A ship mooring control method and system based on motion analysis

Abstract

The application provides a ship mooring control method and system based on motion analysis, and relates to the technical field of ship mooring control. In view of the technical problems that the existing ship mooring control method mainly depends on position feedback, is easy to cause the single pendulum effect, has the resonance risk, has the response lag and has high energy consumption, the application adopts the fusion monitoring of ship motion state of stereovision and laser radar, constructs the mooring big data covering multiple ship types and multiple environment conditions based on the joint simulation of computational fluid dynamics and ship dynamics, optimizes the parameters of the PID control model with the ship displacement, speed and acceleration as inputs through the genetic algorithm, and finally realizes the closed-loop real-time adjustment of the cable tension relying on the high dynamic winch system. The method improves the control precision and stability of the mooring system, and the robustness and environmental adaptability in the face of complex environment.

Description

Technical Field

[0001] This invention relates to the field of ship mooring control technology, and in particular to a ship mooring control method and system based on motion analysis. Background Technology

[0002] Ship mooring is a critical technological aspect in port operations, offshore platform mooring, and nearshore anchoring. Its control precision and reliability directly impact operational safety, efficiency, and equipment lifespan. With the shipping industry moving towards larger scale and automation, and the continuous advancement of green and smart port construction, traditional mooring methods are no longer sufficient to meet the stable mooring requirements under highly dynamic sea conditions. Developing advanced mooring control methods based on real-time motion analysis, possessing environmental adaptability and intelligent decision support, has become an important technological trend for improving port throughput capacity and ensuring the safety of ship operations.

[0003] However, existing ship mooring control methods still have significant shortcomings. First, most methods are based on position control principles. When a ship deviates from its set position, it is pulled back by the restoring force of the cable. This method is prone to pendulum effect, causing the ship to swing continuously around the equilibrium position, which increases the amplitude of motion and reduces mooring stability. Existing technologies lack the ability to accurately perceive and analyze the ship's motion state in real time, usually relying only on position sensors, resulting in response lag and the inability to predict and proactively control the ship's motion trend. Traditional control strategies often rely on empirical parameters and do not fully consider the dynamic coupling effects of environmental loads such as wind, waves, and currents. They also do not establish a mooring database covering multiple ship types and operating conditions. Therefore, the control effect is prone to degradation under complex and variable sea conditions, there is a risk of resonance, and energy consumption is high. In addition, existing methods are also lacking in system integration, adaptive optimization, and fault safety protection, making it difficult to meet the high reliability and collaborative operation requirements of modern and intelligent mooring systems.

[0004] Therefore, there is an urgent need to develop a motion analysis-based ship mooring control method and system with higher stability and robustness. Summary of the Invention

[0005] To address these issues, the present invention provides a ship mooring control method and system based on motion analysis, which overcomes the problems in the prior art where the ship continuously oscillates near the equilibrium position, increasing the amplitude of motion, resulting in poor mooring stability, delayed response, inability to predict and anticipate the ship's motion trend, easy degradation of control effect under complex and variable sea conditions, resonance risk, and high energy consumption.

[0006] To achieve the above objectives, the present invention provides a ship mooring control method based on motion analysis, comprising: S1, Perform a stereoscopic visual scan of the target vessel to obtain monitoring data; S2, perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm; S3, based on ship dynamics theory, constructs a six-degree-of-freedom motion equation and simulates the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform; S4. Establish numerical simulation conditions covering the types and environmental conditions of the target vessel, and generate mooring big data including vessel parameters, environmental parameters, control parameters and mooring effect parameters. S5. Based on the aforementioned mooring big data, establish a PID control model for ship displacement and speed to determine the mooring tension of each cable; S6, The proportional gain coefficient, derivative gain coefficient and integral gain coefficient of the PID control model are optimized by an optimization algorithm, and an optimized PID control model is generated. S7. Based on the optimized PID control model, the mooring tension of each cable is adjusted in real time.

[0007] Furthermore, the monitoring data is fused using stereo vision technology combined with the Kalman filter algorithm to extract the ship's position, speed, and angular velocity as motion state parameters; the SIFT algorithm uses a Gaussian difference pyramid with 6 layers.

[0008] Furthermore, the PID control model is expressed as: ; in, The control force of the i-th cable, This represents the displacement error of the ship relative to its equilibrium position. For the speed of the ship, The integral of the displacement error, This is the proportional gain coefficient. The differential gain coefficient, This is the integral gain coefficient.

[0009] Furthermore, the optimization algorithm is a genetic algorithm with a population size of 50, an iteration count of 100, a crossover probability of 0.8, a mutation probability of 0.05, an initial range of proportional gain coefficient of 5000N / m to 50000N / m, an initial range of differential gain coefficient of 1000N·s / m to 10000N·s / m, and an initial range of integral gain coefficient of 100N / m·s to 1000N / m·s.

[0010] Furthermore, the environmental conditions include wind speed, wave height, and ocean current speed; wherein the simulated wind speed ranges from 5 m / s to 25 m / s, the simulated wave height ranges from 0.5 m to 2.0 m, and the simulated ocean current speed ranges from 0.3 m / s to 1.5 m / s.

[0011] Furthermore, the amount of the mooring big data is no less than 100,000 records, stored in HDF5 format, and covers the optimal control parameters and corresponding mooring effect data under different ship types, different combinations of environmental conditions, and different mooring configurations.

[0012] Furthermore, the mooring tension of each cable is adjusted by a winch system that can adjust the cable tension in real time. The winch system adopts frequency conversion speed regulation technology and is equipped with tension sensors and length encoders, which can monitor the tension and length of the cable in real time.

[0013] Furthermore, an alarm mechanism is triggered when the cable tension reaches the breaking point or when a sensor malfunction is detected.

[0014] The present invention also provides a ship mooring control system based on motion analysis, the system being used to implement any of the aforementioned ship mooring control methods based on motion analysis, the system comprising: The data acquisition module is used to perform stereoscopic vision scanning on the target vessel to obtain monitoring data; The data processing module, connected to the data acquisition module, is used to perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm. The motion simulation module, connected to the data processing module, is used to construct a six-degree-of-freedom motion equation based on ship dynamics theory and to simulate the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform. The working condition database construction module, connected to the motion simulation module, is used to establish numerical simulation working conditions covering the type and environmental condition combination of the target ship, and generate mooring big data including ship parameters, environmental parameters, control parameters and mooring effect parameters. The control model construction module, connected to the working condition database construction module, is used to establish a PID control model for ship displacement and speed based on the mooring big data, so as to determine the mooring tension of each cable; The parameter optimization module, connected to the control model construction module, is used to optimize the proportional gain coefficient, derivative gain coefficient, and integral gain coefficient of the PID control model through optimization algorithms, and generate an optimized PID control model. The cable tension adjustment execution module, connected to the parameter optimization module, is used to adjust the mooring tension of each cable in real time based on the optimized PID control model.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: Firstly, this invention constructs a stereo vision scanning scheme that integrates a multi-camera vision system with lidar, and employs a data fusion algorithm based on extended Kalman filtering and SIFT feature point extraction technology with specific parameters. This solves the problems of traditional mooring monitoring relying on a single position sensor, data being easily interfered with in complex port environments, and insufficient accuracy in perceiving the speed and attitude of ships. It achieves high-precision, high-frequency real-time perception and contour-stable tracking of the six degrees of freedom motion state of moored ships, providing a precise and reliable input foundation for subsequent intelligent motion control, and significantly improving the robustness and environmental adaptability of state perception. Secondly, by establishing a CFD-dynamics co-simulation system covering multiple ship types and environmental conditions to generate over 100,000 pieces of mooring big data, and based on this data, a genetic algorithm was used to globally and automatically optimize the gain coefficients of the PID control model with ship displacement, velocity, and acceleration as inputs. This solved the problems of traditional mooring control relying on fixed position feedback, empirical parameter tuning, inability to adapt to complex sea conditions and different ship types, and being prone to pendulum effect, resonance, and high energy consumption. It realized the transformation of control strategy from empirical fixed to data-driven and adaptive optimization, significantly improving the control accuracy, stability, and energy efficiency of the mooring system. Third, by combining high-precision state perception, big data-driven optimized control, and a fast-response winch system equipped with a vector control frequency converter, a complete intelligent closed loop of perception-decision-execution-safety is formed. This solves the problems of fragmentation of each link, delayed response, and lack of comprehensive safety protection in traditional methods. It realizes millisecond-level active damping control and multi-level safety monitoring of ship motion, and ultimately significantly improves the safety, stability, economy, and intelligence level of ship mooring operations. Attached Figure Description

[0016] Figure 1 A flowchart of a ship mooring control method based on motion analysis is provided for an embodiment of the present invention; Figure 2 This is a structural block diagram of a ship mooring control system based on motion analysis, provided as an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Example 1 like Figure 1 As shown, this invention provides a ship mooring control method based on motion analysis, comprising: S1, Perform a stereoscopic visual scan of the target vessel to obtain monitoring data; S2, perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm; The monitoring data is fused using stereo vision technology combined with the Kalman filter algorithm to extract the ship's position, speed, and angular velocity as motion state parameters; the SIFT algorithm uses a Gaussian difference pyramid with 6 layers.

[0022] In one possible implementation, the visual acquisition unit uses four MV-CH250-10GM high-definition industrial cameras with a resolution of 1920×1080 and a frame rate of 30fps. They are equipped with fixed-focus lenses with a focal length of 25mm, and the sensor's operating wavelength covers the visible light band. To ensure the ability to work at night and in low visibility conditions, two of the cameras are replaced with infrared models from the same series.

[0023] The laser scanning unit uses one RS-LiDAR-M1 laser radar with a ranging range of 0.5-100 meters, a horizontal field of view of 360°, a vertical field of view of 25°, and a scanning frequency of 10Hz.

[0024] The core processing unit uses an industrial computer as an edge computing node, equipped with a 12700H processor, 32GB of memory and a 1TB solid-state drive, and is responsible for real-time data reception, caching and preprocessing.

[0025] Four industrial cameras are rigidly connected, paired up and fixed to pre-cast concrete supports at the quay's edge. The supports are 10 meters high, with a horizontal spacing of 6 meters between cameras within the same pair, and the two pairs of cameras are spaced 20 meters apart along the quayline. The optical axes of all cameras are oriented towards the center of the mooring area, and through pre-calibration, their fields of view are designed to intersect and cover key areas of the target vessel, including its midsection, bow, and stern, forming a stable stereoscopic vision measurement baseline.

[0026] The lidar is installed near the center of gravity of the target ship's deck via a shock-absorbing base, ensuring that its scanning center coincides as much as possible with the ship's center of motion, so as to achieve a 360° scan of the ship's own attitude and its relative distance to the dock. All outdoor equipment, such as cameras and lidar, meets the IP67 protection level and has an operating temperature range of -20℃ to 60℃, which can adapt to the harsh environment commonly encountered in ports, such as rain, fog, and salt spray.

[0027] The camera and lidar are synchronized at the hardware level via the IEEE 1588 protocol, ensuring that the timestamp alignment accuracy of multi-source data is within milliseconds. The acquired raw image data and lidar point cloud data are transmitted to the industrial control computer in real time via gigabit industrial Ethernet, and the network transmission latency is controlled within 10 milliseconds.

[0028] After acquiring the raw monitoring data, a dedicated data processing algorithm is run within the industrial control computer to complete data fusion, state estimation, and feature recognition. Specifically, firstly, stereo matching is performed on the image pairs acquired by the binocular cameras. Combining the camera's intrinsic and extrinsic parameters, the three-dimensional coordinates of feature points on the ship's surface are calculated. Simultaneously, the lidar point cloud data is filtered and segmented to extract point cloud clusters belonging to the ship's structure. The visual 3D point cloud and the lidar point cloud are then registered in a unified world coordinate system with the fixed point at the dock as the origin. An iterative nearest-point algorithm is used for coarse registration, followed by fine registration using feature points.

[0029] For the registered and fused point cloud data, a state estimation model based on the Extended Kalman Filter (EKF) is constructed. The state vector X is defined as: X=[x, y, z, φ, θ, ψ, v x v y v z ω x ω y ω z ]; Where (x, y, z) is the position of the ship's center of gravity, (φ, θ, ψ) are the roll, pitch, and heave angles respectively, and (v x v y v z ) is the linear velocity, (ω) x ω y ω z ) represents angular velocity.

[0030] Using the changes in feature point positions provided by the fused point cloud as observations, recursive filtering is performed through EKF to ultimately output the ship's complete six-degree-of-freedom motion state parameters in real time, including high-precision position, linear velocity, and angular velocity information. This filtering process effectively suppresses interference caused by sensor noise and the loss of individual data.

[0031] The Scale Invariant Feature Transform (SIFT) algorithm is used for feature detection and description of the image. In the specific implementation, the number of layers of the constructed Gaussian difference pyramid is set to 6, and the feature point contrast threshold is set to 0.03, so as to suppress noise response while ensuring feature richness.

[0032] SIFT feature points extracted from consecutive frame images are matched, and mismatches are filtered out by random sample consensus algorithm, so as to reliably track key feature points on the ship contour. The identified image contour feature points are associated with the corresponding features in the aforementioned fused point cloud, providing richer and more stable observation input for the Kalman filter. Especially when visual conditions are acceptable but the lidar may be interfered with by water surface reflection, this visual contour information can play an important supplementary and corrective role.

[0033] This invention addresses the limitations of traditional single-sensor or simple vision systems in complex port environments, such as limited monitoring range, low data reliability, and susceptibility to lighting and weather conditions, by configuring a stereo vision scanning system composed of multiple high-definition industrial cameras and LiDAR, and employing high-precision spatiotemporal synchronization and industrial Ethernet transmission technology. It achieves stable, all-weather, high-coverage, and high-real-time data acquisition of the motion state of moored vessels, improving the environmental adaptability and data acquisition reliability of the monitoring system. Furthermore, by using stereo vision technology combined with an extended Kalman filter algorithm to fuse multi-source monitoring data, it solves the problems of high noise from single data sources and a sharp drop in state estimation accuracy when individual sensors fail. This effectively suppresses data noise and outlier interference, achieving smooth, continuous, and high-precision estimation of the six degrees of freedom motion state of ships, thus improving the motion... The accuracy of parameter extraction and the robustness of the system are improved. By applying the SIFT feature point extraction and matching algorithm with specific parameters and combining it with image sequences for contour tracking, the problems of poor stability and high mismatch rate of traditional feature extraction methods are solved when there is low contrast between the ship and the background, local occlusion, or changes in lighting. Stable and accurate identification and tracking of key features of the ship contour are achieved, improving the reliability of visual observation information and adaptability to complex scenes. By associating and fusing visual contour feature points extracted based on the SIFT algorithm with lidar point clouds, the problem of reduced lidar data quality and decreased state estimation performance under specific conditions such as rain, fog, and water surface reflection is solved. The advantages of multi-sensor information are complemented and redundantly backed up, improving the system's continuous working ability and overall measurement accuracy under adverse weather conditions.

[0034] S3, based on ship dynamics theory, constructs a six-degree-of-freedom motion equation and simulates the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform; In one possible implementation, to achieve accurate prediction of the motion response of moored vessels in complex marine environments, this embodiment constructs a CFD-dynamics co-simulation system, combining the precise calculation of fluid loads by computational fluid dynamics (CFD) with multibody dynamics modeling of the ship. Specifically, the ship is treated as a rigid body, and based on the Newton-Euler equations, its six-degree-of-freedom motion equations are established in the ship's coordinate system, namely sway, roll, heave, pitch, pitch, and bow. ANSYS Fluent is used as the CFD solver, and it is coupled unidirectionally with the ship dynamics model built in MATLAB / Simulink for co-simulation. Each time step is advanced in the dynamics model in Simulink, the current ship motion state is transmitted to Fluent as the boundary condition for dynamic mesh calculation; Fluent then calculates the pressure and viscous forces acting on the hull by the environmental flow field in that instantaneous state, integrates them to obtain the resultant force and resultant moment vector acting on the ship, and returns them to Simulink.

[0035] A fine-grained computational domain was created, encompassing the ship, dock, and a portion of the water area. A high-density structured mesh, with at least 1.5 million elements, was used to subdivide the ship's surface and near-field water area. A boundary layer mesh was applied to the ship's surface to ensure the accuracy of wall function calculations, thereby precisely capturing nonlinear phenomena such as fluid separation and vortices. The SST k-ω turbulence model was selected, and the multiphase flow and wave model employed the VOF (Volume of Fluid) method to track the free surface. Irregular waves were generated at the inlet boundary using fifth-order Stokes wave theory or by inputting wave spectra, such as the JONSWAP spectrum, to simulate real sea conditions.

[0036] A velocity inlet boundary condition is set at the top of the computational domain. A wind speed profile is given to simulate the effect of steady or fluctuating wind on the superstructure of the ship. A velocity inlet is set on the side or bottom of the computational domain to simulate uniform or shear currents. The CFD calculation uses a fixed time step of 0.01 seconds to keep consistent with the control cycle of the dynamic model, ensuring the synchronization of data exchange and the accuracy of the simulation.

[0037] The constructed co-simulation model is then applied to a target ship type, such as a 5000 DWT bulk carrier, under specified environmental conditions. For example, it simulates the ship's time history response from its initial equilibrium position under the combined effects of a steady wind of 15 m / s, an irregular wave with a significant wave height of 1.5 m, and a uniform current of 0.8 m / s. The simulation system outputs the displacement, velocity, and acceleration time history curves for the ship's six degrees of freedom, as well as the motion trajectories of each cable connection point. To verify the model's accuracy, the ship's motion response in the waves obtained from the CFD simulation, such as the heave and pitch response amplitude operators (RAO), can be compared with ship tank model test data or publicly available potential flow theory calculation results. Typically, the prediction error of the model constructed in this embodiment for the main motion modes can be controlled within 15%.

[0038] This invention establishes a six-degree-of-freedom rigid body kinematics and dynamics model of a ship based on the Newton-Euler equations, and uses CFD and system dynamics software for unidirectional coupled co-simulation. This solves the problems of overly simplified environmental load calculations, neglect of fluid nonlinear effects, and dynamic coupling between degrees of freedom in traditional mooring analysis. It achieves refined, high-fidelity numerical simulation of ship motion response in complex marine environments, significantly improving the physical accuracy and reliability of motion prediction. By constructing simulation conditions covering the coupled effects of multiple environmental factors such as wind, waves, and currents, and performing time-domain simulation with a fine time step of 0.01 seconds, it addresses the shortcomings of existing methods in predicting ship motion. The system addresses the issues of insufficient consideration of multiple factors and inadequate dynamic process resolution. It can comprehensively and continuously predict complex motion patterns that ships may exhibit while moored, thereby improving the ability to proactively assess mooring safety risks. By integrating high-precision environmental loads obtained from CFD simulations into the ship dynamics closed-loop simulation model, it solves the problem that early verification of control strategies relies on expensive and time-consuming physical model tests, or can only be verified in simplified models. This provides an efficient, low-cost, and repeatable testing environment for the subsequent design and optimization of PID control models and the generation of mooring big data, reducing the cycle and cost of control system development.

[0039] S4. Establish numerical simulation conditions covering the types and environmental conditions of the target vessel, and generate mooring big data including vessel parameters, environmental parameters, control parameters and mooring effect parameters. Environmental conditions include wind speed, wave height, and ocean current speed; the simulated range for wind speed is 5 m / s to 25 m / s, the simulated range for wave height is 0.5 m to 2.0 m, and the simulated range for ocean current speed is 0.3 m / s to 1.5 m / s.

[0040] The mooring big data contains no fewer than 100,000 records, stored in HDF5 format, and covers the optimal control parameters and corresponding mooring effect data under different ship types, different combinations of environmental conditions, and different mooring configurations.

[0041] In one possible implementation, three representative target ship types are selected as the basic models: a 5000 DWT bulk carrier, approximately 110 meters long and 20 meters wide; a 10000 TEU container ship, approximately 350 meters long and 48 meters wide; and a 50-meter luxury yacht, 50 meters long and 9 meters wide. Detailed parameter profiles are established for each type of vessel, including key ship parameters such as geometric model, mass distribution, moment of inertia, center of gravity position, and waterplane moment of inertia.

[0042] The environmental conditions are systematically discretized into key environmental parameters, forming a gridded combination. Specifically, wind speeds cover 5 levels: 5 m / s, 10 m / s, 15 m / s, 20 m / s, and 25 m / s; wave heights cover 4 levels: 0.5 m, 1.0 m, 1.5 m, and 2.0 m; and ocean current velocities cover 4 levels: 0.3 m / s, 0.7 m / s, 1.1 m / s, and 1.5 m / s. Three to four typical mooring arrangement schemes are pre-set for each type of vessel, including the number of mooring lines, the lead-out angle, and the initial pretension.

[0043] Through full-factor experimental design, theoretically, 9600 basic simulation conditions can be generated by permuting and combining 3 ship types, 5 wind speeds, 4 wave heights, 4 current speeds, 4 mooring configurations, and 10 initial gain groups. By introducing the relative angle changes of wind direction, wave direction, current direction, and ship heading, as well as time-varying conditions such as irregular wave spectra and transient winds, the total number of conditions can be easily expanded to over 100,000, ensuring the scale of big data.

[0044] The automated script drives the CFD-dynamics co-simulation system built in S3 to run all preset working conditions sequentially. The simulation duration for each working condition is 600 seconds of actual time to ensure sufficient dynamic processes are included and steady-state performance can be evaluated. During the simulation, the system automatically collects and records the following four types of core parameters, forming a complete mooring data record: Input parameters: Ship parameters, including ship type designation, main dimensions, mass, and moment of inertia; Environmental parameters, including wind speed, wind direction, wave height, wave direction, current speed, and current direction; Control parameters, including the PID gain coefficient and control cycle used in this simulation.

[0045] Output (effect) parameters: Ship motion response includes statistical characteristics such as the maximum value, root mean square value, and power spectral density of six-degree-of-freedom displacement, velocity, and acceleration. The status of the mooring system, including the maximum, minimum, average, and standard deviation of the tension of each cable; the length and speed of cable winding and unwinding by the winch; Performance indicators include ship positioning accuracy, energy consumption integral, and maximum tension safety factor.

[0046] All generated simulation data are stored in HDF5 (Hierarchical Data Format version 5) format. HDF5 format supports efficient compressed storage and can hierarchically organize multi-dimensional and multi-type data in the form of datasets and groups, which is suitable for the complex structured data in this embodiment.

[0047] At the same time, a relational database, such as SQLite or an index file, is generated to record key metadata for each data entry, such as operating condition ID, ship type, environmental parameter values, and performance index scores. This facilitates the rapid retrieval and extraction of a subset of data that meets specific conditions, such as the optimal control parameters for container ships under wind speeds of 15 m / s and wave heights of 1.5 m, for use in S5 and S6 optimization.

[0048] This invention addresses the shortcomings of existing methods by systematically designing a numerical simulation condition matrix covering three typical ship types and a full range of environmental parameters, including wind speed. This matrix relies on individual empirical conditions, has narrow data coverage, and fails to comprehensively reflect the complex and ever-changing real-world sea conditions and the differences in characteristics between different ship types. It constructs a mooring condition database with over 100,000 records, significantly improving the comprehensiveness and systematic nature of the training and validation data required for control strategy design. Furthermore, by employing the HDF5 hierarchical data format for efficient compression and storage of the generated mooring big data and establishing a metadata-based indexing mechanism, it solves the problems of chaotic data storage, slow read / write speeds, and difficulty in quickly accessing large amounts of simulation data. The system addresses the problem of rapidly retrieving and extracting optimal control parameters under specific operating conditions, enabling efficient organization and management of over 100,000 data records. This significantly improves data utilization and the efficiency of subsequent parameter optimization processes. Furthermore, the established mooring big data platform digitally encapsulates the complex nonlinear relationships between ship dynamics response, environmental loads, and control effects. This solves the problem of control engineers relying on trial and error and struggling to quickly obtain optimal control parameters suitable for specific combinations in the early stages of development. It provides a directly queryable and learnable knowledge base for optimizing the gain coefficients of the PID control model in subsequent steps, greatly reducing the time cost and trial-and-error risk of control parameter tuning.

[0049] S5. Based on the aforementioned mooring big data, establish a PID control model for ship displacement and speed to determine the mooring tension of each cable; S6, The proportional gain coefficient, derivative gain coefficient and integral gain coefficient of the PID control model are optimized by an optimization algorithm, and an optimized PID control model is generated. The PID control model is represented as follows: ; in, The control force of the i-th cable, This represents the displacement error of the ship relative to its equilibrium position. For the speed of the ship, The integral of the displacement error, This is the proportional gain coefficient. The differential gain coefficient, This is the integral gain coefficient.

[0050] The optimization algorithm is a genetic algorithm with a population size of 50, 100 iterations, a crossover probability of 0.8, a mutation probability of 0.05, an initial range of 5000N / m to 50000N / m for the proportional gain coefficient, an initial range of 1000N·s / m to 10000N·s / m for the differential gain coefficient, and an initial range of 100N / m·s to 1000N / m·s for the integral gain coefficient.

[0051] In one possible implementation, the environmental conditions of the target vessel to be controlled, such as a 5000 DWT bulk carrier, as monitored or predicted in real time (e.g., wind speed 15 m / s, wave height 1.5 m, current speed 0.8 m / s), are first matched and searched in the mooring big data constructed by S4. The search mechanism is based on the similarity of environmental parameters, such as Euclidean distance, to quickly filter out all simulation records most similar to the target operating conditions, forming a target training dataset. Each record contains a complete set of control parameters, namely a set of PID gains and their corresponding mooring effects shown in the simulation, such as ship motion, cable tension, and other performance indicators, forming cause-effect paired samples that provide direct basis for model optimization.

[0052] The proportional gain coefficient, differential gain coefficient, and integral gain coefficient together determine the intensity and manner of the system's response to errors, velocity changes, and historical error accumulation.

[0053] The core objective of optimization is to find a set of proportional gain coefficients, differential gain coefficients, and integral gain coefficients that optimize the ship's overall motion performance under this set of parameters. Specifically, a weighted comprehensive evaluation function, which includes the ship's displacement standard deviation, peak cable tension, and system energy consumption integral, is constructed as the fitness function.

[0054] Using real-number encoding, each individual chromosome consists of three genes (proportional gain coefficient, differential gain coefficient, and integral gain coefficient). The initial population is randomly generated within a defined initial search space. The initial range of the proportional gain coefficient is 5000 N / m to 50000 N / m, the initial range of the differential gain coefficient is 1000 N·s / m to 10000 N·s / m, and the initial range of the integral gain coefficient is 100 N / m·s to 1000 N / m·s.

[0055] For each individual in the population, i.e., a set of gain coefficients, these are substituted into the aforementioned PID control model. Using the ship motion data stored in the target training dataset as model input, the theoretical control force sequence is calculated. This control force sequence is then compared with the control performance data recorded in the dataset that best performs in this environment, calculating the fitness score of this set of parameters. A roulette wheel selection algorithm is used to probabilistically select individuals based on their fitness, retaining superior individuals. Arithmetic crossover is performed on the selected parent individuals with a crossover probability of 0.8 to generate new offspring individuals, exploring the parameter combination space. A small random perturbation is applied to some genes of the offspring individuals with a mutation probability of 0.05 to increase population diversity and prevent the algorithm from getting trapped in local optima. This selection, crossover, and mutation process is repeated until the preset number of iterations is reached.

[0056] After the iteration, the combination of gain coefficients corresponding to the individual with the highest fitness is selected as the optimal PID control model parameters for the current target ship under specific environmental conditions. For example, the proportional gain coefficient is 2500 N / m, the derivative gain coefficient is 6000 N·s / m, and the integral gain coefficient is 500 N / m·s.

[0057] This invention establishes a PID control model based on mooring big data, using ship displacement error, velocity, and error integral as direct inputs. This solves the problem of traditional mooring control relying solely on position deviation while neglecting ship velocity and trend, leading to control response lag and susceptibility to pendulum-like overshoot oscillations. It achieves advanced prediction and active damping control of ship motion states, significantly improving the dynamic stability and control accuracy of the mooring system. Furthermore, by applying a genetic algorithm to the global optimization of the PID model's gain coefficients and using paired samples of control parameters and performance indicators from the mooring big data as the fitness evaluation basis, it addresses the heavy engineering dependence of traditional PID parameter tuning. The previous approach, which relied on trial and error based on human experience and was time-consuming and laborious, making it difficult to obtain the global optimal solution, has achieved automated, intelligent, and optimal tuning of the gain coefficient, significantly improving the efficiency and performance ceiling of the control strategy. By utilizing mooring big data covering multiple ship types and operating conditions as the training basis for the optimization algorithm, and by quickly matching and retrieving the nearest optimal solution for specific target ships and real-time environmental conditions, the problem of the fixed parameter PID controller experiencing a sharp decline in control performance and poor robustness when ship type or environmental load changes has been solved. This has enabled the generated optimized PID control model to have the ability to adapt to different operating conditions, significantly improving the versatility and environmental adaptability of the mooring control system.

[0058] S7. Based on the optimized PID control model, the mooring tension of each cable is adjusted in real time.

[0059] The mooring tension of each cable is adjusted by a winch system that can adjust the cable tension in real time. The winch system adopts frequency conversion speed regulation technology and is equipped with tension sensors and length encoders, which can monitor the tension and length of the cable in real time.

[0060] An alarm mechanism is triggered when the cable tension reaches the breaking point or when a sensor malfunction is detected.

[0061] In one possible implementation, each mooring cable is equipped with an independent variable frequency electric winch. For example, an SEW DFV112M4 winch could be used, with a 37kW drive motor whose output speed can be continuously and precisely adjusted within the range of 0-10 r / min. Each winch is controlled by an MM440 series variable frequency drive, which supports vector control mode and enables rapid and precise control of motor torque and speed.

[0062] HBM U9B high-precision tension sensors are installed in the cable guide holes of each cable or at the winch drum bearing housing. These sensors have a measurement range of 0-1000kN and can provide real-time feedback on the actual tension value of the cable. An absolute multi-turn length encoder with a resolution of 1024 lines / turn is installed on the winch output shaft. The cable winding and unwinding length and speed are calculated by measuring the drum rotation angle. The signals from the tension sensors and length encoder are transmitted in real-time to the central controller via a fieldbus, forming a feedback loop for closed-loop control. Control commands are also issued to each frequency converter drive via this bus.

[0063] The central controller reads the real-time tension of each cable and the latest motion status of the ship from the sensors at a cycle of 10ms, including displacement error and speed. This real-time data is then input into the optimized PID control model to calculate the target control force of each cable in the current cycle.

[0064] The target control force is converted into control commands for the winch. Since cable tension is directly related to the output torque of the winch motor, this command is typically a target torque or speed value. After receiving the command, the frequency converter driver, driven by its internal vector control algorithm, precisely adjusts the voltage and frequency output to the motor, enabling the motor to quickly generate the required torque and drive the winch to retrieve or release the cable. At the start of a new control cycle, the system again collects the actual tension feedback from the tension sensor and compares it with the target force of the previous cycle, forming a force control closed loop. Simultaneously, the latest motion status data provided by the ship motion monitor is also used to update the input of the PID model, forming a motion status closed loop. This dual closed-loop structure ensures that the control force accurately tracks the model commands and quickly compensates for environmental disturbances.

[0065] The system monitors the tension of each cable in real time and sets two safety thresholds. The warning threshold is typically set at 60% of the cable breaking strength. When the tension exceeds this value, the system issues a warning on the control interface, but the control logic remains unchanged. The emergency action threshold is set at 95% of the cable breaking strength. Once the tension of any cable reaches or exceeds this threshold, the system immediately triggers the highest priority interruption. The control mode automatically switches from active force control to safety holding mode, sending preset safety commands to all winches to stop or slowly release to prevent further tension increases. Simultaneously, it triggers an audible and visual alarm mechanism and sends an emergency alarm signal to the upper-level monitoring system. It continuously checks the rationality of sensor signals, such as whether the signal is within the range, whether the rate of change is abnormal, and whether communication is interrupted. When any critical sensor, such as a tension sensor or length encoder, is detected to have a persistent fault or signal loss, the system will also trigger an alarm. According to preset strategies, it can downgrade the control mode of the faulty cable or isolate it from the active control network, allowing the remaining healthy cables to work together to maintain basic mooring safety, while simultaneously notifying maintenance personnel.

[0066] This invention solves the problems of slow response, coarse control, and inability to achieve fine force control in traditional mooring winches by employing an electric winch execution system equipped with a vector control type variable frequency drive and combining it with a real-time feedback network consisting of a high-precision tension sensor and a length encoder. It achieves rapid and precise dynamic adjustment of cable tension, thereby efficiently and accurately converting the commands of the optimized control model into physical forces that suppress ship motion, significantly improving the execution fidelity of the entire control loop. By establishing a fast closed-loop control process with a fixed period of 10 milliseconds, the target force command calculated by the optimized PID control model and the actual tension feedback from the sensor are integrated. By comparing and correcting force and ship motion status in real time, the system solves the problems of accumulated control errors and poor system stability caused by execution lag or untimely feedback in traditional open-loop or slow closed-loop control. It achieves millisecond-level response and dynamic compensation to changes in ship motion and cable status, reducing the overall control response time to less than 0.1 seconds, significantly improving the real-time performance and stability of control. By setting multi-level tension safety thresholds based on cable breaking force and implementing real-time monitoring and automatic triggering mechanisms, the system solves the problems of traditional manual monitoring's difficulty in instantly judging and handling tension overload hazards and the susceptibility to cable breakage accidents due to single-point failures. When the tension reaches the breaking force, the system automatically switches to a safety mode and triggers an alarm, transforming overload risk from post-event remediation to pre-event prevention and immediate intervention, significantly reducing the risk of cable breakage and improving the inherent safety of mooring operations. By integrating sensor fault diagnosis and system fault-tolerant logic, the system continuously monitors the rationality and communication status of key sensor signals, resolving the risk of misjudgment and erroneous control commands caused by sensor failure. When a sensor malfunction is detected, the system can promptly issue an alarm and execute preset degradation or isolation strategies, ensuring that the system can still maintain basic safe mooring functions or orderly failure even when some components are abnormal, greatly enhancing the robustness and operational reliability of the entire mooring control system.

[0067] Example 2 like Figure 2 As shown, the present invention provides a ship mooring control system based on motion analysis. The system is used to implement the ship mooring control method based on motion analysis described in any of Embodiment 1. The system includes: The data acquisition module is used to perform stereoscopic vision scanning on the target vessel to obtain monitoring data; The data processing module, connected to the data acquisition module, is used to perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm. The motion simulation module, connected to the data processing module, is used to construct a six-degree-of-freedom motion equation based on ship dynamics theory and to simulate the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform. The working condition database construction module, connected to the motion simulation module, is used to establish numerical simulation working conditions covering the type and environmental condition combination of the target ship, and generate mooring big data including ship parameters, environmental parameters, control parameters and mooring effect parameters. The control model construction module, connected to the working condition database construction module, is used to establish a PID control model for ship displacement and speed based on the mooring big data, so as to determine the mooring tension of each cable; The parameter optimization module, connected to the control model construction module, is used to optimize the proportional gain coefficient, derivative gain coefficient, and integral gain coefficient of the PID control model through optimization algorithms, and generate an optimized PID control model. The cable tension adjustment execution module, connected to the parameter optimization module, is used to adjust the mooring tension of each cable in real time based on the optimized PID control model.

[0068] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A ship mooring control method based on motion analysis, characterized in that, include: S1, Perform a stereoscopic visual scan of the target vessel to obtain monitoring data; S2, perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm; S3, based on ship dynamics theory, constructs a six-degree-of-freedom motion equation and simulates the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform; S4. Establish numerical simulation conditions covering the types and environmental conditions of the target vessel, and generate mooring big data including vessel parameters, environmental parameters, control parameters and mooring effect parameters. S5. Based on the aforementioned mooring big data, establish a PID control model for ship displacement and speed to determine the mooring tension of each cable; S6, The proportional gain coefficient, derivative gain coefficient and integral gain coefficient of the PID control model are optimized by an optimization algorithm, and an optimized PID control model is generated. S7. Based on the optimized PID control model, the mooring tension of each cable is adjusted in real time.

2. The ship mooring control method based on motion analysis according to claim 1, characterized in that, The monitoring data is fused using stereo vision technology combined with Kalman filtering algorithm to extract the ship's position, speed, and angular velocity as motion state parameters; the SIFT feature point extraction algorithm uses a Gaussian difference pyramid with 6 layers.

3. The ship mooring control method based on motion analysis according to claim 1, characterized in that, The PID control model is expressed as follows: ； in, The control force of the i-th cable, This represents the displacement error of the ship relative to its equilibrium position. For the speed of the ship, The integral of the displacement error, This is the proportional gain coefficient. The differential gain coefficient, This is the integral gain coefficient.

4. The ship mooring control method based on motion analysis according to claim 3, characterized in that, The optimization algorithm is a genetic algorithm with a population size of 50, 100 iterations, a crossover probability of 0.8, a mutation probability of 0.05, an initial range of proportional gain coefficient of 5000N / m to 50000N / m, an initial range of differential gain coefficient of 1000N·s / m to 10000N·s / m, and an initial range of integral gain coefficient of 100N / m·s to 1000N / m·s.

5. The ship mooring control method based on motion analysis according to claim 1, characterized in that, The environmental conditions include wind speed, wave height, and ocean current speed; wherein the simulated wind speed range is 5 m / s to 25 m / s, the simulated wave height range is 0.5 m to 2.0 m, and the simulated ocean current speed range is 0.3 m / s to 1.5 m / s.

6. The ship mooring control method based on motion analysis according to claim 1, characterized in that, The mooring big data contains no less than 100,000 records, stored in HDF5 format, and covers the optimal control parameters and corresponding mooring effect data under different ship types, different combinations of environmental conditions, and different mooring configurations.

7. The ship mooring control method based on motion analysis according to claim 1, characterized in that, The mooring tension of each cable is adjusted by a winch system that can adjust the cable tension in real time. The winch system adopts frequency conversion speed regulation technology and is equipped with tension sensors and length encoders, which can monitor the tension and length of the cable in real time.

8. The ship mooring control method based on motion analysis according to claim 7, characterized in that, An alarm mechanism is triggered when the cable tension reaches the breaking point or when a sensor malfunction is detected.

9. A ship mooring control system based on motion analysis, the system being used to implement the ship mooring control method based on motion analysis according to any one of claims 1 to 8, the system comprising: The data acquisition module is used to perform stereoscopic vision scanning on the target vessel to obtain monitoring data; The data processing module, connected to the data acquisition module, is used to perform data fusion on the monitoring data, extract the motion state parameters of the ship, and identify the outline of the target ship through the SIFT feature point extraction algorithm. The motion simulation module, connected to the data processing module, is used to construct a six-degree-of-freedom motion equation based on ship dynamics theory and to simulate the motion response of the target ship under wind, wave, and current environmental loads through a computational fluid dynamics simulation platform. The working condition database construction module, connected to the motion simulation module, is used to establish numerical simulation working conditions covering the type and environmental condition combination of the target ship, and generate mooring big data including ship parameters, environmental parameters, control parameters and mooring effect parameters. The control model construction module, connected to the working condition database construction module, is used to establish a PID control model for ship displacement and speed based on the mooring big data, so as to determine the mooring tension of each cable; The parameter optimization module, connected to the control model construction module, is used to optimize the proportional gain coefficient, derivative gain coefficient, and integral gain coefficient of the PID control model through optimization algorithms, and generate an optimized PID control model. The cable tension adjustment execution module, connected to the parameter optimization module, is used to adjust the mooring tension of each cable in real time based on the optimized PID control model.

Images