Bulk cargo ship dynamic unloading method and system based on multi-address cooperation

Abstract

This invention provides a dynamic unloading method and system for bulk cargo ships based on multi-site collaboration, comprising: Step S1: constructing a terminal operation knowledge base and a yard stockpile knowledge base; Step S2: collecting operation data in real time; Step S3: generating a multi-site list based on the yard stockpile knowledge base, yard inventory and inbound / outbound information, and material usage plan, wherein the multi-site list includes product name and available time for each site; Step S4: generating a multi-site selection sequence scheme for each simultaneously operating conveyor belt system based on the multi-site list and a dynamic site selection model; Step S5: generating an optimal unloading scheme based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume, and multi-site selection sequence scheme through a ship unloading model; Step S6: guiding the unloading operation according to the unloading scheme, dynamically analyzing the deviation between planned and actual volume based on real-time collected operation data and unloading conveyor belt volume, and repeating steps S3-S5 when the deviation exceeds a threshold or a major event is detected.

Description

Technical Field

[0001] This invention relates to the field of bulk cargo terminal loading and unloading and intelligent scheduling technology, specifically to a dynamic unloading method and system for bulk cargo vessels based on multi-site coordination. Background Technology

[0002] Bulk carriers play a crucial role in global commodity trade. Traditional unloading operation planning for bulk carriers often relies on the same port machinery efficiency at the same berth and static ship stowage diagrams and initial unloading plans. It lacks adaptability to various dynamic changes during the operation, such as different port machinery efficiencies, changes in terminal equipment status, and changes in cargo location. Furthermore, the formulation of unloading plans is highly dependent on the personal experience of dispatchers and lacks scientific and quantitative decision support. This results in poor replicability of the plans and difficulty in making optimal decisions in complex situations, leading to interruptions or inefficiencies in unloading operations.

[0003] Meanwhile, the planning of material storage locations (i.e., storage locations, material types, and inventory information) at the terminal is separate from ship unloading operations. This means that when multiple ships and different types of cargo are unloaded at the same time, the real-time acceptance capacity of the rear storage yard, material classification and storage requirements, and conflicts in transfer routes are not fully considered. This can easily lead to storage yard congestion or equipment running idle, which in turn affects the overall operational efficiency of the terminal and the time ships spend in port, resulting in high demurrage fees for ships at the terminal.

[0004] Therefore, there is an urgent need in this field for an intelligent method and system that can respond to changes in real time, deeply integrate ship unloading with yard site management, and dynamically generate optimal or near-optimal conflict-free unloading schemes.

[0005] Patent document CN117592717A discloses a bulk cargo unloading scheme for ship holds, including: obtaining the number of ship holds and the total weight of cargo in each hold, the number of unloading operations in each hold, the allocation ratio corresponding to the number of unloading operations, and the probability of each allocation ratio; and obtaining a feasible unloading scheme for each hold based on the total weight of cargo in each hold, the number of unloading operations, the allocation ratio, and the probability. However, patent document CN117592717A does not consider the availability of conveyor belt systems in the downstream material yard, the types of materials that can be unloaded, the unloading locations, and the adjustment of the number of port machinery during the unloading process for multiple unloading port machinery, nor does it consider the timely dynamic adjustment of the unloading scheme generated by the system after deviations occur during the operation. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide a dynamic unloading method and system for bulk cargo ships based on multi-site coordination.

[0007] According to one aspect of the present invention, a dynamic unloading method for bulk carriers based on multi-site coordination includes:

[0008] Step S1: Construct a knowledge base for dock operations and a knowledge base for stockpiles in the material yard; Step S2: Collect operation data in real time; Step S3: Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plan, generate a multi-location list, which includes the product name and the available time for each material location; Step S4: Based on the multi-address list, generate a multi-address selection sequence scheme for each simultaneously operating conveyor belt system through a dynamic address selection model; Step S5: Based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume and multi-material site selection sequence scheme, generate the optimal unloading scheme through the ship unloading model; Step S6: Guide the unloading operation according to the unloading plan. Based on the real-time collected operation data and unloading conveyor belt volume, dynamically analyze the deviation between the planned volume and the actual volume. When the deviation exceeds the threshold or a major event is detected, repeat steps S3-S5.

[0009] Preferably, the wharf operation knowledge base includes wharf berth water depth, wharf operable vessels, wharf operable types, port machinery operable types and corresponding operational capabilities, port machinery compartment operation capabilities, port machinery operation restriction rules, port machinery operation corresponding to conveyor belt systems, and the relationship between conveyor belt systems and operable types.

[0010] Preferably, the knowledge base of the material yard and stockpile includes: the configuration of the material yard, material strips, storage locations, and stacker, the operating capacity of the stacking equipment, the relationship between the material strips and the stacker, the types of material strips that can be stacked, and the matching of the belt system with the material strips in the material yard.

[0011] Preferably, step S2 includes: Obtain basic ship information, information on the types and quantities of ship compartments, and information on the types and weights of goods unloaded from compartments in the port. Obtain information on the status and maintenance plans of port machinery, conveyor belt systems, and stacker cranes; Acquire information on changes in the port crane's alignment hatches and instantaneous flow rate information of the belt scale within a set time period; Obtain the corresponding dynamic stockpile inventory information from the rear material yard.

[0012] Preferably, in step S4, the material address dynamic selection model is used to perform: Sub-step S4.1: Construct an optimization model based on multiple constraints, where the multiple constraints include: Resource constraints: Multiple ships operate together, each carrying different materials that need to be unloaded, and multiple conveyor belt systems operate in parallel. Spatiotemporal constraints: Each material bar corresponds to one conveyor belt system, and other material bar locations are unavailable during the period when the material bar location is occupied; Operational constraints: Only one type of material can be stored at a single storage location to avoid mixing different types of materials in the same storage location; Sub-step S4.2: Construct a feasible initial solution, wherein the construction principle is to prioritize allocating the most efficient and as many port machines as possible and small-capacity material sites to the most urgent or important tasks at present; Sub-step S4.3: Using neighborhood search, multiple neighborhood solutions are generated near the current solution through preset neighborhood operation operators. New solutions inferior to the current solution are accepted with a predetermined probability to avoid local optima traps and explore the global optimum region. Sub-step S4.4: When a potential optimal solution distribution region is identified, an enhanced local search is initiated. High-density neighborhood exploration is performed within the optimal solution distribution region to locate the local optimum within the optimal solution distribution region. Sub-step S4.5: Based on the current local optimum, repeat sub-steps S4.3 and S4.4 until the iteration termination condition is met, and obtain multiple local optima. Sub-step S4.6: Evaluate each local optimum using the objective function, sort them according to their scores, and select the top-scoring multi-material address selection order schemes.

[0013] Preferably, in sub-step S4.1, the expression for the optimized model is as follows:

[0014]

[0015]

[0016] In the formula, t represents the time period, T is the set of time periods in discrete time points; b is the belt system, B is the set of belt systems; s is the material strip, S is the set of material strips; v is the ship, V is the set of ships; m is the material on ship v. It is the collection of materials on ship v; l It is the stacking address of material bar s. It is the set of stack addresses for material bar s; This is a binary variable indicating whether, at time t, material m from ship v is unloaded onto feed bar s via belt b. This is a binary variable representing the heap address at time t. Is it occupied by the material m of ship v?

[0017] Preferably, in sub-step S4.6, the construction of the objective function includes: determining the optimization objective that maximizes the utilization rate of terminal resources and operational efficiency, wherein, Optimization objectives include: By rationally planning the correspondence between the belt conveyor system and the material location and the operation sequence, the common operation time of multiple belt conveyor systems can be maximized. The expression is:

[0018] In terms of material location selection logic, priority is given to storing material strips with large capacity and numerous existing material piles, as expressed in the following expression:

[0019] Within the same material bar, priority is given to stack addresses with smaller remaining stackable capacity, expressed as:

[0020] Multiple optimization objectives are integrated using a weighted summation method to form the objective function: in, A binary variable indicating whether belt b is actively unloading material from strip s at time t; For heap address l capacity, These are the weighting coefficients.

[0021] Preferably, in step S5, the ship unloading model is used to perform: Sub-step S5.1: Obtain working parameters, which include: the target port machinery's operating capacity for the target product under the target step, the available time for the target port machinery to complete the preceding operations, the number of available port machinery for the target step, the total unloading volume of the target hatch for the target product, the operating volume of the target hatch for the target product in the target step, the number of hatches that need to complete the operations in the target step, the operating hatch range of the target port machinery, the start time of the target step, the operating volume of the target port machinery, the operating time of the target step, and the multi-site selection sequence scheme. Sub-step S5.2: Based on the constraints of non-overlapping and spaced operation of port machinery and the dynamic switching principle of stacking sites, a predetermined number of unloading schemes are randomly generated according to the port machinery-hatch allocation. The dynamic switching principle of stacking sites is based on the multi-site selection order scheme and the current status of the stacking sites. The stacking sites include exhausted stacking sites, receiving stacking sites, and unused stacking sites. The non-overlapping constraint of port machinery means that the working hatch range of each port machinery does not overlap under the target step. The spaced operation constraint means that the interval between the hatches of adjacent port machinery under the target step is a preset number. Sub-step S5.3: By evaluating the preset evaluation items, score each unloading scheme and sort the schemes according to the scoring results; Sub-step S5.4: Expand the unloading scheme by selection-crossover-mutation, where: Select: Retain the solution that satisfies the requirement that the difference between the fore and aft compartments of the ship is less than the set value (i.e., the ship is balanced) and that the port machinery meets the safe operating distance (i.e., there must be at least one hatch between the two port machinery). Crossover: Chromosome exchange occurs by randomly selecting the crossover type. Crossover types are divided into intra-population chromosome exchanges within the upper, middle, and lower stages and inter-population chromosome exchanges within the upper, middle, and lower stages, with single-point crossover. Variation: Random variation is performed according to the single port crane operation procedure and within the scope of the operation hatch. The mutated operation hatch is exchanged with the corresponding conflicting hatch of the port crane. Sub-step S5.5: Repeat sub-steps S5.2 to S5.4 until the set number of iterations or convergence condition is reached, and use the result with the best fitness as the output of the unloading scheme.

[0022] Preferably, in sub-step S5.3, the preset evaluation items include hull balance, total operation time, step integrity, belt system matching, and number of port machinery compartment changes.

[0023] According to one aspect of the present invention, the system according to claim 1, a dynamic unloading system for bulk cargo ships based on multi-site coordination, characterized in that it comprises: Module M1: Constructs a knowledge base for dock operations and a knowledge base for stockpiles in the material yard; Module M2: Real-time acquisition of operation data; Module M3: Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plan, generate a multi-location list, which includes the product name and the available time for each material location; Module M4: Based on a multi-address list, it generates a multi-address selection sequence scheme for each simultaneously operating conveyor belt system through a dynamic address selection model; Module M5: Based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume and multi-material site selection sequence scheme, the optimal unloading scheme is generated through the ship unloading model; Module M6: Guides the unloading operation according to the unloading plan. Based on the real-time collected operation data and unloading conveyor belt volume, it dynamically analyzes the deviation between the planned volume and the actual volume. When the deviation exceeds the threshold or a major event is detected, it repeatedly triggers modules M3-M5.

[0024] Compared with the prior art, the present invention has the following beneficial effects: After the online operation of the dynamic unloading method and system for bulk cargo vessels based on multi-site collaboration, by dynamically selecting the optimal combination of vessel compartments, port machinery, operating systems and material sites, conflicts in the input system of the rear material yard are reduced when multiple vessels are unloading together, the vessel unloading operation time is shortened, and demurrage fees caused by non-productive stays of vessels in port are avoided, thereby improving the operational efficiency of bulk cargo terminals. At the same time, the model can respond in real time to sudden situations such as equipment failure and resource occupation, dynamically replanning to ensure the continuity and stability of operations. Furthermore, it transforms complex, experience-dependent multivariate decision-making problems into a data-driven intelligent optimization solution process, improving the intelligence level of terminal scheduling. Attached Figure Description

[0025] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of a dynamic unloading method for bulk carriers based on multi-site collaboration. Detailed Implementation

[0026] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0027] Example 1: This embodiment discloses a dynamic unloading method for bulk carriers based on multi-site coordination, such as... Figure 1 As shown, it includes: Step S1: Construct a knowledge base for dock operations and a knowledge base for stockpiles in the material yard; Step S2: Collect operation data in real time; Step S3: Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plan, generate a multi-location list, which includes the product name and the available time for each material location; Step S4: Based on the multi-address list, generate a multi-address selection sequence scheme for each simultaneously operating conveyor belt system through a dynamic address selection model; Step S5: Based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume and multi-material site selection sequence scheme, generate the optimal unloading scheme through the ship unloading model; Step S6: Guide the unloading operation according to the unloading plan. Based on the real-time collected operation data and unloading conveyor belt volume, dynamically analyze the deviation between the planned volume and the actual volume. When the deviation exceeds the threshold or a major event is detected, repeat steps S3-S5.

[0028] In this embodiment, the wharf operation knowledge base includes wharf berth water depth, wharf operable vessels, wharf operable types, port machinery operable types and corresponding operational capabilities, port machinery compartment operation capabilities, port machinery operation restriction rules, port machinery operation corresponding belt conveyor systems, and the relationship between belt conveyor system operable types.

[0029] In this embodiment, the knowledge base of the material yard includes: the configuration of the material yard, material strips, storage locations, and stacker, the operating capacity of the stacking equipment, the relationship between the material strips and the stacker, the types of material strips that can be stacked, and the matching of the belt system with the material strips in the material yard.

[0030] In this embodiment, step S2 includes: Obtain basic ship information, information on the types and quantities of ship compartments, and information on the types and weights of goods unloaded from compartments in the port. Obtain information on the status and maintenance plans of port machinery, conveyor belt systems, and stacker cranes; Acquire information on changes in the port crane's alignment hatches and instantaneous flow rate information of the belt scale within a set time period; Obtain the corresponding dynamic stockpile inventory information from the rear material yard.

[0031] In this embodiment, in step S4, the material address dynamic selection model is used to perform: Sub-step S4.1: Based on multiple constraints, construct an optimization model. Multiple constraints include: Resource constraints: Multiple ships operate together, each carrying different materials that need to be unloaded, and multiple conveyor belt systems operate in parallel. Spatiotemporal constraints: Each material bar corresponds to one conveyor belt system, and other material bar locations are unavailable during the period when the material bar location is occupied; Operational constraints: Only one type of material can be stored at a single storage location to avoid mixing different types of materials in the same storage location; Sub-step S4.2: Construct a feasible initial solution, wherein the construction principle is to prioritize allocating the most efficient and as many port machines as possible and small-capacity material sites to the most urgent or important tasks at present; Sub-step S4.3: Using neighborhood search, multiple neighborhood solutions are generated near the current solution through preset neighborhood operation operators. New solutions inferior to the current solution are accepted with a predetermined probability to avoid local optima traps and explore the global optimum region. Sub-step S4.4: When a potential optimal solution distribution region is identified, an enhanced local search is initiated. High-density neighborhood exploration is performed within the optimal solution distribution region to locate the local optimum within the optimal solution distribution region. Sub-step S4.5: Based on the current local optimum, repeat sub-steps S4.3 and S4.4 until the iteration termination condition is met, and obtain multiple local optima. Sub-step S4.6: Evaluate each local optimum using the objective function, sort them according to their scores, and select the top-scoring multi-material address selection order schemes.

[0032] In this embodiment, the expression for optimizing the model in sub-step S4.1 is as follows:

[0033]

[0034]

[0035] In the formula, t represents the time period, T is the set of time periods in discrete time points; b is the belt system, B is the set of belt systems; s is the material strip, S is the set of material strips; v is the ship, V is the set of ships; m is the material on ship v. It is the collection of materials on ship v; l is the stacking location of material strip s. It is the set of stack addresses for material bar s; This is a binary variable indicating whether, at time t, material m from ship v is unloaded onto feed bar s via belt b. This is a binary variable representing the heap address at time t. Is it occupied by the material m of ship v?

[0036] In this embodiment, sub-step S4.6, the construction of the objective function includes: determining the optimization objective that maximizes the utilization rate of terminal resources and operational efficiency, wherein, Optimization objectives include: By rationally planning the correspondence between the belt conveyor system and the material location and the operation sequence, the common operation time of multiple belt conveyor systems can be maximized. The expression is:

[0037] In terms of material location selection logic, priority is given to storing material strips with large capacity and numerous existing material piles, as expressed in the following expression:

[0038] Within the same material bar, priority is given to stack addresses with smaller remaining stackable capacity, expressed as:

[0039] Multiple optimization objectives are integrated using a weighted summation method to form the objective function: in, A binary variable indicating whether belt b is actively unloading material from strip s at time t; For heap address l capacity, These are the weighting coefficients.

[0040] In this embodiment, in step S5, the ship unloading model is used to perform: Sub-step S5.1: Obtain working parameters, which include: the target port machinery's operating capacity for the target product under the target step, the available time for the target port machinery to complete the preceding operations, the number of available port machinery for the target step, the total unloading volume of the target hatch for the target product, the operating volume of the target hatch for the target product in the target step, the number of hatches that need to complete the operations in the target step, the operating hatch range of the target port machinery, the start time of the target step, the operating volume of the target port machinery, the operating time of the target step, and the multi-site selection sequence scheme. Sub-step S5.2: Based on the constraints of non-overlapping and spaced operation of port machinery and the dynamic switching principle of stacking sites, a predetermined number of unloading schemes are randomly generated according to the port machinery-hatch allocation. The dynamic switching principle of stacking sites is based on the multi-site selection order scheme and the current status of the stacking sites. The stacking sites include exhausted stacking sites, receiving stacking sites, and unused stacking sites. The non-overlapping constraint of port machinery means that the working hatch range of each port machinery does not overlap under the target step. The spaced operation constraint means that the interval between the hatches of adjacent port machinery under the target step is a preset number. Sub-step S5.3: By evaluating the preset evaluation items, score each unloading scheme and sort the schemes according to the scoring results; Sub-step S5.4: Expand the unloading scheme by selection-crossover-mutation, where: Select: Retain the solution that satisfies the requirement that the difference between the fore and aft compartments of the ship is less than the set value (i.e., the ship is balanced) and that the port machinery meets the safe operating distance (i.e., there must be at least one hatch between the two port machinery). Crossover: Chromosome exchange occurs by randomly selecting the crossover type. Crossover types are divided into intra-population chromosome exchanges within the upper, middle, and lower stages and inter-population chromosome exchanges within the upper, middle, and lower stages, with single-point crossover. Variation: Random variation is performed according to the single port crane operation procedure and within the scope of the operation hatch. The mutated operation hatch is exchanged with the corresponding conflicting hatch of the port crane. Sub-step S5.5: Repeat sub-steps S5.2 to S5.4 until the set number of iterations or convergence condition is reached, and output the result with the best fitness as the unloading scheme. In this embodiment, in sub-step S5.3, the preset evaluation items include hull balance, total operation time, step completeness, belt system matching, and number of port machinery compartment changes.

[0041] Example 2: Ship unloading operations at bulk cargo terminals involve collaborative work among various terminal units. This requires cooperation from planners, unloader operators, central control dispatchers, and stacker-reclaimer operators to improve overall operational efficiency. The core algorithm involved in this invention is deployed on a backend AI model server. Unloader operators use a tablet app, while planners and dispatchers perform corresponding operations on computer clients.

[0042] 1. Knowledge Base Management Establish a wharf operation knowledge base, collecting knowledge related to wharf unloading operations, such as wharf berth water depth, vessels that can be operated at the wharf, types of vessels that can be operated at the wharf, types of port machinery that can be operated and their corresponding operational capabilities, port machinery compartment handling capabilities, port machinery operation restriction rules, port machinery operation corresponding belt conveyor systems, and the relationship between belt conveyor systems and the types of vessels that can be operated.

[0043] Establish a knowledge base for material yards and stockpiles, collecting knowledge related to the configuration of material yards, stockpiles, storage locations, stackers, stacker operation capabilities, the relationship between stockpiles and stackers, the types of stockpiles that can be stacked, and the matching of belt systems with material yard stockpiles.

[0044] 2. Real-time collection and management of homework performance Obtain basic ship information, information on the types and quantities of ship compartments, and information on the types and weights of cargo unloaded from compartments in the port.

[0045] Obtain information on the status and maintenance plans of equipment such as port machinery, belt conveyor systems, and stacker cranes.

[0046] Obtain information on changes in the port crane's alignment hatches and instantaneous flow information of the belt scale within a set time period.

[0047] Obtain the corresponding dynamic stockpile inventory information from the rear material yard.

[0048] 3. Dynamic Site Selection Model Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plans, a multi-location list is provided, categorized by product name and available time. The dynamic location selection model, based on this multi-location list, considers multi-resource coordination, spatiotemporal constraints, and multiple optimization objectives. While taking into account the complex relationships between ships, materials, conveyor belt systems, and material yard locations, it generates the corresponding location input order for each concurrently operating conveyor belt system. The core model algorithm is as follows: 1) Basic Data Definition

[0049] 2) Model Solving Resource constraints: Multiple ships work together, each carrying different materials that need to be unloaded, and multiple conveyor belt systems operate in parallel.

[0050]

[0051] Spacetime constraints: Each material bar can only correspond to one belt system, and other material bar locations cannot be used while the material bar location is occupied.

[0052]

[0053] Job constraints: Only one type of material can be stored at a single storage site; mixing materials is prohibited.

[0054]

[0055] Prioritize material strips with large capacity and multiple stockpiles, and within the same material strip, prioritize the use of stockpiles with smaller stockpile sizes to improve reuse rate.

[0056]

[0057] Optimization goal: To maximize unloading efficiency, the belt conveyor system should be rationally planned to match material locations and work sequence, maximizing the combined working time of the belt conveyors and ensuring the highest overall terminal efficiency.

[0058]

[0059] Integrating multiple objectives using a weighted summation method: in These are weighting coefficients, reflecting the relative importance of each objective.

[0060] The dynamic site selection model employs a hybrid metaheuristic optimization algorithm that integrates a greedy strategy, neighborhood search, simulated annealing, and local search. Based on the greedy strategy, a feasible initial solution is quickly constructed, prioritizing the allocation of the most efficient and numerous port machinery sites and smaller-capacity sites to the most urgent or important tasks (such as ships carrying materials about to run out of stock). Starting from the initial solution, the model continuously optimizes through an iterative process. Neighborhood search utilizes defined neighborhood operators, such as swap operations (exchanging site assignments between two tasks) and insertion operations (reinserting a task into a different position in the sequence), to generate a series of new solutions (neighborhood solutions) near the current solution. Simulated annealing serves as the core framework, accepting new solutions inferior to the current solution with a certain probability to escape local optima and explore the globally optimal region. When the algorithm discovers a promising solution region during simulated annealing, an enhanced local search is initiated. This search performs a denser and deeper neighborhood exploration within this region to quickly locate the local optimum.

[0061] Finally, the solution space is evaluated using the objective function, and the top-scoring multi-site selection order scheme is selected as the input parameter and passed into the ship unloading model.

[0062] 4. Ship unloading model Based on the port machinery's corresponding operational knowledge base (i.e., the port machinery's unloading capacity for different types of cargo), the ship's basic information, the unloading volume of each compartment of the ship, the unloading belt system, and three material site selection order schemes generated by the material site dynamic selection model, the optimal unloading scheme is generated based on the genetic algorithm.

[0063] 1) Basic Data Definition

[0064] 2) Model Solving The start time for the current step has been determined. Set steps The available port machinery set is available Port machinery is available Available time is Then the steps Start date:

[0065] Port crane - hatch allocation (spacing constraints) Assemble at the hatch for work:

[0066] Allocation rules: Each port crane working hatch range Homework pending, meeting the following requirements: Port machines do not overlap: (The operating hatches of the port machinery do not overlap). Port crane interval operations: (There must be at least one hatch between each port aircraft). This indicates the hatch where the target port machinery is operating.

[0067] Single port crane workload calculation HAECO In the steps (unloading as a whole) (Step) to the hatch workload Allocation based on "capability ratio":

[0068] Step-by-step operation time calculation Scenario 1: No port crane additions or reductions ( constant) step The time is determined by the operating time of the maximum capacity port crane:

[0069] Scenario 2: Port crane additions and subtractions (available) exist (Changes over time) step Duration is Workload .

[0070] Material yard site switching Exhausting heap addresses: ( ) Site acceptance:

[0071] Unused heap address:

[0072] Ship Balance By using a relative difference method to evaluate the feasibility of unloading schemes and to better reflect the balance of the ship's hull after unloading, the current model sets the weight difference between the fore and aft compartments to not exceed 20%.

[0073] Encoding and Scheme Population Generation The unloading schemes are encoded using two-dimensional data. Each row represents whether there is port machinery operation at each hatch in the current step, and each column represents which steps involve port machinery operation at that hatch. The ship is divided into upper, middle, and lower sections according to a "4-4-2" single-hatch unloading ratio. For the hatch currently undergoing port machinery operation in the current step, the value is set to 1. The corresponding workload is calculated, and a set number of unloading scheme populations are generated based on the rule of "random allocation of port machinery and hatches + dynamic switching of stacking locations," denoted as the scheme set. The following is a case study of a ship unloading its cargo in a 9-cabin configuration, initially with three port cranes, which was later reduced to two and one port crane.

[0074]

[0075] Taking the first line as an example, the first line indicates that under the first unloading step, the three port cranes are operating the third, fifth, and seventh hatches respectively.

[0076] Selection-Crossover-Mutation Selection: The tournament selection method prioritizes solutions that satisfy the following conditions: the difference between the fore and aft compartments of the vessel is less than the set value (i.e., the hull is balanced) and the safe operating distance between the port machinery is met (i.e., there must be at least one hatchway between the two port machinery).

[0077] Crossover: Chromosome exchange occurs by randomly selecting the crossover type. Crossover types are divided into intra-population chromosome exchanges at the upper, middle, and lower stages and inter-population chromosome exchanges at the upper, middle, and lower stages, with single-point crossover.

[0078] Variation: Random variation is performed according to the single port crane operation procedure and within the scope of the operation hatch. The mutated operation hatch is then exchanged with the corresponding conflicting hatch of the port crane.

[0079] fitness function The schemes are evaluated by establishing evaluation items in the fitness function. The evaluation items include multiple indicators such as hull balance, total operation time, step completeness, belt system matching, and number of port machinery changes. The schemes are then ranked and screened according to the evaluation results.

[0080] Repeat the above genetic algorithm until the set number of iterations or fitness convergence condition is reached, and use the result with the best fitness as the output of the unloading scheme.

[0081] 5. Dynamic management and control of the entire unloading process Based on the ship unloading model, a ship unloading operation guide is generated to guide the central control dispatch and port machinery operators. Simultaneously, real-time data on conveyor belt flow, port machinery operating hatches, and actual port machinery capacity are collected. The workload, remaining capacity, ship fore-and-aft hold balance deviation, and unloaded and unloadable quantities at each operating hatch are calculated concurrently. The planned and actual quantities are dynamically analyzed and compared with set thresholds. When the deviation exceeds the threshold, an optimization process is dynamically triggered, i.e., based on the latest system status data, the process is re-run to generate a new optimal solution. The optimization process is also triggered when a major event (such as equipment failure or task completion) is detected.

[0082] The present invention also provides a dynamic unloading system for bulk cargo ships based on multi-site coordination. The dynamic unloading system for bulk cargo ships based on multi-site coordination can be implemented by executing the process steps of the dynamic unloading method for bulk cargo ships based on multi-site coordination. That is, those skilled in the art can understand the dynamic unloading method for bulk cargo ships based on multi-site coordination as a preferred embodiment of the dynamic unloading system for bulk cargo ships based on multi-site coordination.

[0083] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0084] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A dynamic unloading method for bulk cargo ships based on multi-site coordination, characterized in that, include: Step S1: Construct a knowledge base for dock operations and a knowledge base for stockpiles in the material yard; Step S2: Collect operation data in real time; Step S3: Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plan, generate a multi-location list, which includes the product name and the available time for each material location; Step S4: Based on the multi-address list, generate a multi-address selection sequence scheme for each simultaneously operating conveyor belt system through a dynamic address selection model; Step S5: Based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume and multi-material site selection sequence scheme, generate the optimal unloading scheme through the ship unloading model; Step S6: Guide the unloading operation according to the unloading plan. Based on the real-time collected operation data and unloading conveyor belt volume, dynamically analyze the deviation between the planned volume and the actual volume. When the deviation exceeds the threshold or a major event is detected, repeat steps S3-S5.

2. The method according to claim 1, characterized in that, The wharf operation knowledge base includes wharf berth water depth, wharf operable vessels, wharf operable types, port machinery operable types and corresponding operational capabilities, port machinery compartment operation capabilities, port machinery operation restriction rules, port machinery operation corresponding belt conveyor systems, and the relationship between belt conveyor system operable types.

3. The method according to claim 1, characterized in that, The knowledge base for material yards and stockpiles includes: the configuration of material yards, material strips, storage locations, and stackers; the operational capacity of stacking equipment; the relationship between material strips and stackers; the types of material strips that can be stacked; and the matching of belt conveyor systems with material strips in the material yard.

4. The method according to claim 1, characterized in that, Step S2 includes: Obtain basic ship information, information on the types and quantities of ship compartments, and information on the types and weights of goods unloaded from compartments in the port. Obtain information on the status and maintenance plans of port machinery, conveyor belt systems, and stacker cranes; Acquire information on changes in the port crane's alignment hatches and instantaneous flow rate information of the belt scale within a set time period; Obtain the corresponding dynamic stockpile inventory information from the rear material yard.

5. The method according to claim 1, characterized in that, In step S4, the material address dynamic selection model is used to perform: Sub-step S4.1: Construct an optimization model based on multiple constraints, where the multiple constraints include: Resource constraints: Multiple ships operate together, each carrying different materials that need to be unloaded, and multiple conveyor belt systems operate in parallel. Spatiotemporal constraints: Each material bar corresponds to one conveyor belt system, and other material bar locations are unavailable during the period when the material bar location is occupied; Operational constraints: Only one type of material can be stored at a single storage location to avoid mixing different types of materials in the same storage location; Sub-step S4.2: Construct a feasible initial solution, wherein the construction principle is to prioritize allocating the most efficient and as many port machines as possible and small-capacity material sites to the most urgent or important tasks at present; Sub-step S4.3: Using neighborhood search, multiple neighborhood solutions are generated near the current solution through preset neighborhood operation operators. New solutions inferior to the current solution are accepted with a predetermined probability to avoid local optima traps and explore the global optimum region. Sub-step S4.4: When a potential optimal solution distribution region is identified, an enhanced local search is initiated. High-density neighborhood exploration is performed within the optimal solution distribution region to locate the local optimum within the optimal solution distribution region. Sub-step S4.5: Based on the current local optimum, repeat sub-steps S4.3 and S4.4 until the iteration termination condition is met, and obtain multiple local optima. Sub-step S4.6: Evaluate each local optimum using the objective function, sort them according to their scores, and select the top-scoring multi-material address selection order schemes.

6. The method according to claim 5, characterized in that, In sub-step S4.1, the expression for optimizing the model is as follows: In the formula, t represents the time period, T is the set of time periods in discrete time points; b is the belt system, B is the set of belt systems; s is the material strip, S is the set of material strips; v is the ship, V is the set of ships; m is the material on ship v. It is the collection of materials on ship v; l It is the stacking address of material bar s. It is the set of stack addresses for material bar s; This is a binary variable indicating whether, at time t, material m from ship v is unloaded onto feed bar s via belt b. This is a binary variable representing the heap address at time t. Is it occupied by the material m of ship v? 7. The method according to claim 5, characterized in that, In sub-step S4.6, the construction of the objective function includes: determining the optimization objective that maximizes the utilization rate of terminal resources and operational efficiency, wherein, Optimization objectives include: By rationally planning the correspondence between the belt conveyor system and the material location and the operation sequence, the common operation time of multiple belt conveyor systems can be maximized. The expression is: In terms of material location selection logic, priority is given to storing material strips with large capacity and numerous existing material piles, as expressed in the following expression: Within the same material bar, priority is given to stack addresses with smaller remaining stackable capacity, expressed as: Multiple optimization objectives are integrated using a weighted summation method to form the objective function: in, A binary variable indicating whether belt b is actively unloading material from strip s at time t; For heap address l capacity, These are the weighting coefficients.

8. The method according to claim 5, characterized in that, In step S5, the ship unloading model is used to perform: Sub-step S5.1: Obtain working parameters, which include: the target port machinery's operating capacity for the target product under the target step, the available time for the target port machinery to complete the preceding operations, the number of available port machinery for the target step, the total unloading volume of the target hatch for the target product, the operating volume of the target hatch for the target product in the target step, the number of hatches that need to complete the operations in the target step, the operating hatch range of the target port machinery, the start time of the target step, the operating volume of the target port machinery, the operating time of the target step, and the multi-site selection sequence scheme. Sub-step S5.2: Based on the constraints of non-overlapping and spaced operation of port machinery and the dynamic switching principle of stacking sites, a predetermined number of unloading schemes are randomly generated according to the port machinery-hatch allocation. The dynamic switching principle of stacking sites is based on the multi-site selection order scheme and the current status of the stacking sites. The stacking sites include exhausted stacking sites, receiving stacking sites, and unused stacking sites. The non-overlapping constraint of port machinery means that the working hatch range of each port machinery does not overlap under the target step. The spaced operation constraint means that the interval between the hatches of adjacent port machinery under the target step is a preset number. Sub-step S5.3: By evaluating the preset evaluation items, score each unloading scheme and sort the schemes according to the scoring results; Sub-step S5.4: Expand the unloading scheme by selection-crossover-mutation, where: Select: Retain the solution that satisfies the difference between the fore and aft compartments of the vessel being less than the set value and the safe operating distance between the port machinery; Crossover: Chromosome exchange occurs by randomly selecting the crossover type. Crossover types are divided into intra-population chromosome exchanges within the upper, middle, and lower stages and inter-population chromosome exchanges within the upper, middle, and lower stages, with single-point crossover. Variation: Random variation is performed according to the single port crane operation procedure and within the scope of the operation hatch. The mutated operation hatch is exchanged with the corresponding conflicting hatch of the port crane. Sub-step S5.5: Repeat sub-steps S5.2 to S5.4 until the set number of iterations or convergence condition is reached, and use the result with the best fitness as the output of the unloading scheme.

9. The method according to claim 5, characterized in that, In sub-step S5.3, the preset evaluation items include hull balance, total operation time, step integrity, belt system matching, and number of port machinery compartment changes.

10. The system according to claim 1, characterized in that, A dynamic unloading system for bulk carriers based on multi-site coordination, characterized in that it includes: Module M1: Constructs a knowledge base for dock operations and a knowledge base for stockpiles in the material yard; Module M2: Real-time acquisition of operation data; Module M3: Based on the material yard knowledge base, material yard inventory and inbound / outbound information, and material usage plan, generate a multi-location list, which includes the product name and the available time for each material location; Module M4: Based on a multi-address list, it generates a multi-address selection sequence scheme for each simultaneously operating conveyor belt system through a dynamic address selection model; Module M5: Based on the terminal knowledge base, ship information, equipment tracking status information, ship compartment volume and multi-material site selection sequence scheme, the optimal unloading scheme is generated through the ship unloading model; Module M6: Guides the unloading operation according to the unloading plan. Based on the real-time collected operation data and unloading conveyor belt volume, it dynamically analyzes the deviation between the planned volume and the actual volume. When the deviation exceeds the threshold or a major event is detected, it repeatedly triggers modules M3-M5.

Images