A method for identifying and managing collision hazards in waterways based on propagation dynamics

Abstract

The application provides a water area collision danger identification and management method based on propagation dynamics, and belongs to the technical field of ship traffic safety. The latitude and longitude information of all ships in a target water area is extracted to form a ship spatial position information set; a propagation dynamics model SEIRS is used to model the multi-ship encounter problem of the target water area based on the ship spatial position information, so as to determine the propagation state of each ship in the target water area; a propagation parameter is introduced to combine the collision danger propagation model of the multi-ship encounter in the target water area, a collision danger propagation system differential equation is constructed, and is solved to obtain an optimal propagation parameter; the optimal propagation parameter value is used to establish a derivative index to represent the collision danger in the target water area; and the ships in the target water area are dispatched according to the derivative index, so that the collision danger management of the target water area is realized. The derivative index comprehensively represents the collision danger of the multi-ship encounter, and the dispatch of the ships based on the derivative index improves the timeliness and reliability of the multi-ship encounter danger management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship traffic safety, in particular to a water collision danger identification and management method based on propagation dynamics. BACKGROUND

[0002] With the continuous development of global economy, the number of ships as the main tool for transporting goods at sea has also increased significantly. Busy ship traffic will lead to more crowded ships in the water area, and ships will face more frequent encounters and more complex and variable encounter forms, which makes it more likely for ships to collide. At the same time, for water areas with busy and complex ship traffic, the existence of collision danger in the water area is a dynamic and continuous process, which is not suddenly generated and disappeared, but gradually propagated through ships in the water area. In order to keep the safety of ship traffic in the water area stable and continuous, ship traffic supervisors not only need to pay attention to the state of the water collision danger, but also need to understand the potential and propagation ability of the water collision danger, so as to manage the ships in the water area.

[0003] The existing method indirectly assesses the water collision danger based on non-accident information, including directly establishing the relationship between the water collision danger and the ship traffic variables through mathematical analysis expression method, and applying machine learning and deep learning technology to the quantification and prediction of water collision danger. However, the above water collision danger identification method focuses more on collision danger assessment or analysis at a certain moment or time period, and characterizes the water collision danger from the perspective of space-time distribution, but pays insufficient attention to the dynamic change mechanism of the collision danger between different moments. Ignoring the dynamic propagation process of the collision danger under the condition of multi-ship interaction, it is difficult to reveal how the collision danger continues to spread and evolve among ships, thereby having significant limitations in grasping the overall situation and systematic risk causes of the water collision danger, especially in the current complex water environment with high traffic density and busy ship traffic. Such static assessment is difficult to support the dynamic and accurate risk management needs.

[0004] Therefore, there is an urgent need for a water collision danger identification and management method based on propagation dynamics. SUMMARY

[0005] Therefore, the present application provides a water collision danger identification and management method based on propagation dynamics, which integrates the propagation mechanism of water collision danger modeling, and describes the propagation ability, peak characteristics and danger intensity of collision danger in the target water area. This method has the characteristics of dynamics, mechanism and predictability, thereby providing more scientific and effective theoretical and methodological support for the prospective monitoring, strategy formulation and active prevention and control of maritime collision danger.

[0006] To this end, the present application provides the following technical solutions:

[0007] A water collision danger identification and management method based on propagation dynamics, comprising:

[0008] Receiving the latitude and longitude information of all ships in the target water area to form a ship spatial position information set; modeling the multi-ship encounter problem of the target water area based on the ship spatial position information set through the SEIRS propagation dynamics model to obtain a collision danger propagation model for multi-ship encounter in the target water area; determining the propagation state of each ship in the target water area based on the collision danger propagation model; introducing a propagation parameter and combining the collision danger propagation model for multi-ship encounter in the target water area to construct a collision danger propagation system differential equation; solving the collision danger propagation system differential equation based on the propagation state of each ship in the target water area to obtain an optimal propagation parameter; using the optimal propagation parameter value to establish a derived index to represent the collision danger in the target water area; and scheduling the ships in the target water area according to the derived index to achieve collision danger management of the target water area.

[0009] Further, the determination of the propagation state of each ship in the target water area based on the collision danger propagation model comprises:

[0010] The ship spatial positions in the target water area are clustered using the DBSCAN spatial clustering method to obtain a plurality of clustering clusters, each clustering cluster being a multi-ship encounter cluster; the space occupied by the multi-ship encounter cluster is a multi-ship encounter area; the overall space range formed by the multi-ship encounter cluster being enlarged by 2 times along the radial direction is the overall influence range; the annular area between the overall influence range and the multi-ship encounter cluster is the multi-ship encounter influence area; the area outside the multi-ship encounter area and the multi-ship encounter influence area in the target water area is the non-multi-ship encounter influence area; the ships in the multi-ship encounter area are infected ships; the ships in the multi-ship encounter influence area are exposed ships; the ships that leave the multi-ship encounter area and enter the multi-ship encounter influence area are recovered ships; and the ships in the non-multi-ship encounter influence area are susceptible ships.

[0011] Further, the collision danger propagation system differential equation is:

[0012]

[0013]

[0014] wherein, N represents the number of susceptible ships, E represents the number of exposed ships, I represents the number of infected ships, R represents the number of recovered ships; N represents the total number of ships, i.e., the sum of the number of ships in the four propagation states; the propagation parameter includes: β is the infection rate, is the latent conversion rate, is the recovery rate and is the immune loss rate; system parameters: denotes the entry rate parameter and denotes the exit rate parameter.

[0015] Further, the solving the collision danger propagation system differential equation obtains the optimal propagation parameter, comprising: integrating the collision danger propagation system differential equation to generate a simulation sequence of the propagation state variable; comparing the simulation sequence result with the historical data result in error, and adopting the fmincon algorithm based on constraint optimization to fit the propagation parameter most consistent with the system propagation characteristics as the optimal propagation parameter value.

[0016] Further, the use of the optimal propagation parameter value establishes a derivative index to represent the collision danger in the target water area, comprising: representing the danger propagation ability by the basic reproduction number; representing the danger peak characteristic by the maximum infection proportion; representing the danger intensity level by the cumulative infection load.

[0017] Further, the basic reproduction number:

[0018]

[0019] wherein, denotes the basic reproduction number.

[0020] Further, the maximum infection proportion:

[0021]

[0022] wherein, denotes the number of infected ships at time, denotes the total number of ships at time, is a time label, denotes the last time label in the detection period of the target water area; denotes the maximum infection proportion.

[0023] Further, the cumulative infection load:

[0024]

[0025] wherein, denotes the total length of the detection period of the target water area, denotes the cumulative infection load.

[0026] Further, according to the derived index, the ship scheduling in the target water area is carried out, including: when the basic regeneration number, the maximum infection ratio or the cumulative infection load is greater than a preset threshold, the ship outside the target water area is limited to enter the target water area and the ship inside the target water area is reminded to pay attention to the encounter situation and adjust the heading and speed in time.

[0027] A water area collision danger identification and management device based on propagation dynamics, comprising: a ship state determination unit, a propagation parameter acquisition unit, a derived index calculation unit and a ship scheduling unit.

[0028] The ship state determination unit receives the latitude and longitude information of all ships in the target water area to form a ship space position information set; based on the ship space position information set, a propagation dynamics model SEIRS is used to model the multi-ship encounter problem of the target water area to obtain a collision danger propagation model of the multi-ship encounter in the target water area; and based on the collision danger propagation model, the propagation state of each ship in the target water area is determined.

[0029] The propagation parameter acquisition unit introduces the propagation parameter, combines the collision danger propagation model of the multi-ship encounter in the target water area, constructs a collision danger propagation system differential equation, and solves the collision danger propagation system differential equation based on the propagation state of each ship in the target water area to obtain the optimal propagation parameter.

[0030] The derived index calculation unit uses the optimal propagation parameter value to establish a derived index to represent the collision danger in the target water area.

[0031] The ship scheduling unit schedules the ships in the target water area according to the derived index to realize the collision danger management of the target water area.

[0032] The advantages and positive effects of the present application are as follows:

[0033] The method introduces the propagation parameter, combines the propagation dynamics model to model the multi-ship encounter problem of the target water area, obtains the optimal propagation parameter, uses the optimal propagation parameter to establish a derived index to comprehensively represent the collision danger of the multi-ship encounter, and schedules the ships based on the derived index, thereby improving the timeliness and reliability of the multi-ship encounter danger management.

[0034] 1. According to the relationship between the ship and the multi-ship encounter, the state of the ship in the water area is accurately and carefully divided. This state division not only can refine the relationship between the ship in the water area and the danger, but also can provide support for the establishment of the subsequent collision danger propagation system.

[0035] 2, The method introduces the propagation dynamics model SEIRS into the research of maritime traffic, constructs a dynamic propagation system of collision danger in water area, and constructs a differential equation model thereof, obtains key parameters of the propagation system through numerical integration and parameter optimization fitting, accurately describes state change of a ship individual in the system, and lays a theoretical foundation for comprehensively mastering the situation of collision danger in water area.

[0036] 3, The method constructs derived indexes capable of characterizing collision danger in water area from multiple dimensions based on the constructed differential equation of the collision danger propagation system and the fitted propagation parameters, which can not only characterize the dangerous state of a target water area, but also measure the collision danger propagation potential of the water area, so that the ship traffic management can be more comprehensive. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0038] Figure 1 The flow chart of the water collision danger identification and management method based on propagation dynamics in embodiment 1;

[0039] Figure 2 The structure diagram of the water collision danger identification and management device based on propagation dynamics in embodiment 1;

[0040] Figure 3 The multi-ship encounter cluster in embodiment 2;

[0041] Figure 4 The influence range of the multi-ship encounter cluster in embodiment 2;

[0042] Figure 5 The state of each ship in the target water area in embodiment 2;

[0043] Figure 6 The state transition process in the collision danger propagation system and the propagation parameters between states in embodiment 2;

[0044] Figure 7 The ship entering and exiting diagram of the collision danger propagation system in embodiment 2;

[0045] Figure 8 The diagram of the derived indexes changing with time in embodiment 2. DETAILED DESCRIPTION

[0046] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application, so that those skilled in the art can better understand the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of the present application.

[0047] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units need not be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0048] The present application provides a water collision danger identification and management method based on propagation dynamics. First, according to the actual mechanism of ship encounter collision danger, the water collision danger system is mapped into a propagation dynamics model, and the state of the ship in the water is defined. Then, according to the divided ship state, the collision danger propagation system differential equation based on propagation dynamics is established, and the propagation parameters are obtained by numerical integration and parameter optimization technology fitting. According to the obtained propagation parameters, the derived index which can describe the characteristics of water collision danger from multiple dimensions is constructed, and the collision danger situation of the target water area is comprehensively evaluated. Finally, the derived index is used for propagation scheduling in the target water area.

[0049] A water collision danger identification and management method based on propagation dynamics, comprising the following steps:

[0050] S1, extracting the latitude and longitude information of all ships in the target water area to form a ship spatial position information set; and modeling the multi-ship encounter problem of the target water area based on the ship spatial position information to determine the state of each ship in the target water area.

[0051] S2, introducing propagation parameters to combine the collision danger propagation model of multi-ship encounter in the target water area to construct a collision danger propagation system differential equation.

[0052] S3, solving the collision danger propagation system differential equation to obtain the optimal propagation parameters.

[0053] S4, establishing a derived indicator using the optimal propagation parameter.

[0054] S5, guiding ship scheduling according to the derived indicator to realize ship management.

[0055] Embodiment 1

[0056] When ships approach each other in a water area, collision danger arises, and the ships will face the possibility of a collision accident. If the above behavior occurs between multiple ships, collision danger in the multi-ship encounter process will arise, which will affect each ship in the encounter. Considering the complex multi-ship encounter form, limited collision avoidance space, and increasing difficulty of collision avoidance, it is the main element of collision danger at the water level. Any ship approaching or entering a multi-ship encounter will be affected, resulting in obvious collision danger. Therefore, in combination with the above Figure 1 As shown in FIG. 1, a water collision danger identification and management method based on propagation dynamics is used to reduce collision danger, including the following steps:

[0057] S1, extracting the latitude and longitude information of all ships in the target water area to form a ship spatial position information set; and modeling the multi-ship encounter problem of the target water area based on the ship spatial position information to determine the state of each ship in the target water area.

[0058] The propagation process of collision danger in the multi-ship encounter of the target water area is modeled by a propagation dynamics model. In this embodiment, the SEIRS (Susceptible-Exposed-Infectious-Recovered-Susceptible) model is used. Among them, S represents the susceptible, E represents the exposed, I represents the infected, and R represents the recovered.

[0059] In this embodiment, according to the ship spatial position information set, the DBSCAN spatial clustering method is used to cluster the ship spatial positions at each time in the target water area, and each clustering cluster obtained by clustering is used as a multi-ship encounter cluster.

[0060] The space occupied by the multi-ship encounter cluster is the multi-ship encounter area; the overall space range formed by the multi-ship encounter cluster expanding outward by 2 times along the radial direction is the total impact range; the annular area between the total impact range and the multi-ship encounter cluster is the multi-ship encounter impact area; the area outside the multi-ship encounter area and the multi-ship encounter impact area in the target water area is the non-multi-ship encounter impact area; the ships in the multi-ship encounter area are the infected ships; the ships in the multi-ship encounter impact area are the exposed ships; the ships that leave the multi-ship encounter area and enter the multi-ship encounter impact area are the recovered ships; the ships in the non-multi-ship encounter impact area are the susceptible ships.

[0061] For a ship in the target water area, it may experience the process from susceptible to exposed to infected to recovered, and the recovered ship in the multi-ship encounter will also change back to the initial susceptible under certain conditions, forming a closed loop.

[0062] S2, introduce the propagation parameter to combine the collision danger propagation model of multi-ship encounter in the target water area, and construct the collision danger propagation system differential equation:

[0063]

[0064]

[0065] wherein, represents the number of susceptible ships, represents the number of exposed ships, represents the number of infected ships, represents the number of recovered ships; represents the total number of ships, that is, the sum of the number of ships in four propagation states; considering the transfer mechanism between the propagation states, the propagation parameters between the propagation states are defined, and the propagation parameters include: is the infection rate, is the latent conversion rate, is the recovery rate, and is the immune loss rate; Since the ship traffic in the water area is not a closed system, new ships will enter and exit the encounter water area at any time, therefore, the system parameters are introduced: represents the entering rate parameter, and represents the exiting rate parameter.

[0066] S3, solve the collision danger propagation system differential equation to obtain the optimal propagation parameter.

[0067] 1) Based on the propagation state of each ship at each time, the number of ships in each propagation state at each time is calculated by combining the collision danger propagation system differential equation.

[0068] 2) According to the number of ships in each propagation state at each time, the water area collision danger propagation differential equation model is fitted, and the propagation parameter is estimated: in this embodiment, the ode45 algorithm based on explicit Runge-Kutta is used to integrate and solve the collision danger propagation system differential equation, to generate a simulation sequence of state variables; the simulation sequence result is compared with the historical data result, and the fmincon algorithm based on constraint optimization is used to fit the propagation parameter most consistent with the system propagation characteristics as the optimal propagation parameter value.

[0069] S4, establish a multi-dimensional representation derived index of water area collision danger with optimal propagation parameter value, represent the danger propagation ability, danger peak characteristic and danger intensity level in the target water area, and realize more comprehensive representation of the target water area collision danger situation.

[0070] 1) Represent the danger propagation ability by the basic reproduction number, which is defined as the expected number of newly infected ships caused by a single infected ship in a ship population where all individuals are susceptible. Considering the non-closed nature of the ship traffic system, i.e. the number of ships in the system is not fixed but there are ships entering and exiting at any time, the formula of the basic reproduction number is:

[0071]

[0072] When >1, the ship encounter situation in the current water area will lead to the spread and diffusion of collision danger in the ship population, and the larger this value, the faster the spread and diffusion.

[0073] When <1, the collision danger in the current water area tends to gradually dissipate, and even if there is a ship encounter, it is difficult to form the spread and propagation of collision danger.

[0074] 2) Represent the danger peak characteristic by the maximum infection ratio, which is defined as the maximum value of the ratio of the number of ships in the infected state to the total number of ships at that time in the entire propagation process of collision danger in the water area. The calculation formula is:

[0075]

[0076] Wherein, represents the number of infected ships at , represents the total number of ships at , is the time label, which is mapped to the specific time stamp in specific calculation, represents the last time label in the detection period of the target water area; the maximum infection ratio is closer to 1, indicating that the instantaneous peak of the collision danger of the system is higher.

[0077] 3) Represent the danger propagation ability by the cumulative infection load, which is defined as how many "ship-time units" of multi-ship encounter state the water area collision danger system has experienced in the entire propagation period. The calculation formula is:

[0078]

[0079] Wherein, ​​represents the total length of the detection period of the target water area, represents the cumulative infection load. The cumulative infection load can reflect the persistence and total intensity of the overall dangerous state of the ship in a certain period, is used to measure the dangerous carrying pressure of the water area in this stage, and is an important basis for evaluating the overall dangerous intensity level of the water area.

[0080] S5, guiding ship scheduling according to the derived index, realizing ship management.

[0081] Embodiment 2

[0082] Based on the method in embodiment 1, the method is further illustrated in a certain target water area:

[0083] S1, extracting ship spatial position information in the target water area from various types of navigation sensors, including timestamp , ship movement business identification code , ship longitude , ship latitude , and constructing a ship spatial position information set of the target water area . The ship spatial position information set of the ship at time t is as follows:

[0084]

[0085] According to the ship spatial position information set, the DBSCAN spatial clustering method is used to cluster the ship spatial position in the target water area, and each clustering cluster obtained by clustering is regarded as a multi-ship encounter cluster.

[0086] 1) The division of the target water area, as shown in Figure 3 and Figure 4 :

[0087] The space occupied by the multi-ship encounter cluster is the multi-ship encounter area; the overall space range formed by the multi-ship encounter cluster expanding outward by 2 times in the radial direction is the overall influence range; the annular area between the overall influence range and the multi-ship encounter cluster is the multi-ship encounter influence area, and the area outside the multi-ship encounter area and the multi-ship encounter influence area in the target water area is the non-multi-ship encounter influence area.

[0088] 2) Map the ship collision problem in the target water area to SEIRS, and the state of each ship is as shown in Figure 5 :

[0089] The ships in the water area are regarded as the research group, and the collision danger is regarded as the epidemic; the transmission states include: susceptible, exposed, infected and recovered. The mechanism of multi-ship encounter generating collision danger is combined with the principle of transmission dynamics, and the state of each ship in the collision danger transmission model is defined from the perspective of multi-ship encounter: the ships in the multi-ship encounter area are infected ships, which have been infected by the multi-ship encounter “virus” and can infect other ships with collision danger from the perspective of epidemic transmission dynamics. The ships in the multi-ship encounter influence area are exposed ships, which have been infected by the multi-ship encounter “virus” but cannot infect other ships with collision danger from the perspective of epidemic transmission dynamics.

[0090] The ships in the multi-ship encounter influence area are recovered ships, which have been recovered from collision danger by sailing away from the surrounding ships from the perspective of epidemic transmission dynamics.

[0091] The ships in the non-multi-ship encounter influence area are susceptible ships, which have not been infected by the multi-ship encounter “virus” from the perspective of epidemic transmission dynamics.

[0092] According to the ship state definition, the transmission state of each ship in the target water area at each moment is identified to form a ship transmission state label information set , the ship The ship transmission state label information set at a certain moment is shown as follows:

[0093]

[0094] S2, according to the transmission state of the ship state determination unit, the transmission state of the ship in the collision danger transmission system is divided, the transmission parameters between each transmission state are defined by considering the transfer mechanism between each transmission state, and the differential equation of the collision danger transmission system is established. As shown in Figure 6 , the differential equation of the collision danger transmission system:

[0095]

[0096]

[0097]

[0098]

[0099] wherein, S(t) represents the number of susceptible ships, E(t) represents the number of exposed ships, I(t) represents the number of infected ships, R(t) represents the number of recovered ships. denotes the total number of ships, i.e. the sum of the number of ships in the four propagation states; the propagation parameters include: is the infection rate, is the latent transformation rate, is the recovery rate and is the immunity loss rate; the system parameters include: denotes the entry rate parameter and denotes the exit rate parameter. In this embodiment, the entry rate parameter and the exit rate parameter can be determined by the traffic situation in the water area, and the entry and exit processes of the ships are shown in FIG. 3. Figure 7

[0100] S3, based on the collision risk propagation system differential equation, the number of ships in each propagation state at each time is calculated according to the propagation state of each ship at each time. The optimal propagation parameters in the collision risk propagation system are fitted by numerical integration and parameter optimization.

[0101] According to the ship propagation state label information set obtained in the ship state determination unit, the number of ships in each propagation state at each time is calculated as the time propagation state set .

[0102]

[0103] wherein, denotes the number of susceptible ships at time , denotes the number of exposed ships at time , denotes the number of infected ships at time , denotes the number of recovered ships at time . The time series of each propagation state , , and are generated.

[0104] Based on the propagation state set at each time, the collision risk propagation differential equation model of the water area is fitted, and the propagation parameters are estimated. Numerical integration and parameter optimization are used to fit the propagation parameters and the system parameters. First, the ode45 algorithm based on explicit Runge-Kutta is used to integrate and solve the SEIRS differential equation, the initial values of each state and the candidate values of each propagation parameter and system parameter are given, and the simulation sequence of the model state variable is generated. Second, the simulation sequence is compared with the propagation state set at each time ​The obtained actual propagation state sequence is compared for error, and a target function is defined as the sum of the mean square errors of each state variable at the observation time point. Finally, the fmincon algorithm based on constraint optimization is used to search for the optimal parameter combination to minimize the error between the simulation results and the actual observation, thereby inversely fitting the propagation parameter values that best match the system propagation characteristics, including the infection rate, latent transformation rate, recovery rate, immunity loss rate, entry rate, and exit rate.

[0105] S4, calculate the collision risk propagation system derived index, multi-dimensional representation of the collision risk situation in the water area:

[0106] After the key propagation parameters of the collision risk propagation system are calculated by the propagation parameter acquisition unit, these key propagation parameters are used to construct derived indexes that can depict and represent the collision risk situation in the water area from different dimensions, including a derived index that can represent the water area risk propagation ability, i.e., the basic reproduction number , a derived index that can represent the water area risk peak characteristics, i.e., the maximum infection ratio , and a derived index that can represent the water area risk intensity level and scale, i.e., the cumulative collision risk load . For a preset detection period in the target water area, the basic reproduction number, the maximum infection ratio, and the cumulative collision risk load can quantitatively represent the propagation ability, peak characteristics, and intensity level of the collision risk in the water area. Further, the collision risk situation information set of the water area at the detection period can be generated.

[0107] S5, in this embodiment, the detection results of the target water area in the preset detection period are as shown in Tables 1 and Figure 8 Table 1 is the collision risk propagation system derived indexes and normalized values in a 4-hour period, the normalization adopts the min-max normalization, and the normalization coefficient can be obtained through historical data analysis and calculated and dynamically adjusted based on the total number of ships.

[0108] Table 1

[0109]

[0110] Figure 8The detection results of 6 consecutive detection periods of the target water area show that the basic reproduction number of the second period data increases significantly, indicating that the collision danger may spread and propagate, and the cumulative infection load reaches a peak at the same time, indicating that the dangerous state has accumulated to the critical level. At this time, the system automatically triggers management intervention, intervenes through means such as limiting the entry of new ships into the water area and VHF broadcast reminding the ships in the water area to pay attention to the encounter situation. Subsequently, the related ship actively adjusts the speed or heading, reducing the complexity of multi-ship interaction. The intervention effect of the above regulatory intervention is verified in the subsequent period. If no action is taken, the danger index will further rise to a relatively higher value. After the actual intervention, the basic reproduction number falls to the safety threshold, and the cumulative infection load and the maximum infection proportion of the auxiliary index also decrease, so that the collision in the target water area is effectively alleviated and controlled.

[0111] Embodiment 3

[0112] The application also provides a water area collision danger identification and management device based on propagation dynamics, as shown in the figure, comprising a ship state judgment unit, a propagation parameter acquisition unit, a derived index calculation unit and a ship scheduling unit. Figure 2

[0113] 1. Ship state judgment unit: receive the latitude and longitude information of all ships in the target water area to form a ship spatial position information set; model the multi-ship encounter problem in the target water area based on the ship spatial position information set through the SEIRS propagation dynamics model to obtain a collision danger propagation model for multi-ship encounter in the target water area; determine the propagation state of each ship in the target water area based on the collision danger propagation model.

[0114] 2. Propagation parameter acquisition unit: introduce propagation parameters to combine the collision danger propagation model for multi-ship encounter in the target water area to construct a collision danger propagation system differential equation; solve the collision danger propagation system differential equation based on the propagation state of each ship in the target water area to obtain the optimal propagation parameter.

[0115] 3. Derived index calculation unit: use the optimal propagation parameter value to establish a derived index to represent the collision danger in the target water area.

[0116] 4. Ship scheduling unit: schedule the ships in the target water area according to the derived index to realize the collision danger management of the target water area. When the basic reproduction number, the maximum infection proportion or the cumulative infection load is greater than the preset threshold, the ships outside the target water area are limited to enter the target water area and the ships in the target water area are reminded to pay attention to the encounter situation and adjust the heading and speed in time.

[0117] In summary, the derived index provided by the method supports the detection, strategy formulation and active prevention and control of maritime collision danger, effectively alleviating the collision danger in the target water area.

[0118] ​It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying and managing water collision hazards based on propagation dynamics, characterized in that, include: Receive latitude and longitude information of all ships in the target waters to form a set of ship spatial location information; Based on the spatial location information set of ships, the SEIRS propagation dynamics model is used to model the multi-ship encounter problem in the target waters, and a collision hazard propagation model for multi-ship encounters in the target waters is obtained. The propagation status of each vessel in the target waters is determined based on a collision hazard propagation model; By introducing propagation parameters and combining them with a collision hazard propagation model of multiple vessels encountering each other in the target waters, a differential equation for the collision hazard propagation system is constructed. The optimal propagation parameters are obtained by solving the differential equation of the collision hazard propagation system based on the propagation status of each vessel in the target water area. Optimal propagation parameter values ​​are used to establish derived indices to characterize collision hazards in the target water area; By using derived indicators to schedule vessels within the target waters, collision risk management can be achieved.

2. The method according to claim 1, characterized in that, The determination of the propagation status of each vessel in the target waters based on the collision hazard propagation model includes: The DBSCAN spatial clustering method is used to cluster the spatial positions of ships in the target waters, resulting in multiple clusters, each of which is a multi-ship encounter cluster. The space occupied by the multi-ship encounter cluster is the multi-ship encounter area; The entire spatial range formed by expanding outward by twice the radial direction from the multi-ship encounter cluster is the total area of ​​influence; the annular area between the total area of ​​influence and the multi-ship encounter cluster is the multi-ship encounter influence area. The area outside the multi-vehicle encounter zone and the multi-vehicle encounter influence zone in the target waters is the non-multi-vehicle encounter influence zone. Ships within the multi-ship encounter area were infected. Vessels within the area affected by multiple vessel encounters are considered exposed vessels; Vessels that leave the multi-vessel encounter area and enter the multi-vessel encounter influence area are considered to be recovering vessels; Vessels not located in areas affected by multiple vessel encounters are considered vulnerable vessels.

3. The method according to claim 1, characterized in that, The differential equation of the collision hazard propagation system: in, Indicates the number of susceptible vessels. Indicates the number of exposed ships. Indicates the number of infected ships. Indicates the number of ships restored; This represents the total number of ships, which is the sum of the number of ships in all four propagation states; propagation parameters include: For infection rate, For latent conversion rate, For recovery rate and For immune loss rate; System parameters: Indicates the entry rate parameter and This indicates the exit rate parameter.

4. The method according to claim 1, characterized in that, The process of solving the differential equations of the collision hazard propagation system to obtain the optimal propagation parameters includes: The differential equations of the collision hazard propagation system are solved by integration to generate a simulated sequence of propagation state variables; The simulated sequence results are compared with historical data results to find the optimal propagation parameter values ​​by using the constrained optimization-based fmincon algorithm to fit the propagation parameters that best match the propagation characteristics of the system.

5. The method according to claim 1, characterized in that, The method of establishing derived indices to characterize collision hazards in target waters using optimal propagation parameter values ​​includes: The basic reproduction number characterizes the ability to spread a hazard. The peak danger profile is characterized by the highest infection rate. The level of danger is characterized by the cumulative infection load.

6. The method according to claim 5, characterized in that, The basic reproduction number: in, This represents the basic reproduction number.

7. The method according to claim 5, characterized in that, The maximum infection rate: in, Indicates in The number of infected ships at any given time. Indicates in The total number of ships at any given time. For time tags, This indicates the last time stamp within the period during which the target water area was monitored. This indicates the maximum infection rate.

8. The method according to claim 5, characterized in that, The cumulative infection load: in, This indicates the total length of the time period during which the target water area is monitored. This indicates the cumulative infection load.

9. The method according to claim 5, characterized in that, Vessel scheduling within the target waters is based on derived indicators, including: When the basic reproduction number, maximum infection rate, or cumulative infection load exceeds a preset threshold, vessels outside the target waters are restricted from entering the target waters, and vessels within the target waters are reminded to be aware of the encounter situation and adjust their course and speed in a timely manner.

10. A device for identifying and managing water collision hazards based on propagation dynamics, characterized in that, include: Ship status determination unit, propagation parameter acquisition unit, derived index calculation unit, and ship scheduling unit; The vessel status determination unit receives the latitude and longitude information of all vessels in the target water area, forming a set of vessel spatial location information. Based on the spatial location information set of ships, the SEIRS propagation dynamics model is used to model the multi-ship encounter problem in the target waters, and the collision hazard propagation model of multi-ship encounter in the target waters is obtained; based on the collision hazard propagation model, the propagation state of each ship in the target waters is determined. The propagation parameter acquisition unit introduces propagation parameters and combines them with a collision hazard propagation model of multiple ships encountering each other in the target waters to construct a collision hazard propagation system differential equation; and solves the collision hazard propagation system differential equation based on the propagation state of each ship in the target waters to obtain the optimal propagation parameters. The derived index calculation unit uses the optimal propagation parameter value to establish a derived index characterizing the collision risk in the target water area; The vessel scheduling unit schedules vessels within the target waters based on derived indicators to achieve collision risk management within the target waters.

Images