Hydrogen energy unmanned ship cooperative optimization management system based on artificial intelligence

Abstract

The application relates to the technical field of unmanned ships, in particular to a hydrogen energy unmanned ship cooperative optimization management system based on artificial intelligence, which comprises an energy management hub unit, an intelligent flight path planning unit, a multi-ship cooperative decision unit, an energy-carbon coupling monitoring unit, a dynamic task allocation unit, an autonomous fault disposal unit and a shipborne equipment linkage unit; through the cooperation of the energy management hub and the energy-carbon coupling monitoring unit, the system realizes multi-source dynamic optimization of hydrogen energy, lithium batteries and renewable energy and accurate management of the whole chain carbon footprint, relies on intelligent flight path planning and multi-ship cooperative decision, combines an improved ant colony algorithm and distributed robust optimization, significantly improves the cluster path planning and conflict avoidance capability in a complex environment, and with the aid of a dynamic task allocation and autonomous fault disposal mechanism, the system fault tolerance and redundancy reconstruction level are enhanced while the efficient execution of the task is ensured.

Description

Technical Field

[0001] This invention relates to the field of unmanned vessel technology, specifically to a collaborative optimization management system for hydrogen-powered unmanned vessels based on artificial intelligence. Background Technology

[0002] Hydrogen-powered unmanned vessels are intelligent watercraft that are powered by hydrogen fuel and have autonomous navigation capabilities. Their core advantages are "zero emissions" and "long endurance"—the hydrogen fuel cell reaction only produces water, making it environmentally friendly. It also has high energy density and can support thousands of kilometers of navigation on a single refueling. It can perform tasks such as marine monitoring, border patrol, and scientific research exploration, reducing manpower costs and safety risks.

[0003] Existing hydrogen-powered unmanned surface vessel technologies have significant limitations. In terms of energy management, traditional systems mostly use a single hydrogen fuel cell for power supply, lacking a dynamic multi-energy coordination mechanism. This results in a fixed output ratio of hydrogen storage tank, lithium battery, and fuel cell, making it difficult to adapt to energy consumption fluctuations under complex operating conditions. Furthermore, the lack of coupling with carbon cost factors prevents low-carbon scheduling. Regarding multi-vessel coordination, existing solutions mostly rely on centralized control architectures. Cluster decision-making is susceptible to communication delays and interference from the marine environment, and distributed optimization capabilities are insufficient, leading to low task allocation efficiency and lagging collaborative collision avoidance strategies. In terms of carbon monitoring, carbon footprint tracking is mostly limited to end-of-life emissions and does not cover the carbon flow distribution across the entire hydrogen production, storage, and transportation chain. Carbon intensity assessment models are coarse and cannot support accurate low-carbon trajectory planning. Summary of the Invention

[0004] The purpose of this invention is to provide an artificial intelligence-based collaborative optimization management system for hydrogen-powered unmanned vessels to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a collaborative optimization management system for hydrogen-powered unmanned vessels based on artificial intelligence. The system includes an energy management central unit, an intelligent trajectory planning unit, a multi-vessel collaborative decision-making unit, an energy-carbon coupling monitoring unit, a dynamic task allocation unit, an autonomous fault handling unit, and a shipboard equipment linkage unit.

[0006] Preferably, the energy management central unit includes a hydrogen fuel cell control submodule, a composite energy storage management submodule, a multi-source acquisition and fusion submodule, an operating condition prediction submodule, an output optimization submodule, and a thermal management submodule. The output optimization submodule dynamically allocates the output ratio of the fuel cell, the hydrogen storage tank, and the lithium battery.

[0007] Preferably, the power allocation formula for the output optimization submodule is as follows:

[0008] P 总 =P FC +P H2 +P BAT ;

[0009] In the formula: P 总 P represents the total output power of the system. FC For the output power of the hydrogen fuel cell, P H2 P represents the dynamic energy release power of the hydrogen storage tank. BAT This refers to the output power of the lithium battery.

[0010] Preferably, the energy-carbon coupling monitoring unit includes a carbon emission tracing submodule, a carbon flow calculation submodule, a carbon efficiency assessment submodule, a low-carbon scheduling submodule, a carbon trading interface submodule, and an emission reporting submodule. The carbon flow calculation submodule is based on carbon emission flow theory to accurately track the carbon flow distribution throughout the entire process from hydrogen production, storage and transportation to energy conversion. The carbon efficiency assessment submodule quantifies the carbon emission intensity per unit voyage mileage.

[0011] Preferably, the carbon emission intensity formula of the carbon efficiency assessment submodule is as follows:

[0012] C 强度 =D 航行 / C 总 ;

[0013] In the formula: C 强度 C represents the carbon intensity per unit of voyage. 总 D represents the total carbon emissions over the entire flight. 航行 This represents the actual voyage distance.

[0014] Preferably, the intelligent trajectory planning unit includes an environmental perception submodule, a threat assessment submodule, an energy consumption constraint submodule, a trajectory generation submodule, a real-time correction submodule, and a cooperative collision avoidance submodule. The trajectory generation submodule employs an improved ant colony optimization algorithm. The energy consumption constraint submodule transforms the power prediction provided by the energy management central unit into a hard constraint for path search. The path planning constraint formula for the energy consumption constraint submodule is as follows:

[0015] E 约束 =E 预测 −ΔE 安全 ;

[0016] In the formula: E 约束 E is the upper limit of the hard constraint on energy consumption for trajectory planning. 预测 The power prediction value ΔE provided by the operating condition prediction submodule 安全 The threshold for safe redundancy energy consumption.

[0017] Preferably, the multi-vessel collaborative decision-making unit includes a cluster topology management submodule, an information consensus submodule, a strategy synchronization submodule, a task negotiation submodule, a distributed optimization submodule, and a human-computer interaction submodule, wherein the distributed optimization submodule applies the distributed bar optimization method.

[0018] Preferably, the dynamic task allocation unit includes a task parsing submodule, a capability assessment submodule, a benefit calculation submodule, a conflict detection submodule, a bidding decision submodule, and an allocation confirmation submodule. The benefit calculation submodule introduces Nash negotiation theory to construct a multi-objective benefit function that integrates task completion, energy consumption, and time cost.

[0019] The multi-objective benefit function of the benefit calculation submodule is shown below:

[0020] U=α·η 任务 +β·η 能效 −γ·C 碳 ;

[0021] In the formula: U represents the overall benefit of task execution, η 任务 η represents the task completion rate. 能效 For energy efficiency, C 碳 α represents the carbon cost coefficient, and β and γ are the weighting coefficients, respectively.

[0022] Preferably, the autonomous fault handling unit includes a health prediction submodule, an anomaly diagnosis submodule, a risk deduction submodule, a contingency plan retrieval submodule, a self-healing control submodule, and a redundancy reconstruction submodule. The risk deduction submodule uses an improved Benders decomposition algorithm to perform rapid fault impact analysis, and the redundancy reconstruction submodule draws on the logic of a bidirectional solid-state power controller.

[0023] Preferably, the shipborne equipment linkage unit includes a propulsion control submodule, a payload management submodule, a communication relay submodule, a navigation and positioning submodule, an execution drive submodule, and a status feedback submodule.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0025] This invention system, through the collaboration of an energy management center and an energy-carbon coupling monitoring unit, achieves multi-source dynamic optimization of hydrogen energy, lithium batteries, and renewable energy, as well as precise management of the entire carbon footprint. Relying on intelligent trajectory planning and multi-vessel collaborative decision-making, combined with improved ant colony algorithms and distributed robust optimization, it significantly enhances the cluster path planning and conflict avoidance capabilities in complex environments. With the help of dynamic task allocation and autonomous fault handling mechanisms, it enhances the system's fault tolerance and redundancy reconstruction level while ensuring efficient task execution, comprehensively improving the intelligent, low-carbon, and highly reliable operation capabilities of hydrogen-powered unmanned vessels. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the system in this invention;

[0027] Figure 2 System schematic diagram of the energy management central unit;

[0028] Figure 3 System schematic diagram of the carbon coupling monitoring unit;

[0029] Figure 4 System schematic diagram of the intelligent trajectory planning unit;

[0030] Figure 5 System schematic diagram of a multi-vessel collaborative decision-making unit;

[0031] Figure 6 System schematic diagram for the dynamic task allocation unit;

[0032] Figure 7 System schematic diagram of the autonomous fault handling unit;

[0033] Figure 8 This is a system schematic diagram of the shipborne equipment linkage unit. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] This invention relates to an AI-based collaborative optimization management system for hydrogen-powered unmanned surface vessels. The system comprises an energy management central unit, an intelligent trajectory planning unit, a multi-vessel collaborative decision-making unit, an energy-carbon coupling monitoring unit, a dynamic task allocation unit, an autonomous fault handling unit, and a shipboard equipment linkage unit. The energy management central unit is responsible for optimizing and scheduling the entire process of hydrogen production, storage, and consumption, providing continuous power to the vessels. The intelligent trajectory planning unit generates globally optimal navigation paths based on the marine environment and mission objectives. The multi-vessel collaborative decision-making unit acts as the intelligent brain of the vessel group, enabling autonomous coordination of group behavior. The energy-carbon coupling monitoring unit tracks the system's carbon footprint in real time and executes low-carbon scheduling. The dynamic task allocation unit... The system efficiently redistributes cluster tasks based on real-time situational awareness. The autonomous fault handling unit predicts and responds to internal and external risks to ensure mission reliability. The shipboard equipment linkage unit is the final execution layer of commands, controlling the ship's navigation and operations. Each unit collaborates with the AI ​​algorithm model through a data bus. The energy management central unit provides energy consumption constraints for the trajectory planning unit. The multi-ship collaborative decision-making unit sets collaborative rules for the dynamic task allocation unit. The energy-carbon coupling monitoring unit provides carbon emission cost factors for the energy management central unit. The autonomous fault handling unit provides safety redundancy strategies for the shipboard equipment linkage unit, forming a closed-loop intelligent management system from energy optimization and path planning to cluster collaborative execution.

[0036] The energy management central unit includes a hydrogen fuel cell control submodule, a composite energy storage management submodule, a multi-source acquisition and fusion submodule, an operating condition prediction submodule, an output optimization submodule, and a thermal management submodule. The multi-source acquisition and fusion submodule integrates status data from renewable energy sources such as wind and solar power, hydrogen fuel cells, and energy storage systems. The operating condition prediction submodule predicts future energy consumption trends based on the flight plan. The output optimization submodule dynamically allocates the output ratio of fuel cells, hydrogen storage tanks, and lithium batteries. The thermal management submodule improves overall energy efficiency through the recovery and utilization of waste heat from electrochemical reactions. The output optimization submodule of this unit is communicatively connected to the low-carbon scheduling submodule of the energy-carbon coupling monitoring unit, receiving real-time carbon costs to optimize the output strategy. At the same time, its operating condition prediction submodule outputs power prediction data to the energy consumption constraint submodule of the intelligent trajectory planning unit.

[0037] The power allocation formula for the output optimization submodule is shown below:

[0038] P 总 =P FC +P H2 +P BAT ;

[0039] In the formula: P 总 P represents the total output power of the system (unit: kW). FC P represents the output power of a hydrogen fuel cell (unit: kW). H2 The dynamic energy release power of the hydrogen storage tank (unit: kW), P BAT The formula represents the output power of the lithium battery (unit: kW). It describes the power coordination and distribution relationship between the fuel cell, hydrogen storage tank, and lithium battery in the energy management central unit, ensuring that the total power meets the navigation requirements.

[0040] The carbon-energy coupling monitoring unit includes a carbon emission tracing submodule, a carbon flow calculation submodule, a carbon efficiency assessment submodule, a low-carbon scheduling submodule, a carbon trading interface submodule, and an emission reporting submodule. The carbon flow calculation submodule is based on carbon emission flow theory to accurately track the carbon flow distribution throughout the entire process from hydrogen production, storage and transportation to energy conversion. The carbon efficiency assessment submodule quantifies the carbon emission intensity per unit of voyage. The low-carbon scheduling submodule generates operating strategies that take carbon costs into account. The carbon flow calculation submodule of this unit communicates with the multi-source acquisition and fusion submodule of the energy management central unit to acquire energy data. The output of its low-carbon scheduling submodule serves as a key input to the output optimization submodule in the energy management central unit, guiding the system to operate in a low-carbon manner.

[0041] The carbon emission intensity formula for the carbon efficiency assessment submodule is shown below:

[0042] C 强度 =D 航行 / C 总 ;

[0043] In the formula: C强度 Carbon emission intensity per unit of voyage (unit: gCO2 / nm), C 总 Total carbon emissions for the entire flight (unit: gCO2), D 航行 The formula represents the actual voyage distance (in km). It quantifies the carbon efficiency assessment logic of the carbon coupling monitoring unit during the voyage and is directly related to the low-carbon scheduling strategy.

[0044] The intelligent trajectory planning unit includes an environmental perception submodule, a threat assessment submodule, an energy consumption constraint submodule, a trajectory generation submodule, a real-time correction submodule, and a cooperative collision avoidance submodule. The trajectory generation submodule uses an improved ant colony optimization algorithm to plan a globally optimal path that satisfies multiple constraints in three-dimensional space. The energy consumption constraint submodule transforms the power prediction provided by the energy management central unit into hard constraints for path search. The cooperative collision avoidance submodule of this unit communicates with the strategy synchronization submodule of the multi-ship cooperative decision-making unit to receive cluster cooperative instructions, ensuring that the trajectory is conflict-free. Its planned trajectory data is synchronously sent to the navigation and positioning submodule of the shipborne equipment linkage unit as the basis for execution.

[0045] The path planning constraint formula for the energy consumption constraint submodule is as follows:

[0046] E 约束 =E 预测 −ΔE 安全 ;

[0047] In the formula: E 约束 The upper limit of hard constraints on energy consumption for trajectory planning (unit: kWh), E 预测 Power prediction values ​​(unit: kWh) provided for the operating condition prediction submodule, ΔE 安全 The safety redundancy energy consumption threshold (unit: kWh) is represented by this formula, which reflects how the energy consumption constraint submodule in the intelligent trajectory planning unit transforms energy prediction data into hard boundary conditions for path search.

[0048] The multi-vessel collaborative decision-making unit includes a cluster topology management submodule, an information consensus submodule, a strategy synchronization submodule, a task negotiation submodule, a distributed optimization submodule, and a human-computer interaction submodule. The distributed optimization submodule applies the bibliometric optimization method to address the uncertainties of the marine environment and communication, and solves the optimal collaborative strategy for the cluster. The strategy synchronization submodule of this unit communicates with the collaborative collision avoidance submodule of the intelligent trajectory planning unit and issues collaborative commands. Its task negotiation submodule interacts with the bidding decision submodule of the dynamic task allocation unit to jointly determine the final task allocation scheme.

[0049] The dynamic task allocation unit includes a task analysis submodule, a capability assessment submodule, a benefit calculation submodule, a conflict detection submodule, a bidding decision submodule, and an allocation confirmation submodule. The benefit calculation submodule introduces Nash negotiation theory to construct a multi-objective benefit function that integrates task completion, energy consumption, and time cost. The allocation confirmation submodule of this unit communicates with the task negotiation submodule of the multi-ship collaborative decision-making unit to report the allocation results. Its task analysis submodule receives the real-time shipboard resource status returned by the status feedback submodule of the shipboard equipment linkage unit to realize dynamic allocation.

[0050] Multi-objective benefit function of the benefit calculation submodule:

[0051] U=α·η 任务 +β·η 能效 −γ·C 碳 ;

[0052] In the formula: U represents the overall benefit of task execution, η 任务 η represents the task completion rate (on a scale of 0 to 1). 能效 Energy efficiency (unit: kW·h / nm), C 碳 The carbon cost coefficient (unit: yuan / gCO2) is given by α, β, and γ, which are weighting coefficients (∑=1). This formula reflects the application of Nash negotiation theory in dynamic task allocation units, and balances task completion, energy efficiency and carbon cost through multi-objective weighted balancing.

[0053] The autonomous fault handling unit includes a health prediction submodule, an anomaly diagnosis submodule, a risk simulation submodule, a contingency plan retrieval submodule, a self-healing control submodule, and a redundancy reconfiguration submodule. The risk simulation submodule uses an improved Benders decomposition algorithm for rapid fault impact analysis. The redundancy reconfiguration submodule borrows the logic of a bidirectional solid-state power controller to achieve redundant switching between the energy and propulsion systems. The anomaly diagnosis submodule of this unit is communicatively connected to the status feedback submodule of the shipborne equipment linkage unit to obtain fault symptoms. Its self-healing control submodule directly sends commands to the propulsion control submodule of the shipborne equipment linkage unit to perform fault-tolerant operations.

[0054] The shipborne equipment linkage unit includes a propulsion control submodule, a payload management submodule, a communication relay submodule, a navigation and positioning submodule, an execution drive submodule, and a status feedback submodule. The execution drive submodule receives trajectory instructions from the intelligent trajectory planning unit and collaborative instructions from the multi-ship collaborative decision-making unit, and controls the propellers and steering gears to complete actions. The status feedback submodule of this unit is communicatively connected to the health prediction submodule of the autonomous fault handling unit and continuously uploads equipment health data. Its navigation and positioning submodule provides high-precision position information to the real-time correction submodule of the intelligent trajectory planning unit to achieve closed-loop navigation.

[0055] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A collaborative optimization management system for hydrogen-powered unmanned vessels based on artificial intelligence, characterized in that: The system includes an energy management central unit, an intelligent trajectory planning unit, a multi-ship collaborative decision-making unit, an energy and carbon coupling monitoring unit, a dynamic task allocation unit, an autonomous fault handling unit, and a shipboard equipment linkage unit.

2. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The energy management central unit includes a hydrogen fuel cell control submodule, a composite energy storage management submodule, a multi-source acquisition and fusion submodule, an operating condition prediction submodule, an output optimization submodule, and a thermal management submodule. The output optimization submodule dynamically allocates the output ratio of the fuel cell, hydrogen storage tank, and lithium battery.

3. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 2, characterized in that: The power allocation formula for the output optimization submodule is as follows: P 总 =P FC +P H2 +P BAT ; In the formula: P 总 P represents the total output power of the system. FC For the output power of the hydrogen fuel cell, P H2 P represents the dynamic energy release power of the hydrogen storage tank. BAT This refers to the output power of the lithium battery.

4. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The energy-carbon coupling monitoring unit includes a carbon emission tracing submodule, a carbon flow calculation submodule, a carbon efficiency assessment submodule, a low-carbon scheduling submodule, a carbon trading interface submodule, and an emission reporting submodule. The carbon flow calculation submodule is based on carbon emission flow theory to accurately track the carbon flow distribution throughout the entire process from hydrogen production, storage and transportation to energy conversion. The carbon efficiency assessment submodule quantifies the carbon emission intensity per unit of voyage.

5. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 4, characterized in that: The carbon emission intensity formula for the carbon efficiency assessment submodule is shown below: C 强度 =D 航行 / C 总 ; In the formula: C 强度 C represents the carbon intensity per unit of voyage. 总 D represents the total carbon emissions over the entire flight. 航行 This represents the actual voyage distance.

6. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The intelligent trajectory planning unit includes an environmental perception submodule, a threat assessment submodule, an energy consumption constraint submodule, a trajectory generation submodule, a real-time correction submodule, and a cooperative collision avoidance submodule. The trajectory generation submodule employs an improved ant colony optimization algorithm. The energy consumption constraint submodule transforms the power prediction provided by the energy management central unit into hard constraints for path search. The path planning constraint formula for the energy consumption constraint submodule is as follows: E 约束 =E 预测 −ΔE 安全 ; In the formula: E 约束 E is the upper limit of the hard constraint on energy consumption for trajectory planning. 预测 The power prediction value ΔE provided by the operating condition prediction submodule 安全 The threshold for safe redundancy energy consumption.

7. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The multi-vessel collaborative decision-making unit includes a cluster topology management submodule, an information consensus submodule, a strategy synchronization submodule, a task negotiation submodule, a distributed optimization submodule, and a human-computer interaction submodule. The distributed optimization submodule applies the distributed bar optimization method.

8. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The dynamic task allocation unit includes a task parsing submodule, a capability assessment submodule, a benefit calculation submodule, a conflict detection submodule, a bidding decision submodule, and an allocation confirmation submodule. The benefit calculation submodule introduces Nash negotiation theory to construct a multi-objective benefit function that integrates task completion, energy consumption, and time cost. The multi-objective benefit function of the benefit calculation submodule is shown below: U=a·h 任务 +b·h 能效 −γ·C 碳 ; In the formula: U represents the overall benefit of task execution, η 任务 η represents the task completion rate. 能效 For energy efficiency, C 碳 α represents the carbon cost coefficient, and β and γ are the weighting coefficients, respectively.

9. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The autonomous fault handling unit includes a health prediction submodule, an anomaly diagnosis submodule, a risk deduction submodule, a contingency plan retrieval submodule, a self-healing control submodule, and a redundancy reconstruction submodule. The risk deduction submodule uses an improved Benders decomposition algorithm to perform rapid fault impact analysis, and the redundancy reconstruction submodule draws on the logic of a bidirectional solid-state power controller.

10. The AI-based collaborative optimization management system for hydrogen-powered unmanned vessels according to claim 1, characterized in that: The shipborne equipment linkage unit includes a propulsion control submodule, a payload management submodule, a communication relay submodule, a navigation and positioning submodule, an execution drive submodule, and a status feedback submodule.

Images