Container shipping method and device

Abstract

The invention belongs to the technical field of automatic wharfs, and particularly provides a container shipping method and device.The container shipping method comprises the steps that S1, ship stowage information of a ship is obtained, and a container weight grading range is set based on the ship stowage information; s2, obtaining container pre-allocation information of the pre-shipping containers, and clustering and grouping the pre-shipping containers based on the container pre-allocation information and the container weight grading range; s3, determining a shipping sequence of the grouped containers based on the ship stowage information and the container pre-allocation information; and S4, wharf resource information is obtained, and the grouped containers are sequentially loaded into the ships through wharf transportation instruments on the basis of the wharf resource information and the ship loading sequence. According to the container shipping method, the container shipping efficiency can be effectively improved while the stability and safety of the ship are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated terminals, and in particular to a container loading method and device. Background Art

[0002] In modern container terminal operations, the development and implementation of loading plans significantly impacts vessel operational efficiency and navigation safety. Currently, the industry generally relies on manual experience to formulate and implement loading plans. The primary challenge lies in the difficulty of coordinating multiple complex constraints. These constraints require simultaneous consideration of at least twelve core parameters, encompassing three key categories: container physical attributes (such as size, weight, and hazardous materials classification); key vessel loading parameters (including slot restrictions, stability, and strength requirements); and terminal operational resources (involving quay crane configuration and truck scheduling). This manual process is inefficient and prone to errors. Furthermore, manual loading methods struggle to precisely control loading quality and effectively achieve the optimal combination of key technical indicators, such as longitudinal strength, transverse balance, and stability height, directly impacting vessel navigation safety and seaworthiness. Furthermore, due to the lack of intelligent optimization methods for quay crane conflicts and truck routing, traditional methods often result in low equipment resource utilization. Equipment idle waiting time often accounts for over 30% of total operating time, severely restricting the terminal's overall throughput capacity and operational efficiency. Summary of the Invention

[0003] In view of this, the present invention provides a container loading method and device, which can effectively improve the container loading efficiency while improving the stability and safety of the ship.

[0004] To solve at least one of the above technical problems, the present invention adopts the following technical solutions:

[0005] In a first aspect, the present invention provides a method for loading a container onto a ship, comprising:

[0006] Step S1: Obtain the ship's loading information, and set the container weight classification range based on the ship's loading information;

[0007] Step S2: obtaining container pre-allocation information of the pre-shipping containers, and clustering the pre-shipping containers based on the container pre-allocation information and the container weight classification range;

[0008] Step S3: Determine the loading order of the grouped containers based on the ship loading information and the container pre-allocation information;

[0009] Step S4: Acquire terminal resource information, and based on the terminal resource information and the loading sequence, use terminal transport equipment to load the grouped containers into the ship in sequence.

[0010] In one embodiment of the present invention, the ship loading information includes ship load information, bay information and ship type information, and step S1 includes:

[0011] Based on the ship's load information, the dynamic range is used to set the ship's container weight classification range.

[0012] In one embodiment of the present invention, the container pre-allocation information includes container type information, weight information, and transportation information. Step S2 includes:

[0013] Step S21: Based on the weight information of each pre-shipping container, the pre-shipping containers are weight-classified using a container weight classification range to obtain a weight grade of each pre-shipping container;

[0014] Step S22: constructing a group attribute set based on the container pre-allocation information and weight grade, where the group attribute set includes static attributes, dynamic attributes, and business attributes;

[0015] Step S23: Based on the grouping attribute set, the pre-shipping containers are grouped using a clustering algorithm.

[0016] In one embodiment of the present invention, step S23 includes:

[0017] Step S231: Based on the ship type information and the group attribute set, a genetic algorithm is used to assign weights to the attributes in the group attribute set, and a plurality of attribute weight combinations matching the ship type information are generated;

[0018] Step S232: clustering the pre-shipped containers using a clustering algorithm based on attribute weight combinations to generate multiple container grouping schemes;

[0019] Step S233: construct an evaluation matrix based on the container grouping scheme, and comprehensively evaluate multiple container grouping schemes based on the evaluation matrix to obtain the optimal grouping scheme;

[0020] Step S234: group the pre-shipping containers based on the optimal grouping solution.

[0021] In one embodiment of the present invention, step S3 includes:

[0022] Step S31: establishing a three-dimensional spatial model of the ship based on the ship type information and the shell position information;

[0023] Step S32: Based on the three-dimensional spatial model of the ship, the container pre-allocation information, and the optimal grouping scheme, a three-dimensional spatial constraint solver is used to perform constraint analysis to obtain the optimal loading position of each grouped container in the optimal grouping scheme;

[0024] Step S33: Determine the loading order of the grouped containers based on the optimal loading position, bay information, and ship type information of each grouped container.

[0025] In one embodiment of the present invention, step S32 includes:

[0026] Step S321: performing location matching for each grouped container based on the three-dimensional ship model and container pre-allocation information to obtain a preliminary loading location for each grouped container;

[0027] Step S322: Based on the preliminary loading position, ship type information, bay information, and container pre-allocation information of each grouped container, a three-dimensional space constraint solver is used to perform constraint analysis;

[0028] Step S323: Adjust the preliminary loading position based on the constraint analysis result to obtain the optimal loading position for each grouped container.

[0029] In one embodiment of the present invention, the terminal transportation equipment includes quay cranes and container trucks, and the terminal resource information includes a three-dimensional operation envelope model of the quay cranes and a dynamic scheduling network for container trucks. Step S4 includes:

[0030] Step S41: determining the operating envelope of each quay crane on the wharf based on the quay crane three-dimensional operating envelope model;

[0031] Step S42: determining the dispatch status of container trucks used to transport containers at the terminal based on the container truck dynamic dispatch network;

[0032] Step S43: Dynamic planning is performed based on the operating envelope of each quay crane, the dispatching status of the container trucks, and the loading sequence of the grouped containers to obtain the optimal transportation path for each grouped container;

[0033] Step S44: Based on the loading sequence and the optimal transport path for each grouped container, the container truck and the quay crane sequentially load the grouped containers into the ship.

[0034] In one embodiment of the present invention, step S43 includes:

[0035] Step S431: Based on the operation envelope of each quay crane, a matching algorithm is used to assign the grouped containers to each quay crane in the loading order, and a loading task list of the containers to be transported by each quay crane is generated;

[0036] Step S432: Based on the scheduling status of each container truck, the tasks in the loading task list are assigned to each container truck, and the optimal transportation path for each grouped container is planned in real time based on the scheduling position of each container truck.

[0037] In one embodiment of the present invention, step S431 includes:

[0038] A spatiotemporal conflict matrix is ​​constructed based on the operation envelope of each quay crane and the loading task list. The spatiotemporal conflict matrix is ​​used to detect conflicts among the quay cranes, and the loading task list is dynamically adjusted when conflicts exist.

[0039] A second aspect of the present invention further provides a container loading device, comprising:

[0040] The weight classification module is used to obtain the ship's loading information and set the container weight classification range based on the ship's loading information;

[0041] A clustering and grouping module is used to obtain container pre-allocation information of pre-shipping containers and cluster and group the pre-shipping containers based on the container pre-allocation information and the container weight classification range;

[0042] The loading sequence determination module is used to determine the loading sequence of the grouped containers based on the ship loading information and the container pre-allocation information;

[0043] The loading module is used to obtain terminal resource information, and based on the terminal resource information and the loading sequence, the terminal transportation equipment loads the grouped containers into the ship in sequence.

[0044] The above technical solution of the present invention has at least one of the following beneficial effects:

[0045] The container loading method of the present invention clusters and groups containers based on their weight classification ranges and pre-allocation information, and then rationally arranges the loading sequence in conjunction with vessel stowage information. This method effectively improves container loading efficiency, while also preventing overloading or vessel instability, thereby enhancing navigation safety and stability. Furthermore, by acquiring terminal resource information and enabling terminal transport equipment to sequentially load the grouped containers onto the vessel based on this information and the loading sequence, this method achieves orderly scheduling of terminal transport equipment, reduces waiting time and repetitive work, and thus improves overall loading efficiency, increasing the utilization rate of loading and unloading equipment and enhancing automated operation capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of a container loading method according to one embodiment of the present invention;

[0047] Figure 2 This is a flow chart of clustering and grouping pre-shipping containers based on container pre-allocation information and container weight classification ranges in one embodiment of the present invention;

[0048] Figure 3 This is a flow chart of grouping pre-shipped containers using a clustering algorithm based on a grouping attribute set in one embodiment of the present invention;

[0049] Figure 4 A flowchart of determining the loading order of grouped containers in one embodiment of the present invention;

[0050] Figure 5 A flowchart of obtaining the optimal loading position of each grouped container in an optimal grouping solution in one embodiment of the present invention;

[0051] Figure 6 A flow chart of enabling terminal transport equipment to sequentially load grouped containers into a ship in one embodiment of the present invention;

[0052] Figure 7 The figure is a schematic structural diagram of a container loading device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0054] Reference Manual Figure 1 , which shows the process of a container loading method provided by an embodiment of the present invention, which can be applied to a container loading device. Among them, the container loading device can be, but is not limited to, various servers, personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be an independent server or a server cluster or distributed system composed of multiple servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs) and big data and artificial intelligence platforms. Specifically, Figure 1 As shown, the container loading method may include the following steps S1 to S4:

[0055] Step S1: Obtain ship loading information of a ship, and set a container weight classification range based on the ship loading information.

[0056] In an embodiment of the present invention, the ship loading information may include ship load information, bay information and ship type information. The fully loaded tonnage of the ship can be obtained based on the ship load information, and the container weight classification range of the ship can be dynamically set using the dynamic interval. Specifically, the dynamic interval refers to a numerical interval for dynamically adjusting the container weight classification threshold according to the fully loaded tonnage D of the ship. According to the fully loaded tonnage D of different ships, they can be divided into light tonnage segment, medium tonnage segment and large tonnage segment. For each tonnage segment, a corresponding function rule is set to calculate the container weight classification range corresponding to ships of different tonnages. The dynamic interval consists of multiple function segments, and its formula is:

[0057]

[0058] Among them, W i is the classification threshold for the i-th weight class (such as light box, medium box, heavy box), f i To define the sub-interval of the ship's fully loaded tonnage Mapping function on . Function segment f i This function can be a linear, polynomial, or other nonlinear function, and its specific form is determined by the ship's structural strength and stability requirements, and is not limited by the present invention. That is, for ships with larger fully laden tonnages, due to their relatively sufficient structural strength margin and stability reserve, the upper limit of the light container weight, the upper limit of the medium container weight range, and the lower limit of the heavy container weight are set accordingly. Conversely, for ships with smaller fully laden tonnages, to effectively control the overall center of gravity height and avoid the risk of local structural overload, the upper limit of the light container weight, the upper limit of the medium container weight range, and the lower limit of the heavy container weight are set accordingly. For example, if a ship with a fully laden tonnage of 20,000 tons has greater structural strength and loading redundancy, the container weight classification range can be set to less than 6 tons for light containers, 6 to 12 tons for medium containers, and more than 12 tons for heavy containers. For a ship with a fully laden tonnage of 5,000 tons, to control the center of gravity height and local overload risk, the container weight classification range can be adjusted to less than 3 tons for light containers, 3 to 6 tons for medium containers, and more than 6 tons for heavy containers.

[0059] Step S2: Obtain container pre-allocation information of the pre-shipping containers, and cluster the pre-shipping containers based on the container pre-allocation information and the container weight classification range.

[0060] In the embodiment of the present invention, the container pre-allocation information includes box type information, weight information, and transportation information. Based on the weight information of each container in the container pre-allocation information and the container weight classification range set in step S1, a corresponding weight grade can be assigned to each pre-shipped container, and a multi-dimensional grouping attribute set can be further constructed based on other container pre-allocation information. Then, based on the grouping attribute set, a clustering algorithm can be used to group the pre-shipped containers. In one possible embodiment, in conjunction with the reference specification, Figure 2 , the step S2 may include the following steps S21 to S23:

[0061] Step S21: Based on the weight information of each pre-shipping container, the pre-shipping containers are weight-classified using the container weight classification range to obtain the weight grade of each pre-shipping container.

[0062] In this embodiment, each pre-shipped container can be classified as a light container, medium container, or heavy container according to the weight classification range set in step S1 and the preset classification boundary values. Specifically, the corresponding weight value can be found using the weight information of each container in the container pre-allocation information, and the weight value can be mapped to the classification interval of the set weight classification range, so that each container obtains the corresponding weight classification identifier. This provides a weight reference basis for subsequent grouping.

[0063] Step S22: construct a group attribute set based on the container pre-allocation information and weight grade, where the group attribute set includes static attributes, dynamic attributes, and business attributes.

[0064] In this embodiment, after obtaining the weight grade of each pre-shipping container, a grouping attribute set for grouping analysis of each pre-shipping container can be further constructed based on the box type information and transportation information in the container pre-allocation information, combined with the weight grade. Specifically, each pre-shipping container grouping attribute set can include three categories: static attributes, dynamic attributes, and business attributes. Among them, static attributes are basic feature information in the box type information of each pre-shipping container that does not change during the transportation process, such as the box type information based on the ISO 6346 standard, dangerous goods classification information (such as the IMDG code), and refrigerated container identification information corresponding to each pre-shipping container; dynamic attributes are information such as the weight grade and oversize characteristics corresponding to each pre-shipping container; and business attributes are the destination port code, transit priority, and special requirements of the shipping company corresponding to each pre-shipping container in the transportation information. Therefore, by constructing a grouping attribute set, it is convenient to cluster and group each pre-shipping container according to the grouping attribute set.

[0065] Step S23: Based on the grouping attribute set, the pre-shipping containers are grouped using a clustering algorithm.

[0066] In the embodiment of the present invention, after the construction of the group attribute set is completed, the pre-shipped containers can be further grouped using a clustering algorithm based on the attribute characteristics of each container and the ship type information. Figure 3 , the step S23 may include the following steps S231 to S234:

[0067] Step S231: Based on the ship type information and the group attribute set, a genetic algorithm is used to assign weights to the attributes in the group attribute set, and a plurality of attribute weight combinations matching the ship type information are generated.

[0068] In the embodiment of the present invention, the ship type information can be used to reflect the structural parameters and stability parameters of the ship. In order to make the clustering of containers conform to the ship type, a feature space can be first constructed based on the grouping attribute set of each pre-shipped container, and the attribute of each pre-shipped container can be represented in vector form, for example, x p =[x1 (p) ,x2 (p) ,…,x d (p) ], where x i (p) is the value or code of the p-th pre-loaded container in the i-th attribute dimension, for example, it can be the box type, weight grade, etc. However, due to the different attention paid to the attributes of each dimension by different ship types, the importance of each attribute in the grouping attribute set of pre-loaded containers is not the same. For example, in some ship types, the weight grade is more important than the box type information, while in some ship types, the dangerous goods category information is more important than the weight grade. Therefore, multiple weight strategies can be generated in advance, such as the weight grade priority strategy and the box type priority strategy. Furthermore, in order to characterize the relative importance of each dimension attribute in each weight strategy in the subsequent similarity calculation, that is, to express the proportion of the attribute in the grouping judgment of the corresponding weight strategy, multiple groups of attribute weight combinations w = [w1, w2,…, w d ], where w i It is used to indicate the contribution of the i-th attribute dimension to the overall similarity. The weight w in each attribute weight combination is i The value of can be randomly set according to the predetermined value rule. The predetermined value rule is that the more important attribute in the corresponding weight strategy has a higher value, and the opposite value has a lower value. Each weight w i The value of is limited to the interval [0,1], and the sum of the weights of all attributes meets the normalization condition, that is,

[0069] Furthermore, the multiple sets of attribute weight combinations w corresponding to each weight strategy can be used as the initial population of the genetic algorithm for iteration, and each set of attribute weight combinations w corresponds to a chromosome individual in the initial population. During the iteration process, the population can be continuously updated and evolved through operators such as selection, crossover, and mutation to generate new attribute weight combinations. After generation, each set of attribute weight combinations can be evaluated based on the fitness function that matches the ship type information, so as to select the optimal attribute weight combination corresponding to each weight strategy. Specifically, the parent generation can be first selected from the current population based on the fitness, and then the two parent attribute weight combination vectors can be linearly cross-combined to form a new offspring attribute weight combination. The offspring attribute weight combination has a new weight in the i-th attribute dimension. for:

[0070]

[0071] Among them, a is the weight fusion coefficient, is the i-th weight component of the first parent, is the i-th weight component of the second parent. Then, the weight component w in the weight combination of the offspring attributes can be i Mutation is performed, and the weight component w after mutation is i for:

[0072] w i :=w i +δ,δ~N(0,σ 2 )

[0073] Among them, δ is a random disturbance, σ is a variation intensity parameter, δ has a mean of 0 and a variance of σ 2 Normal distribution, that is, δ~N(0,σ 2 After the new generation of combinations after mutation is normalized, it can enter the next round of fitness evaluation until the optimal fitness value in the population exceeds the set threshold and then the optimal attribute weight combination corresponding to the weight strategy is output.

[0074] Step S232: clustering the pre-shipped containers using a clustering algorithm based on the attribute weight combination to generate multiple container grouping schemes.

[0075] In this embodiment, after obtaining multiple attribute weight combinations corresponding to different weighting strategies, a weighted attribute similarity model can be constructed based on each attribute weight combination w, and pre-shipped containers can be clustered based on this model. Specifically, for any two containers q and p, the weighted attribute similarity distance between them is:

[0076]

[0077] in, is the distance function of the i-th attribute dimension. This distance function can be expressed using the standardized Euclidean difference to characterize the difference between the attributes of two containers q and p. The containers can then be clustered using a density clustering algorithm based on the weighted similarity distance. Specifically, a neighborhood can be defined for the pre-shipped container q, which contains all containers p whose similarity distance to the pre-shipped container is less than a threshold ε, that is:

[0078] N ε (q) = {x (p) |Dis(q, p)≤ε}

[0079] If the number of containers in the neighborhood of the pre-loaded container q is not less than the minimum preset value, the pre-loaded container q can be marked as a core container q, and then all containers p in the neighborhood of the core container q are visited one by one recursively, and then the core container q and the containers p in the neighborhood are formed into a cluster c. If the container p is also a core container, that is, the number of containers in its neighborhood is not less than the minimum preset value, all containers in the area of ​​the container p can be further visited, and the unassigned containers are also added to the cluster c. Then continue the above steps until all density-reachable and density-connected containers are visited and included in the cluster c. All containers that cannot be density-reachable from any core container, if they are not core containers themselves, are marked as noise points. By traversing all core containers and completing the cluster expansion process, a set of cluster groups C can be output:

[0080] C={c1,c2,···,c k}

[0081] The cluster group C is a container grouping scheme, and each cluster c in the cluster group is a group.

[0082] Step S233: construct an evaluation matrix based on the container grouping schemes, and perform a comprehensive evaluation on multiple container grouping schemes based on the evaluation matrix to obtain an optimal grouping scheme.

[0083] In an embodiment of the present invention, an optimal value scheme and a worst value scheme can be constructed respectively based on all evaluation indicators, and each candidate container grouping scheme is measured for distance with the two benchmark schemes in a multidimensional indicator space at the same time, and the positive distance between each candidate container grouping scheme and the optimal value scheme and the negative distance between each candidate container grouping scheme and the worst value scheme are calculated. Then, the degree to which the container grouping scheme is close to the optimal value scheme in terms of overall indicator performance is obtained through a pre-set closeness coefficient. The higher the closeness coefficient, the better the overall performance of the container grouping scheme, and then all container grouping schemes are ranked. Specifically, a set of representative evaluation indicators can be first extracted for each container grouping scheme, and the evaluation indicators include but are not limited to container type consistency, dangerous goods consistency, destination port consistency, etc. An evaluation matrix R can be constructed based on each container grouping scheme and its corresponding evaluation indicators:

[0084] R=[r ij ] m*n

[0085] Among them, m is the number of container grouping schemes, n is the number of evaluation indicators, r ij is the score of the i-th solution on the j-th evaluation index. Based on this evaluation matrix, the weighted decision matrix v can be further constructed ij :

[0086]

[0087] Among them, z j is the decision weight of the jth evaluation indicator, which can be set based on actual needs. Then, based on the weighted decision matrix v ij Obtain the forward distance between each candidate container grouping solution and the optimal solution and the negative distance from the worst solution Specifically, the forward distance The formula is:

[0088]

[0089] Negative distance The formula is:

[0090]

[0091] Then we can base it on the forward distance and negative distance Get the closeness coefficient A of each container grouping scheme i :

[0092]

[0093] When A i The closer it is to 1, the better the overall performance of the container grouping scheme is, and the closeness coefficient A of each container grouping scheme can be calculated. i Sort all container grouping schemes to obtain the optimal grouping scheme.

[0094] Step S234: group the pre-shipping containers based on the optimal grouping solution.

[0095] Step S3: Determine the loading order of the grouped containers based on the ship loading information and the container pre-allocation information.

[0096] In an embodiment of the present invention, a three-dimensional spatial model of a ship can be constructed using the ship type information in the ship loading information, and then the three-dimensional spatial model of the ship, the container pre-allocation information and the optimal grouping scheme can be used to perform multiple constraint analysis using a three-dimensional spatial constraint solver to obtain the optimal loading position of each grouped container, wherein the three-dimensional spatial constraints may include but are not limited to structural strength restrictions, stability requirements, center of gravity distribution, isolation of hazardous materials, and layout of refrigerated units. Then, based on the optimal loading position and ship type information of each grouped container, a loading sequence that conforms to the ship type structure, loading rules and operation logic can be generated. In one possible embodiment, in conjunction with the reference specification, Figure 4 , the step S3 may include the following steps S31 to S33:

[0097] Step S31: Establish a three-dimensional spatial model of the ship based on the ship type information and the shell position information.

[0098] In this embodiment, the ship type information can be used to characterize the overall geometric shape of the hull and the boundaries of the available loading area, such as the ship's length, width, draft, number of deck layers, hatch cover structure, and loading line restrictions; while the bay information can characterize the structure of the loading units on the ship, such as the cabin division, row and column numbers, layer numbers, and the physical dimensions, support capacity, and load limit of each bay. Based on the ship type information and bay information, a three-dimensional coordinate system can be used to grid and discretely represent the space on the ship to establish a three-dimensional spatial model of the ship. Each grid cell can represent a standard container space, and the attributes of each grid cell record the structural affiliation (e.g., cabin / deck), location identification, size specifications, and constraints such as support capacity and load limit. In this way, the spatial layout structure of the ship can be fully reflected.

[0099] Step S32: Based on the three-dimensional spatial model of the ship, the container pre-allocation information and the optimal grouping scheme, a three-dimensional spatial constraint solver is used to perform constraint analysis to obtain the optimal loading position of each grouped container in the optimal grouping scheme.

[0100] In this embodiment, based on the established three-dimensional ship model, combined with the pre-allocation information and optimal grouping scheme of the containers, the container positions of each group can be matched, and the three-dimensional space constraint solver can be used to perform constraint analysis to obtain the optimal loading position of each grouped container, thereby achieving unified optimization of loading safety and operation efficiency. In one possible embodiment, in conjunction with the reference specification, Figure 5 , the step S32 may include the following steps S321 to S323:

[0101] Step S321: perform container location matching on each grouped container based on the three-dimensional spatial model of the ship and the container pre-allocation information to obtain a preliminary loading position of each grouped container.

[0102] In this embodiment, preliminary slot matching can be performed on the grouped containers based on the constructed three-dimensional ship model and each container's pre-assigned container information. Specifically, the size category of each container, such as a 20-foot container or a 40-foot container, can be determined based on the container type information in the pre-assigned container information. The system then searches the three-dimensional ship model for available grid cells that can accommodate the container and assigns a preliminary loading location to each container.

[0103] Step S322: Based on the preliminary loading position, ship type information, bay information and container pre-allocation information of each grouped container, a three-dimensional space constraint solver is used to perform constraint analysis.

[0104] In this embodiment, the three-dimensional spatial constraint solver can be used to perform constraint analysis on preset key loading constraints, which include longitudinal structural strength constraints, transverse load balance constraints, and center of gravity height control constraints. First, a load distribution function of the ship in the longitudinal direction can be constructed based on the ship type information and the bay position information. The load distribution function can be used to reflect the unit linear mass of the longitudinal projection of the container at all preliminary loading positions in the longitudinal direction of the ship. Specifically, the hull can be divided into several longitudinal unit intervals based on the ship type information and the bay position information. For example, each bay position is used as a basic unit, and each interval is represented by a center coordinate x i In each interval, the total weight of the container assigned to the bay can be calculated based on the initial loading position. i and convert it into load density per unit length q(x i ), load density per unit length q(x i ) is calculated as:

[0105]

[0106] Where Δx is the length of the interval in the longitudinal direction. Further, the load density per unit length q(x i ) Construct the load distribution function {x i ,q(x i )}. Then, based on the load distribution function and the longitudinal structural mechanics model of the hull in the ship type information, the longitudinal shear force V(x) and bending moment M(x) at each longitudinal section of the ship can be calculated. The calculation formula of the longitudinal shear force V(x) is:

[0107]

[0108] The calculation formula of bending moment M(x) is:

[0109]

[0110] Where x is a certain cross-sectional position in the longitudinal direction of the ship, L is the total length of the ship, q(s) is the load density per unit length in the longitudinal direction of the ship, and d(s) is the integral length of the cross-sectional area. Based on the bending moment M(x), the bending moment curve of the ship when each container is at the initial loading position can be constructed. Then, based on the ship type information, the ship's allowable bending moment limit curve M is obtained to calibrate the maximum safe load capacity of each longitudinal position. lim (x i ), by comparing point by point, determine whether there is any position x i Satisfies the inequality:

[0111] |M(x i )|>M lim (x i )

[0112] If the inequality is satisfied, it means that when each container is at the initial loading position, the bending moment at that position has exceeded the limit, and there is a risk of structural instability or yield. The corresponding position of this position can be marked as a structural load conflict position, waiting for subsequent adjustment of the position of the container at this position.

[0113] Furthermore, after completing the longitudinal structural strength constraint analysis, a transverse loading balance analysis can be performed on each transverse row of the ship, i.e., the same row number in the bay, based on the ship type information, preliminary loading position, and bay information to evaluate whether the moment balance requirements are met in the transverse direction. Specifically, each transverse row can be first regarded as being divided into two symmetrical areas, the port area and the starboard area, with the centerline of the hull as the axis of symmetry. For each transverse row r, the weight information of all allocated containers on the left and right sides of the row can be counted in turn, and the overturning moment can be calculated in combination with the transverse position of each container position relative to the centerline, i.e., the transverse force arm. The overturning moment M r The calculation formula is:

[0114]

[0115] in, is the weight of the i-th container in the port area of ​​row r, is the lateral moment arm of the i-th container position in the port area of ​​row r relative to the centerline, is the weight of the jth container in the starboard slot in row r, is the lateral moment arm of the jth container position in the starboard area of ​​row r relative to the centerline, n L and n R The number of containers in the port and starboard areas in row r are respectively. Then, the preset allowable transverse unbalance moment threshold can be obtained based on the ship type information. And judge by the following inequality:

[0116]

[0117] If this inequality holds, row r is laterally unbalanced when the containers are initially loaded, which can cause the ship to heel or roll during navigation. This row can be marked as unbalanced, pending subsequent adjustments to the container positions.

[0118] Furthermore, after completing the lateral load balance constraint analysis, the center of gravity height control constraint analysis can also be performed on the ship to ensure that the ship's longitudinal stability safety requirements are met. Specifically, the ship can be divided into multiple vertical loading units based on the ship type information, for example, by cabin number or deck level, and each allocated container can be regarded as a discrete mass point with fixed mass and spatial position based on the initial loading position, and the total center of gravity height H can be calculated. cg , total center of gravity height H cg The calculation formula is:

[0119]

[0120] Among them, W i is the weight of the i-th container on the vertical loading unit, h i is the vertical height of the center of gravity of the container relative to the ship's hull reference plane, such as the cargo hold floor, and N is the total number of containers on the vertical loading unit. Then, based on the ship's model information, we can further obtain the distance KB from the ship's reference plane to the center of buoyancy and the distance BM from the ship's center of buoyancy to the initial metacentric center. This allows us to obtain the initial metacentric height GM of each container in its initial loading position. The calculation formula for the initial metacentric height GM is:

[0121] GM=KB+BM-H cg

[0122] Then, the initial stability height GM of each container at the initial loading position is compared with the stability critical value GM preset in the ship type information. min For comparison:

[0123] GM <GM min

[0124] If the above inequality holds, it indicates that the ship's longitudinal stability is insufficient when the containers are in their initial loading positions. The vertical loading unit can be marked as a high-risk stacking area, pending subsequent adjustments to the positions of the containers in that vertical loading unit.

[0125] Step S323: Adjust the preliminary loading position based on the constraint analysis result to obtain the optimal loading position for each grouped container.

[0126] In this embodiment, after performing constraint analysis on the longitudinal structural strength constraints, lateral load balance constraints, and center of gravity height control constraints, the initial loading positions of each grouped container can be dynamically adjusted based on the analysis results to obtain an optimal loading configuration that meets loading safety, operational efficiency, and loading and unloading sequence requirements. Specifically, the bending moment curve can be referenced to first redistribute containers in a given location to adjacent bays with higher structural margins until the bending moment falls below the allowable bending moment limit. Secondly, the positions of containers in unbalanced configurations can be swapped, for example, by swapping containers of the same weight class in the same row to reduce the lateral moment imbalance until the lateral moment is balanced. The positions of containers in high-risk stacking areas can then be adjusted, prioritizing the movement of medium or heavy containers from upper bays to lower levels or within the hold, while shifting light containers upwards as appropriate to free up space in lower locations without compromising stability. These steps can be repeated, adjusting the container positions until all three-dimensional spatial constraints are met. The position of each container that meets these three-dimensional spatial constraints is then used as the optimal loading position for each grouped container.

[0127] Step S33: Determine the loading order of the grouped containers based on the optimal loading position, bay information, and ship type information of each grouped container.

[0128] In this embodiment, the loading order of the grouped containers can be determined based on the optimal loading position of each grouped container, in the order from bottom to top, from the cabin to the deck, and from the bow to the stern. Specifically, the containers can be logically sorted based on the coordinate information of the optimal loading position of the container in the three-dimensional model, such as the layer number, the bay number, and the row and column number. First, hierarchical scheduling is carried out according to the vertical level where the cabin is located, and the bottom container of the cabin is loaded first to ensure the overall stacking stability and the supporting capacity of the lower layer. Within the same vertical layer, the order of loading the stern first and then the bow is further arranged according to the longitudinal position of the bay to avoid the rear-loaded container blocking the front operation path. In the horizontal direction, the containers can be stacked in order based on the hatch opening direction, loading channel layout and other operational constraints in the ship type information to ensure the accessibility of the terminal transportation equipment and maximize the operation efficiency.

[0129] Step S4: Acquire terminal resource information, and based on the terminal resource information and the loading sequence, use terminal transport equipment to load the grouped containers into the ship in sequence.

[0130] In an embodiment of the present invention, terminal transportation equipment includes quay cranes and container trucks, and terminal resource information includes a three-dimensional operation envelope model of the quay cranes and a dynamic scheduling network for container trucks. After obtaining the loading order of the grouped containers, the quay cranes and container trucks can be scheduled based on the dynamic resource information to complete the orderly loading of containers from the yard to the ship. In one possible embodiment, Figure 6 As shown, step S4 may include the following steps S41 to S44:

[0131] Step S41: Determine the operating envelope range of each quay crane on the wharf based on the quay crane three-dimensional operating envelope model.

[0132] In this embodiment, a 3D operational envelope model of each quay crane can be first obtained to determine the operational range covered by each quay crane under the current berth layout. This 3D operational envelope model is a structured representation of the quay crane's operational spatial capabilities and can include parameters such as the crane's maximum reach, setback limit, lifting height, lowering depth, spreader horizontal trajectory range, and operating track length. Based on these parameters, combined with the vessel's berthing position, berth length and width, and the safe operating intervals between adjacent quay cranes, an operational envelope for each quay crane can be constructed—that is, the operational area covered by each quay crane in three dimensions.

[0133] Specifically, the ship's bay coordinate system can be used as the reference coordinate system to perform spatial mapping on the maximum operating leading and trailing edges, and upper and lower limits of lifting of each quay crane, thereby determining the operable bay range and corresponding floor height range of each quay crane.

[0134] Step S42: Determine the dispatch status of container trucks used to transport containers at the terminal based on the container truck dynamic dispatch network.

[0135] In this embodiment, the dispatching status information of all container trucks available for transportation operations in the terminal can be obtained in real time based on the dynamic dispatching network for container trucks. The dynamic dispatching network for container trucks refers to a dynamic dispatching network composed of data such as the location information, task queue, operation status, path traffic conditions, and yard-quay crane operation instructions of container trucks. The dynamic dispatching network for container trucks can realize global status tracking and task allocation optimization of container trucks by linking with the terminal's dispatching platform, vehicle management system (TOS), and onboard positioning equipment of container trucks (such as GPS, RTLS, etc.). Specifically, based on the dynamic dispatching network for container trucks, the current position of each container truck, the current execution task (such as loading, standby, return, etc.), path congestion and other indicators can be comprehensively analyzed to calculate its dispatching priority and expected available time period.

[0136] Step S43: Perform dynamic planning based on the operating envelope of each quay crane, the dispatching status of the container truck, and the loading sequence of the grouped containers to obtain the optimal transportation path for each grouped container.

[0137] Dynamic planning can be performed by humans according to certain rules or it can be performed automatically.

[0138] In a possible embodiment, step S43 may include the following steps S431 and S432:

[0139] Step S431: Based on the operation envelope of each quay crane, a matching algorithm is used to distribute the grouped containers to each quay crane in the loading order, and a loading task list of the containers to be transported by each quay crane is generated.

[0140] In this embodiment, based on the operation envelope of each quay crane, we can first screen out the candidate quay crane set that can be assigned to each container to be loaded, build a feasible matching relationship between the container task set and the quay crane operation set, and build a matching cost matrix based on parameters such as operation distance, resource load, and scheduling priority. Specifically, we can first build the matching cost C caused by assigning the i-th container task to the j-th quay crane. ij , matching cost C ij The calculation formula is:

[0141] C ij =α·D ij +β·L j +γ·P i

[0142] Among them, D ij L is the lateral operating distance between the target position of the i-th container and the current position of the j-th quay crane; j is the current assigned task volume of the jth quay crane, which can reflect its workload; P iis the order of the i-th container; α, β and γ are preset weight factors, which can be used to regulate the balance between different scheduling objectives and the matching cost C ij The smaller the cost, the better. The cost values ​​of all tasks and quay cranes are then aggregated to construct a cost matrix C of size n*m. Each element in the matrix is ​​a real number, representing the matching cost corresponding to the task assignment, where n is the number of container loading tasks to be assigned, and m is the number of quay cranes. Once the cost matrix is ​​constructed, the Hungarian algorithm can be used to find the optimal matching solution. This involves first normalizing each row and column of the matrix so that the minimum cost element is located near the diagonal. A minimal set of lines covering all zero elements is then constructed, assigning a quay crane to each container task. If coverage is not complete, meaning that zero-cost matching cannot be found for all tasks, the uncovered areas are added or subtracted, gradually optimizing the current matching until the optimal solution is achieved. In cases where the number of quay cranes and tasks is inconsistent, virtual tasks or virtual quay crane nodes can be introduced to fill the gap. Once the solution is complete, the optimal matching cost results can be used to determine the container task list required for each quay crane. Each task list can include information such as the container number to be loaded by the quay crane, the target container location, the loading sequence, and the start and end time windows for the operation. This ensures the reasonable distribution of the quay crane operating load.

[0143] Furthermore, a spatiotemporal conflict matrix can be constructed based on the operation envelope of each quay crane and the loading task list. The spatiotemporal conflict matrix can be used to detect conflicts between each quay crane and dynamically adjust the loading task list when conflicts exist. Specifically, the operation envelope of each quay crane can be the set of bay areas that it can cover within a certain period of time. This set B j (t) can be expressed as:

[0144]

[0145] Among them, [t,t+Δk j ] is the available operating time period of quay crane j, and They represent the bay area number intervals of the current left and right operating ranges of quay crane j, Δt j The theoretical empirical time required for quay crane j to complete the current task. Based on the set of bay areas that each quay crane can cover within a certain period of time, a two-dimensional spatiotemporal conflict matrix C can be constructed to describe whether there is a conflict in the operating area between quay cranes. Each element C in the matrix ij It is used to indicate whether any two quay cranes have cross-operations in a specific operation period. For example, B j (t) and B i(t) intersect, that is, the operating time periods of quay crane i and quay crane j overlap at least partially, and the operating bay ranges (lateral spatial areas) of quay crane i and quay crane j also overlap, C ij If it is 1, it means that the quay crane j and quay crane i will collide, otherwise C ij If the value is 0, it means that quay crane j and quay crane i will not collide. In the event of a conflict, the two conflicting tasks can be first identified and staggered based on their corresponding operation time windows, bay positions, and the minimum lateral safety distance between the quay cranes to avoid collision.

[0146] Step S432: Based on the scheduling status of each container truck, the tasks in the loading task list are assigned to each container truck, and the optimal transportation path for each grouped container is planned in real time based on the scheduling position of each container truck.

[0147] In this embodiment, real-time planning can be performed by humans according to certain rules or automatically. A scheduling cost function can be constructed to evaluate the cost of trucks performing tasks in the loading task list, and to achieve optimal allocation with the lowest cost. Specifically, a truck set K including all trucks at the terminal can be constructed, where K = {k1, k2, ..., k m Where m is the number of container trucks, and each container truck can be assigned k based on its dispatch status. j Set the corresponding state, each truck k j The state can include the current position p j , current task load L j And idle state. The container trucks with idle state "idle" can be filtered out from the container truck set K to perform the tasks in the loading task list, and the loading time of each container truck k is calculated. j Current location p j Calculate the distance D to the estimated target task starting point ij , where i is the number of the task in the loading task list, and then construct the scheduling cost function C ij :

[0148] C i j=α·D ij +β·L j +γ·W i

[0149] Among them, C ij is the total cost of assigning the i-th task to the j-th truck, and the current task load L j Can be used as the cost function C ij The load balancing item is used to avoid congestion or resource overload caused by frequent scheduling of some container trucks. i=i represents the urgency of the i-th task, such as tasks close to the shipping deadline or tasks with higher priority. This can be set based on on-site operational conditions. α, β, and γ are preset weighting factors that can be used to balance different scheduling objectives. All available trucks and pending tasks are then aggregated to construct a cost matrix. Based on this cost matrix, the task-truck combination with the lowest cost is selected for allocation. For example, if the number of tasks and trucks is close, the Hungarian algorithm can be used to achieve minimum cost matching. If there are far more tasks than trucks, a priority ranking method can be used to select the truck with the lowest current cost for each high-priority task. In this embodiment, tasks for containers with earlier loading orders receive higher priority. The status of each truck is updated after each round of allocation, but this is not a limitation of the present invention. After assigning tasks from the shipping task list to each truck, a shortest path algorithm can be used to plan the optimal transportation route for each grouped container in real time. This real-time planning can be performed manually according to specific rules or automatically. Specifically, the road network within the terminal can be abstracted as a weighted directed graph. The cost of each edge in the weighted directed graph can include geometric distance, traffic congestion status, and restricted operating area information. Based on the weighted directed graph, each path segment in the terminal can be given a dynamically adjusted traffic weight w ij (t):

[0150] w ij (t) = d ij ·(1+θ·ρ ij (t)+η·κ ij (t))

[0151] Among them, w ij (t) is the terminal road node v at time t i to v j The edge weight value of node v i to v j The traffic weight of the path between them. The lower the weight, the better the traffic condition of the road. ij is the geometric distance of the road segment, ρ ij (t) is the real-time congestion factor of the current path, such as traffic density, vehicle frequency, etc., κ ij (t) is the restricted traffic factor, for example, whether the road is under construction, prohibited, or has scheduling priority, and θ and η are the empirical weight coefficients of the past traffic status of the terminal. ij (t), a weighted directed graph can be constructed based on the paths in the terminal and the corresponding traffic weights of each path. Then, a search can be performed on the weighted directed graph to obtain the container truck k j From the current position p jThe path with the lowest total traffic weight between the starting point and the destination is the shortest path after considering traffic conditions and restrictions. This can effectively improve transportation efficiency.

[0152] Step S44: Based on the loading sequence and the optimal transport path for each grouped container, the container truck and the quay crane sequentially load the grouped containers into the ship.

[0153] In this embodiment, for fully automated terminals using unmanned container trucks and automated quay cranes, the loading task list, container truck path, and quay crane operation instructions can be automatically sent to the relevant control terminals in a standard communication protocol format (such as JSON, PROFINET, or Modbus) by connecting to the dispatching control center (such as the TOS system, AGV control system, and quay crane control system). Specifically, the system will package the starting position, target bay position, path planning trajectory, and time window parameters of each unmanned container truck to generate a control instruction stream, and maintain real-time communication with the dispatching server through the on-board control unit to achieve closed-loop control of vehicle acceleration, deceleration, steering, obstacle avoidance, and precise parking. At the same time, for automated quay cranes, the system generates a control sequence for quay crane lifting, height adjustment, trolley operation, and spreader rotation based on the loading task sequence, bay position coordinates, and spreader information, and pushes it to the quay crane controller for execution in real time. For semi-automated terminals using manually driven container trucks and manually operated quay cranes, instructions can be provided to dispatchers and operation drivers in the form of operation instructions. Specifically, information can be displayed via a central control terminal, an operation interface, or a mobile operation terminal, including each truck's dispatch task number, container number, specific locations of pickup and drop-off points, recommended travel routes, and estimated arrival times. Drivers can then navigate the interface to complete the handling operation. For manual quay crane operators, the system provides loading instructions for each task, including target container position, spreader type, lifting time window, and queue order. Operators manually control the equipment to carry out the operation based on the dispatch recommendations. This is not a limitation of the present invention.

[0154] In other embodiments of the present invention, loading plans can be dynamically adjusted based on actual operational needs and on-site resource availability. Specifically, real-time resource status information for the current terminal can be obtained. If this information changes, such as a change in the number of available quay cranes or container trucks, the operation plan can be regenerated based on a heuristic search. Specifically, the objective function is to minimize total operation delay, resulting in an updated task allocation scheme for quay cranes and container trucks, as well as task time windows. The specific implementation details of the above method are common knowledge and will not be elaborated here.

[0155] In summary, the container loading method of the present invention effectively improves container loading efficiency by clustering containers based on weight classification ranges and pre-allocation information, and then rationally arranges the loading sequence in conjunction with ship stowage information. This prevents overloading or ship instability, thereby enhancing navigation safety and stability. Furthermore, by obtaining terminal resource information and then enabling terminal transport equipment to sequentially load the grouped containers onto the ship based on this information and the loading sequence, this method achieves orderly scheduling of terminal transport equipment, reduces waiting time and repetitive work, thereby improving overall loading efficiency, and enhancing the utilization rate and automation capabilities of loading and unloading equipment.

[0156] Second, as Figure 7 As shown, the present invention further provides a container loading device 700, which may include:

[0157] The weight classification module 710 is used to obtain the ship's loading information and set the container weight classification range based on the ship's loading information;

[0158] Clustering and grouping module 720, which is used to obtain container pre-allocation information of pre-shipping containers and cluster and group the pre-shipping containers based on the container pre-allocation information and the container weight classification range;

[0159] The loading sequence determining module 730 is used to determine the loading sequence of the grouped containers based on the ship loading information and the container pre-allocation information;

[0160] The loading module 740 is used to obtain terminal resource information and, based on the terminal resource information and the loading sequence, enable the terminal transportation equipment to load the grouped containers into the ship in sequence.

[0161] It should be noted that the devices provided in the above embodiments are only illustrated by the division of the above functional modules when implementing their functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the devices provided in the above embodiments and the corresponding method embodiments are based on the same concept. The specific implementation process is detailed in the corresponding method embodiments and will not be repeated here.

[0162] Unless otherwise defined, the technical or scientific terms used in the present invention shall have the usual meanings understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0163] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A container loading method, characterized in that: include: Step S1: Obtain ship loading information of a ship, and set a container weight classification range based on the ship loading information; Step S2: obtaining container pre-allocation information of the pre-shipping containers, and clustering the pre-shipping containers based on the container pre-allocation information and the container weight classification range; Step S3: determining the loading order of the grouped containers based on the ship loading information and the container pre-allocation information; Step S4: Acquire terminal resource information, and based on the terminal resource information and the loading sequence, use terminal transportation equipment to load the grouped containers into the ship in sequence.

2. The container loading method according to claim 1, characterized in that: The ship loading information includes ship load information, bay information and ship type information, and the step S1 includes: Based on the ship's load information, a dynamic interval is used to set the ship's container weight classification range.

3. The container loading method according to claim 2, characterized in that: The container pre-allocation information includes container type information, weight information, and transportation information. Step S2 includes: Step S21: Based on the weight information of each pre-shipping container, the pre-shipping containers are weight-classified using the container weight classification range to obtain a weight grade of each pre-shipping container; Step S22: constructing a group attribute set based on the container pre-allocation information and the weight grade, wherein the group attribute set includes static attributes, dynamic attributes, and business attributes; Step S23: grouping the pre-shipping containers using a clustering algorithm based on the grouping attribute set.

4. The container loading method according to claim 3, characterized in that: The step S23 includes: Step S231: Based on the ship type information and the group attribute set, a genetic algorithm is used to assign weights to the attributes in the group attribute set, and a plurality of attribute weight combinations matching the ship type information are generated; Step S232: clustering the pre-shipped containers using a clustering algorithm based on the combination of multiple attribute weights to generate multiple container grouping schemes; Step S233: constructing an evaluation matrix based on the container grouping scheme, and comprehensively evaluating multiple container grouping schemes based on the evaluation matrix to obtain an optimal grouping scheme; Step S234: grouping the pre-shipping containers based on the optimal grouping solution.

5. The container loading method according to claim 4, characterized in that: The step S3 comprises: Step S31: establishing a three-dimensional space model of the ship based on the ship type information and the shell position information; Step S32: Based on the three-dimensional spatial model of the ship, the container pre-allocation information, and the optimal grouping scheme, a three-dimensional spatial constraint solver is used to perform constraint analysis to obtain an optimal loading position for each grouped container in the optimal grouping scheme; Step S33: Determine the loading order of the grouped containers based on the optimal loading position of each grouped container, the container position information, and the ship type information.

6. The container loading method according to claim 5, characterized in that: The step S32 includes: Step S321: performing location matching for each grouped container based on the three-dimensional ship model and the container pre-allocation information to obtain a preliminary loading location for each grouped container; Step S322: Based on the preliminary loading position of each grouped container, the ship type information, the bay position information, and the container pre-allocation information, a three-dimensional space constraint solver is used to perform constraint analysis; Step S323: Adjust the preliminary loading position based on the constraint analysis result to obtain the optimal loading position for each grouped container.

7. The container loading method according to claim 6, characterized in that: The terminal transportation equipment includes quay cranes and container trucks, and the terminal resource information includes a three-dimensional operation envelope model of the quay cranes and a dynamic scheduling network of container trucks. Step S4 includes: Step S41: determining the operating envelope range of each quay crane on the wharf based on the quay crane three-dimensional operating envelope model; Step S42: determining the dispatching status of the container trucks for transporting containers at the terminal based on the container truck dynamic dispatching network; Step S43: performing dynamic planning based on the operating envelope of each quay crane, the dispatching status of the container trucks, and the loading sequence of the grouped containers to obtain the optimal transportation path for each grouped container; Step S44: Based on the loading sequence and the optimal transport path of each grouped container, the container truck and the quay crane sequentially load the grouped containers into the ship.

8. The container loading method according to claim 7, characterized in that: The step S43 includes: Step S431: Based on the operation envelope of each quay crane, a matching algorithm is used to assign the grouped containers to each quay crane in the loading order, and a loading task list of the containers to be transported by each quay crane is generated; Step S432: Based on the scheduling status of each container truck, the tasks in the loading task list are assigned to each container truck, and the optimal transportation path for each grouped container is planned in real time based on the scheduling position of each container truck.

9. The container loading method according to claim 8, characterized in that: The step S431 includes: A spatiotemporal conflict matrix is ​​constructed based on the operation envelope of each quay crane and the loading task list, and the spatiotemporal conflict matrix is ​​used to perform conflict detection on each quay crane, and the loading task list is dynamically adjusted when a conflict exists.

10. A container loading device, characterized in that: include: A weight classification module, the weight classification module is used to obtain ship loading information of the ship and set a container weight classification range based on the ship loading information; a clustering and grouping module, configured to obtain container pre-allocation information of the pre-shipping containers, and cluster and group the pre-shipping containers based on the container pre-allocation information and the container weight classification range; a loading sequence determining module, configured to determine a loading sequence of the grouped containers based on the ship stowage information and the container pre-allocation information; The loading module is used to obtain terminal resource information and, based on the terminal resource information and the loading sequence, enable terminal transportation equipment to load the grouped containers into the ship in sequence.

Images