CN-121185299-B - Ship route multi-target optimization model for composite sea condition scene

CN121185299BCN 121185299 BCN121185299 BCN 121185299BCN-121185299-B

Abstract

The invention discloses a multi-target optimization model of a ship route facing a composite sea condition scene, which relates to the technical field of ship route management and comprises the following steps of utilizing marine monitoring equipment, a sensor network and ship self equipment to collect ship navigation state data and marine environment data in real time and combining historical data to establish a motion model of a ship under different sea conditions; according to the invention, by constructing a floating obstacle dynamic attitude model, dynamic changes such as rolling, pitching and the like of the floating obstacle under the action of wind waves are captured in real time, and a time-varying risk field is constructed by combining with an uncertainty ellipse of a track. Compared with the traditional static obstacle processing mode, the model can accurately quantify the real-time change of the dynamic windward area, the resistance coefficient and the influence range of the obstacle, avoid the route planning error caused by neglecting the dynamic characteristics of the obstacle, update the motion state of the obstacle in real time through the dynamic attitude vector, enable the ship to avoid the risk area expanded by the change of the attitude of the obstacle in advance, and obviously reduce the collision probability.

Inventors

WANG YUCHUAN
YANG TAONING
LI RUIBIN
ZHU JUNCHI
FAN ZHENMING
LU LIXIN
LIU BOHAO
SUN YONGQIANG

Assignees

交通运输部水运科学研究所

Dates

Publication Date: 20260512
Application Date: 20250916

Claims (8)

1. The ship route multi-target optimization model for the composite sea condition scene is characterized by comprising the following steps of: the method comprises the steps of collecting ship navigation state data and marine environment data in real time by using marine monitoring equipment, a sensor network and ship self equipment, and establishing a motion model of a ship under different sea conditions by combining historical data; Taking a starting point and a destination as input, and generating an initial route by combining basic information of the ocean map; judging whether an obstacle exists on the initial route, if present: judging whether the obstacle is a floating obstacle or not, and determining: The method comprises the steps of obtaining shape information of an obstacle, constructing a floating obstacle dynamic attitude model, calculating a dynamic windward area, a resistance coefficient and an influence range radius of the floating obstacle, and constructing a time-varying risk field by combining a track uncertainty ellipse; Constructing an objective function according to the time-varying risk field and combining course and course variation, and calculating to obtain an optimal course; The dynamic attitude model of the floating obstacle is determined according to the following relation: ; wherein: Is the attitude vector of the floating obstacle; The rolling angle of the floating obstacle at the moment t; a pitch angle of the floating obstacle at the time t; the rolling angular velocity of the floating obstacle at the moment t; The pitch angle speed of the floating obstacle at the moment t; The risk field is constructed by the following specific steps: calculating the dynamic windward area and the dynamic resistance coefficient of the obstacle based on the established dynamic attitude model; determining the radius of the influence range of the floating obstacle according to the dynamic windward area and the resistance coefficient; Track uncertainty caused by the change of the obstacle posture is used for constructing a track uncertainty ellipse; and constructing a time-varying risk field by combining the calculated dynamic influence range radius R (t) of the floating obstacle and the track uncertainty ellipse parameter.
2. The composite sea condition scenario-oriented ship route multi-objective optimization model according to claim 1, wherein the running state of the floating obstacle is obtained, whether a fixed obstacle exists in a risk field of the floating obstacle is detected, and if so: The floating barrier and the fixed barrier have collision risk, potential collision points are constructed at the overlapping positions of the edges of the fixed barrier and the risk fields of the floating barrier, the collision direction of the floating barrier at each collision point is obtained, and the comprehensive collision risk area of the floating barrier is calculated; and calculating a new risk field of the floating obstacle.
3. The composite sea situation scenario-oriented ship route multi-objective optimization model according to claim 2, wherein the calculating the comprehensive collision risk area of the floating obstacle comprises the following specific steps: Collecting marine monitoring satellite images, shore radar monitoring data, ship automatic identification system data and underwater sonar detection data information; Carrying out space position matching and fusion on the fixed obstacle information detected by different data sources based on a geographic coordinate system, eliminating repeated detection and errors, and constructing an accurate fixed obstacle position database; Acquiring a real collision point according to the potential collision point, and calculating the collision direction of the collision instant floating obstacle; Training a machine learning model by utilizing historical collision data, current floating obstacle and fixed obstacle state and environmental condition data, inputting relevant characteristics of a collision point as a model, and outputting collision probability of the collision point; the method comprises the steps of obtaining structural design drawings and material specification information of fixed obstacles and floating obstacles, establishing a three-dimensional structure model, simulating stress and strain distribution of the obstacles in the collision process, determining weak parts and damage modes of the structure, and calculating total energy released in the collision process according to the mass and speed of the floating obstacles and the relative motion state of the floating obstacles in the collision moment; according to the damage degree, residual buoyancy and gravity center position change of the obstacle after collision, combining a ocean circulation model and real-time ocean current and tide data, predicting a drift path of the obstacle in the ocean, and further obtaining a physical collision risk area; The method comprises the steps of taking obstacle states at the moment of collision and marine environment parameters as boundary conditions, calculating local water flow speed field and pressure field changes caused by collision, establishing a ship motion mathematical model, taking the local water flow speed field and pressure field changes as external interference force, inputting the external interference force into the ship motion mathematical model, calculating the navigational speed and course deviation of a ship in affected water flow, and evaluating the influence degree of the water flow changes on ship control by combining the ship type, loading condition and control performance parameters; carrying out space-time alignment on the calculation results of the physical collision risk area and the fluid dynamics influence area, and merging the physical collision risk area and the fluid dynamics influence area by adopting a weighted fusion algorithm; According to the influence degree of different areas on ship navigation safety, different weights are given to a physical collision risk area and a fluid dynamics influence area, and a comprehensive collision risk field is generated.
4. The composite sea condition scenario-oriented ship route multi-objective optimization model according to claim 2, wherein the calculating of the new risk field of the floating obstacle comprises the following specific steps: Overlapping and fusing the comprehensive collision risk area of the floating obstacle and the time-varying risk field; weighting the risk level of the comprehensive collision risk area and the time-varying risk field; And constructing a new risk field for potential threat to ship navigation according to the weighted result.
5. The multi-objective optimization model for ship route facing composite sea condition scenario according to claim 2, wherein the specific steps of detecting whether there is a fixed obstacle in the risk field of the floating obstacle are as follows: for satellite images and underwater sonar images, extracting the outline edges of the obstacle by using an edge detection algorithm, acquiring shape characteristics by using a shape descriptor, and judging whether the shape characteristics are fixed obstacles or not: Aiming at radar and AIS data, adopting a Kalman filtering algorithm to carry out smoothing treatment on position and speed data of an obstacle, and analyzing a motion trail of the obstacle; if the track has no obvious change for a long time and the relative position with the submarine topography is fixed, the fixed obstacle is determined.
6. A composite sea situation scenario-oriented ship route multi-objective optimization model according to claim 3, wherein the obtaining real collision points according to potential collision points comprises the following specific steps: Dividing a time-varying risk field of a floating barrier into small grids in a three-dimensional space, and setting a unique number and a space coordinate for each small grid; Calculating a risk value of each small grid according to a dynamic attitude model, a dynamic windward area, an influence range radius and a track uncertainty ellipse of the floating obstacle, wherein the higher the risk value is, the larger the collision probability of the area is; discretizing the edge of the fixed obstacle into a plurality of points, and judging whether the points fall into a high-risk small grid of a floating obstacle risk field one by one: if a point falls into the high-risk small grid, determining the point as a potential collision point; And judging whether the potential collision points are actually collided or not by combining the motion trail of the floating obstacle in a future period of time, and if the motion trail of the floating obstacle is intersected at the point, determining that the potential collision points are actually collided, wherein the potential collision points are actually collided.
7. The multi-objective optimization model for a ship route facing a composite sea condition scenario according to claim 3, wherein the calculating the collision direction of the collision instant floating obstacle comprises the following specific steps: Simulating the collision process of the floating obstacle and the fixed obstacle at the collision point, and calculating the resultant force direction of the floating obstacle at the moment of collision; considering the influence of environmental factors such as ocean currents, waves and the like on the collision direction, the calculated resultant force direction is adjusted by establishing an environmental factor correction model, and the corrected resultant force direction is the collision direction.
8. A composite sea situation scenario-oriented ship route multi-objective optimization model according to claim 3, wherein the determining the impact range of the debris splashing comprises the following specific steps: Based on a discrete unit method, dispersing a floating barrier and a fixed barrier into a plurality of units, and simulating the crushing and splashing conditions of the units in the collision process; Setting connection strength and fracture criteria between units, and determining the units which are broken according to total energy released in the collision process and structural stress distribution; the method comprises the steps of combining the pushing action of ocean currents and waves on fragments, establishing a fragment motion equation, and calculating the motion trail of the fragments in an ocean environment; The motion trail of the fragments in the marine environment is simulated for a plurality of times through a fragment motion equation, the occurrence probability of the fragments in different areas is counted, a fragment splashing probability distribution cloud chart is drawn, and the influence range of the fragments splashing is determined.

Description

Ship route multi-target optimization model for composite sea condition scene Technical Field The invention relates to the technical field of ship route management, in particular to a ship route multi-target optimization model for a composite sea condition scene. Background The accuracy and safety of ship route planning are one of core technologies in the navigation field, and directly influence shipping efficiency, fuel consumption and navigation safety. In a complex marine environment, sea conditions such as stormy waves, ocean currents, tides and obstacles such as floating ice, abandoned ships and floating devices, which are faced by ships, present dynamic change characteristics, and the traditional route planning model is difficult to comprehensively adapt to real-time risk assessment requirements under the complex sea conditions. In the prior art, a static obstacle treatment mode is generally adopted in ship route planning, a floating obstacle is simplified into a fixed form or a uniform motion model, and dynamic attitude changes such as rolling, pitching and the like of the obstacle under the action of wind waves are ignored. This simplification results in an inability to accurately quantify real-time fluctuations in the frontal area, drag coefficient, and impact range of the obstacle, thereby leading to risk misjudgement in route planning. The traditional model does not consider the track uncertainty caused by the attitude change of the floating obstacle, a dynamically expanded risk area is difficult to construct, and the ship possibly faces collision risks due to untimely avoidance, so how to construct the adaptive route planning which considers the dynamic characteristics of the floating obstacle and realizes multi-objective optimization becomes a problem to be solved in the technical field of the current ship route management. Disclosure of Invention The invention aims to solve the problems that the traditional model does not consider track uncertainty caused by the change of the attitude of a floating obstacle, a dynamically-expanded risk area is difficult to construct and a ship possibly faces collision risks due to untimely avoidance, and provides a novel ship route multi-target optimization model method and a system thereof for a composite sea condition scene. In order to achieve the purpose, the invention adopts the following technology to face the ship route multi-objective optimization model of the composite sea situation scene, and comprises the following steps: the method comprises the steps of collecting ship navigation state data and marine environment data in real time by using marine monitoring equipment, a sensor network and ship self equipment, and establishing a motion model of a ship under different sea conditions by combining historical data; Taking a starting point and a destination as input, and generating an initial route by combining basic information of the ocean map; judging whether an obstacle exists on the initial route, if present: judging whether the obstacle is a floating obstacle or not, and determining: The method comprises the steps of obtaining shape information of an obstacle, constructing a floating obstacle dynamic attitude model, calculating a dynamic windward area, a resistance coefficient and an influence range radius of the floating obstacle, and constructing a time-varying risk field by combining a track uncertainty ellipse; and constructing an objective function according to the time-varying risk field and combining course and course change, and calculating to obtain an optimal course. Further, the floating obstacle dynamic attitude model is determined according to the following relation: wherein: θ (t) is the rolling angle of the floating obstacle at the moment t; Omega θ (t) is the rolling angular velocity of the floating obstacle at the moment t; the pitch angle rate of the floating obstacle at time t. Further, the risk field is constructed by the following specific steps: calculating the dynamic windward area and the dynamic resistance coefficient of the obstacle based on the established dynamic attitude model; determining the radius of the influence range of the floating obstacle according to the dynamic windward area and the resistance coefficient; Track uncertainty caused by the change of the obstacle posture is used for constructing a track uncertainty ellipse; and constructing a time-varying risk field by combining the calculated dynamic influence range radius R (t) of the floating obstacle and the track uncertainty ellipse parameter. Further, the running state of the floating obstacle is obtained, whether a fixed obstacle exists in a risk field of the floating obstacle is detected, and if the fixed obstacle exists: The floating barrier and the fixed barrier have collision risk, potential collision points are constructed at the overlapping positions of the edges of the fixed barrier and the risk fields of the floating barrier, the collision dire