A method and system for modular assembly task planning of an offshore vessel-bridge under dynamic sea conditions

Abstract

The application provides a kind of offshore boat bridge modular assembly task planning method and system under dynamic sea state, relate to the technical field of ocean engineering automation and task planning.The method adopts three-layer architecture of hierarchical state space planning+physical field potential energy guidance+rolling time domain re-planning, through HTN task atomization decomposition, four-dimensional space-time interference tensor construction, stability entropy quantification operation risk, NSGA-III multi-objective solution and dynamic regret value feedback re-planning key technology, realizes the accurate coupling of floating box module splicing topological constraint and dynamic change of sea state, effectively solves the pain points such as poor environmental adaptability, multi-objective conflict difficult to balance of traditional method, finally achieves the technical effect of shortening the total operation time and no collision accident, significantly improves the safety, robustness and space-time utilization rate of offshore boat bridge assembly, and can be expanded to the automation assembly scene of large-scale modular facilities such as artificial island and offshore base.

Description

Technical Field

[0001] This invention relates to the field of marine engineering automation and mission planning technology, specifically to a method and system for mission planning of modular assembly of offshore pontoon bridges under dynamic sea conditions. Background Technology

[0002] Offshore pontoon bridges, as key facilities for the rapid transport of supplies between ships and shorelines, play an irreplaceable role in scenarios such as naval replenishment and emergency rescue. Their assembly process requires the coordination of dozens of pontoon modules, powered pontoon bridges, and locking mechanisms, involving complex task logic and multi-agent scheduling, placing extremely high demands on the safety, dynamic adaptability, and efficiency of planning methods.

[0003] Currently, task planning for modular assembly of offshore pontoon bridges mainly relies on static Gantt charts or traditional job shop scheduling (JSP) algorithms. However, in actual marine environments, these traditional methods have significant technical shortcomings: First, they lack environmental coupling. Traditional planning only focuses on the pre-existing logical constraints between tasks (such as the order of module assembly), without considering the physical limitations imposed by sea conditions such as waves and currents on specific assembly actions. This leads to a disconnect between the planned scheme and the actual operating environment, easily causing operational risks. Second, they lack the ability to handle dynamic uncertainties. Sea conditions have significant time-varying characteristics, with frequent extreme situations such as sudden gusts and swells. Static planning schemes cannot respond to environmental changes in real time, easily causing accidents such as module collisions and connector damage during assembly, affecting the continuity of operations. Finally, it is difficult to balance multiple conflicting objectives. Traditional methods often unilaterally pursue assembly speed, resulting in insufficient operational stability; if safety is overemphasized, the project duration will be significantly extended, failing to achieve a dynamic optimal trade-off between efficiency and safety.

[0004] In summary, existing task planning methods are insufficient to meet the actual needs of modular assembly of offshore pontoon bridges under dynamic sea conditions. There is an urgent need for an efficient planning method that can couple environmental data, quantify operational risks, and dynamically adjust planning strategies to address the aforementioned technical challenges. Summary of the Invention

[0005] This invention addresses the technical problems of traditional offshore pontoon bridge planning methods, which lack environmental coupling, fail to consider the physical constraints of sea conditions on splicing operations, leading to a disconnect between the plan and the actual operating environment; static planning cannot cope with the dynamic uncertainties brought about by time-varying sea conditions, easily causing accidents such as module collisions and connector damage; and it is difficult to balance the multi-objective conflict between "assembly speed" and "operational stability" and achieve dynamic optimal balance. It provides a task planning system that can satisfy the hard logical constraints of the module splicing topology, respond to changes in sea condition data in real time, and dynamically adjust the assembly sequence by quantifying operational risks, achieving an optimal trade-off between efficiency and safety.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for planning the modular assembly of offshore pontoon bridges under dynamic sea conditions, the method comprising: The top-level task objective of pontoon bridge assembly is analyzed, and the top-level task objective is decomposed into a sequence of atomic tasks with logical dependencies based on the hierarchical task network (HTN); wherein, the pontoon bridge is an offshore pontoon bridge composed of multiple floating box modules and shore connectors. For atomic tasks, a four-dimensional spatiotemporal grid is established for the operational sea area, the motion envelope of each pontoon module under dynamic sea conditions is calculated, and a dynamic interference potential energy tensor reflecting spatial conflicts and time occupancy is constructed. By combining wave spectrum data with the hydrodynamic response characteristics of the floating box module, a stability entropy evaluation function is constructed to quantify the physical operational risks of each atomic task within a time window; With minimizing the total project duration and minimizing the cumulative stability entropy as multiple optimization objectives, an evolutionary algorithm is used to search for the optimal task execution sequence that satisfies topological constraints and interference constraints; The system monitors the actual progress of the task in real time, calculates the dynamic regret value, and triggers local replanning when the dynamic regret value exceeds a preset threshold.

[0007] On the other hand, the present invention also provides a task planning system for modular assembly of offshore pontoon bridges under dynamic sea conditions. The system includes a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the system is triggered to execute the above-mentioned task planning method for modular assembly of offshore pontoon bridges under dynamic sea conditions.

[0008] Compared with the prior art, the beneficial effects of the present invention are: 1. Significantly improved operational safety: By quantifying operational risks through stability entropy, the algorithm can dynamically match task difficulty with sea conditions, execute high-difficulty locking tasks during stable windows, and arrange low-risk tasks during adverse sea conditions, with no accidents such as module collisions or connector damage throughout the process.

[0009] 2. Strong dynamic adaptability and robustness: The dynamic regret value feedback mechanism can respond to sudden changes in sea state with minimal computational cost, avoid the lag caused by global recalculation, ensure the continuous and stable operation of the planning system, and eliminate the need to passively wait for a fault to occur before making adjustments.

[0010] 3. Optimization of work efficiency and time and space utilization: The four-dimensional spatiotemporal interference tensor supports parallel and close operation of multiple modules. Combined with the third-generation non-dominated sorting genetic algorithm (NSGA-III), it achieves the optimal balance between total project time and risk. Compared with traditional methods, the total project time is shortened, and it can be extended to the automated assembly scenarios of large modular facilities such as artificial islands and offshore bases.

[0011] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0012] Figure 1 This is a flowchart of the task planning method for modular assembly of offshore pontoon bridges under dynamic sea conditions according to the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings, so as to more clearly understand the purpose, features and advantages of this invention. It should be understood that the embodiments shown in the drawings are not intended to limit the scope of this invention, but are only for illustrating the essential spirit of the technical solutions of this invention. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0014] Unless the context requires otherwise, throughout the specification and claims, the word “comprising” and its variations, such as “including” and “having”, shall be understood to have an open, inclusive meaning, that is, to be interpreted as “including, but not limited to”.

[0015] Throughout this specification, references to "an embodiment" or "an embodiment" indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Therefore, the appearance of "in an embodiment" or "an embodiment" in various places throughout the specification does not necessarily refer to the same embodiment. Furthermore, a particular feature, structure, or characteristic may be combined in any manner in one or more embodiments.

[0016] The singular forms “a” and “the” used in this specification and the appended claims include plural references unless otherwise expressly stated herein. It should be noted that the term “or” is generally used to mean “and / or” unless otherwise expressly stated herein.

[0017] In the following description, in order to clearly demonstrate the structure and working method of the present invention, a number of directional terms will be used. However, terms such as "front", "back", "left", "right", "outside", "inside", "outward", "inward", "up", and "down" should be understood as convenient terms and not as limiting terms.

[0018] The implementation details of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following content is only for the convenience of understanding the implementation details and is not necessary for implementing this solution.

[0019] The purpose of this invention is to address the shortcomings of the aforementioned background technologies by employing a three-layer architecture of "hierarchical state-space planning + physical field potential energy guidance + rolling temporal replanning," combined with stability entropy constraints and a dynamic regret feedback mechanism, to achieve efficient and accurate planning for the modular assembly of offshore pontoon bridges. Figure 1 As shown, the specific technical solution for a task planning method for modular assembly of offshore pontoon bridges under dynamic sea conditions is as follows: Step 1: Task atomization decomposition based on HTN; Based on the design drawings of the offshore platform, the top-level task objective of "assembling the pontoon bridge" is recursively decomposed into a sequence of atomic tasks with logical dependencies, such as "towing", "aligning", and "locking", using the Hierarchical Task Network (HTN), and a task dependency graph is output. The assembly pontoon bridge is a simple offshore pontoon bridge composed of multiple pontoon modules and several shore connectors, such as a simple offshore pontoon bridge composed of 6 pontoon modules (numbered M1-M6) and 1 shore connector (numbered C1).

[0020] Specifically, first, input the top-level task. The ultimate goal is to complete the orderly assembly of multiple floating box modules and shore connectors to form a complete pontoon bridge channel.

[0021] Next, a hierarchical task network (HTN) is used for recursive decomposition, defining the mapping relationship between core task methods and atomic tasks:

[0022]

[0023] Similarly, the task of connecting all pontoon modules and connecting the pontoon modules to the beach connector is decomposed. "Tow_X" means that the tugboat tows the pontoon module X to the target area, "Align_X" means that the attitude of the pontoon module X is adjusted to meet the splicing accuracy requirements (position deviation ≤ 0.5m, angle deviation ≤ 1°), and "Lock_X" means that the mechanical locking mechanism is activated to fix the pontoon module.

[0024] Then, the task dependency graph is output, represented by a Directed Acyclic Graph (DAG), where the nodes are the atomic tasks mentioned above. (like" Tow_M 2” Align_M 2"), directed edges represent the pre-existing constraints between atomic tasks (such as... (That is, the drag task of M2 can only be started after M1 is locked), forming a set of feasible task chains that satisfy the topological logic constraints.

[0025] Step 2: Construct a four-dimensional spacetime interference tensor; A four-dimensional spatiotemporal grid is established for the operational sea area. The motion envelope of each pontoon module under dynamic sea conditions is calculated, and a dynamic interference potential energy tensor reflecting spatial conflict and time occupation is constructed.

[0026] Specifically, first, define the operational sea area grid (x, y) and the discrete time axis t, where x and y cover the area required for the assembly operation, and t matches the sea state prediction time domain.

[0027] Next, for each atomic task Based on the hydrodynamic model and real-time sea state data, the motion envelope radius at different times t is calculated. The expression is:

[0028] in, The physical dimensions of the floating box module. The drift coefficient, For real-time significant wave height; for example, when hour, This indicates that the spatial safety envelope of the floating box module at this moment is a circular area with a radius of 4m centered at the mission execution center.

[0029] Then, construct the dynamic interference potential energy tensor. Traverse the spatiotemporal envelope of all atomic tasks: if atomic tasks Envelope region and atomic task The envelope region is in the same spacetime coordinate. If there is overlap (i.e., the overlap value is greater than 1), then the tensor value at that coordinate is set as the collision penalty value. This creates a soft constraint penalty for tasks that conflict with the time and space.

[0030] Step 3: Calculate the stability entropy of the job window; By combining wave spectrum data with the hydrodynamic response characteristics of the floating box module, a stability entropy evaluation function is constructed to quantify the physical operational risks of each atomic task within a time window.

[0031] Specifically, retrieve the currently executing atomic task from the preset database. The corresponding amplitude response operator (RAO) curve of the floating box module This curve characterizes the attitude response amplitude of the pontoon module under the action of waves at different frequencies.

[0032] Extracting the wave spectrum at time t Response variance is calculated through convolution operations. The response variance reflects the intensity of the movement of the floating box module at that moment.

[0033] Response variance Substitute into the stability entropy calculation formula:

[0034] in, The limiting angle for the operation; Through stability entropy Quantify tasks The physical operation risk performed at time t. For example, when the sea state is calm ( When it is smaller), A smaller value indicates a lower operational risk; when sea conditions deteriorate ( (increase) A surge in the value indicates that this is not a suitable time to perform a high-difficulty splicing task.

[0035] Step 4: Solve using a multi-objective genetic algorithm; The third-generation non-dominated sorting genetic algorithm NSGA-III is employed, with the optimization objectives of minimizing the total project duration and minimizing the cumulative stability entropy. Combined with spatiotemporal interference tensor constraints, it searches for Pareto optimal solutions and outputs the optimal task instruction sequence with timestamps. NSGA-III is an evolutionary algorithm for multi-objective optimization problems. It uses non-dominated sorting to select non-dominated solutions and introduces a reference point mechanism to maintain solution diversity. It can efficiently search for Pareto optimal solutions in multi-objective conflicts such as "completion time" and "stability entropy," adapting to the task planning requirements under complex constraints.

[0036] Specifically, chromosome encoding is performed first: an integer encoding method based on task priority is used, the chromosome length is equal to the total number of atomic tasks, and the encoded value is the task number. For example, chromosome "1,3,2,5,4,..." represents a task. Prioritize execution, then... , wait.

[0037] Define the objective function: Objective function 1: That is, to minimize the total project duration, using the latest completion time of all atomic tasks as the evaluation metric; Objective function 2: That is, minimizing the cumulative stability entropy, which reflects the total operational risk of the entire assembly process.

[0038] Constraint handling: When calculating fitness during chromosome decoding, the interference potential tensor penalty from step 2 is superimposed. ,but If we set the value to infinity, we can rule out the possibility that this sequence is the optimal solution.

[0039] The improved third-generation non-dominated sorting genetic algorithm (NSGA-III) is run, and through selection, crossover, and mutation operations, iterates 50 times to search for a Pareto optimal solution set, ultimately outputting a set of optimal task instruction sequences with timestamps. Clearly define the start and end times for each atomic task.

[0040] Step 5: Dynamic regret value feedback and replanning.

[0041] Real-time monitoring of actual task progress, calculation of dynamic regret value, and when the dynamic regret value continues... K The number of control cycles exceeded the preset threshold. At that time, local replanning is triggered.

[0042] Specifically, the operation status is monitored in real time, and the actual progress is obtained through sensor data from the tugboat and pontoon modules (such as...). At that time, it was completed (Task), and predict the optimal cost of the remaining tasks. (Optimal project duration and total risk based on the current sea conditions).

[0043] Assuming the original plan continues... Based on the original timestamps of the remaining tasks and current sea state changes, predict their final cost. .

[0044] Calculate dynamic regret value:

[0045] in, To maintain the projected final cost, This represents the optimal cost for the remaining tasks at the current moment.

[0046] when continuous K The number of control cycles exceeded the preset threshold. If the percentage is 20%, it means that the environmental change has caused the original plan to deteriorate by more than 20%, triggering a local replanning. The process returns to step 4 to recalculate the optimal task instruction sequence for the remaining tasks. If the threshold is not exceeded, the original plan continues to be executed.

[0047] Next, the implementation effect of the present invention will be verified, as follows: (1) Verification of working conditions The verification process is set as a dynamic sea state process: the initial sea state is level 3 ( , , From the 4th hour onwards, the sea state gradually increased to level 5. , , This simulates scenarios of deteriorating sea conditions that may be encountered in actual operations.

[0048] (2) Verification results Using the method of this invention: by monitoring stability entropy, the deterioration of sea conditions is predicted in the 4th hour. The high-difficulty "Lock_M3" and "Lock_M4" tasks are automatically moved to the 3rd-4th hour (the window of stable sea conditions) for completion. During the period of severe sea conditions from the 4th to the 8th hour, low-risk "Tow_M5" and "Tow_M6" tasks and equipment maintenance tasks are arranged. The total working time is 11.5 hours. There are no accidents such as module collisions or connector damage throughout the process. The safety and efficiency of the operation have both met expectations.

[0049] Compared with the traditional greedy algorithm, which forcibly executes the splicing task according to a fixed sequence when the sea conditions deteriorate, resulting in 3 module collision interruptions, extending the total working time to 14 hours and incurring equipment maintenance costs, the superiority of the method of this invention is verified.

[0050] Through the above specific implementation steps, the present invention realizes the precise planning of modular assembly of offshore pontoon bridges under dynamic sea conditions, effectively balancing operational efficiency and safety risks. It can be directly applied to actual scenarios such as naval replenishment and emergency rescue, and can also be extended to the automated assembly planning of large modular facilities such as artificial islands and offshore bases.

[0051] In some embodiments, the present invention also provides a task planning system for modular assembly of offshore pontoon bridges under dynamic sea conditions. The system includes a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the system is triggered to execute the above-described task planning method for modular assembly of offshore pontoon bridges under dynamic sea conditions.

[0052] This invention proposes a task planning method and system for modular assembly of offshore pontoon bridges under dynamic sea conditions. It adopts a three-layer architecture of "hierarchical state space planning + physical field potential energy guidance + rolling time domain replanning". Through key technologies such as HTN task atomization decomposition, four-dimensional spatiotemporal interference tensor construction, stability entropy quantification of operational risks, NSGA-III multi-objective solution and dynamic regret value feedback replanning, it achieves precise coupling between the topological constraints of pontoon module splicing and dynamic sea state changes. It effectively solves the pain points of traditional methods such as poor environmental adaptability and difficulty in balancing multi-objective conflicts. Finally, it achieves the technical effect of shortening the total operation time and eliminating collision accidents, significantly improving the safety, robustness and spatiotemporal utilization of offshore pontoon bridge assembly. It can be extended to the automated assembly scenarios of large modular facilities such as artificial islands and offshore bases.

[0053] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the invention by those skilled in the art without departing from the spirit and essence of the invention. Such modifications or substitutions should all fall within the scope of the invention, or any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the invention should be covered within the protection scope of the invention. Therefore, the protection scope of the invention should be determined by the scope of the claims.

Claims

1. A method for planning the modular assembly of offshore pontoon bridges under dynamic sea conditions, characterized in that, The method includes: The top-level task objective of pontoon bridge assembly is analyzed, and the top-level task objective is decomposed into a sequence of atomic tasks with logical dependencies based on the hierarchical task network (HTN); wherein, the pontoon bridge is an offshore pontoon bridge composed of multiple floating box modules and shore connectors. For atomic tasks, a four-dimensional spatiotemporal grid of the operational sea area is established, the motion envelope of each pontoon module under dynamic sea conditions is calculated, and a dynamic interference potential energy tensor reflecting spatial conflicts and time occupancy is constructed. Specifically, the following steps are taken: First, the operational sea area grid (x,y) and discrete time axis t are defined, where x and y cover the area required for assembly operations, and t matches the sea state prediction time domain. Next, for each atomic task Based on the hydrodynamic model and real-time sea state data, the motion envelope radius at different times t is calculated. The expression is: ； in, The physical dimensions of the floating box module. The drift coefficient, Real-time significant wave height; Then, construct the dynamic interference potential energy tensor. Traverse the spatiotemporal envelope of all atomic tasks: if atomic tasks Envelope region and atomic task The envelope region is in the same spacetime coordinate. If there is overlap, the tensor value at that coordinate is set as the collision penalty value. This creates a soft constraint penalty for tasks that conflict with the time and space; By combining wave spectrum data with the hydrodynamic response characteristics of the pontoon module, a stability entropy evaluation function is constructed to quantify the physical operational risk of each atomic task within a time window; among other things, the currently executing atomic task is retrieved from a preset database. The corresponding amplitude response operator curve of the floating box module This curve characterizes the attitude response amplitude of the pontoon module under the action of waves at different frequencies; Extracting the wave spectrum at time t Response variance is calculated through convolution operations. The response variance reflects the intensity of the movement of the floating box module at time t; response variance Substitute into the stability entropy calculation formula: ； in, The limiting angle for the operation; Through stability entropy Quantify tasks The physical operation risk performed at time t; With minimizing the total project duration and minimizing the cumulative stability entropy as multiple optimization objectives, an evolutionary algorithm is used to search for the optimal task execution sequence that satisfies topological constraints and interference constraints; The system monitors the actual progress of the operation in real time, calculates the dynamic regret value, and triggers local replanning when the dynamic regret value exceeds a preset threshold. Specifically, it monitors the operation status in real time by acquiring actual progress data from sensors on the tugboat and pontoon modules, and predicts the optimal cost for the remaining tasks. ; Continue with the original plan, and predict the final cost based on the original timestamps of the remaining tasks and current sea state changes. ; Calculate dynamic regret value : ； in, To maintain the projected final cost, The optimal cost for the remaining tasks at the current moment; when continuous K The number of control cycles exceeded the preset threshold. When this occurs, a local replanning is triggered, recalculating the optimal sequence of task instructions for the remaining tasks; if the threshold is not exceeded, the originally planned tasks continue to be executed; where... K It is a natural number.

2. The method according to claim 1, characterized in that, The top-level task objective of the pontoon bridge assembly is decomposed into a sequence of atomic tasks with logical dependencies based on a hierarchical task network (HTN), specifically including: First, input the top-level task. The ultimate goal is to complete the orderly assembly of multiple floating box modules and beach connectors to form a complete pontoon bridge channel. Next, a hierarchical task network (HTN) is used for recursive decomposition, defining the mapping relationship between core task methods and atomic tasks: ； ； The task of completing the connections between all pontoon modules and between the pontoon modules and the beach connectors is decomposed, among which, Indicates tugboat towing pontoon module To the target area, Indicates adjustment of the float module The orientation is adjusted to meet the requirements for splicing accuracy. This indicates that the mechanical locking mechanism has been activated to secure the float module. Indicates the beach connector number; Then, output the task dependency graph, represented by a directed acyclic graph (DAG), where nodes in the DAG are atomic tasks. Directed edges are atomic tasks The pre-existing constraints between them form a set of task chains that satisfy the topological logic constraints.

3. The method according to claim 1, characterized in that, The stability entropy Quantify tasks The risks of physical operations performed at time t include: When response variance Hours indicates calm sea conditions. A smaller value indicates lower operational risk; when An increase indicates a deterioration in sea conditions. An increase in the value indicates that this is not a suitable time to perform a high-difficulty task.

4. The method according to claim 3, characterized in that, The method, which uses minimizing the total project duration and minimizing the cumulative stability entropy as multiple optimization objectives, employs an evolutionary algorithm to search for the optimal task execution sequence that satisfies topological and interference constraints. Specifically, this includes: First, chromosome encoding is performed: an integer encoding method based on task priority is adopted, the chromosome length is equal to the total number of atomic tasks, and the encoded value is the task number; Then, the objective function is defined: Objective function 1: This represents minimizing the total project duration, using the latest completion time of all atomic tasks. As evaluation indicators; Objective function 2: , which represents minimizing the cumulative stability entropy, reflects the total operational risk of the entire assembly process; Constraint handling: When calculating fitness during chromosome decoding, an interference potential tensor penalty is superimposed. ,but Let it be infinity; Finally, the third-generation non-dominated sorting genetic algorithm is run, and through selection, crossover, and mutation operations, iterates 50 times to search for the Pareto optimal solution set, ultimately outputting a set of optimal task instruction sequences with timestamps. Specify the start time of each atomic task. With end time .

5. A modular assembly task planning system for offshore pontoon bridges under dynamic sea conditions, the system comprising a memory for storing computer program instructions and a processor for executing program instructions, characterized in that, When the computer program instructions are executed by the processor, the system is triggered to execute the task planning method for modular assembly of offshore pontoon bridges under dynamic sea conditions as described in any one of claims 1 to 4.

Images