A simulation system and method for automatic identification and dynamic adaptation of a sea import channel of a port group

Abstract

This invention provides an automatic identification and dynamic adaptation simulation system and method for maritime import channels in port clusters, comprising: a multi-source data fusion processing module for constructing a multi-source data fusion database; an automatic channel identification and classification module for obtaining five-level channel identification results; a channel-cargo matching module for matching cargo types with channels and obtaining matching results; a dynamic adaptation simulation module for simulating the entire process of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution based on the five-level channel identification results and matching results, and obtaining simulation results; and a visualization and decision support module for calculating channel saturation based on the simulation results, generating a five-level penetrating heat map, quantitative indicators, and optimization suggestions. This invention provides precise decision support for port cluster planning and port-industry collaboration.

Description

Technical Field

[0001] This invention belongs to the field of port simulation technology, specifically relating to an automatic identification and dynamic adaptation simulation system and method for port cluster maritime import channels. Background Technology

[0002] With the integrated development of global industrial, supply, and transportation chains, port clusters' maritime import channels, as the core carriers for transporting bulk commodities (iron ore, coal, etc.), directly impact logistics scheduling efficiency, port resource utilization, and enterprise terminal unloading coordination capabilities due to their identification accuracy, classification rationality, and simulation adaptability. Currently, port channel identification and logistics simulation technologies have been applied to some extent in port operations, mainly focusing on the following aspects: (1) Channel identification technology: Existing technologies are mostly based on AIS ship trajectory data to realize route extraction and channel identification within a single port or between regional ports. They mainly use simple trajectory clustering or statistical methods, focusing on the linear channel identification of "port-to-port", and have not formed a multi-level, full-link identification system.

[0003] (2) Channel classification method: Traditional channel classification is mostly based on the purpose of the waterway (such as general waterway and special waterway), without combining the characteristics of the cargo and the needs of enterprises for refined classification. It is impossible to accurately distinguish between special channels for bulk materials and regular multi-cargo channels, resulting in waste of special channel resources and frequent congestion in regular channels.

[0004] (3) Application of simulation models: Existing simulation models mostly focus on local simulation of port berths and waterways, using fixed parameter settings, without linking terminal enterprises, storage yards and other nodes, lacking dynamic adaptability, and without detailed quantitative calculation formulas. The simulation results have low consistency with the actual logistics scenario, making it difficult to support accurate decision-making.

[0005] Based on the current state of technology, existing port cluster maritime import channel identification and simulation technologies have the following significant shortcomings, failing to meet the needs of full-chain, refined, and dynamic management: (1) Incomplete identification hierarchy: It only covers channels between single ports or regional ports, without extending to global shipping routes and terminal enterprises / yards. It lacks a full-link hierarchy system of "global-regional-port cluster-single port-target enterprise", resulting in a disconnect between global shipping route layout and enterprise terminal unloading needs.

[0006] (2) Insufficient intelligence in channel identification: There is a lack of quantitative automated identification formulas and algorithms, and the identification relies heavily on manual assistance. The identification accuracy is low and the efficiency is poor, and it is impossible to achieve automatic classification and dynamic updating of channel types.

[0007] (3) The simulation model is not refined enough: the simulation process lacks detailed quantitative calculation formulas, the parameters such as ship generation, operation efficiency, and channel load are set in a relatively coarse manner, and the parameters are not adaptively adjusted by combining multi-source dynamic data, resulting in insufficient accuracy and reliability of the simulation results.

[0008] To address the shortcomings of existing technologies and solve technical problems such as incomplete identification levels, low intelligence, and insufficient refinement of simulation models in port cluster maritime import channels, this invention aims to achieve accurate identification of channels across the entire supply chain, cargo-oriented classification, refined dynamic simulation, and collaborative scheduling among enterprises. This will enhance the intelligence and refinement of port cluster channel management and ensure a stable supply of bulk commodities. Summary of the Invention

[0009] This invention aims to construct a five-level channel identification system of "global-regional-port cluster-single port-target enterprise / yard" to achieve automated and accurate identification of channels across the entire chain, realize dynamic simulation of all aspects and multi-scale processes from global shipping routes to enterprise unloading, improve the accuracy and reliability of simulation results, achieve deep linkage between channel identification, simulation and enterprise needs, establish a dynamic adaptation mechanism, and provide accurate decision support for port cluster planning and port-industry collaboration.

[0010] To achieve the above objectives, the present invention provides the following solution: An automatic identification and dynamic adaptation simulation system for maritime import channels in port clusters includes: The multi-source data fusion processing module is used to collect and preprocess multi-source data from the port and build a multi-source data fusion database. The channel automatic identification and classification module is used to construct a five-level channel system of "global-regional-port cluster-single port-target enterprise / yard" based on the multi-source data fusion database and using a hierarchical progressive logic, to automatically identify channels at each level and obtain the five-level channel identification results; The channel-cargo matching module is used to match cargo types with channels based on the five-level channel identification results to obtain matching results; The dynamic adaptive simulation module is used to simulate the entire process of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution based on the five-level channel identification results and matching results, and obtain simulation results. The visualization and decision support module is used to calculate channel saturation based on the simulation results, generate a five-level penetration heat map, quantitative indicators and optimization suggestions, and complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

[0011] Preferably, the multi-source data fusion processing module includes: The data acquisition unit is used to collect multi-source data from the port, including static data, dynamic real-time data, and business demand data. The static data includes historical global AIS vessel trajectory data, port terminal attribute data, and basic information of the target enterprise. The dynamic real-time data includes vessel AIS trajectory data, meteorological and hydrological data, waterway status data, and yard inventory data. The business demand data includes enterprise logistics demand plans and port operation plans. The data preprocessing unit is used to remove outliers, fill in missing values, and standardize the port multi-source data to obtain standardized port multi-source data. The data fusion unit is used to establish the mapping relationship between the static data, the dynamic real-time data and the business requirement data, and to construct a multi-source data fusion database by using data correlation calculation and multi-source data weighted fusion.

[0012] Preferably, the automated channel identification and classification module includes: The global layer channel identification unit is used to extract the main channels for transoceanic imports worldwide. It adopts an improved route clustering algorithm and calculates route similarity based on ship AIS track data to obtain the global layer channel identification results. The regional layer channel identification unit is used to obtain the radiation range of the regional port cluster based on the global layer channel identification results, calculate the cargo flow correlation between ports, and obtain the regional layer channel identification results. The port cluster channel identification unit is used to calculate the importance of channels between ports in the port cluster, identify primary and secondary channels, and obtain the port cluster channel identification results. The single-port level channel identification unit is used to calculate the operational correlation between berths, storage yards, and gates based on single-port terminal attribute data, divide the internal channels of a single port, and obtain single-port level channel identification results. The enterprise / yard level access identification unit is used to calculate the transportation correlation between the port / yard and the target enterprise, identify enterprise-dedicated access channels and public access channels, and obtain enterprise / yard level access identification results.

[0013] Preferably, the channel-cargo matching module includes: The channel-cargo matching unit is used to calculate the channel-cargo matching degree based on the five-level channel identification results, by using the proportion of cargo flow of preset cargo types on the channel and the proportion of the channel's operational capability to adapt to the preset cargo types, and to obtain the matching result. The channel classification unit is used to automatically identify key channels and regular channels for target cargo based on preset judgment thresholds, and to establish a dynamic update mechanism to periodically recalculate the matching degree and update the channel classification.

[0014] Preferably, the dynamic adaptation simulation module includes: The ship generation unit is used to calculate the ship arrival probability based on historical AIS ship trajectory data and port operation plans, and generate ship arrival sequences using a Poisson distribution model. The arrival decision unit is used to determine the berthing plan based on the ship arrival sequence, through the berth, waterway, yard and enterprise acceptance status, using a comprehensive decision function combined with independent quantitative judgment formulas for each dimension. The loading and unloading operation unit is used to dynamically correct the rated loading and unloading efficiency based on weather conditions and equipment status, obtain the actual loading and unloading efficiency, and calculate the loading and unloading operation time. The yard transfer unit is used to calculate the real-time inventory and capacity utilization of the yard; Enterprise delivery unit, used to calculate enterprise delivery timeliness and unloading utilization rate; The dynamic adjustment unit is used to adaptively correct the ship arrival sequence, loading and unloading efficiency, yard transshipment rhythm and enterprise delivery plan based on preset constraints and by switching priority calculation formulas. The multi-scale simulation unit is used to simulate global shipping routes, regional port cluster channel traffic, ship density in each channel within the port cluster, and single-port operations and enterprise unloading processes based on preset simulation granularity, and output simulation indicators.

[0015] Preferably, the visualization and decision support module includes: The channel saturation calculation unit is used to calculate the channel saturation based on the actual freight volume and design capacity of the preset channel; The heatmap visualization unit is used to generate a five-level penetration heatmap based on the channel saturation, and supports switching between global, regional, port cluster, single port, and enterprise levels. The decision generation unit is used to generate optimization suggestions based on the simulation indicators.

[0016] This invention also provides an automatic identification and dynamic adaptation simulation method for maritime import channels in port clusters, and the system includes: Collect and preprocess multi-source port data to construct a multi-source data fusion database; Based on the multi-source data fusion database, a five-level channel system of "global-regional-port cluster-single port-target enterprise / yard" is constructed using a hierarchical and progressive logic. The system automatically identifies channels at each level and obtains the identification results of the five-level channels. Based on the five-level channel identification results, the goods type and channel are matched to obtain the matching results; Based on the five-level channel identification results and matching results, a full-process simulation of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution is conducted to obtain simulation results. Based on the simulation results, the channel saturation is calculated, and a five-level penetration heat map, quantitative indicators, and optimization suggestions are generated to complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

[0017] Preferred methods for constructing a multi-source data fusion database include: The port collects multi-source data, which includes static data, dynamic real-time data, and business demand data. The static data includes historical global AIS vessel trajectory data, port terminal attribute data, and basic information of the target enterprise. The dynamic real-time data includes vessel AIS trajectory data, meteorological and hydrological data, waterway status data, and yard inventory data. The business demand data includes enterprise logistics demand plans and port operation plans. Outlier removal, missing value completion, and data standardization are performed on the port multi-source data to obtain standardized port multi-source data. Establish a mapping relationship between the static data, the dynamic real-time data, and the business requirement data, and construct a multi-source data fusion database by using data correlation degree calculation and multi-source data weighted fusion.

[0018] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention revolves around a method and system for intelligent identification and multi-level dynamic adaptive simulation of port cluster maritime import channels. It establishes a dynamic collaborative mechanism compatible with multi-source data fusion, five-level channel identification, accurate cargo matching, and enterprise terminal needs, breaking through the limitations of incomplete channel identification levels and low intelligence in traditional technologies. It creates a five-level automated channel identification system, channel-cargo matching model, and multi-level refined dynamic simulation method for the entire port cluster chain, solving the collaborative optimization problems of inaccurate channel classification, coarse simulation model parameters, poor dynamic adaptability, and insufficient enterprise linkage in traditional technologies. It develops modular and configurable dynamic adaptive simulation algorithms and visual decision support modules to achieve full-process control of "automated channel identification - accurate cargo matching - refined simulation process - intuitive decision support", and constructs a full-cycle port cluster maritime import channel identification and dynamic simulation system compatible with global shipping route layout, regional port cluster collaboration, single port operation scheduling, and enterprise terminal unloading needs. This system enhances the intelligence and precision of port cluster channel management, optimizes channel resource allocation, ensures a stable supply of bulk commodities, and strengthens the adaptability of port clusters and regional logistics to various dynamic changes. It effectively supports core port operations and planning functions such as port cluster channel identification, cargo classification, logistics process simulation, channel scheduling optimization, and enterprise collaboration. Attached Figure Description

[0019] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the system structure according to an embodiment of the present invention; Figure 2 This is a simulation logic diagram for ship arrival in an embodiment of the present invention; Figure 3 This is a logic diagram for the five-level channel automated identification in an embodiment of the present invention; Figure 4 This is a flowchart of the channel-cargo matching calculation in an embodiment of the present invention; Figure 5 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1: like Figure 1 As shown, an automatic identification and dynamic adaptation simulation system for maritime import channels in port clusters includes: a multi-source data fusion processing module, an automatic channel identification and classification module, a channel-cargo matching module, a dynamic adaptation simulation module, and a visualization and decision support module.

[0024] A multi-source data fusion processing module is used to collect and preprocess port multi-source data to construct a multi-source data fusion database; a further implementation method is that the multi-source data fusion processing module includes: The data acquisition unit is used to collect multi-source port data, which includes static data, dynamic real-time data, and business demand data. Static data includes global AIS historical ship trajectory data, port terminal attribute data (berth type, design draft, loading and unloading equipment, operational capacity, etc.), and basic information of target enterprises (unloading capacity, yard inventory limit, etc.). Dynamic real-time data includes ship AIS trajectory data (ship position, heading, cargo volume, cargo type, etc.), meteorological and hydrological data, waterway status data, and yard inventory data. Business demand data includes enterprise logistics demand plans and port operation plans. The data preprocessing unit is used to remove outliers, fill in missing values, and standardize the port multi-source data to obtain standardized port multi-source data.

[0025] In this embodiment, outlier removal (taking AIS trajectory data as an example): outlier trajectory points are removed using the 3σ criterion, and the calculation formula is as follows: in, Let n be the standard deviation of the coordinates of the trajectory points, and n be the number of trajectory points. For the first i The longitude (or latitude) of each trajectory point. The average longitude (or latitude) of all trajectory points; when the coordinates of the trajectory points satisfy... When an outlier occurs, it is identified as an anomaly and removed.

[0026] Missing value completion (taking yard inventory data as an example): Missing inventory data is completed using linear interpolation. The calculation formula is as follows: in, For the missing inventory data at time k, This is the inventory data at time k-1. This is the inventory data at time k+1. , , These are the corresponding times.

[0027] Data standardization: Standardize data of different magnitudes to the [0,1] interval for easier subsequent calculations. The calculation formula is as follows: in, For standardized data, The original data, The minimum value of the original data. This represents the maximum value of the original data.

[0028] The data fusion unit is used to establish the mapping relationship between static data, dynamic real-time data and business requirement data. It uses data correlation calculation and multi-source data weighted fusion to build a multi-source data fusion database.

[0029] In this embodiment, the data correlation calculation (reflecting the mapping relationship between multiple types of data) uses the cosine similarity formula to calculate the correlation between two types of data, quantifying the relevance of different data sources, and serving as the core criterion for determining the mapping relationship. The formula is as follows: in, The correlation between the two types of data (range [0,1]) is the degree of correlation between the two types of data. The closer the value is to 1, the stronger the correlation between the two types of data and the closer the mapping relationship. This represents a type of standardized data (such as AIS ship trajectory standardized data). This represents another type of standardized data (such as standardized data of dock attributes). , The first two types of data are respectively i There are n feature components, where n is the dimension of the data features.

[0030] For the cross-mapping of the three types of data, the correlation between AIS data and terminal attribute data is calculated sequentially. The correlation between AIS data and enterprise demand data The correlation between dock attribute data and enterprise demand data Only retain the correlation. Establish an effective mapping relationship between the data pairs.

[0031] Multi-source data weighted fusion formula (forming a unified database): Based on the degree of correlation, weights are assigned to multi-source data that satisfy the mapping relationship, and weighted fusion is performed to generate unified fused data features. After all the fused data is summarized, a multi-source data fusion database is formed. The formula is as follows: in, Let i be the feature value of the fused data; , , These are the standardized values ​​of the i-th feature in the AIS data, port attribute data, and enterprise demand data, respectively. , , These are the fusion weights for the three types of data, obtained by normalizing the data correlation, and satisfying the following conditions. The specific weight calculation is as follows: The above formula enables precise correlation and weighted fusion of dynamic data, static data, and business requirement data, forming a multi-source data fusion database with unified structure and consistent data, providing high-quality data support for subsequent five-level channel identification and refined simulation.

[0032] The automated channel identification and classification module is used to construct a five-level channel system—"global-regional-port cluster-single port-target enterprise / yard"—based on a multi-source data fusion database and employing a hierarchical, progressive logic. It then automates the identification of channels at each level to obtain the five-level channel identification results. For example... Figure 3 As shown.

[0033] A further implementation method includes a channel automatic identification and classification module comprising: The global layer channel identification unit is used to extract the main transoceanic import channels worldwide. It employs an improved route clustering algorithm, calculating route similarity based on ship AIS track data to achieve highly accurate route clustering and global layer channel identification. Specifically, the core improvement is to optimize the Dynamic Time Warping (DTW) algorithm to address the unique characteristics of transoceanic shipping, and simultaneously implement an adaptive clustering threshold mechanism. The specific formula is as follows: ① Calculation of route trajectory similarity (using Dynamic Time Warping (DTW) algorithm): Existing technologies that directly use general dynamic time warping algorithms to calculate route similarity have four major drawbacks: First, the use of planar Euclidean distance cannot adapt to the spherical coordinate characteristics of global transoceanic routes, resulting in large calculation deviations in high-latitude shipping areas; second, the equal-weight matching rules do not distinguish the importance differences between critical sections such as choke channels and hub port channels and non-critical sections in the open ocean, which can easily lead to misjudgments due to mismatches in core channels; third, they only consider spatial coordinates and do not incorporate semantic features of ship navigation such as speed and heading, which can easily lead to abnormal matches that do not conform to navigation patterns; fourth, there are no path constraints specific to shipping scenarios, and alignment paths are prone to jumps, resulting in clustering results that do not conform to actual shipping logic.

[0034] To address the shortcomings of existing technologies, a customized improvement to the dynamic time warping algorithm for shipping scenarios is proposed. A weighted constraint-based dynamic time warping method is used to solve for the optimal temporal alignment path between two trajectory sequences using dynamic programming, minimizing the cumulative distance after alignment. Where A and B are two AIS trajectory sequences to be compared, and m and n are the number of points in the two trajectories, respectively. , These are the first and second trajectories of trajectories A and B, respectively. i, j The coordinates of the trajectory points w ( i,j ) is a fusion dynamic weighting coefficient based on the importance of the flight segment and the flight characteristics (value range [0,1]). DTW ( A,B The optimal alignment cumulative distance between two trajectories; the smaller the value, the higher the similarity between the two routes.

[0035] ② Clustering threshold determination: Set a similarity threshold θ ( θ =0.3 (can be adjusted according to the actual scenario), when the two routes DTW When the value is ≤ θ, the routes are classified as the same type. After clustering, routes with a sample size of ≥ 50 are extracted as the main global transoceanic import channels, and the global hub ports passed through by the routes are identified.

[0036] The regional-level channel identification unit is used to obtain the radiation range of a regional port cluster based on the global-level channel identification results, calculate the cargo flow correlation between ports, and obtain the regional-level channel identification results. In this embodiment, based on the global-level identification results, the radiation range of the regional port cluster is defined, the cargo flow correlation between ports is calculated, and regional-level channels are identified. The calculation formula is as follows: in, For the port i With the port j The correlation of goods flow For the port i to port j Annual freight volume, For the port i Total annual freight volume; set correlation threshold. β ( β =0.2), when ≥ β At that time, determine the port i With the port j There are regional-level channels between them.

[0037] The port cluster channel identification unit is used to calculate the importance of channels between ports within a port cluster, identify primary and secondary channels, and obtain the port cluster channel identification results. In this embodiment, the calculation formula for calculating the importance of channels between ports within a port cluster and identifying primary and secondary channels is as follows: in, For the port i With the port j The importance of the channels between them (value range [0,1]). α Weighting coefficients ( α =0.7, focusing on freight volume). This represents the maximum freight volume across all channels within the port cluster. For channel ij The opening time of navigation, The shortest navigation time for all channels within the port cluster; As the main passage, This is a secondary channel.

[0038] The single-port level channel identification unit is used to calculate the operational correlation between berths, storage yards, and gates based on single-port terminal attribute data, and to delineate internal channels within the single port to obtain single-port level channel identification results. In this embodiment, based on single-port terminal attribute data, the operational correlation between berths, storage yards, and gates is calculated, and internal channels within the single port are delineated. The calculation formula is as follows: in, berth p With the storage yard (or gate) q The degree of relevance of the tasks berth p Arrive at the storage yard (or gate). q Number of assignments per year berth p Total number of assignments per year; At that time, determine the berth. p With the storage yard (or gate) q There are fixed operating channels between them.

[0039] The enterprise / yard level access identification unit is used to calculate the transportation correlation between the port / yard and the target enterprise, identify enterprise-dedicated access channels and public access channels, and obtain the enterprise / yard level access identification results. In this embodiment, the calculation of the transportation correlation between the port (yard) and the target enterprise, and the identification of enterprise-dedicated access channels and public access channels, are performed using the following formula: in, For enterprises e With storage yard q Transportation correlation, For enterprises e From the storage yard q Annual delivery volume For enterprises e Total annual delivery volume; At that time, it was determined to be an enterprise e A dedicated channel is provided; otherwise, it is a public transfer channel.

[0040] The channel-cargo matching module is used to match cargo types with channels based on the five-level channel recognition results, and obtain the matching results; for example... Figure 4 As shown, in a further embodiment, the channel-cargo matching module includes: The channel-cargo matching unit is used to calculate the channel-cargo matching degree based on the five-level channel identification results, by considering the proportion of cargo flow for preset cargo types on the channel and the proportion of operational capabilities of the channel to adapt to preset cargo types, and to obtain the matching result; specifically, the channel-cargo matching degree is calculated as follows: in, For channel c With goods h The matching degree (value range [0,1]), γ is the weighting coefficient (γ=0.8, focusing on cargo flow). For channel c Goods delivery h Annual freight volume, For channel c Total annual freight volume For channel c Suitable goods h The ability to perform tasks For channel c Total operational capacity.

[0041] The channel classification unit automatically identifies key channels and regular channels for target cargo types based on preset judgment thresholds, and establishes a dynamic update mechanism to periodically recalculate the matching degree and update the channel classification. Specifically, the channel classification judgment is as follows: ①Target Category Key Channel: When a certain category of goods exists h ,satisfy Furthermore, when the channel connects to a dedicated terminal and directly links to the target enterprise, it is marked as a key channel for the target cargo. ② Regular channel: When no type of goods h satisfies When a channel can accommodate multiple types of goods, it is marked as a regular channel.

[0042] (3) Dynamic update: recalculated every quarter The channel classification is updated based on the calculation results to ensure the accuracy of the classification.

[0043] The dynamic adaptive simulation module is used to simulate the entire process of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution based on the five-level channel identification and matching results, obtaining simulation results. It adopts a modular and configurable design to construct a multi-scale, full-process refined simulation model. Each module is equipped with detailed quantitative calculation formulas to achieve full-process simulation from global shipping routes to enterprise unloading. A further implementation method is that the dynamic adaptive simulation module includes: The ship generation unit is used to calculate the ship arrival probability based on historical AIS ship trajectory data and port operation plans, employing a Poisson distribution model to generate a ship arrival sequence; specifically, the ship arrival rate is calculated as follows: in, This refers to the vessel arrival rate (vessels / day). The total freight volume (tons) during the simulation period. This refers to the average cargo capacity per vessel (tons / vessel). The number of operating days (days) within the simulation period.

[0044] Ship arrival probability calculation (Poisson distribution): in, t The simulation time window (days) is used. n for t The number of ships arriving within a given time period, where e is a natural constant ( e ≈2.718), P ( N ( t )= n )for t Arrive within the time limit n The probability of a ship.

[0045] The arrival decision unit is used to determine the berthing plan based on the vessel arrival sequence, considering berth, channel, yard, and enterprise acceptance status. It utilizes a comprehensive decision function combined with independent quantitative judgment formulas for each dimension. Figure 2 As shown.

[0046] Specifically, the formula and logic are as follows: ①Comprehensive decision function (integrating judgment results from various dimensions): , in, The result is a comprehensive decision-making outcome (1 = reliable berthing, 0 = unreliable berthing). This is the berth condition determination coefficient. This is the navigational qualification coefficient for the waterway. This is the determination coefficient for the storage yard capacity. The four coefficients are the enterprise acceptance criteria. All four coefficients are either 1 (condition met) or 0 (condition not met). The decision result is 1 only when all coefficients are 1.

[0047] ②Independent quantification formulas for each dimension: a. Berth status determination formula (quantification of idle status): in, berth p Current idle time (hours) The estimated loading and unloading time for the vessel (in hours, which can be estimated using the loading and unloading operation time formula mentioned above) is used to determine that the berth is available for occupancy when the berth's idle time is not less than the estimated loading and unloading time.

[0048] b. Navigation Channel Determination Formula (Water Depth Constraint Quantification): in, The actual water depth of the channel (m). The ship's draft (m). For safety margin water depth (m, usually taken as 0.5-1.0m), To supplement the wave influence factor (0.1-0.3m, based on real-time meteorological data) with the excess water depth (m), the accuracy of the judgment is improved.

[0049] c. Formula for determining storage yard capacity (quantification of inventory adequacy): in, This represents the maximum storage capacity of the storage yard (in tons). This represents the current real-time inventory (tons) at the storage yard. This refers to the ship's cargo capacity (tons). k The inventory redundancy coefficient (with a value of 1.05-1.1, reserving a small amount of redundancy space) is set to ensure that the storage yard has sufficient capacity to accommodate ship cargo.

[0050] d. Enterprise Acceptance Judgment Formula (Quantification of Acceptance Capacity): in, This represents the company's maximum daily unloading capacity (tons / day). This refers to the ship's cargo capacity (tons). The estimated loading and unloading time (hours) of the vessel is converted to days (divided by 24). If the enterprise's daily unloading capacity is not less than the vessel's average daily unloading volume, the enterprise is deemed acceptable.

[0051] The loading and unloading unit is used to dynamically adjust the rated loading and unloading efficiency based on weather conditions and equipment status to obtain the actual loading and unloading efficiency and calculate the loading and unloading operation time; specifically, the calculation formula is as follows: ① Loading and unloading efficiency calculation: in, Actual loading and unloading efficiency (tons / hour). Rated loading and unloading efficiency (tons / hour). Meteorological impact coefficients (sunny day = 1.0, cloudy day = 0.9, rainy day = 0.7, heavy rain = 0.3). The equipment availability factor is (good condition = 1.0, minor fault = 0.8, major fault = 0.2).

[0052] ② Calculation of loading and unloading operation time: in, This refers to the loading and unloading operation time (in hours). The number of loading and unloading equipment (units) put into operation.

[0053] The yard transfer unit is used to calculate real-time yard inventory and capacity utilization; specifically, the calculation formula is as follows: ① Real-time inventory calculation at the storage yard: in, S ( t )for t Real-time stockpile inventory (tons). S ( t -1) is t Stockpile inventory (tons) at time -1. for t The amount of goods entering the storage yard at any given time (in tons). for t The amount (in tons) of goods leaving the storage yard at any given time.

[0054] ② Calculation of storage yard capacity utilization: in, This represents the storage yard capacity utilization rate (value range [0,1]). The maximum storage capacity of the storage yard (tons); when At that time, the storage yard was determined to be at full capacity, triggering an early warning.

[0055] The enterprise delivery unit is used to calculate the enterprise's delivery timeliness and unloading utilization rate; specifically, the calculation formulas for enterprise delivery timeliness and unloading utilization rate are as follows: ① Delivery time calculation: in, L represents delivery time (hours), and L represents the transportation distance (km) from the port yard to the enterprise. The average speed of the transport vehicle (km / h). Waiting time (in hours) for companies to receive and unload goods.

[0056] ② Calculation of enterprise unloading utilization rate: in, The enterprise's unloading utilization rate (value range [0,1]). This represents the actual amount (tons) unloaded by the company. This represents the company's maximum daily unloading capacity (tons).

[0057] The dynamic adjustment unit is used to adaptively correct ship arrival sequences, loading and unloading efficiency, yard transshipment rhythm, and enterprise delivery plans based on preset constraints and by using a switching priority calculation formula. Specifically, when constraints are triggered during simulation (such as yard full capacity or channel congestion), the unit automatically calculates the adjustment scheme. Taking channel switching as an example, the switching priority calculation formula is as follows: in, This represents the channel switching priority (value range [0,1]). For the saturation of the alternative channels, The required time (in hours) for switching. To assess the utilization rate of the storage yard corresponding to the alternative access routes, The weighting coefficients (summing up to 1) determine the channel with the highest priority as the switching target.

[0058] The multi-scale simulation unit is used to simulate global shipping routes, regional port cluster channel traffic flow, vessel density within each channel of a port cluster, and single-port operations and enterprise unloading processes based on a preset simulation granularity, and outputs simulation indicators. The simulation granularity includes: ① Macro-scale: Simulate global shipping routes and regional port cluster channel traffic to calculate the overall utilization rate of the channel network. The formula is as follows: ,in For channel c Design capacity; ②Mesoscale: Ship density in each channel within the simulated port complex, calculated using the formula: ,in This represents the number of ships on channel c. Let c be the length of channel c (km); ③ Microscale: Simulate single-port operations and enterprise unloading processes, and output indicators such as berth utilization rate and loading and unloading efficiency (see the corresponding module for formulas).

[0059] The visualization and decision support module is used to calculate channel saturation based on simulation results, generate five-level penetration heat maps, quantitative indicators, and optimization suggestions, and complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

[0060] A further implementation method includes a visualization and decision support module comprising: The channel saturation calculation unit is used to calculate channel saturation based on the actual freight volume and design capacity of the preset channel; specifically, channel saturation calculation (core evaluation indicator): in, For channel c The saturation (value range [0,1]). For channel c Actual freight volume (tons). For channel c Design capacity (tons); Saturation determination criteria: (Overload, red) (High load, orange) (Medium load, yellow) (Low load, green).

[0061] The heatmap visualization unit is used to generate a five-level penetration heatmap based on channel saturation, and supports switching between global, regional, port cluster, single port, and enterprise levels. The decision generation unit is used to generate optimization suggestions based on simulation indicators. Specifically, based on the calculation results of simulation indicators, when a certain channel... When [the company] needs to increase loading and unloading equipment or optimize shipping routes, suggestions such as "increase loading and unloading equipment" or "optimize shipping routes" are provided. At the same time, suggestions such as "adjust the unloading rhythm and improve delivery efficiency" are provided.

[0062] In summary, the present invention provides a method for constructing a five-level channel identification system of "global-regional-port cluster-single port-target enterprise / yard" through multi-source data fusion and hierarchical progressive identification logic, and for automatically identifying and classifying maritime import channels based on channel-cargo matching quantification formula, thereby achieving accurate identification of key channels and regular channels for target cargo.

[0063] The present invention provides a method for adaptively correcting simulation module parameters such as ship arrival sequence, loading and unloading efficiency, yard transshipment rhythm, and enterprise delivery plan based on changes in multi-dimensional constraints such as waterway navigation conditions, berth occupancy status, yard capacity threshold, and enterprise inventory demand during simulation operation, through preset dynamic adjustment rules, thereby achieving multi-scale dynamic simulation of the entire process.

[0064] The present invention provides a method for visualizing and supporting the decision-making of port cluster import channel status by calculating and integrating multi-dimensional indicators such as channel saturation, utilization efficiency, and congestion risk based on simulation operation data. It generates a heat map that can penetrate the five-level channel system and outputs targeted optimization suggestions such as channel expansion priority, berth dynamic allocation scheme, and enterprise replenishment strategy adjustment.

[0065] Example 2: like Figure 5As shown, the present invention also provides an automatic identification and dynamic adaptation simulation method for port cluster maritime import channels, applying the system of Embodiment 1, including: Collect and preprocess multi-source port data to construct a multi-source data fusion database; Based on a multi-source data fusion database, a five-level channel system of "global-regional-port cluster-single port-target enterprise / yard" is constructed using a hierarchical and progressive logic. The system automatically identifies channels at each level and obtains the identification results of the five-level channels. Based on the five-level channel recognition results, the goods type and channel are matched to obtain the matching results; Based on the five-level channel identification results and matching results, a full-process simulation of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution was conducted to obtain simulation results. Based on the simulation results, the channel saturation is calculated, and a five-level penetration heat map, quantitative indicators, and optimization suggestions are generated to complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

[0066] A further implementation method involves constructing a multi-source data fusion database, including: Collect multi-source port data, which includes static data, dynamic real-time data, and business requirement data. Static data includes global AIS ship trajectory historical data, port terminal attribute data, and basic information of target enterprises. Dynamic real-time data includes ship AIS trajectory data, meteorological and hydrological data, waterway status data, and yard inventory data. Business requirement data includes enterprise logistics demand plans and port operation plans. Outlier removal, missing value completion, and data standardization are performed on multi-source port data to obtain standardized multi-source port data. Establish mapping relationships between static data, dynamic real-time data, and business requirement data, and construct a multi-source data fusion database by using data correlation calculation and weighted fusion of multi-source data.

[0067] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. An automatic identification and dynamic adaptation simulation system for maritime import channels in port clusters, characterized in that, include: The multi-source data fusion processing module is used to collect and preprocess multi-source data from the port and build a multi-source data fusion database. The channel automatic identification and classification module is used to construct a five-level channel system of "global-regional-port cluster-single port-target enterprise / yard" based on the multi-source data fusion database and using a hierarchical progressive logic, to automatically identify channels at each level and obtain the five-level channel identification results; The channel-cargo matching module is used to match cargo types with channels based on the five-level channel identification results to obtain matching results; The dynamic adaptive simulation module is used to simulate the entire process of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution based on the five-level channel identification results and matching results, and obtain simulation results. The visualization and decision support module is used to calculate channel saturation based on the simulation results, generate a five-level penetration heat map, quantitative indicators and optimization suggestions, and complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

2. The system according to claim 1, characterized in that, The multi-source data fusion processing module includes: The data acquisition unit is used to collect multi-source data from the port, including static data, dynamic real-time data, and business demand data. The static data includes historical global AIS vessel trajectory data, port terminal attribute data, and basic information of the target enterprise. The dynamic real-time data includes vessel AIS trajectory data, meteorological and hydrological data, waterway status data, and yard inventory data. The business demand data includes enterprise logistics demand plans and port operation plans. The data preprocessing unit is used to remove outliers, fill in missing values, and standardize the port multi-source data to obtain standardized port multi-source data. The data fusion unit is used to establish the mapping relationship between the static data, the dynamic real-time data and the business requirement data, and to construct a multi-source data fusion database by using data correlation calculation and multi-source data weighted fusion.

3. The system according to claim 2, characterized in that, The automated channel identification and classification module includes: The global layer channel identification unit is used to extract the main channels for transoceanic imports worldwide. It adopts an improved route clustering algorithm and calculates route similarity based on ship AIS track data to obtain the global layer channel identification results. The regional layer channel identification unit is used to obtain the radiation range of the regional port cluster based on the global layer channel identification results, calculate the cargo flow correlation between ports, and obtain the regional layer channel identification results. The port cluster channel identification unit is used to calculate the importance of channels between ports in the port cluster, identify primary and secondary channels, and obtain the port cluster channel identification results. The single-port level channel identification unit is used to calculate the operational correlation between berths, storage yards, and gates based on single-port terminal attribute data, divide the internal channels of a single port, and obtain single-port level channel identification results. The enterprise / yard level access identification unit is used to calculate the transportation correlation between the port / yard and the target enterprise, identify enterprise-dedicated access channels and public access channels, and obtain enterprise / yard level access identification results.

4. The system according to claim 1, characterized in that, The channel-cargo matching module includes: The channel-cargo matching unit is used to calculate the channel-cargo matching degree based on the five-level channel identification results, by using the proportion of cargo flow of preset cargo types on the channel and the proportion of the channel's operational capability to adapt to the preset cargo types, and to obtain the matching result. The channel classification unit is used to automatically identify key channels and regular channels for target cargo based on preset judgment thresholds, and to establish a dynamic update mechanism to periodically recalculate the matching degree and update the channel classification.

5. The system according to claim 2, characterized in that, The dynamic adaptation simulation module includes: The ship generation unit is used to calculate the ship arrival probability based on historical AIS ship trajectory data and port operation plans, and generate ship arrival sequences using a Poisson distribution model. The arrival decision unit is used to determine the berthing plan based on the ship arrival sequence, through berth, waterway, yard and enterprise acceptance status, using a comprehensive decision function combined with independent quantitative judgment formulas for each dimension; The loading and unloading operation unit is used to dynamically correct the rated loading and unloading efficiency based on weather conditions and equipment status, obtain the actual loading and unloading efficiency, and calculate the loading and unloading operation time. The yard transfer unit is used to calculate the real-time inventory and capacity utilization of the yard; Enterprise delivery unit, used to calculate enterprise delivery timeliness and unloading utilization rate; The dynamic adjustment unit is used to adaptively correct the ship arrival sequence, loading and unloading efficiency, yard transshipment rhythm and enterprise delivery plan based on preset constraints and by switching priority calculation formulas. The multi-scale simulation unit is used to simulate global shipping routes, regional port cluster channel traffic, ship density in each channel within the port cluster, and single-port operations and enterprise unloading processes based on preset simulation granularity, and output simulation indicators.

6. The system according to claim 5, characterized in that, The visualization and decision support module includes: The channel saturation calculation unit is used to calculate the channel saturation based on the actual freight volume and design capacity of the preset channel; The heatmap visualization unit is used to generate a five-level penetration heatmap based on the channel saturation, and supports switching between global, regional, port cluster, single port, and enterprise levels. The decision generation unit is used to generate optimization suggestions based on the simulation indicators.

7. A method for automatic identification and dynamic adaptation simulation of maritime import channels in port clusters, using the system described in any one of claims 1-6, characterized in that, include: Collect and preprocess multi-source port data to construct a multi-source data fusion database; Based on the multi-source data fusion database, a five-level channel system of "global-regional-port cluster-single port-target enterprise / yard" is constructed using a hierarchical and progressive logic. The system automatically identifies channels at each level and obtains the identification results of the five-level channels. Based on the five-level channel identification results, the goods type and channel are matched to obtain the matching results; Based on the five-level channel identification results and matching results, a full-process simulation of ship arrival, loading and unloading operations, yard transshipment, and enterprise distribution is conducted to obtain simulation results. Based on the simulation results, the channel saturation is calculated, and a five-level penetration heat map, quantitative indicators, and optimization suggestions are generated to complete the automatic identification and dynamic adaptation simulation of the port cluster's maritime import channels.

8. The method according to claim 7, characterized in that, Methods for constructing a multi-source data fusion database include: The port collects multi-source data, which includes static data, dynamic real-time data, and business demand data. The static data includes historical global AIS vessel trajectory data, port terminal attribute data, and basic information of the target enterprise. The dynamic real-time data includes vessel AIS trajectory data, meteorological and hydrological data, waterway status data, and yard inventory data. The business demand data includes enterprise logistics demand plans and port operation plans. Outlier removal, missing value completion, and data standardization are performed on the port multi-source data to obtain standardized port multi-source data. Establish a mapping relationship between the static data, the dynamic real-time data, and the business requirement data, and construct a multi-source data fusion database by using data correlation degree calculation and multi-source data weighted fusion.

Images