Scheme S2). Index establishment unit: Numbers, classifies, and structurally stores the generated multiple overall layout schemes, forms an overall layout set A = {S1, S2} containing all layout schemes, and stores them for future reference.

[0139] Scale optimization module: As Figure 4The layout mode generation module and the data acquisition module are connected to the scale optimization module together, covering two functional units: parameter update and model solution. Based on a specific implementation case, the parameter update unit randomly selects the overall layout S1 from the overall layout set, and at the same time imports the corresponding port yard feature data and facility and equipment performance parameters to dynamically update the relevant parameters in formulas (1) to (6). The model solution unit constructs an optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal based on the updated parameters, and uses solution methods such as genetic algorithms to calculate the optimal equipment scale configuration plan. The solution result is to build 12 quay cranes, configure 20 cantilevered ARMGs, and 40 AGVs. The output result is fed back to the simulation model construction module[[ID=~]] [[ID=~]]

[0140] Simulation model construction module: As[[ID=~]] Figure 4 shown, this simulation model construction module is connected to the data acquisition module, the layout mode generation module, and the scale optimization module, covering three functional units: parameter update, model construction, and dynamic simulation. Based on a specific implementation case, the parameter update unit updates the relevant parameters of simulation models M1 and M2 respectively according to the yard feature data, the parameters of facilities and equipment, the overall layout set, and the scale optimization results of each facility and equipment. The model construction unit respectively establishes corresponding simulation models M1 and M2 for the solutions S1 and S2 of the overall layout set A in the layout mode generation module. Construct an interaction relationship network between agents (as[[ID=~]] Figure 3 shown), M1 determines 12 quay cranes, configures 20 cantilevered ARMGs, and 40 AGVs; M2 determines 12 quay cranes, configures 20 non-cantilever ARMGs, and 40 L-AGVs, and the number of skilled worker agents is set to 100. The dynamic simulation unit simulates the loading and unloading operations of the cold and dangerous goods yard of the automated terminal for one year. The skilled worker agents adopt the Lévy flight random movement model based on the Mantegna algorithm to simulate the random behavior of skilled workers in a two-dimensional space, which conforms to the mathematical descriptions in formulas (7) to (9), and dynamically reflects the operation flow of the mixed operation of manual and unmanned equipment. Finally, various port operation data are output[[ID=~]] [[ID=~]]

[0141] Comprehensive evaluation module: As[[ID=~]] Figure 4 shown, this comprehensive evaluation module is connected to the simulation model construction module to obtain various port operation data. The comprehensive evaluation module is based on an evaluation index system, covering three functional units: data preprocessing, weight assignment (entropy weight method), and multi-index comprehensive ranking (using the TOPSIS method). Based on a specific implementation case, the data preprocessing unit embeds formulas (10) and formulas (19) to (27) to complete the preprocessing of the data. The weight assignment unit embeds formulas (11) to (13) to complete the weight assignment for each index by the entropy weight method. The multi-index comprehensive ranking unit embeds formulas (14) to (17) to support the parallel evaluation of multiple simulation models. Finally, the closeness index of each simulation model is output, and the calculated value for M1 is[[ID=~]] The value is 0.755, for M2 The value is 0.245.

[0142] Judgment module: As Figure 4 shown, this judgment module is connected to the comprehensive evaluation module and includes two functional units: comparison decision-making and feedback control. Based on specific implementation cases, the comparison decision-making unit inputs the closeness indexes of each simulation model: The value is 0.755, for M2 value is 0.245, and automatically identifies the optimal model M1 with the highest closeness. The overall layout corresponding to M1 is the overall layout S1 adopted by the scale optimization module, and it is considered that the scale of facilities and the overall layout are optimal. The system terminates the iteration and realizes the dynamic collaborative optimization of the layout and the scale of facilities.

[0143] The above embodiments only represent the implementation manners of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that for those skilled in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. An automated terminal cold and dangerous goods yard layout simulation method considering mixed manual and unmanned operations, characterized in that The layout simulation method for cold and dangerous goods yards in automated terminals includes the following steps: In the first step, collect the characteristic data of the yard to be built in the automated terminal and the parameters of the facilities and equipment in the automated terminal to be built. In the second step, generate an overall layout set; based on the characteristic data of the yard to be built in the automated terminal and the parameters of the facilities and equipment in the automated terminal to be built obtained in the first step, generate n overall layouts to form an overall layout set A. Step 3: Construct and solve the optimization model in the first stage; arbitrarily select an overall layout S from the set A of overall layouts generated in Step 2 i , construct the mathematical model in the first stage, obtain the optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal, and use the genetic algorithm to solve the optimization model in the first stage; Step 4: Build the physical model for the second stage, i.e., the simulation model of the cold and dangerous goods yard in the automated terminal, denoted as the simulation model set B = {M1, M2, M3…M i …M n}, where i ∈ 1......n; In the fifth step, evaluate each simulation model; first, construct an evaluation index system for the layout of cold and dangerous goods yards in the automated port area, select the required port operation data according to the evaluation index system, and use the entropy weight method and TOPSIS method for processing to obtain the comprehensive scores of each simulation model respectively; note that the i-th simulation model is the i-th evaluation object. Step 6, determine the optimal layout; according to what is obtained in Step 5 evaluate and compare different simulation models, where 0 ≤ δ i ≤ 1, and sort according to the numerical value. The larger the numerical value, the closer it is to the optimal level.

2. The automated terminal cold and dangerous goods yard layout simulation method considering manual and unmanned mixed operations according to claim 1, wherein In the first step: The characteristic data of the yard to be built in the automated terminal mainly includes: geographical location, available land depth, terrain conditions, natural conditions, main goods, number of dock working days, estimated annual throughput, predicted ship type, estimated number of technical workers, and total port construction budget. The parameters of the facilities and equipment in the automated terminal to be built are considered separately for horizontal transportation equipment, yard cranes, and quay cranes; the specific parameters considered for each type of facility and equipment include: unit price of the facility and equipment, operation efficiency, unit-hour energy consumption, endurance, and maximum acceleration.

3. The automated terminal cold dangerous goods yard layout simulation method considering manual and unmanned mixed operation according to claim 1, wherein The second step is specifically as follows: Step 2.1, based on the characteristic data of the yard to be built in the automated terminal obtained in the first step, determine whether the automated terminal to be built can be built into a vertical, horizontal, or U-shaped yard layout, and determine the available yard layout methods. Step 2.2, based on the parameters of the facilities and equipment in the automated terminal to be built in the first step, determine the availability of various facilities and equipment. Step 2.3, combine the facility equipment and the yard layout method to form an overall layout set A = {S1, S2, S3... S i ... S n}, where i ∈ 1...... n.

4. A simulation method for the layout of a cold and dangerous goods yard in an automated terminal considering mixed manual and unmanned operations according to claim 1, characterized in that The third step is specifically as follows: Step 3.1, determine the decision variables and key parameters of the optimization model. The decision variables include: the number of quay cranes ; the number of yard cranes ; the number of horizontal transportation equipment ; The key parameters include: the handling efficiency of a single quay crane ; the handling efficiency of a single yard crane ; the transportation efficiency of a single horizontal transportation device ; the equipment unit price is divided into the unit price of yard cranes , the unit price of quay cranes , the unit price of horizontal transportation equipment , the total construction budget of the to-be-built automated terminal is G, where \(n\) is a positive integer, the number of days of terminal operation , and the expected annual throughput is Q; Step 3.2, construct the objective function of the optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal; ensure the highest operation efficiency U, maximize the throughput of the system bottleneck link, and ensure the balance of the collaborative operation ability of quay cranes, yard cranes, and horizontal transportation tools. （1）， Step 3.2, construct the constraint conditions of the optimization model for the scale of facilities and equipment in the cold and dangerous goods yard of the automated terminal. Step 3.3, use the genetic algorithm to solve the optimization model of the scale of facilities and equipment in the cold dangerous goods yard of the automated terminal, and obtain the optimal scale of each facility and equipment corresponding to the overall layout S i The corresponding optimal scale of facilities and equipment.

5. The automated terminal cold and dangerous goods yard layout simulation method considering manual and unmanned mixed operation according to claim 4, characterized in that The constraint conditions in Step 3.2 are specifically as follows: （2）， （3）， （4）， （5）， ， Formula (2) means that the total purchase cost of quay cranes, yard cranes, and horizontal transportation equipment cannot exceed the total port construction budget G. Formulas (3) to (5) ensure that the combined handling capacities of quay cranes, yard cranes, and horizontal transportation equipment at the terminal meet or exceed the requirements of the annual throughput Q; that is, the handling capacity of each piece of equipment should at least be able to handle the average daily cargo volume. , ensuring the operation efficiency and service level of the terminal. Formula (6) is a non-negative integer constraint.

6. The automated terminal cold and dangerous goods yard layout simulation method considering manual and unmanned mixed operation according to claim 4, characterized in that The fourth step is specifically as follows: Step 4.1, in the simulation software, model all the overall layouts in A respectively; the overall layout S i Model it as the corresponding simulation model M in the simulation software i ; S i The facilities and equipment in are determined as the facility and equipment agents in M i ; Add the skilled worker agents to the simulation model; traverse the overall layout set A = {S1, S2, S3... S i ... S n}, and correspondingly construct the simulation model set B = {M1, M2, M3… M i … M n}; Step 4.2, determine the agent scale; each simulation model uses the same scale of facilities and equipment, and its scale is the optimal scale of facilities and equipment obtained by solving in the third step; determine the scale of agents such as technical workers according to the estimated number of technical workers in the characteristic data of the yard to be built in the automated terminal in the first step. Step 4.3, construct the agent behavior mode; in the simulation model, the facility and equipment agents use the random function built into the simulation software to generate the random behavior mode of the agents; for the technical worker agents, the Levy flight is used to describe the random movement behavior, and the description formula is as follows: The position of the technical worker is described by two-dimensional coordinates (xp, yp), and its position change is expressed as: （7）， （8）， where μxp and μyp are the drift coefficients in the xp and yp directions; σxp and σyp are the diffusion coefficients in the xp and yp directions; , are the increments of independent Wiener processes; Generate the step size of Levy flight using the method based on the Mantegna algorithm , and the probability density function of its step size s is as follows: （9）， Among them, β is the distribution index, ranging from (0, 2], which controls the heavy-tailed characteristics of the step size, and D is the normalization constant; Step 4.4, the model is set to simulate the loading and unloading operations of the cold and dangerous goods yard in the automated terminal for one year, and output the average operating speed of the horizontal transportation equipment, the average operating speed of the yard crane, the average speed of the quay crane, the labor cost, the construction cost, the utilization rate of the horizontal transportation equipment, the utilization rate of the storage space, the utilization rate of the yard crane, the utilization rate of the quay crane, the total loading and unloading time of the horizontal transportation equipment, the total loading and unloading time of the yard crane, the total loading and unloading time of the quay crane, the average charging time of the horizontal transportation equipment, the cargo throughput, the proportion of automated facilities, the total energy consumption of the refrigerated containers, and when the horizontal transportation equipment or the yard crane identifies that there are technical workers on the front path, the position coordinates and driving speeds of both.

7. The automated terminal cold and dangerous goods yard layout simulation method considering manual and unmanned mixed operations according to claim 6, characterized in that, The specific content of the fifth step is as follows: Step 5.1, construct an evaluation index system for the layout of the cold and dangerous goods yard in the automated port area; the evaluation system consists of evaluation indicators in six dimensions, and the evaluation indicators include automation level, cost investment, resource utilization, operation efficiency, human-machine interaction safety, and green energy conservation; the evaluation indicators are divided into benefit-type indicators and cost-type indicators. The indicators for dealing with the cost investment part are regarded as cost-type indicators, and the rest of the indicators are regarded as benefit-type indicators; Step 5.2, calculate the comprehensive collision time T; the comprehensive collision time T in the evaluation index system is the human-machine interaction safety evaluation index and is calculated separately. Specifically: The parameters are set as follows: When the horizontal transportation equipment or the yard crane recognizes that there is a skilled worker on the front path, the projection coordinates of the position of the horizontal transportation equipment or the yard crane on the equipment movement trajectory are: , and the projection coordinates of the two-dimensional coordinates (xp, yp) of the position of the skilled worker on the equipment movement trajectory are: ; The traveling speed of the horizontal transportation equipment or the yard crane is v, which is a vector, and the positive direction is the direction of this speed; the traveling speed of the skilled worker is v r , the maximum acceleration of the facility equipment is a, the lane width is taken as L; when the equipment recognizes a person, the relative distance between the person and the equipment is expressed as ; The comprehensive collision time T, the calculation method is as follows: （19）， （20）， （21）， （22）， （23）， （24）， （25）， （26）， （27）， Among them, the coefficient μ is a conversion coefficient, aiming to unify the two states of human-computer interaction; Describe the relative positions of people and devices; When describing the two critical states, it is the distance between the device and the person; y describes the distance that the vehicle decelerates during the time when a person crosses the road; The meaning of the comprehensive collision time T is as follows: when on the same straight line at this time = 1, in fact, the formula after TTC transformation is used to describe the relative safety of human - machine interaction; when That is, when the person and the device are moving in the same direction, if 0 > That is, the person is in front of the vehicle and the distance between the person and the vehicle is less than the braking distance of the vehicle, then it is considered that there is a potential danger; when At this time, that is, when the person and the device are moving towards each other or in opposite directions. If moving towards each other , there is danger; if moving in opposite directions , keep safe; When when their running directions are perpendicular to each other = 0, and in fact, the relative safety of human - machine interaction is described by the formula after PET deformation: When it is considered dangerous; when or it is considered very safe; Then the value range of the comprehensive collision time T shown is as follows: When it is in this state, it is considered that the human-machine interaction is in a safe state, and the larger the value of T, the safer it is; If it is in this state, it is considered that there is a potential danger in the human-machine interaction; Step 5.3, select the required port operation data according to the evaluation index system, and then perform dimensionless processing on the required operation data and the comprehensive collision time T obtained in Step 5.

2. The specific process is as follows: （10）， Among them, X max is the maximum value of the j-th index of the evaluation system; X min is the minimum value of the j-th index of the evaluation system; x ij is the preprocessed data; X ij is the original value in the evaluation system; Step 5.4, calculate the information entropy and weights of the required operation data and the comprehensive collision time T obtained in Step 5.2: ， Among them, is the information entropy of the j-th index, is the weight of the j-th index; is for the i-th simulation model M i , that is, the probability of occurrence of the j-th index of the i-th evaluation object; n represents n evaluation objects, that is, all simulation models {M1, M2, M3... M i ... M n} in set B; m represents there are m evaluation indexes; Step 5.5 Calculate the positive ideal solution and the negative ideal solution ; where represents the optimal value combination of each index; represents the worst value combination of each index: （14）， （15）， }（16）， （17）， Among them, represents the positive ideal solution value of the first index; represents the positive ideal solution value of the nth index; represents the negative ideal solution value of the first index; represents the negative ideal solution value of the nth index; represents the jth index; J1 represents the set of benefit-type indicators; For the benefit-type indicator, take the maximum value of this indicator among all the schemes; For the benefit-type indicator, take the minimum value of this indicator among all the schemes; represents the set of cost-type indicators; For the cost-type indicator, take the minimum value of this indicator among all the schemes; For the cost-type indicator, take the maximum value of this indicator among all the schemes; Step 5.6, using to represent the i-th simulation model M i to the distance to the positive ideal solution; , representing the i-th simulation model M i to the distance to the negative ideal solution, calculate the relative closeness degree of the i-th evaluation object to the positive and negative ideal solutions , the formula is as follows: （18）。 8. A simulation method for the layout of a cold and dangerous goods yard in an automated terminal considering manual and unmanned mixed operations according to claim 7, characterized in that The specific content of the sixth step is as follows: If the overall layout is S i The corresponding M i is the closeness of the largest automated terminal cold and dangerous goods yard simulation model, then it is considered that M i The scale of the facilities and equipment adopted by the model and the overall layout are optimal; If the overall layout is S i The corresponding M i is not the closeness degree For the simulation model of the cold and dangerous goods yard in the largest automated terminal, the closeness degree The largest model is M k , then take M k The corresponding S k as the overall layout selected in the third step, and then complete the third step to the sixth step in sequence until both the scale of facilities and equipment and the overall layout are optimal.

9. An automated terminal cold and dangerous goods yard layout simulation system considering mixed manual and unmanned operations, characterized in that, Implement the layout simulation method of the cold and dangerous goods yard in the automated terminal according to any one of claims 1-8 through the layout simulation system of the cold and dangerous goods yard in the automated terminal, including a data acquisition module, a layout mode generation module, a scale optimization module, a simulation model construction module, a comprehensive evaluation module, and a judgment module.

10. The automated terminal cold and dangerous goods yard layout simulation system considering manual and unmanned mixed operations according to claim 9, characterized in that, The layout simulation system of the cold and dangerous goods yard in the automated terminal is specifically: The data acquisition module: used to obtain the characteristic data of the yard to be built in the automated terminal and the parameters of the facilities and equipment of the automated terminal to be built from the port environment, and preprocess the data; The layout mode generation module: accesses the data acquisition module, inputs the yard feature data of the port and the performance data of the port facilities and equipment. The layout mode generation module outputs structured data containing the overall layout set A = {S1, S2, S3,..., S i ,..., S n} based on the yard layout determination rules and the equipment layout specifications; The scale optimization module: Connects to the layout mode generation module and the data collection module, randomly selects an overall layout mode S from the layout mode generation module i , inputs the yard feature data of the port and the performance data of the port facilities and equipment into the optimization model for the scale of the cold and dangerous goods yard facilities and equipment in the automated terminal, updates the model parameters, and solves the optimization model; finally outputs the scale optimization results of each facility and equipment and feeds them back to the simulation model construction module; The simulation model construction module: Connects to the data acquisition module, layout mode generation module, and scale optimization module, inputs the yard feature data, parameters of facilities and equipment, the overall layout set, and the scale optimization results of each facility and equipment, updates the relevant parameters of the simulation model, and constructs different simulation models M1, M2, M3... M i ... M n ; Outputs various port operation data; The comprehensive evaluation module: accesses the simulation model construction module to obtain various port operation data; based on the evaluation index system, supports the parallel evaluation of multiple simulation models, and finally outputs the closeness index of each simulation model; The judgment module: accesses the comprehensive evaluation module and includes two functional units of comparison decision-making and feedback control; inputs the closeness index of each simulation model, sorts and selects different overall layout and facility equipment scale schemes; automatically identifies the optimal model with the highest closeness; realizes the dynamic collaborative optimization of the layout and facility scale.

Images