Ship mooring state monitoring and early warning method

Abstract

The invention relates to the technical field of ship early warning, and discloses a ship mooring state monitoring and early warning method, which comprises the following steps: collecting environmental characteristic data, stability data of ships in a target mooring water area and stranded cable data from the target mooring water area according to a fixed time interval in a first preset time period; calculating a stability score of the mooring ship in the target mooring water area according to the stability data; constructing a wind-water topology based on all ships in the target mooring water area, and generating corresponding standard environment data for each mooring ship based on the wind-water topology and the environment characteristic data; training based on the standard environment data and the stranded cable data of each mooring ship in a first preset time period and the corresponding stability score to obtain a stability prediction model; and in a second preset time period, dynamically managing the length and tension of each stranded cable based on the stability prediction model in combination with an optimization algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of ship early warning, and more specifically, to a ship mooring state monitoring and early warning method. Background Art

[0002] Ship mooring refers to the process of fixing the hull of a ship by means of cables, mooring equipment, etc. after the ship is moored at a pier, buoy or anchorage to maintain its stability and prevent drifting or accidental movement.

[0003] However, ship mooring is affected by many factors such as tides, wind speed and direction, current speed and direction, etc., which leads to dynamic changes in the ship's mooring state. Existing technologies are difficult to cope with complex and nonlinear dynamic changes, resulting in unstable state of the ship when moored, affecting transportation operations. Summary of the invention

[0004] The present invention provides a ship mooring status monitoring and early warning method, which solves the technical problems raised in the background technology.

[0005] The present invention provides a ship mooring status monitoring and early warning method, comprising:

[0006] A data collection module, used to collect environmental characteristic data, stability data of the vessel in the target mooring waters and cable winch data of the target mooring waters at fixed time intervals within a first preset time period;

[0007] Environmental characteristic data include: water flow direction and velocity, wind direction and velocity, and water surface height;

[0008] Stability data include: acceleration parameters, angular velocity parameters, inclination parameters, GPS parameters, driving speed, draft depth and driving direction;

[0009] Cable laying data include: length and tension of bow and stern cables, bow and stern return cables and bow and stern spring cables;

[0010] A data analysis module for calculating a stability score for a moored vessel in a target mooring water area based on the stability data;

[0011] A data correction module is used to construct a Feng Shui topology based on all the ships in the target mooring waters, and generate corresponding standard environmental data for each moored ship based on the Feng Shui topology and environmental characteristic data;

[0012] A model building module, used for training a stability prediction model based on standard environmental data, cable-twisting data and corresponding stability scores of each moored vessel within a first preset time period;

[0013] The cable management module is used to dynamically manage the length and tension of each cable based on the stability prediction model combined with the optimization algorithm within a second preset time period.

[0014] Furthermore, the stability score is calculated using the following formula: ;

[0015] in, represents the stability score of the moored vessel, , and represents the first stability weight, the second stability weight and the third stability weight of the ship, , and All are not 0, and the sum is 1. represents the acceleration parameter of the moored vessel, represents the angular velocity parameter of the moored vessel, Represents the inclination parameter of a moored vessel.

[0016] Furthermore, the Feng Shui topology is constructed, including:

[0017] Load a simulated water surface, which includes the water surface boundary and water area;

[0018] Configure the water area based on environmental characteristic data: water flow direction and velocity, as well as wind direction and velocity;

[0019] Configure the water surface boundary based on the target mooring area: shape, size and first hydraulic rigidity function;

[0020] The position coordinates of each vessel are obtained based on the GPS parameters, and each vessel is mapped as a topological point in the water area based on the position coordinates;

[0021] For each topological point, a rigid influence range is configured; based on the ship's speed, draft and direction, the corresponding rigid influence range is configured with a wind rigidity function and a second hydraulic rigidity function to obtain a Feng Shui topology.

[0022] Furthermore, the water area configuration is based on the environmental characteristic data, including:

[0023] The simulated water surface is constructed based on a matrix structure, the matrix structure includes a first channel and a second channel, and the first channel and the second channel both include the same number of colliders;

[0024] The first channel and the second channel of the matrix structure are both defined with the same water surface boundary, and the water surface boundary is divided into a power source boundary and a rigid source boundary; wherein the power source boundary and the rigid source boundary are both composed of a number of colliders, and the water area is composed of the remaining colliders;

[0025] For each collider at the boundary of the power source in the first channel, the hydraulic collision direction and hydraulic collision speed are configured based on the water flow direction and water flow speed; for each collider at the boundary of the power source in the second channel, the wind collision direction and wind collision speed are configured based on the wind direction and wind speed;

[0026] In the first channel or the second channel, if collider i is adjacent to any collider with a collision direction and a collision speed, and is located in the collision direction of the corresponding collider, the first updated collision direction and the first updated collision speed of collider i are obtained, and the calculation formula is as follows: ; ;

[0027] in, represents the first updated collision direction of collider i, represents the first updated collision velocity of collider i, represents the collision velocity of collider i, represents the collision direction of collider i, represents the collision speed of collider i+1, Indicates the collision direction of collider i+1;

[0028] Among them, i and i+1 are the indexes of the colliders in the first channel or the second channel, and the initial collision speed and initial collision direction of the colliders in the water area in the first channel and the second channel are both 0.

[0029] Furthermore, the water surface boundary configuration based on the target mooring waters includes:

[0030] Get the shape and size of the dock boundary and water boundary of the target mooring waters, based on the shape and size of the dock boundary and water boundary, and at a preset scale , which are mapped as rigid source boundaries and dynamic source boundaries in the first channel and the second channel of the matrix structure, respectively. >0;

[0031] Wherein, the first hydraulic rigidity function is configured for the collider in the rigid source boundary;

[0032] In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to a collider in any rigid source boundary, and the corresponding collider is located in the collision direction of collider i, then the second updated collision direction and second updated collision speed of collider i are obtained based on the first hydraulic rigidity function. The calculation formula of the first hydraulic rigidity function is as follows: ; ;

[0033] in, represents the second updated collision direction of collider i, represents the second updated collision velocity of collider i, represents the angle between the collision direction of collider i and the dock boundary, Represents the first velocity attenuation coefficient.

[0034] Furthermore, each vessel is mapped as a topological point in the water area based on the position coordinates, including: obtaining the GPS coordinates of each vessel, calculating the relative coordinates of the GPS coordinates and the target mooring water area , based on the relative coordinates, the corresponding ship is mapped to the first channel and the second channel of the matrix structure, specifically: replace the first channel and the second channel in the matrix structure Line The colliders of a column are topological points.

[0035] Furthermore, for each node in the water area, a rigid influence range is configured, including defining the rigid influence range as an ellipse:

[0036] The center point of the ellipse is: the topological point;

[0037] The semi-major axis of the ellipse is: ;in, represents the length of the semi-major axis of the ellipse, represents the empirical coefficient, , Indicates the preset zoom ratio. Indicates the draft of the vessel. Indicates the speed of the vessel;

[0038] The semi-minor axis of the ellipse is: ;in, represents the length of the semi-minor axis of the ellipse;

[0039] The inclination angle of the semi-major axis of the ellipse is: the direction of travel of the ship;

[0040] Based on the coverage area of ​​the corresponding ellipse in the first channel and the second channel of the matrix structure, the colliders of the corresponding areas in the first channel and the second channel are used as the rigid influence range of the corresponding topological point;

[0041] The rigid influence range is configured with a second hydraulic rigidity function in the first channel, and the rigid influence range is configured with a wind rigidity function in the second channel;

[0042] In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the third updated collision direction and the third updated collision speed of collider i are obtained based on the second hydraulic rigidity function. The calculation formula of the second hydraulic rigidity function is as follows: ; ;

[0043] in, Indicates the third updated collision direction, represents the third updated collision velocity, Represents the angle between the water flow direction and the semi-major axis of the ellipse, represents the second velocity attenuation coefficient;

[0044] In the second channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the fourth updated collision direction and fourth updated collision speed of collider i are obtained based on the wind rigidity function. The calculation formula of the wind rigidity function is as follows: ; ;

[0045] in, Indicates the fourth updated collision direction, represents the fourth updated collision velocity, represents the angle between the wind direction and the semi-major axis of the ellipse, Indicates the third speed attenuation coefficient.

[0046] Furthermore, the corresponding standard environment data is generated, including:

[0047] Step 81, at each moment in the first preset time period, construct a Feng Shui topology of the corresponding moment;

[0048] Step 82, input the water flow direction and water flow velocity and wind direction and wind speed at the corresponding time for the power source boundaries of the first channel and the second channel of the matrix structure in the Feng Shui topology, obtain the collision direction and collision velocity of the adjacent collision sub-regions of the rigid influence range of the topological point corresponding to the moored vessel in the first channel and the second channel, and calculate the corrected water flow direction and corrected water flow velocity. The calculation formula is as follows: ; ;

[0049] in, Indicates the correction of water flow direction. Indicates the corrected water flow rate. It represents the Euclidean distance between the mth collider adjacent to the rigid influence range in the first channel and the corresponding topological point. It represents the collision velocity of the mth collider adjacent to the rigid influence range in the first channel. Indicates the collision direction of the mth collider adjacent to the rigid influence range in the first channel, where m represents the index of the collider adjacent to the rigid influence range in the first channel;

[0050] Step 83, repeating step 82 in the second channel to obtain a corrected wind direction and a corrected wind speed;

[0051] Step 84, replacing the environmental characteristic data based on the corrected water flow direction and corrected water flow velocity as well as the corrected wind direction and corrected wind velocity, to obtain the corresponding standard environmental data of the moored vessel.

[0052] Furthermore, the stability prediction model is trained, including:

[0053] In the first preset time period, the standard environment data and the cable-twisting data of the j-th ship at the k-th moment are combined and normalized to construct a feature vector, the feature vector is used as sample data, and the stability score of the j-th ship at the k-th moment is used as a sample label to train a stability prediction model;

[0054] Among them, the stability prediction model is constructed based on a 1D convolutional neural network, and the weight parameters and bias parameters of the 1D convolutional neural network are reversely updated through the mean square error loss function.

[0055] Further, dynamic management includes:

[0056] Step 101, at the s-th moment in the second preset time period, obtaining environmental characteristic data at the s-th moment, and generating standard environmental data of the target moored vessel;

[0057] Step 102, initializing and generating a number of individual populations that meet the constraint conditions;

[0058] Step 103, the individual codes are constructed based on the length and tension of the bow and stern cables, the bow and stern return cables, and the bow and stern spring cables, and are expressed as: ;in, Respectively represent the lengths of bow and stern cables, bow and stern return cables, and bow and stern spring cables. Respectively represent the tension of bow and stern cables, bow and stern return cables and bow and stern spring cables;

[0059] Step 104, the constraint conditions are: the lengths of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the corresponding cable lengths, and the tensions of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the preset tension thresholds;

[0060] Step 105, merging the standard environment data with the code of each individual and normalizing them to obtain the genetic vector of the individual;

[0061] Step 106, obtaining a stability score of each genetic vector based on the stability prediction model;

[0062] Step 107, if the stability score is greater than or equal to the preset stability threshold, the code of the corresponding individual is retained, otherwise the code of the corresponding individual is used as the parent generation for crossover and mutation;

[0063] Step 108, repeating steps 105, 106 and 107 for a preset number of times, then outputting the code of the individual corresponding to the highest stability score;

[0064] If the highest stability score is less than the preset stability threshold, the mooring warning of the target vessel is activated and the transportation work is stopped;

[0065] Otherwise, based on the corresponding individual code, the length and tension of each stranded cable at the sth moment are obtained.

[0066] The beneficial effects of the present invention are:

[0067] 1. By collecting environmental and ship status data in real time and combining genetic algorithms to dynamically optimize each mooring cable, scientific management and regulation of the mooring status can be achieved, ensuring that the ship remains stable in a dynamic environment or issuing danger warnings, significantly improving the ship's mooring safety and transportation efficiency, and reducing the risks caused by environmental fluctuations.

[0068] 2. By quantifying the mutual influence between ships in the moored waters, a standard environmental data generation model based on Feng Shui topology was constructed, so that each ship can obtain targeted standard environmental parameters based on global environmental parameters. The stability prediction model trained based on these corrected data provides a reliable basis for the dynamic optimization and early warning of the ship's cable winding. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a flow chart of a ship mooring status monitoring and early warning method of the present invention;

[0070] Figure 2 It is a schematic diagram of the simulated water surface of the present invention. DETAILED DESCRIPTION

[0071] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0072] like Figure 1-2 As shown, a ship mooring status monitoring and early warning method comprises:

[0073] A data collection module, used to collect environmental characteristic data, stability data of the vessel in the target mooring waters and cable winch data of the target mooring waters at fixed time intervals within a first preset time period;

[0074] Environmental characteristic data include: water flow direction and velocity, wind direction and velocity, and water surface height;

[0075] Stability data include: acceleration parameters, angular velocity parameters, inclination parameters, GPS parameters, driving speed, draft depth and driving direction;

[0076] Cable laying data include: length and tension of bow and stern cables, bow and stern return cables and bow and stern spring cables;

[0077] A data analysis module for calculating a stability score for a moored vessel in a target mooring water area based on the stability data;

[0078] A data correction module is used to construct a Feng Shui topology based on all the ships in the target mooring waters, and generate corresponding standard environmental data for each moored ship based on the Feng Shui topology and environmental characteristic data;

[0079] A model building module, used for training a stability prediction model based on standard environmental data, cable-twisting data and corresponding stability scores of each moored vessel within a first preset time period;

[0080] The cable management module is used to dynamically manage the length and tension of each cable based on the stability prediction model combined with the optimization algorithm within a second preset time period.

[0081] In one embodiment of the present invention, the stability score is calculated using the following formula: ;

[0082] in, represents the stability score of the moored vessel, , and represents the first stability weight, the second stability weight and the third stability weight of the ship, , and All are not 0, and the sum is 1. represents the acceleration parameter of the moored vessel, represents the angular velocity parameter of the moored vessel, Represents the inclination parameter of a moored vessel.

[0083] Specifically, by calculating the stability score of the ship, the key stability data collected by the ship within the first preset time period are weighted and integrated. Specifically, the score reflects the stability of the ship in a dynamic state through the acceleration parameter, angular velocity parameter and inclination parameter at a moment. Each parameter corresponds to a weight (the first, second and third stability weights), which are not zero and their sum is 1, thereby ensuring that each parameter contributes to the comprehensive score and avoiding the situation where a parameter is ignored due to a zero weight. This method accurately quantifies the stability of the ship. The acceleration indicates the intensity of the ship's movement, the angular velocity indicates the turning and swaying characteristics of the ship, and the inclination indicates the overall tilt state of the hull. By taking a reasonable weighted average of these three indicators, the stability score obtained reflects the stability of the ship under complex sea conditions.

[0084] In one embodiment of the present invention, constructing a Feng Shui topology includes:

[0085] Load a simulated water surface, which includes the water surface boundary and water area;

[0086] Configure the water area based on environmental characteristic data: water flow direction and velocity, as well as wind direction and velocity;

[0087] Configure the water surface boundary based on the target mooring area: shape, size and first hydraulic rigidity function;

[0088] The position coordinates of each vessel are obtained based on the GPS parameters, and each vessel is mapped as a topological point in the water area based on the position coordinates;

[0089] For each topological point, a rigid influence range is configured; based on the ship's speed, draft and direction, the corresponding rigid influence range is configured with a wind rigidity function and a second hydraulic rigidity function to obtain a Feng Shui topology.

[0090] Specifically, the simulated environment is used to accurately reflect the interaction of ships so as to correct the subsequent environmental data. First, this method loads a simulated water surface, which is composed of a water surface boundary and an internal water area. Then, the flow direction and velocity of the water flow and the direction and velocity of the wind are configured for the water area using environmental characteristic data. At the same time, the water surface boundary is configured according to the actual situation of the target mooring water area (such as the shape and size of the dock or shoreline), and the first hydraulic rigidity function is introduced to reflect the physical characteristics of the boundary. Subsequently, the method uses GPS data to obtain the location information of each ship and maps each ship as a topological point in the water area. For each topological point, according to the speed, draft and direction of the ship, the rigid influence range is further configured, and the wind rigidity function and the second hydraulic rigidity function are set within the rigid influence range to quantify the impact of environmental forces on the ship. Through this method, the Feng Shui topological model can clearly describe the dynamic interaction between each ship and the environment under complex sea conditions, which provides a basis for the subsequent correction of environmental data.

[0091] In one embodiment of the present invention, configuring the water area based on environmental characteristic data includes:

[0092] The simulated water surface is constructed based on a matrix structure, the matrix structure includes a first channel and a second channel, and the first channel and the second channel both include the same number of colliders;

[0093] The first channel and the second channel of the matrix structure are both defined with the same water surface boundary, and the water surface boundary is divided into a power source boundary and a rigid source boundary; wherein the power source boundary and the rigid source boundary are both composed of a number of colliders, and the water area is composed of the remaining colliders;

[0094] For each collider at the boundary of the power source in the first channel, the hydraulic collision direction and hydraulic collision speed are configured based on the water flow direction and water flow speed; for each collider at the boundary of the power source in the second channel, the wind collision direction and wind collision speed are configured based on the wind direction and wind speed;

[0095] In the first channel or the second channel, if collider i is adjacent to any collider with a collision direction and a collision speed, and is located in the collision direction of the corresponding collider, the first updated collision direction and the first updated collision speed of collider i are obtained, and the calculation formula is as follows: ; ;

[0096] in, represents the first updated collision direction of collider i, represents the first updated collision velocity of collider i, represents the collision velocity of collider i, represents the collision direction of collider i, represents the collision speed of collider i+1, Indicates the collision direction of collider i+1;

[0097] Among them, i and i+1 are the indexes of the colliders in the first channel or the second channel, and the initial collision speed and initial collision direction of the colliders in the water area in the first channel and the second channel are both 0.

[0098] Specifically, this method proposes to use a matrix structure to construct a simulated water surface. The matrix consists of two channels (the first channel and the second channel), and each channel consists of several colliders. In both channels, the same water surface boundary is defined, and the water surface boundary is further divided into a power source boundary and a rigid source boundary, where both types of boundaries are composed of multiple colliders, and the water area itself is composed of the remaining colliders. Then, for each collider of the power source boundary in the first channel, the hydraulic collision direction and hydraulic collision speed are configured for it according to the collected water flow direction and flow velocity; and in the second channel, the wind direction and wind speed are used to configure the wind collision direction and wind collision speed for the corresponding collider, thereby introducing the characteristics of water flow and wind flow in the actual environment into the simulation environment.

[0099] It should be noted that this method specifies the interactive update mechanism of colliders in the matrix structure: in the first channel or the second channel, if a collider i is adjacent to any collider with a defined collision direction and collision speed, and the adjacent collider is located in the collision direction of collider i, then according to the preset calculation formula, collider i will obtain the first updated collision direction and the first updated collision speed. This method simulates the impact of water flow or wind force on the surrounding environment during the transmission process, so that the simulated water surface can gradually form a dynamic state that reflects the propagation of actual physical forces. It is worth noting that the initial collision speed and collision direction of all colliders are set to 0. Only under the action of external parameters and after continuous updates, can a collision state that conforms to the actual situation be gradually established.

[0100] In one embodiment of the present invention, configuring the water surface boundary based on the target mooring waters includes:

[0101] Get the shape and size of the dock boundary and water boundary of the target mooring waters, based on the shape and size of the dock boundary and water boundary, and at a preset scale , which are mapped as rigid source boundaries and dynamic source boundaries in the first channel and the second channel of the matrix structure, respectively. >0;

[0102] Wherein, the first hydraulic rigidity function is configured for the collider in the rigid source boundary;

[0103] In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to a collider in any rigid source boundary, and the corresponding collider is located in the collision direction of collider i, then the second updated collision direction and second updated collision speed of collider i are obtained based on the first hydraulic rigidity function. The calculation formula of the first hydraulic rigidity function is as follows: ; ;

[0104] in, represents the second updated collision direction of collider i, represents the second updated collision velocity of collider i, represents the angle between the collision direction of collider i and the dock boundary, Represents the first velocity attenuation coefficient.

[0105] Specifically, by obtaining the shape and size of the dock boundary and the water boundary in the target mooring water area, and mapping these actual boundaries to the first channel and the second channel in the simulated water surface matrix structure at a preset scaling ratio, a rigid source boundary and a power source boundary are formed respectively. For each collider in the rigid source boundary, a first hydraulic rigidity function is also configured to reflect the influence of the boundary on the water flow and its propagation characteristics. In the first channel, if the collision direction and collision velocity of a certain collider i are not 0, and the collider is adjacent to any collider in the rigid source boundary and the collider of the rigid boundary is located in the collision direction of the collider i, the collider i is updated using the first hydraulic rigidity function, and its second updated collision direction and second updated collision velocity are calculated. The update formula takes into account the angle between the original collision direction of the collider i and the dock boundary and the first velocity attenuation coefficient. This method helps to more accurately transmit and reflect the influence of the dock boundary on the water flow, so that the Feng Shui topology can truly simulate the environment.

[0106] In one embodiment of the present invention, each vessel is mapped as a topological point in the water area based on the position coordinates, including: obtaining the GPS coordinates of each vessel, calculating the relative coordinates of the GPS coordinates and the target mooring water area , based on the relative coordinates, the corresponding ship is mapped to the first channel and the second channel of the matrix structure, specifically: replace the first channel and the second channel in the matrix structure Line The colliders of a column are topological points.

[0107] Specifically, by obtaining the GPS coordinates of each ship and calculating its relative coordinates with the target mooring waters (existing technology), the actual position information is converted into accurate position data in the simulation environment. This allows the distribution of each ship in the actual mooring waters to be reflected in the simulated waters. Next, using these relative coordinates, the colliders at the corresponding row and column positions in the first and second channels of the matrix structure are replaced with topological points, thereby constructing the real spatial distribution of ships on the simulated water surface.

[0108] In one embodiment of the present invention, a rigid influence range is configured for each node in the water area, including defining the rigid influence range as an ellipse:

[0109] The center point of the ellipse is: the topological point;

[0110] The semi-major axis of the ellipse is: ;in, represents the length of the semi-major axis of the ellipse, represents the empirical coefficient, , Indicates the preset zoom ratio. Indicates the draft of the vessel. Indicates the speed of the vessel;

[0111] The semi-minor axis of the ellipse is: ;in, represents the length of the semi-minor axis of the ellipse;

[0112] The inclination angle of the semi-major axis of the ellipse is: the direction of travel of the ship;

[0113] Based on the coverage area of ​​the corresponding ellipse in the first channel and the second channel of the matrix structure, the colliders of the corresponding areas in the first channel and the second channel are used as the rigid influence range of the corresponding topological point;

[0114] The rigid influence range is configured with a second hydraulic rigidity function in the first channel, and the rigid influence range is configured with a wind rigidity function in the second channel;

[0115] In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the third updated collision direction and the third updated collision speed of collider i are obtained based on the second hydraulic rigidity function. The calculation formula of the second hydraulic rigidity function is as follows: ; ;

[0116] in, Indicates the third updated collision direction, represents the third updated collision velocity, Represents the angle between the water flow direction and the semi-major axis of the ellipse, represents the second velocity attenuation coefficient;

[0117] In the second channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the fourth updated collision direction and fourth updated collision speed of collider i are obtained based on the wind rigidity function. The calculation formula of the wind rigidity function is as follows: ; ;

[0118] in, Indicates the fourth updated collision direction, represents the fourth updated collision velocity, represents the angle between the wind direction and the semi-major axis of the ellipse, Indicates the third speed attenuation coefficient.

[0119] Specifically, a rigid influence range is first configured for each node in the water area, and the rigid influence range is defined as an ellipse. The center point of the ellipse is the topological point obtained by the previous mapping, which reflects the actual position of the ship in the water area. The length of the semi-major axis of the ellipse is determined by factors such as the empirical coefficient, the preset scaling ratio, the draft depth and the speed of the ship, while the semi-minor axis provides the size parameter of the ellipse in the other direction; in addition, the inclination angle of the ellipse is set to the direction of travel of the ship, so that the rigid influence range can accurately reflect the dynamic influence of the ship in a specific direction. Based on the area covered by this ellipse, in the first channel and the second channel of the matrix structure, the corresponding collider is used as the rigid influence range of the topological point, and the second hydraulic rigidity function is configured in the first channel, and the wind rigidity function is configured in the second channel.

[0120] Specifically, the updating process of the surrounding colliders is based on the rigid influence range. In the first channel, if the initial collision direction and collision speed of a collider i are not 0, and a neighboring collider is located in this direction, the system will calculate the third updated collision direction and the third updated collision speed of the collider i based on the second hydraulic rigidity function. This update takes into account the angle between the water flow direction and the semi-major axis of the ellipse and the preset second speed attenuation coefficient. Similarly, in the second channel, if the collider i meets the corresponding conditions, the fourth updated collision direction and the fourth updated collision speed of the collider i are obtained based on the wind rigidity function. This process combines the angle between the wind direction and the semi-major axis of the ellipse and the third speed attenuation coefficient. In this way, the Feng Shui topology can more carefully simulate the changes in mechanical parameters in the environment around the ship to obtain standard environmental data.

[0121] In one embodiment of the present invention, generating corresponding standard environment data includes:

[0122] Step 81, at each moment in the first preset time period, construct a Feng Shui topology of the corresponding moment;

[0123] Step 82, input the water flow direction and water flow velocity and wind direction and wind speed at the corresponding time for the power source boundaries of the first channel and the second channel of the matrix structure in the Feng Shui topology, obtain the collision direction and collision velocity of the adjacent collision sub-regions of the rigid influence range of the topological point corresponding to the moored vessel in the first channel and the second channel, and calculate the corrected water flow direction and corrected water flow velocity. The calculation formula is as follows: ; ;

[0124] in, Indicates the correction of water flow direction. Indicates the corrected water flow rate. It represents the Euclidean distance between the mth collider adjacent to the rigid influence range in the first channel and the corresponding topological point. It represents the collision velocity of the mth collider adjacent to the rigid influence range in the first channel. Indicates the collision direction of the mth collider adjacent to the rigid influence range in the first channel, where m represents the index of the collider adjacent to the rigid influence range in the first channel;

[0125] Step 83, repeating step 82 in the second channel to obtain a corrected wind direction and a corrected wind speed;

[0126] Step 84, replacing the environmental characteristic data based on the corrected water flow direction and corrected water flow velocity as well as the corrected wind direction and corrected wind velocity, to obtain the corresponding standard environmental data of the moored vessel.

[0127] Specifically, the Feng Shui topology constructed above is used to generate the standard environmental data corresponding to each ship, providing accurate input for the subsequent stability prediction model. Specifically, within the first preset time period (step 81), the system constructs the current Feng Shui topology at each moment, reflecting the dynamic interaction between the ship and its surrounding environment. Then, in step 82, by inputting the current water flow direction and flow velocity into the first channel of the matrix structure in the Feng Shui topology, and inputting the wind direction and wind speed into the second channel, the collision direction and collision velocity of the adjacent collision sub within the rigid influence range around the corresponding topological point of each moored ship are obtained, and the original water flow data is corrected using a preset calculation formula to obtain the corrected water flow direction and water flow velocity. Subsequently, in step 83, the wind direction and wind speed are processed in the second channel using a method similar to the first channel to obtain the corrected wind direction and wind speed. Finally, in step 84, the corrected parameters replace the original environmental feature data to generate the standard environmental data of each moored ship at the current moment.

[0128] In one embodiment of the present invention, a test vessel is established in the target mooring waters, and a dedicated water flow direction and water flow velocity, as well as wind direction and wind speed detection device is carried out to detect the local area of ​​the test vessel to obtain the corresponding parameters. Then, by calculating the loss of the actually collected parameters and the correction parameters of the Feng Shui topology (such as the mean square error loss), the attenuation coefficient in the Feng Shui topology is reversely optimized.

[0129] In one embodiment of the present invention, training to obtain a stability prediction model includes:

[0130] In the first preset time period, the standard environment data and the cable-twisting data of the j-th ship at the k-th moment are combined and normalized to construct a feature vector, the feature vector is used as sample data, and the stability score of the j-th ship at the k-th moment is used as a sample label to train a stability prediction model;

[0131] Among them, the stability prediction model is constructed based on a 1D convolutional neural network, and the weight parameters and bias parameters of the 1D convolutional neural network are reversely updated through the mean square error loss function.

[0132] Specifically, a stability prediction model is constructed based on the standard environmental data and cable-twisting data collected within the first preset time period. Specifically, at each moment, for each moored vessel, the standard environmental data processed by the data correction module is merged with the corresponding cable-twisting data, and normalized to generate a feature vector. This feature vector reflects the environmental state and cable-twisting condition of the vessel at that time. At the same time, the system uses the stability score of the vessel at that moment as a sample label, thus forming a complete set of input-output data pairs. Subsequently, these sample data are used to train a stability prediction model based on a 1D convolutional neural network. During the training process, the mean square error is used as the loss function, and the weights and bias parameters in the convolutional neural network are continuously updated through back propagation, so that the model can capture the mapping relationship between environmental changes, cable-twisting conditions and vessel stability.

[0133] In one embodiment of the present invention, dynamic management includes:

[0134] Step 101, at the s-th moment in the second preset time period, obtaining environmental characteristic data at the s-th moment, and generating standard environmental data of the target moored vessel;

[0135] Step 102, initializing and generating a number of individual populations that meet the constraint conditions;

[0136] Step 103, the individual codes are constructed based on the length and tension of the bow and stern cables, the bow and stern return cables, and the bow and stern spring cables, and are expressed as: ;in, Respectively represent the lengths of bow and stern cables, bow and stern return cables, and bow and stern spring cables. Respectively represent the tension of bow and stern cables, bow and stern return cables and bow and stern spring cables;

[0137] Step 104, the constraint conditions are: the lengths of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the corresponding cable lengths, and the tensions of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the preset tension thresholds;

[0138] Step 105, merging the standard environment data with the code of each individual and normalizing them to obtain the genetic vector of the individual;

[0139] Step 106, obtaining a stability score of each genetic vector based on the stability prediction model;

[0140] Step 107, if the stability score is greater than or equal to the preset stability threshold, the code of the corresponding individual is retained, otherwise the code of the corresponding individual is used as the parent generation for crossover and mutation;

[0141] Step 108, repeating steps 105, 106 and 107 for a preset number of times, then outputting the code of the individual corresponding to the highest stability score;

[0142] If the highest stability score is less than the preset stability threshold, the mooring warning of the target vessel is activated and the transportation work is stopped;

[0143] Otherwise, based on the corresponding individual code, the length and tension of each stranded cable at the sth moment are obtained.

[0144] Specifically, the ship's mooring state is optimized in real time by using the stability prediction model through the dynamic management process, so as to determine the optimal length and tension of each mooring cable (including bow and stern cables, bow and stern reverse cables and bow and stern spring cables) at each moment. First, in step 101, after collecting environmental feature data at each moment, the standard environmental data of the target ship is generated using the previously constructed data correction method; next, in step 102, the system initializes and generates a population of individuals that meet physical constraints (such as the length of the mooring cable does not exceed the actual cable length, and the tension does not exceed the preset threshold). Each individual expresses the specific length and tension parameters of all mooring cables through the encoding of step 103. On this basis, step 105 merges and normalizes the encoding of each individual with the current standard environmental data to form the genetic vector of the individual; then, through step 106, the trained stability prediction model is used to score each genetic vector to reflect the stability of the ship under the corresponding parameter combination. In step 107, the system selects individuals whose stability scores reach or exceed the preset threshold, and crosses and mutates the individuals that do not meet the requirements as parents, iterates and optimizes (step 108) until the preset number of times is reached, and then outputs the individual code with the highest stability score. If the highest score is still lower than the stability threshold, the system will trigger a mooring warning and stop the transportation operation; otherwise, the optimal length and tension of each mooring cable at the current moment are determined based on the final individual code. The whole process realizes real-time control and risk warning of the mooring state in a complex dynamic environment, ensuring the safety and operational stability of the ship in the moored state.

[0145] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A ship mooring status monitoring and early warning method, characterized in that: include: A data collection module, used to collect environmental characteristic data, stability data of the vessel in the target mooring waters and cable winch data of the target mooring waters at fixed time intervals within a first preset time period; Environmental characteristic data include: water flow direction and velocity, wind direction and velocity, and water surface height; Stability data include: acceleration parameters, angular velocity parameters, inclination parameters, GPS parameters, driving speed, draft depth and driving direction; Cable laying data include: length and tension of bow and stern cables, bow and stern return cables and bow and stern spring cables; A data analysis module for calculating a stability score for a moored vessel in a target mooring water area based on the stability data; A data correction module is used to construct a Feng Shui topology based on all the ships in the target mooring waters, and generate corresponding standard environmental data for each moored ship based on the Feng Shui topology and environmental characteristic data; A model building module, used for training a stability prediction model based on standard environmental data, cable-twisting data and corresponding stability scores of each moored vessel within a first preset time period; The cable management module is used to dynamically manage the length and tension of each cable based on the stability prediction model combined with the optimization algorithm within a second preset time period.

2. A ship mooring status monitoring and early warning method according to claim 1, characterized in that: Calculate the stability score using the following formula: ； in, represents the stability score of the moored vessel, , and represents the first stability weight, the second stability weight and the third stability weight of the ship, , and All are not 0, and the sum is 1. represents the acceleration parameter of the moored vessel, represents the angular velocity parameter of the moored vessel, Represents the inclination parameter of a moored vessel.

3. A ship mooring status monitoring and early warning method according to claim 1, characterized in that: Build Feng Shui topology, including: Load a simulated water surface, which includes the water surface boundary and water area; Configure the water area based on environmental characteristic data: water flow direction and velocity, as well as wind direction and velocity; Configure the water surface boundary based on the target mooring area: shape, size and first hydraulic rigidity function; The position coordinates of each vessel are obtained based on the GPS parameters, and each vessel is mapped as a topological point in the water area based on the position coordinates; For each topological point, a rigid influence range is configured; based on the speed, draft and direction of the ship, the corresponding rigid influence range is configured with a wind rigidity function and a second hydraulic rigidity function to obtain a Feng Shui topology.

4. A ship mooring status monitoring and early warning method according to claim 3, characterized in that: Water area configuration based on environmental characteristic data, including: The simulated water surface is constructed based on a matrix structure, the matrix structure includes a first channel and a second channel, and the first channel and the second channel both include the same number of colliders; The first channel and the second channel of the matrix structure are both defined with the same water surface boundary, and the water surface boundary is divided into a power source boundary and a rigid source boundary; wherein the power source boundary and the rigid source boundary are both composed of a number of colliders, and the water area is composed of the remaining colliders; For each collider at the boundary of the power source in the first channel, the collision direction and collision speed are configured based on the water flow direction and water flow speed; for each collider at the boundary of the power source in the second channel, the collision direction and collision speed are configured based on the wind direction and wind speed; In the first channel or the second channel, if collider i is adjacent to any collider with a collision direction and a collision speed, and is located in the collision direction of the corresponding collider, the first updated collision direction and the first updated collision speed of collider i are obtained, and the calculation formula is as follows: ； ； in, represents the first updated collision direction of collider i, represents the first updated collision velocity of collider i, represents the collision velocity of collider i, represents the collision direction of collider i, represents the collision speed of collider i+1, Indicates the collision direction of collider i+1; Among them, i and i+1 are the indexes of the colliders in the first channel or the second channel, and the initial collision speed and initial collision direction of the colliders in the water area in the first channel and the second channel are both 0.

5. A ship mooring status monitoring and early warning method according to claim 4, characterized in that: The water surface boundary configuration based on the target mooring waters includes: Get the shape and size of the dock boundary and water boundary of the target mooring waters, based on the shape and size of the dock boundary and water boundary, and at a preset scale , which are mapped as rigid source boundaries and dynamic source boundaries in the first channel and the second channel of the matrix structure, respectively. >0; Wherein, the first hydraulic rigidity function is configured for the collider in the rigid source boundary; In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to a collider in any rigid source boundary, and the corresponding collider is located in the collision direction of collider i, then the second updated collision direction and second updated collision speed of collider i are obtained based on the first hydraulic rigidity function. The calculation formula of the first hydraulic rigidity function is as follows: ； ； in, represents the second updated collision direction of collider i, represents the second updated collision velocity of collider i, represents the angle between the collision direction of collider i and the dock boundary, Represents the first velocity attenuation coefficient.

6. A ship mooring status monitoring and early warning method according to claim 5, characterized in that: Map each ship as a topological point in the waters based on the location coordinates, including: obtaining the GPS coordinates of each ship, calculating the relative coordinates of the GPS coordinates and the target mooring waters , based on the relative coordinates, the corresponding ship is mapped to the first channel and the second channel of the matrix structure, specifically: replace the first channel and the second channel in the matrix structure Line The colliders of a column are topological points.

7. A ship mooring status monitoring and early warning method according to claim 6, characterized in that: For each node in the water area, configure the rigid sphere of influence, including defining the rigid sphere of influence as an ellipse: The center point of the ellipse is: the topological point; The semi-major axis of the ellipse is: ;in, represents the length of the semi-major axis of the ellipse, represents the empirical coefficient, , Indicates the preset zoom ratio. Indicates the draft of the vessel. Indicates the speed of the vessel; The semi-minor axis of the ellipse is: ;in, represents the length of the semi-minor axis of the ellipse; The inclination angle of the semi-major axis of the ellipse is: the direction of travel of the ship; Based on the coverage area of ​​the corresponding ellipse in the first channel and the second channel of the matrix structure, the colliders of the corresponding areas in the first channel and the second channel are used as the rigid influence range of the corresponding topological point; The rigid influence range is configured with a second hydraulic rigidity function in the first channel, and the rigid influence range is configured with a wind rigidity function in the second channel; In the first channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the third updated collision direction and the third updated collision speed of collider i are obtained based on the second hydraulic rigidity function. The calculation formula of the second hydraulic rigidity function is as follows: ； ； in, Indicates the third updated collision direction, represents the third updated collision velocity, Represents the angle between the water flow direction and the semi-major axis of the ellipse, represents the second speed attenuation coefficient; In the second channel, if the collision direction and collision speed of collider i are not 0, collider i is adjacent to any collider in the rigid influence range, and the corresponding collider is located in the collision direction of collider i, then the fourth updated collision direction and fourth updated collision speed of collider i are obtained based on the wind rigidity function. The calculation formula of the wind rigidity function is as follows: ； ； in, Indicates the fourth updated collision direction, represents the fourth updated collision velocity, represents the angle between the wind direction and the semi-major axis of the ellipse, Indicates the third speed attenuation coefficient.

8. A ship mooring status monitoring and early warning method according to claim 7, characterized in that: Generate corresponding standard environment data, including: Step 81, at each moment in the first preset time period, construct a Feng Shui topology of the corresponding moment; Step 82, input the water flow direction and water flow velocity and wind direction and wind speed at the corresponding time for the power source boundaries of the first channel and the second channel of the matrix structure in the Feng Shui topology, obtain the collision direction and collision velocity of the adjacent collision sub-regions of the rigid influence range of the topological point corresponding to the moored vessel in the first channel and the second channel, and calculate the corrected water flow direction and corrected water flow velocity. The calculation formula is as follows: ； ； in, Indicates the correction of water flow direction. Indicates the corrected water flow rate. It represents the Euclidean distance between the mth collider adjacent to the rigid influence range in the first channel and the corresponding topological point. It represents the collision velocity of the mth collider adjacent to the rigid influence range in the first channel. Indicates the collision direction of the mth collider adjacent to the rigid influence range in the first channel, where m represents the index of the collider adjacent to the rigid influence range in the first channel; Step 83, repeating step 82 in the second channel to obtain a corrected wind direction and a corrected wind speed; Step 84, replacing the environmental characteristic data based on the corrected water flow direction and corrected water flow velocity as well as the corrected wind direction and corrected wind velocity, to obtain the corresponding standard environmental data of the moored vessel.

9. A ship mooring status monitoring and early warning method according to claim 8, characterized in that: The stability prediction model obtained by training includes: In the first preset time period, the standard environment data and the cable-twisting data of the j-th ship at the k-th moment are combined and normalized to construct a feature vector, the feature vector is used as sample data, and the stability score of the j-th ship at the k-th moment is used as a sample label to train a stability prediction model; Among them, the stability prediction model is constructed based on a 1D convolutional neural network, and the weight parameters and bias parameters of the 1D convolutional neural network are reversely updated through the mean square error loss function.

10. A ship mooring status monitoring and early warning method according to claim 9, characterized in that: Dynamic management, including: Step 101, at the s-th moment in the second preset time period, obtaining environmental characteristic data at the s-th moment, and generating standard environmental data of the target moored vessel; Step 102, initializing and generating a number of individual populations that meet the constraint conditions; Step 103, the individual codes are constructed based on the length and tension of the bow and stern cables, the bow and stern return cables, and the bow and stern spring cables, and are expressed as: ;in, Respectively represent the lengths of bow and stern cables, bow and stern return cables, and bow and stern spring cables. Respectively represent the tension of bow and stern cables, bow and stern return cables and bow and stern spring cables; Step 104, the constraint conditions are: the lengths of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the corresponding cable lengths, and the tensions of the bow and stern cables, the bow and stern return cables and the bow and stern spring cables are all less than or equal to the preset tension thresholds; Step 105, merging the standard environment data with the code of each individual and normalizing them to obtain the genetic vector of the individual; Step 106, obtaining a stability score of each genetic vector based on the stability prediction model; Step 107, if the stability score is greater than or equal to the preset stability threshold, the code of the corresponding individual is retained, otherwise the code of the corresponding individual is used as the parent generation for crossover and mutation; Step 108, repeating steps 105, 106 and 107 for a preset number of times, then outputting the code of the individual corresponding to the highest stability score; If the highest stability score is less than the preset stability threshold, the mooring warning of the target vessel is activated and the transportation work is stopped; Otherwise, based on the corresponding individual code, the length and tension of each stranded cable at the sth moment are obtained.

Images