A method for allocating ship berths in a water service area based on navigable bearing capacity

Abstract

The present invention discloses a method for allocating ship berthing in a water service area based on navigable bearing capacity, which relates to the field of allocation technology. It includes dividing the water service area into multiple dynamically enabled elastic berth partitions and dynamically adjusting the partition enabling status; collecting the characteristics of each ship in the formation using a hierarchical coding structure, generating a dynamic priority score by combining fuel urgency and service requirements, and constructing a ship feature coding matrix; collecting the environmental data of the berth in the current and future periods, constructing a time series environmental matrix, and outputting a berth availability period table; combining the ship feature coding matrix and the time series environmental matrix, calculating the berth path loss value, constructing a conflict prediction vector, and generating a dynamic matching loss score; generating a scheduling decision sequence through multiple traversals based on the matching loss score, the conflict prediction vector, and the berth availability period table. The present invention significantly improves the resource utilization efficiency and scheduling stability of the water service area.

Description

Technical Field

[0001] The present invention relates to the field of distribution technology, and in particular to a method for allocating ship berths in a water service area based on navigable bearing capacity. Background Art

[0002] In the context of the continuous expansion of the water transportation system and the continuous penetration of intelligent technologies, as an important replenishment and rest node during the operation of ships, the intelligent level of management and dispatching in water service areas has become one of the key factors affecting ship operation efficiency and navigable safety levels. With the intensive development of inland waterway shipping, coastal transportation, and feeder shipping networks, water service areas are facing increasingly complex multi-ship berthing demands and high-dynamic environmental interferences, such as influencing factors like tidal fluctuations, sudden weather, and flow velocity changes. In this context, the traditional ship berthing allocation method based on static berth division and fixed dispatching rules has become difficult to effectively handle typical situations such as multiple ships arriving simultaneously, diverse priorities, and different service requirements. In recent years, some studies have begun to explore ship berthing optimization methods based on AIS track analysis, berth reservation mechanisms, or multi-objective optimization algorithms, but there are still obvious limitations in terms of adaptability to dynamic environments and balanced dispatching of regional bearing capacity.

[0003] When dealing with the berth allocation problem in complex navigable scenarios, the existing technologies mainly have the following deficiencies: First, the allocation of berth resources is still mostly based on static division, lacking the ability to dynamically adapt to environmental changes, resulting in some berths being in a state of "available but inaccessible" during actual peak periods, causing resource waste; Second, most existing methods do not establish a systematic ship feature coding system, and cannot quantify key dispatching parameters such as remaining fuel, task urgency, and formation attributes of ships and incorporate them into the allocation decision-making, resulting in a lack of differentiation and real-time nature in dispatching responses; Third, although some existing methods have introduced path cost functions for matching optimization, they generally do not fully consider key elements such as path conflict prediction and berth navigable bearing capacity, resulting in frequent changes in the allocation of high-bearing berths, affecting the overall dispatching stability and operation safety.

[0004] Therefore, a ship berthing allocation scheme in a water service area based on navigable bearing capacity is needed. Summary of the Invention

[0005] In view of the problems existing in the existing handling of the berth allocation problem in complex navigable scenarios, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to achieve precise dispatching of different types of ships and preferential guarantee of high-bearing berths.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In the first aspect, the present invention provides a method for allocating ship berthing in a water service area based on navigation carrying capacity, which includes dividing the water service area into a plurality of dynamically enabled flexible berth partitions, integrating real-time tide, meteorological data and historical parameters to form an activatable berth set, and dynamically adjusting the partition activation status; using a hierarchical coding structure to collect the characteristics of each ship in the formation, and generating a dynamic priority score in combination with fuel urgency and service demand to construct a ship characteristic coding matrix; collecting environmental data of the berth in the current and future periods, constructing a time series environmental matrix, predicting the future accessibility and environmental interference of the berth, and outputting a berth availability time period table; combining the ship characteristic coding matrix and the time series environmental matrix, calculating the berth path loss value, constructing a conflict prediction vector based on a preset time window, and generating a dynamic matching loss score; performing multiple traversals based on the matching loss score, the conflict prediction vector and the berth availability time period table to generate a scheduling decision sequence, giving priority to ensuring the scheduling stability of high-carrying capacity berths, and achieving efficient ship-berth matching.

[0009] As a preferred scheme of the method for allocating ship berthing in water service areas based on navigation carrying capacity described in the present invention, the division of the flexible berth zones includes: dividing the existing berths in the water service area into a number of berth basic units that can be activated independently or used in conjunction with each other according to physical spacing, water area morphology and accessibility of berthing channels.

[0010] As a preferred solution of the method for allocating ships at water service areas based on navigation carrying capacity described in the present invention, the hierarchical coding structure includes: collecting basic parameters of each ship in the formation according to the ship identity, including draft, main dimensions of the hull, maneuverability level and expected docking time, and constructing a basic feature unit vector of the ship; using a hierarchical coding structure, logically nesting the basic feature unit vector and the dynamic indicator layer, the dynamic indicator layer including fuel urgency, task timeliness and individual service demand factors.

[0011] As a preferred solution of the method for allocating ship berthing in water service areas based on navigation carrying capacity described in the present invention, the construction of the ship feature coding matrix includes: constructing a dynamic priority score based on a weighted combination of various factors in the dynamic indicator layer, and injecting it into the ship feature coding vector as a feature weight parameter; arranging the feature coding vectors of all ships in the formation in order to form a ship feature coding matrix for allocation calculation.

[0012] As a preferred embodiment of the ship berthing allocation method for a water service area based on navigable carrying capacity according to the present invention, the construction of the time-series environment matrix includes: collecting the environmental data of the berth in the current and preset future periods, encoding each moment according to the timestamp, and constructing a time-series data stream of the original environment; and constructing a time-series environment matrix based on the environmental data stream, where the rows represent time segments and the columns correspond to environmental factors.

[0013] As a preferred embodiment of the ship berthing allocation method for a water service area based on navigable carrying capacity according to the present invention, the berth availability period table includes: judging whether the index combination of each time segment in the time-series environment matrix meets the berth safe berthing condition, and outputting a sequence of future berth accessibility; calculating the environmental interference degree level of each time period by comparing the interference index in each time segment with the historical disturbance feature template; and generating a berth availability period table by combining the accessibility sequence and the interference degree level.

[0014] As a preferred embodiment of the ship berthing allocation method for a water service area based on navigable carrying capacity according to the present invention, the calculation of the berth path loss value is as follows: based on the draft depth, hull length, and maneuverability level in the ship feature coding matrix, extracting the flow velocity gradient and wind pressure fluctuation intensity in the corresponding time window of the berth in the time-series environment matrix, and constructing a berth path disturbance factor vector; and calculating the berth path loss value according to the path disturbance factor vector in combination with the ship maneuvering redundancy margin.

[0015] As a preferred embodiment of the ship berthing allocation method for a water service area based on navigable carrying capacity according to the present invention, the generation of the matching loss score includes: adopting a sliding time window method to extract the state switching rate of each berth in a continuous time period from the time-series environment matrix, generating a corresponding conflict prediction vector for characterizing the channel overlap probability; weighting and fusing the berth path loss value and the conflict prediction vector to construct a multi-objective loss function, and outputting a dynamic matching loss score matrix for describing the adaptation degree of each ship at different berths and time periods.

[0016] As a preferred embodiment of the ship berthing allocation method for a water service area based on navigable carrying capacity according to the present invention, the scheduling decision sequence includes: the first round of traversal: traversing all berths, matching the berth availability period of each ship within the estimated arrival time window, and outputting a set of candidate berths ; the second round of traversal: sorting the set of candidate berths according to the navigable carrying capacity of the berths, and extracting a set of berths higher than the set threshold ; the third round of traversal: for the set of berths For each berth j, obtain the matching loss score. With the goal of minimizing the loss, select the top k berths with low losses as the candidate set. ; Fourth round of traversal: For the candidate set For each berth, calculate the conflict prediction vector, and comprehensively select the berth with the highest scheduling score; Stop iterating when reaching the maximum number of iterations, and output the matching result for each ship.

[0017] The beneficial effects of the present invention are as follows: Through the matching score mechanism, the present invention realizes the precise scheduling of different types of ships and the priority guarantee of high-capacity berths, integrates environmental factors such as tides and meteorology, dynamically activates the berth set and adjusts the enabled state, and then through multiple rounds of traversal and comprehensive scoring mechanism, outputs a scheduling decision sequence with the least loss and conflict avoidance, significantly improving the resource utilization efficiency and scheduling stability of the water service area, and adapting to the high-dynamic ship berthing requirements in the future intelligent shipping scenario. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic flow chart of a method for allocating ship berths in a water service area based on navigable carrying capacity.

[0020] Figure 2 It is a schematic flow chart for generating the matching loss score of a method for allocating ship berths in a water service area based on navigable carrying capacity.

[0021] Figure 3 It is a schematic flow chart of a scheduling decision sequence of a method for allocating ship berths in a water service area based on navigable carrying capacity. Detailed Embodiments

[0022] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the detailed embodiments of the present invention with reference to the drawings of the specification.

[0023] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Second, the so-called "one embodiment" or "embodiment" herein refers to specific features, structures or characteristics that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or selectively exclusive embodiments from other embodiments.

[0025] As mentioned in the above background art, when the prior art deals with the berth allocation problem in complex navigation scenarios, there are mainly the following deficiencies: First, the berth resource allocation is still mostly based on static division, lacking the ability to dynamically adapt to environmental changes, resulting in some berths being in the state of "available but inaccessible" during the actual peak period, causing resource waste; Second, most of the existing methods do not establish a systematic ship feature coding system, and cannot quantify key scheduling parameters such as the remaining fuel, task urgency, and formation attributes of ships and integrate them into the allocation decision-making, resulting in a lack of differentiation and real-time in the scheduling response; Third, although some methods have introduced path cost functions for matching optimization, they generally do not fully consider key factors such as path conflict prediction and berth navigation carrying capacity, resulting in frequent changes in the allocation of high-capacity berths, affecting the overall scheduling stability and operation safety. Therefore, a ship docking allocation scheme for a water service area based on navigation carrying capacity is needed.

[0026] Figure 1 It is a flowchart of a ship docking allocation method for a water service area based on navigation carrying capacity according to an embodiment of the present invention. As Figure 1 shown, in the ship docking allocation method for a water service area based on navigation carrying capacity, it includes: S1: Divide the water service area into multiple dynamically enabled elastic berth partitions, integrate real-time tide, meteorological data and historical parameters to form an activatable berth set, and dynamically adjust the partition enabling state; S2: Adopt a hierarchical coding structure to collect the characteristics of each ship in the formation, and generate a dynamic priority score in combination with fuel urgency and service requirements to construct a ship feature coding matrix; S3: Collect the environmental data of the berth at the current and future periods, construct a time-series environmental matrix, predict the future accessibility and environmental interference degree of the berth, and output a berth availability time table; S4: Combine the ship feature coding matrix and the time-series environmental matrix, calculate the berth path loss value, construct a conflict prediction vector based on a preset time window, and generate a dynamic matching loss score; S5: Generate a scheduling decision sequence through multiple traversals based on the matching loss score, conflict prediction vector and berth availability time table, and give priority to ensuring the scheduling stability of high-capacity berths to achieve efficient matching of ships and berths.

[0027] In one embodiment of the present invention, step S1 specifically includes dividing the water service area into multiple dynamically enabled elastic berth partitions, integrating real-time tide, meteorological data and historical parameters, dynamically calculating the carrying capacity coefficient, generating a partition state matrix, and real-time adjusting the partition enabling state.

[0028] In the actual operation and management of water service areas, the physical distance between berths determines whether the berths are physically possible to be activated at the same time. When two berths are close to each other, especially when the lateral spacing is insufficient, interference may occur due to the turning space required for ship entry and exit operations. Therefore, the present invention first calculates the minimum effective distance in the horizontal and vertical directions based on the spatial relationship between each berth and the adjacent berth based on the physical coordinate data between the berths. The morphology of the water area is also used as one of the judgment bases. For example, some water areas are curved, narrowed, or covered with aquatic plants, which affects the continuity and predictability of ship passage path planning. The accessibility of the berth channel includes but is not limited to factors such as the degree of connectivity between the berth and the main channel and the number of heading adjustments required to enter the berth.

[0029] Specifically, the existing berths in the water service area are divided into several berth basic units that can be used independently or in conjunction with each other based on the physical distance between them, the water area morphology and the accessibility of the berthing channel. After completing the above basic space analysis, historical data is further retrieved to analyze which berths have been used jointly during peak hours, and the frequency, duration and ship type distribution of combined use are counted. This information not only reflects the functional compatibility between berths, but also reveals the flexible activation strategies that can be implemented in the management process. Combined with these data, the berths are divided into berth basic units.

[0030] Preferably, the tidal height, water flow direction and wind force level of each berth basic unit at its current location are collected, and the impact level is marked respectively. In combination with the berth orientation and surrounding obstruction, different berth units are given differentiated navigation interference labels to distinguish the degree of interference of the water area with the ship's berthing operation.

[0031] Specifically, tidal height, water flow direction and wind force level are key environmental factors that affect the safe berthing of ships. If the tidal height is too high or too low, it will affect the height difference between the ship and the dock facilities, thereby increasing the risk of berthing; if the water flow direction and the berth direction are inversely related, the ship will need to apply more power to adjust the course during the process of entering and leaving the berth; strong winds may cause the hull to shift sideways and deflect, resulting in unstable berthing. Therefore, the present invention deploys edge sensing equipment such as water level meters, current meters and meteorological stations to collect the above environmental parameters in real time, and classifies them according to the berthing grade specifications issued by the industry (such as wind force levels above level 4 are not suitable for berthing). For example, when the tide of a berth changes too fast at the current time and the direction is facing the main wind direction, its "navigation interference label" is marked as "high", which means that this berth currently has a high degree of uncertainty in terms of navigation safety. If the berth is located on the inner side of the water, has natural obstructions (such as islands, harbor walls, etc.), and is oriented parallel to the mainstream direction, the berth may be assigned a "low interference label" and have a higher activation priority in the scheduling model.

[0032] Preferably, the berth occupancy data of each basic berth unit in the recent cycle is retrieved, and the ship type distribution, berthing time, and berth-to-berth transfer situation are recorded. A daily load trend is formed to reflect the adaptability of the basic berth unit to diverse ships and its potential saturation level.

[0033] It should be noted that the carrying capacity of a berth depends not only on its physical size and environmental conditions but also on its historical usage frequency and the compatible ship type structure. In this invention, the load trend is set as the key indicator to describe the dynamic utilization rate of the berth per unit. This indicator consists of three types of data: First is the ship type distribution, including length, draft, and type (such as cargo ship, oil tanker, passenger ship, etc.), which is used to judge the versatility of the berth; second is the berthing time, that is, the average occupancy duration of each ship at the berth, reflecting the load intensity of the berth in the time dimension; third is the berth-to-berth transfer situation, which counts whether different ships tend to move from one berth to another, revealing the usage coherence or scheduling priority between berths.

[0034] By constructing a three-dimensional relationship diagram of "berth - ship type - time" and unfolding it in chronological order on a daily basis, the load trend curve of each berth can be obtained. If a certain berth is saturated for consecutive days or shows obvious incompatibility with specific large ship types, it should be prioritized to be avoided during the generation of dynamic berth zoning. On the contrary, if a certain berth shows high adaptability, low congestion, and good transfer connection ability, it can be preferentially marked as an elastic and available area.

[0035] Preferably, by combining the real-time collected tide and meteorological data, berth units with prominent unfavorable berthing conditions are filtered, and berths with good recent load records and low interference levels are preferentially selected from the remaining areas as the currently available zoning to form a dynamic berth set and mark the available level; for the currently available berth zoning, distribution density and water area streamline analysis are carried out. If there are phenomena of zoning aggregation or streamline conflicts, according to the historical load records and interference labels, the available area is adjusted and supplemented from the standby berths, or some low-priority areas are returned to the standby state to achieve dynamic balance adjustment of the available berth area.

[0036] It should be noted that when enabling berths, not only the environment and usage history need to be considered, but also the overall water flow line (including water flow direction and ship turning path) and the uniformity of spatial distribution should be taken into account. If the enabled berths are concentrated on one side of the same water area, or there are phenomena such as crossing or blocking of berthing paths, it will greatly increase the ship scheduling and safety risks. Therefore, based on the two-dimensional water area map overlaid with the enabled berth area, the present invention analyzes the enabled berth density and the conflict area of the flow line direction in real time, and proposes deployment suggestions. Specifically, if it is found that there are 3 or more high-level enabled berths in the current DBS concentrated in the same block and their main passage paths interfere with each other, the system automatically calls the standby berth set, re-evaluates according to the interference level, load trend and environmental adaptability, selects the optimal alternative area from it, supplements it to the DBS, and adjusts the original conflict area to a low enabled level or a temporarily frozen state. This mechanism can achieve the dynamic balance and spatial optimization of berth enabling, and effectively relieve the water traffic pressure during peak hours.

[0037] In an embodiment of the present invention, step S2 specifically includes collecting the characteristics of each ship in the formation using a hierarchical coding structure, generating a dynamic priority score by combining the fuel urgency and service requirements, and constructing a ship characteristic coding matrix. The aim is to solve the problem of insufficient differential treatment of ships in traditional berth management. Especially in the scenario where multiple ships cooperate to enter and leave the water area, existing methods often only sort according to the berthing order or preset tasks, lacking a comprehensive consideration of the real-time ship state, task urgency and resource consumption, resulting in low resource allocation efficiency, untimely scheduling response, and even long waiting times for high-priority ships.

[0038] Preferably, the hierarchical coding structure includes: collecting basic parameters of each ship in the formation according to the ship identity identifier, including draft depth, main hull dimensions, maneuverability level and estimated berthing duration, and constructing a ship basic characteristic unit vector; using a hierarchical coding structure to logically nest the basic characteristic unit vector with the dynamic index layer, and the dynamic index layer includes fuel urgency, task timeliness and individual service demand factors.

[0039] Specifically, when collecting the basic characteristics of ships, the present invention considers four categories of key physical indicators: First, the draft depth, which directly determines the minimum requirement of the ship for the berth water depth and is an important basis for judging the berth adaptability; second, the main dimensions of the hull, including the ship length, ship width, etc., which are important bases for judging the physical accommodation capacity of the berth; third, the maneuverability level, which is usually quantified by the minimum turning radius, deceleration ability and berthing assistance system level of the ship, and this parameter has a great impact on the berth safety and operability; fourth, the expected docking duration, which is an important basis for formulating the port operation plan and berth rotation strategy. In conventional berth management, these parameters are often statically recorded and only used for pre-match and manual allocation. However, in the present invention, these parameters are incorporated into the feature unit vector to provide underlying support for subsequent hierarchical coding and dynamic priority decision-making.

[0040] Preferably, the construction of the ship feature coding matrix includes: constructing a dynamic priority score based on the weighted combination of various factors in the dynamic index layer, and injecting it as a feature weight parameter into the ship feature coding vector; arranging the feature coding vectors of all ships in the formation in sequence to form a ship feature coding matrix for allocation calculation. To avoid the problems of dimension chaos and feature interference caused by the "splicing process" of basic parameters and dynamic states in traditional methods, the present invention introduces a hierarchical coding structure. This structure takes the above-mentioned basic feature unit vector as the first layer of coding, and accesses the dynamic index layer through logical nesting on it. Among them, the dynamic index layer introduces three categories of parameters closely related to the ship operation state, task timeliness and resource constraints: The first is the fuel emergency degree, which is calculated based on the ratio between the current fuel reserve of the ship and the estimated remaining fuel consumption of the task, reflecting the urgency of the ship's continuous navigation ability without berthing; the second is the task timeliness, such as whether it is a cold chain transportation ship, whether it is performing an emergency material transportation task, etc., and task time limit weights can be assigned; the third is the individual service demand factor, such as additional tasks such as crew shift plans, shore power access requirements, and living supply requirements, which are also structurally embedded in the index system to form a complete dynamic weight vector.

[0041] That is, the ship feature coding adopts a combination of static and dynamic two-layer indicators, and the specific method is as follows:

[0042] The first layer is the static structural characteristics of the ship, such as ship length, ship width, draft depth and type; the second layer is the dynamic indicators, including maneuverability, risk level and scheduling priority. When coding, these two groups of indicators will be first spliced into a whole vector in a predetermined order, and each group of indicators is multiplied by a weighting factor to highlight its importance in the final decision.

[0043] In the process of constructing the ship feature coding matrix, the present invention proposes a weighted scoring mechanism based on the dynamic index layer. Specifically, for each ship, after the corresponding dynamic index layer factors are standardized, a dynamic priority score is calculated according to a preset simple weighted calculation combination formula. This score reflects the current comprehensive urgency degree of ship berthing.

[0044] Finally, the complete feature coding vectors of each ship are composed of basic units and nested dynamic indexes, and are arranged in matrix form according to the identity order to form a ship feature coding matrix. Each row in the matrix corresponds to the coding result of a ship, and the column vectors contain its structured various parameters.

[0045] It can be seen that the present invention establishes a flexible, accurate and extensible ship feature modeling scheme by introducing structured hierarchical coding, dynamic priority scoring and linkage rules. Compared with the traditional static berth matching mechanism, this scheme can more accurately reflect the priority of ship service requirements and berth adaptability in practical applications, significantly improve the intelligent level of scheduling decision-making and resource utilization efficiency, and is especially suitable for water service areas with complex tidal conditions and dynamic formations.

[0046] In an embodiment of the present invention, step S3 specifically includes: collecting the environmental data of the berth in the current and future periods, constructing a time-series environmental matrix, predicting the future accessibility and environmental interference degree of the berth, and outputting a berth availability period table.

[0047] Since traditional berth scheduling usually makes matching decisions based on static or short-term estimated berth states, it is difficult to cope with sudden or periodic environmental changes such as tidal changes, wind and wave impacts, and visibility reduction. In the formation operation or large ship scheduling scenarios with extremely high requirements for berthing safety, ignoring the future environmental variability will greatly increase the berthing risk, and even lead to berth vacancy or scheduling conflicts. Therefore, the present invention starts from the perspective of environmental perception and prediction, establishes a systematic data processing and evaluation mechanism, and provides a highly reliable future environmental basis for berth scheduling. The dynamic change of the berth environment will directly affect the accessibility of the ship (for example: the ship cannot complete the berthing or departure actions under the conditions of strong crosswind or extremely low visibility). Therefore, relying only on the current state data for scheduling and lacking the prediction of the future situation are likely to cause decision lag and resource waste. In the present invention, the collected data not only includes "current state data", but also clearly proposes to cover the "preset future period", such as the meteorological and hydrological prediction values for the next 6 hours, 12 hours or 24 hours, so as to support dynamic prediction and medium-term planning.

[0048] Conventional acquisition schemes usually rely on a single sensor or static meteorological sources, such as port meteorological stations, tide gauges, etc., and cannot form high-frequency, structured continuous environmental flow data. In our solution, a multi-source perception fusion mechanism is introduced. The data sources include: high-frequency current meters deployed in the berth area (collecting sea current rate and direction), automatic weather stations (wind speed, wind direction, visibility), laser wave height radar (wave measurement), water level gauges (tide level), and weather model API interfaces (obtaining future time series forecasts). The environmental data from different sources are synchronized and archived through a unified timestamp mechanism. Each type of data is tagged with accurate sampling frequencies and spatial coordinate labels to ensure the temporal continuity and spatial consistency of the data. Through this structured acquisition method, the key environmental factors affecting berth availability can be comprehensively covered, providing sufficient data support for subsequent matrix modeling and analysis.

[0049] Preferably, the construction of the time-series environmental matrix includes: collecting multi-source environmental data such as the current velocity, wind direction, wave height, water level, and visibility of the berth in the current and preset future periods, and encoding each moment according to the timestamp to construct an original environmental time-series data stream; based on the environmental data stream, constructing a time-series environmental matrix, where the rows represent time segments and the columns correspond to environmental factors, forming a multi-variable time-series data structure under a unified metric.

[0050] Among them, the key steps in constructing this matrix include two aspects. First is the timestamp encoding, and second is data alignment and filling. Timestamp encoding means marking each piece of collected environmental data with a unique time point number for unified management. To ensure the consistency of time granularity, the present invention standardizes all timestamps, uniformly rounding down to the nearest 10-minute node to form a standardized time index sequence. Next, the data alignment operation is carried out.

[0051] Preferably, the berth availability period table includes: introducing a berth environmental threshold vector, judging whether the berth safety berthing condition is met according to the index combinations of each time segment in the time-series environmental matrix, and outputting a future reachability sequence of the berth; calculating the environmental interference degree level of each time period by comparing the interfering indicators (such as instantaneous wind speed gradient, flow direction mutation frequency) in each time segment with the historical disturbance feature template; combining the reachability sequence and the interference degree level to generate a berth availability period table for subsequent matching and scheduling calls.

[0052] It should be noted that berth accessibility refers to whether a ship can complete berthing or unberthing operations under safe conditions within a specific time period. The core lies in whether it meets the environmental threshold requirements for the safe use of the berth. For this purpose, the present invention introduces a berth environmental threshold vector, which is jointly determined by artificial experience and historical data and represents the maximum acceptable range of various environmental parameters during berth operations. On this basis, for each time segment in the time-series environmental matrix, the corresponding parameter group is compared item by item with the threshold vector. If all parameters are within the threshold range, it is considered that the time period has "accessibility" and is marked as 1; otherwise, it is marked as 0. The following berth future accessibility sequence is formed: Accessibility sequence: [1, 1, 0, 0, 1, 1, 1, 0, …].

[0053] On the other hand, the evaluation of environmental interference degree mainly considers the mutation frequency and perturbation amplitude of environmental elements. Although these factors do not necessarily immediately affect berthing safety, they will increase the operation complexity and failure probability. The present invention uses interference indexes as basic features, such as instantaneous wind speed gradient (the value of wind speed change per unit time), flow direction mutation frequency (the number of flow direction changes per unit time), etc. Specifically, the present invention matches the environmental change trend of each time period with the historical perturbation template, and uses time-series analysis algorithms such as Euclidean distance and dynamic time warping (DTW) to calculate the degree of difference, so as to judge the perturbation level of this time period. The interference degree level is divided into three levels: low interference, medium interference, and high interference, and the corresponding values are 1, 2, and 3 respectively. For example, when the wind speed changes by more than 1.5 m / s within 10 minutes and the flow direction changes by more than 45°, it is judged as a high interference level. Combining the analysis results of the above two dimensions, that is, the future accessibility sequence and the interference degree level sequence, the present invention generates a final berth availability time table. This time table not only represents which time periods can be berthed, but also quantifies the environmental interference intensity within each time period, providing decision-making support for subsequent scheduling.

[0054] In an embodiment of the present invention, step S4 is to fuse the static ship characteristics with the dynamically changing environmental elements, quantify the degree of obstruction and potential risks of the berth to ship berthing within a specific time period, so as to provide a scientific and quantifiable adaptation basis for berth scheduling. In traditional berth matching, the coupling relationship between ship individual characteristics and environmental perturbations is often ignored, resulting in the scheduling scheme being unable to accurately respond to the actual berthing difficulty. The present invention makes the matching process more targeted and dynamic through the linkage modeling of path loss value and conflict prediction.

[0055] Preferably, the calculation of the berth path loss value is performed by extracting the flow velocity gradient and wind pressure fluctuation intensity within the corresponding time window of the berth from the time series environment matrix based on the draft depth, hull length, and maneuverability level in the ship feature coding matrix, constructing a berth path disturbance factor vector; and calculating the berth path loss value according to the path disturbance factor vector in combination with the ship maneuvering redundancy margin.

[0056] Specifically, in actual operation, in order to map these ship features to the environmental complexity of the berth path, it is necessary to extract the disturbance information related to the berth channel from the time series environment matrix and construct a so-called berth path disturbance factor vector. The core elements of this vector include, but are not limited to: flow velocity gradient, wind pressure fluctuation intensity, water flow direction change rate, and wave height change frequency. These indicators can be obtained through differential and statistical analysis of the real-time data of environmental sensors within a sliding window. For example, the flow velocity gradient can be calculated by dividing the average flow velocity difference between the front and rear time slices by the time interval.

[0057] Next, the berth path disturbance factor vector is fused with the ship maneuvering ability to calculate the berth path loss value. The loss value can be understood as the additional maneuvering effort that the ship needs to expend during the berthing process, and its magnitude depends on the difference between the environmental disturbance degree and the ship's redundant maneuvering ability. For example, if a ship has a high maneuvering redundancy margin (i.e., it can withstand certain wind pressure and water flow changes without affecting maneuvering), then the loss value is relatively small in an environment with large wind pressure fluctuations; conversely, if the maneuvering redundancy margin is low, even a medium-level disturbance will significantly increase the berthing difficulty. The following function can be used to model the loss value:

[0058] ;

[0059] where, is the weight of the disturbance factor, is the intensity of the th disturbance term, is the tolerance threshold of the ship in the th disturbance dimension, is used to eliminate negative values to avoid excessive punishment, represents the number of all key segments or key nodes involved in the current ship planning path.

[0060] Preferably, the generation of the matching loss score includes: adopting a sliding time window method to extract the state switching rate of each berth from the time series environment matrix within a continuous time period, generating a corresponding conflict prediction vector to characterize the channel overlap probability; weighting and fusing the berth path loss value and the conflict prediction vector to construct a multi-objective loss function, and outputting a dynamic matching loss score matrix to describe the adaptation degree of each ship at different berths and time periods.

[0061] Exemplarily, the construction algorithm for the channel overlap probability is as follows: Perform trajectory distribution density statistics on all navigation channels within the geographical area where the berth path is located and the reachable time window. Through historical or real-time AIS trajectory data, use the Kernel Density Estimation (KDE) method to model the channel overlap degree at each moment and construct a conflict prediction vector.

[0062] Specifically, in order to predict the berthing conflicts in different time periods, it is also necessary to construct a conflict prediction vector based on a preset time window. The so-called time window refers to a sliding time interval of a fixed length, which is used to model the short-term trend of environmental state changes. For example, a time window with a length of 15 minutes can be set, and it slides 5 minutes each time, so as to cover multiple consecutive windows within the next hour. In each time window, the switching rate of the berth state within this time period (such as the transfer frequency of idle → pre-berthing → occupied) will be extracted, and its overlap degree within the entire area will be statistically calculated. This process needs to specifically identify the spatial channel overlap areas, that is, those parts where the navigation paths of the berths cross or interfere with each other, and establish a spatial mapping relationship between the berths through historical trajectory data and the channel map. The conflict prediction vector can be defined as: , where represents the probability of conflict between the berth path and other berth paths in the -th time window, which is comprehensively weighted and calculated from the historical traffic density, scheduling frequency, and berthing operation overlap degree of the channel overlap area. represents the number of discrete time points within the scheduling evaluation period starting from the estimated arrival time of the ship.

[0063] Finally, the system will perform weighted fusion on the berth path loss value and the conflict prediction vector to construct a dynamic multi-objective matching loss function, thereby outputting a dynamic matching loss score. This scoring matrix is the evaluation result of the adaptability of the ship under different time periods and different berth combinations, which is used to support the selection of the objective function of the scheduling optimization algorithm. The scoring function uses a simple weighted calculation of the loss value and the conflict prediction value. Among them, the weight factor is used to balance the influence of the path loss and the conflict risk. The higher the score, the more unsatisfactory the combination (that is, the greater the loss). The scheduling algorithm will preferentially avoid high-scoring combinations among the optional berths, thereby improving the safety and smoothness of berthing.

[0064] To sum up, this step realizes the dynamic quantification of the berth path loss through the fine modeling of ship attributes and environmental disturbances, and combines the conflict probability prediction to construct a matching scoring system for multi-objective optimization, significantly improving the intelligent level of berth scheduling. This method not only has high real-time performance, but also can adapt to various types of port environments and ship queues of different scales, which is an important supplement and optimization direction for traditional berth scheduling algorithms.

[0065] In an embodiment of the present invention, step S5 specifically includes designing a collaborative allocation rule based on the matching loss matrix and the navigation carrying capacity index of each berth, constructing a berth allocation plan with the minimum matching loss, giving priority to ensuring the scheduling stability of high-capacity berths, and achieving efficient ship-berth matching.

[0066] Exemplarily, the calculation of the navigation carrying capacity index is through

[0067] Preferably, the scheduling decision sequence includes: the first round of traversal: traversing all berths, matching the available berth time periods of each ship within the estimated arrival time window, and outputting the set of candidate berths ; the second round of traversal: sorting the set of candidate berths according to the navigation carrying capacity of the berths, and extracting the set of berths higher than the set threshold ; the third round of traversal: for each berth j in the set of berths , obtaining the matching loss score, and selecting the top k berths with low loss as the candidate set with the goal of minimum loss ; the fourth round of traversal: for each berth in the candidate set , calculating the conflict prediction vector, and comprehensively selecting the berth with the maximum scheduling score; stop iterating when reaching the maximum number of iterations, and output the matching result for each ship.

[0068] Exemplarily, the first round of traversal is to screen the set of candidate berths based on berth availability. Specifically: for each ship to berth, obtain its estimated arrival time window; traverse all berths and query the berth availability time period table; if the berth is available within the ship's arrival time window, keep it; otherwise, eliminate the berth; output the set of candidate berths for the ship ; the second round of traversal is to perform regional optimization based on the navigation carrying capacity index. Specifically: for each berth in the set of candidate berths , obtain its navigation carrying capacity index; sort the berths in descending order according to the navigation carrying capacity value; screen out the high-capacity berths that meet the threshold and construct the set of berths ; if the set of berths is empty, fallback to the berths with sub-optimal bearing capacity in the set of candidate berths to avoid scheduling failure; the third round of traversal is to perform local optimization screening based on the matching loss score. Specifically: obtain the matching loss score corresponding to each berth in the set of berths , sort the matching loss scores in ascending order with the goal of minimum loss; select the top k berths with low loss as the candidate set , as the preferred alternative; the fourth round of traversal is to introduce the conflict prediction vector to control the distribution stability. Specifically: obtain the conflict prediction vector of each berth in the candidate set , and construct the following comprehensive scheduling score function:

[0069] ;

[0070] in, is the berth navigation carrying capacity, Score the matching loss, is the conflict penalty factor, is a small smoothing factor to prevent division by zero errors, For berth, For ships.

[0071] For each candidate set Berth y calculation in , the berth corresponding to the maximum value is taken as the final recommended berth.

[0072] Finally, a matching triple is output for the ship, including the ship, berth and time, and the berth usage status table is updated: the berth is marked as occupied during the time period.

[0073] It can be seen that the matching and screening mechanism of the present invention can achieve efficient adaptation between ships and berths under a variety of resource and scheduling constraints. This process helps to significantly improve the rationality and fairness of berth allocation, reduce ship waiting time, and reduce scheduling delays caused by conflicts or resource incoordination. In addition, the solution can improve the stability of the overall navigation order, alleviate local overload problems, and ensure the smoothness and continuity of port operations. Ultimately, the port throughput efficiency is improved and the resource utilization rate is maximized, providing solid support for intelligent scheduling in complex traffic environments.

[0074] In summary, the present invention achieves accurate scheduling of different types of ships and priority guarantee of high-load berths through a matching scoring mechanism, integrates environmental factors such as tides and meteorology, dynamically activates berth sets and adjusts the enabled status, and then outputs a scheduling decision sequence with minimum loss and conflict avoidance through multiple rounds of traversal and comprehensive scoring mechanisms. Ultimately, it improves the resource utilization efficiency and scheduling stability of water service areas, and adapts to the high-dynamic ship berthing needs in future smart shipping scenarios.

[0075] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for allocating ship berths in a water service area based on navigable bearing capacity, characterized in that: Including: Dividing the water service area into multiple dynamically enabled elastic berth zones, integrating real-time tide, meteorological data and historical parameters to form an activatable berth set, and dynamically adjusting the zone activation status; Collecting the characteristics of each ship in the formation using a hierarchical coding structure, and generating a dynamic priority score by combining fuel urgency and service demand to construct a ship characteristic coding matrix; Collecting the current and future cycle environmental data of the berth, constructing a time-series environmental matrix, predicting the future accessibility and environmental interference degree of the berth, and outputting a berth availability time period table; Combining the ship characteristic coding matrix and the time-series environmental matrix, calculating the berth path loss value, constructing a conflict prediction vector based on a preset time window, and generating a dynamic matching loss score; Generating a scheduling decision sequence through multiple traversals based on the matching loss score, conflict prediction vector and berth availability time period table, giving priority to ensuring the scheduling stability of high-capacity berths, and achieving efficient ship-berth matching; The berth availability time period table includes: judging whether the berth safety berthing condition is met according to the index combination of each time segment in the time-series environmental matrix, and outputting a berth future accessibility sequence; calculating the environmental interference degree level of each time period by comparing the interference index and the historical disturbance feature template in each time segment; combining the accessibility sequence and the interference degree level to generate a berth availability time period table; The generation of the matching loss score includes: adopting a sliding time window method to extract the state switching rate of each berth in the continuous time period from the time-series environmental matrix to generate a corresponding conflict prediction vector for characterizing the channel overlap probability; weighting and fusing the berth path loss value and the conflict prediction vector to construct a multi-objective loss function, and outputting a dynamic matching loss score matrix for describing the adaptation degree of each ship at different berths and time periods; The described scheduling decision sequence includes: The first round of traversal: Traverse all berths, match the berth availability periods of each ship within the estimated arrival time window, and output the set of candidate berths ; The second round of traversal: Sort the set of candidate berths according to the navigation carrying capacity of the berths, and extract the set of berths higher than the set threshold ; The third round of traversal: For each berth j in the set of berths , obtain the matching loss score, and select the top k berths with low loss as the candidate set with the goal of minimizing the loss ; The fourth round of traversal: For each berth in the candidate set , calculate the conflict prediction vector, and comprehensively select the berth with the largest scheduling score; Stop the iteration when the maximum number of iterations is reached, and output the matching results for each ship.

2. The method for allocating ship berths in a water service area based on navigable bearing capacity according to claim 1, wherein: The division of the elastic berth zone includes: dividing the existing berths in the water service area into several berth basic units that can be independently or jointly enabled according to the physical distance, water area form and access to the berthing channel.

3. The method for allocating ship berths in a water service area based on navigable bearing capacity according to claim 1, wherein: The hierarchical coding structure includes: collecting basic parameters of each ship in the formation according to the ship identity identification, including draft depth, main hull dimensions, maneuverability level and expected berthing duration, and constructing a ship basic characteristic unit vector; adopting a hierarchical coding structure to logically nest the basic characteristic unit vector with the dynamic index layer, and the dynamic index layer includes fuel urgency, task timeliness and individual service demand factors.

4. The method for allocating ship berths in a water service area based on navigable bearing capacity according to claim 3, characterized in that: The construction of the ship characteristic coding matrix includes: constructing a dynamic priority score based on the weighted combination of each factor in the dynamic index layer, and injecting it as a characteristic weight parameter into the ship characteristic coding vector; arranging the characteristic coding vectors of all ships in the formation in sequence to form a ship characteristic coding matrix for distribution calculation.

5. The method for allocating ship berths in a water service area based on navigable bearing capacity according to claim 1, characterized in that: The construction of the time-series environmental matrix includes: collecting the environmental data of the berth in the current and preset future cycles, and encoding each moment according to the timestamp to construct a time-series data stream of the original environment; constructing a time-series environmental matrix based on the environmental data stream, where the rows represent time segments and the columns correspond to environmental elements.

6. The method for allocating ship berths in a water service area based on navigable bearing capacity according to claim 1, wherein: The calculation of the berth path loss value extracts the flow velocity gradient and wind pressure fluctuation intensity within the corresponding time window of the berth in the time series environment matrix based on the draft, hull length, and maneuverability level in the ship feature coding matrix, and constructs a berth path disturbance factor vector; according to the path disturbance factor vector, the berth path loss value is calculated in combination with the ship control redundancy margin.

Images