Marine distress target search aid decision-making method, equipment and medium

Abstract

The invention provides an auxiliary decision-making method and device for searching a maritime distress target and a medium, and relates to the technical field of maritime affairs, and the method comprises the steps: obtaining the marine meteorological environment forecast information of a distress area and a nearby sea area, and determining a search sea area; calculating a sea area risk level and a search action emergency degree of the search sea area; obtaining a search resource; establishing a search resource decision model, and combining the search action emergency degree, the sea area risk level and the search resources to obtain an optimal search resource planning scheme; and dividing the search sea area into sub-areas according to the sea area risk level, obtaining a working sea area of the search power and determining an optimal global path plan of the search power in combination with the optimal search resource planning scheme. According to the scheme, the search action is balanced in three aspects of search speed, search cost and search safety, and the efficiency of the marine search action is improved.

Description

Technical Field

[0001] The present application relates to the field of maritime technology, and in particular to a method, device and medium for assisting decision-making in searching for targets in distress at sea. Background Art

[0002] With the rapid development of the marine economy, the frequency of marine accidents has also increased, and search operations for marine distress accidents have become an important part of marine safety. Generally speaking, the goal of search operation planning is often to shorten the search time as much as possible, keep the search resource cost as low as possible, and make the search safety as high as possible under the premise of successful rescue. For marine distress accidents of different urgency, the focus of search plan evaluation should also be different. However, when planning marine search operations, existing plans usually only consider the search time or search resource cost, and the factors considered are relatively single. When comparing search plans, there is a lack of a quantitative evaluation method. In general, search operation planning and resource scheduling for marine accidents should consider more factors, and existing plans cannot comprehensively plan marine search operations. Summary of the invention

[0003] The purpose of the present invention is to provide a method, device and medium for assisting decision-making in searching for targets in distress at sea in order to solve the problem that existing maritime search schemes cannot comprehensively plan maritime search operations.

[0004] The above-mentioned purpose of the present application is achieved through the following technical solutions: S1: Obtain the marine meteorological environment forecast information of the distress area and nearby sea areas, and determine the search area; S2: Calculate the sea risk level of the search area and the urgency of the search operation; S3: Acquire search resources; establish a search resource decision model, combine the urgency of the search operation, the risk level of the sea area and the search resources, and obtain the best search resource planning scheme based on search time, search cost and search safety; S4: Divide the search sea area into sub-areas according to the sea risk level, combine the best search resource planning scheme, obtain the working sea area of ​​the search force and determine the optimal global path planning of the search force.

[0005] An electronic device comprises a processor, a memory, a user interface and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes a method, device and medium for auxiliary decision-making for searching for distressed targets at sea.

[0006] A computer-readable storage medium stores instructions, and when the instructions are executed, a method, device and medium for auxiliary decision-making for searching for distressed targets at sea are executed.

[0007] The beneficial effects of the technical solution provided by this application are: 1. Comprehensively considering the risks of the search and rescue sea area and the urgency of the search and rescue operation, the search resource decision-making plan is reasonably planned based on the search time, search cost, and search safety. Specific algorithms are introduced to quickly allocate search and rescue resources to adapt to the dynamically changing environment. A comprehensive and integrated search operation planning system is built to improve the efficiency of maritime search operation plan planning and resource utilization.

[0008] 2. The present invention provides a method, device and medium for assisting decision-making in searching for targets in distress at sea. The method includes a search action risk and urgency assessment module, a search resource planning module and a search path planning module, and can provide a comprehensive planning scheme for searching for accidents in distress at sea. Among them, the variables used in the search action urgency module include factors such as the number of people in distress, the type of accident in distress, and sea temperature, and the urgency level is quantified by using the empirical evaluation method and the analytic hierarchy process; the search resource planning module aims to minimize the search action duration, minimize the search resource cost, and maximize the search action safety, establish a search resource planning model, change the search time, search cost, and search safety weight distribution in the search scheme planning model based on the search action urgency level, and use the index evaluation normalization process to evaluate the search scheme; the search path planning module includes a module for determining the overall search area and a global search method decision module. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present application will be further described below with reference to the accompanying drawings and embodiments, in which: Figure 1 is a step diagram in an embodiment of the present application; Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0010] In order to have a clearer understanding of the technical features, purposes and effects of the present application, the specific implementation methods of the present application are now described in detail with reference to the accompanying drawings.

[0011] Embodiments of the present application provide a method, device and medium for assisting decision-making in searching for a target in distress at sea.

[0012] Please refer to Figure 1 , Figure 1 This is a step diagram of a method, device and medium for assisting decision-making in searching for distressed targets at sea in an embodiment of the present application, including: S1: Obtain the marine meteorological environment forecast information of the distress area and nearby sea areas, and determine the search area; S2: Calculate the sea risk level of the search area and the urgency of the search operation; S3: Acquire search resources; establish a search resource decision model, combine the urgency of the search operation, the risk level of the sea area and the search resources, and obtain the best search resource planning scheme based on search time, search cost and search safety; S4: Divide the search sea area into sub-areas according to the sea risk level, combine the best search resource planning scheme, obtain the working sea area of ​​the search force and determine the optimal global path planning of the search force.

[0013] Step S1 includes: S11: Establish a wind-wave-current coupling numerical prediction model to obtain the marine meteorological environment forecast information of the distress area and nearby sea areas through the wind-wave-current coupling numerical prediction model; The wind-wave-current coupled numerical prediction model includes: atmosphere module, ocean wave module and ocean current module; As an embodiment, the wind-wave-current coupling numerical prediction model is composed of an atmospheric module, an ocean wave module and an ocean current module. The model comprehensively considers the atmospheric, ocean wave and ocean current modules to simulate the complex dynamic processes in the marine environment. The atmospheric module predicts the wind field (wind speed and direction), provides a power source on the ocean surface, and provides a driving force for the evolution of waves and currents; the ocean wave module calculates the generation, propagation, development and breaking process of ocean waves, and couples with the wind field and ocean currents; the ocean current module simulates the tides and ocean current fields in the ocean, and considers the changes in ocean currents caused by factors such as wind and temperature and salinity. The wind-wave-current coupling numerical prediction model is constructed based on WRF (Weather Research and Forecasting) -FVCOM (Finite-Volume Coastal Ocean Model) -SWAVE (Simulating WAVEs), and a deep learning correction model can be further used to optimize the forecast results.

[0014] As an embodiment, the WRF-FVCOM-SWAVE coupling model is a coupling system that integrates meteorological, oceanic and wave models. WRF (Weather Research and Forecasting Model) is a high-resolution numerical weather forecast model, mainly used for meteorological forecasting and climate research. It can simulate various physical processes in the atmosphere, such as temperature, humidity, wind speed, etc. FVCOM (Finite Volume Ocean Model) is a finite volume ocean model suitable for ocean flow simulation in complex terrain. It can consider the influence of tides, wind and other factors on ocean flow. SWAVE (Nearshore Wave Model) is a wave model used to simulate the area from deep sea to nearshore, which can simulate the generation, propagation and attenuation of waves. The WRF-FVCOM-SWAVE coupling model realizes the linkage simulation of wind, waves and currents through data interaction between modules. The WRF model provides atmospheric boundary conditions, such as wind speed, temperature and air pressure, which are passed to the FVCOM and SWAVE models. The sea surface temperature and ocean flow information calculated by the FVCOM model are fed back to WRF to affect the simulation of the atmosphere. The wave elements calculated by the SWAVE model are passed to the FVCOM model, which uses them to calculate the ocean currents under the influence of radiation stress and sea surface stress; then, the FVCOM model calculates the ocean currents and water surface elevation results and passes them to the SWAVE model for wave calculations in the next time step. WRF, FVCOM, and SWAVE achieve two-way nested dynamic coupling through shared boundaries and state variables (wind field, flow field, wave stress), which can efficiently simulate the interaction process of wind, waves, and currents, and provide reliable marine meteorological environment forecast information.

[0015] S12: Build a deep learning correction model; obtain historical actual marine meteorological environment observation data and historical marine meteorological environment forecast information; The deep learning correction model is trained through historical marine meteorological environment forecast information and historical actual marine meteorological environment observation data; Correct the marine meteorological environment forecast information through the trained deep learning correction model; As an embodiment, the deep learning correction model optimizes the forecast results through the deep learning model to further improve the model accuracy. This process uses the errors generated by the existing observation data and the numerical model to correct the forecast output of the model so that the forecast results are closer to the actual observations. The error between the historical numerical forecast data and the observation data is used as the target variable, the wind-wave-current parameters are used as the input, and the input (wind-wave-current parameters) is combined with the target (model error) to train a deep learning model that can predict and correct the errors of the wind-wave-current coupled numerical forecast model.

[0016] As an embodiment, a deep learning correction model is trained based on sample data; the sample data includes sample input data and sample output data, the sample input data includes sample flow characteristic parameters, sample wind characteristic parameters, and sample wave characteristic parameters, the sample output data includes sample flow characteristic parameter errors, sample wind characteristic parameter errors, and sample wave characteristic parameter errors, and the sample errors are obtained by subtracting the corresponding sample output data from the actual measured parameters of the sample.

[0017] S13: Based on the prior information of the distress target at sea and the corrected marine meteorological environment forecast information, a deep learning model of the initial distribution of the distress target at sea based on the convolutional long short-term memory network algorithm is constructed and trained; the prior information of the distress includes: the planned navigation route of the distressed ship at sea, the basic information of the distressed ship at sea, the basic information of the people who fell into the water at sea, and the feedback information of the passing ships; Obtain real-time prior information on distress; predict the real-time prior information on distress through the deep learning model of initial distribution of distress targets at sea, and obtain the initial distribution set of targets; As an embodiment, prior information on distress refers to prior knowledge for predicting a distress event based on existing data and experience before the distress event occurs, and also includes the planned sailing route of a ship in distress at sea, basic information of a ship in distress at sea, basic information of people who fall into the water at sea, feedback information from passing ships, etc.

[0018] As an embodiment, the deep learning model of the initial distribution of targets in distress at sea is a prediction model based on historical data and deep learning algorithms, which is used to predict the possible location distribution of targets in distress. This method trains the model by inputting a large amount of historical distress event data (including the last reported location, the last reported time, the type of distress accident, the number of people in distress, etc.) and prior distress information into the model, and optimizes the model parameters by minimizing the error between the predicted position and the actual initial distribution position. The cross-validation method is used to evaluate the generalization ability of the model to ensure its performance on new data. The trained model can be used for actual prediction, giving the initial distribution under specific conditions, and helping the search operation to determine the potential initial distribution of targets. The output initial distribution is visualized as a probability density map, showing the potential risks in different areas, which is used for initial positioning in the search operation. Among them: the deep learning model of the initial distribution of targets in distress at sea is trained based on the initial distribution sample data of targets in distress at sea; the initial distribution sample data of targets in distress at sea includes distribution sample input data and distribution sample output data, the distribution sample input data is historical distress event data and prior distress information, including the last reported position, the last reported time, the type of distress accident, and the number of people in distress, and the distribution sample output data includes the initial distribution position of targets in distress at sea.

[0019] S14: Based on wind-induced drift, flow-induced drift and wave-induced drift, a drift trajectory prediction model for targets in distress at sea is constructed, and based on the initial distribution set of the target, the drift trajectory is predicted to obtain the final distribution set; As an embodiment, the drift trajectory prediction model of the target in distress at sea is used to predict the drift path of the target in distress in the marine environment. The model predicts the movement trajectory of the target in a specific time period based on dynamic factors such as ocean currents, wind speed, waves, and the characteristics of the target itself, thereby helping search personnel to more accurately determine the current and future positions of the target and improve the search efficiency. Using the Lagrangian drift model, factors such as ocean currents, wind, and waves are converted into the motion equation of the target to derive its possible drift trajectory. Wind-induced drift refers to considering the drift effect of wind force on the target. The effect of wind will cause the change of drift speed, and the modeling is performed by considering wind speed and wind direction. Flow-induced drift refers to considering the influence of the direction and speed of the ocean current on the drift path, obtaining ocean current data from the wind-wave-current coupling model, and calculating the actual movement of the target in the flow environment. Wave-induced drift refers to considering the periodic pressure exerted by waves on the drifting target, obtaining ocean wave data from the wind-wave-current coupling model, and calculating the influence of ocean waves on the position of the target in distress.

[0020] S15: Based on the final distribution set and in combination with the convex hull algorithm, determine the search area.

[0021] As an embodiment, the final distribution set refers to the set of final positions or states of distressed targets (such as floating objects, ships or personnel) in the ocean generated by the initial distribution deep learning model of distressed targets at sea and the drift trajectory prediction model of distressed targets at sea. The convex hull algorithm is used to calculate the convex hull of a set of points, that is, the smallest convex polygon containing all predicted final distribution points.

[0022] As an embodiment, the predicted distribution points are , , .according to Coordinates sort the point set if The coordinates are the same, then according to Sort the coordinates. Traverse the sorted point set from left to right. For each point, check whether the current point, the previous point, and the second-to-last point form a right turn (i.e., the three points are collinear or turned outward). If a right turn is formed, remove the second-to-last point; if a right turn is not formed, add the current point to the lower convex hull and gradually build the convex hull of the lower half. After the convex hull of the lower half is built, traverse the sorted point set in reverse, repeat the construction process of the lower convex hull, and build the upper convex hull. Finally, merge the lower convex hull and the upper convex hull to form a complete convex hull, and use the convex hull as the search area.

[0023] Step S2 includes: S21: Obtaining distress information of the target in distress at sea; the distress information includes: the last reported position, the last reported time, the type of distress accident, the number of people in distress and the sea temperature; S22: Based on the distress information, combined with the empirical evaluation method and the analytic hierarchy process, the urgency of the search operation is quantitatively evaluated to obtain the urgency of the search operation; S23: Obtain marine meteorological environmental data, and calculate the marine risk index by combining the empirical evaluation method and the analytic hierarchy process; the marine meteorological environmental data include: sea surface wind speed, wave height, sea visibility and current speed; S24: Based on the K-means clustering method and combined with the sea area risk index, the search sea area is divided into different sub-sea areas according to the risk level to obtain the sea area risk level of the search sea area; Risk levels include: extreme risk, high risk, medium risk, low risk and no risk.

[0024] As an embodiment, the comprehensive hierarchical analysis method is a multiple-level analysis model for different types of marine ships, taking into account the comprehensive influence of single or multiple factors of marine meteorological environmental data such as sea surface wind speed, wave height, sea visibility, current speed, etc. When calculating the sea area risk index, multiple hierarchical analysis models are used to calculate the search sea area risk index respectively, and the search sea area risk level is determined according to the highest risk level principle; Among them, the K-means clustering method clusters each point in the search sea area according to the risk index, determines the center of each cluster, and divides the sea areas with similar risk levels into the same category, thereby realizing a graded assessment of the sea area risk level.

[0025] Step S22 includes: S221: The type of distress incident, number of people in distress, and sea temperature of the distress information are set as evaluation indicators, that is, ; Set the evaluation level set to ={extreme urgency, high urgency, moderate urgency, slight urgency}; Table 1 Classification of urgency of search operations

[0026] S222: Use the empirical evaluation method to construct an urgency evaluation matrix; Set evaluation indicators The degree of membership of each element in the evaluation set V , Indicates the evaluation index The evaluation result of the kth level is The proportion of the number of people who have been evaluated to the total number of people who have been evaluated, that is, the evaluation index The kth level has the comment the extent of; Depend on Construct a single evaluation index evaluation set :

[0027] in Indicates The parameters are divided into m levels, Characterization The probability of the kth urgency level; ; ; S223: Use the analytic hierarchy process to determine the weight set of evaluation indicators ; Evaluate indicators for urgency , construct the judgment matrix as follows:

[0028] Among them, the judgment matrix has , , The nature of; elements Indicates the impact factor on the urgency of the search operation and the relative importance of As an example, The value of adopts the 1~9 scale method; the specific scale value and meaning are shown in Table 2.

[0029] Table 2 Scale values ​​and meanings of judgment matrix elements

[0030] Calculate the product of each row of the judgment matrix , ; calculate The cube root of , ; Vector Normalized to , ; Calculate the largest eigenvalue, ; Use consistency indicators , compare CI with the average random consistency index RI to test whether the judgment matrix has satisfactory consistency and obtain the consistency test result; As an embodiment, for matrices of order 1 to 9, RIs are shown in Table 3 respectively.

[0031] Table 3 Average random consistency index of 1-9 order matrices

[0032] According to the results of the consistency test, determine the weight set of evaluation indicators , represents the importance ranking weight of the three evaluation indicators of sea temperature, type of distress, and number of people in distress to the urgency of the search operation; S224: Perform comprehensive evaluation based on the weight set of evaluation indicators to obtain the search urgency probability; Search Urgency Probability Matrix ,in In planning this search operation, The probability of occurrence of each urgency level; represents a single evaluation metric evaluation set; S225: Calculate the urgency of the search operation based on the search urgency probability; Set a risk level ratio threshold L. If the ratio of the largest element to the second largest element in the search urgency probability matrix A is greater than the threshold L, the urgency corresponding to the largest element in the search urgency probability matrix A is used as the final search action urgency. If the ratio of the largest element to the second largest element in the search urgency probability matrix A is less than the threshold L, the final search action urgency is determined by the median of each urgency in A.

[0033] Step S23 includes: S231: Set the marine meteorological environment data as the evaluation index, that is, ; Set the risk assessment level set to ; Table 4 Classification of search force navigation search risk assessment levels

[0034] S232: Repeat steps S222 to S225 to obtain the marine risk index, as follows: Adopting the empirical evaluation method, construct the urgency evaluation matrix; The analytic hierarchy process is used, combined with the urgency evaluation matrix, to determine the weight set of evaluation indicators; A comprehensive evaluation is performed based on the weight set of evaluation indicators to obtain the search risk level probability of all grid points in the search area; By searching the risk level probability, the risk index of all grid points in the search sea area is calculated, that is, the sea area risk index.

[0035] Step S24 includes: S241: Determine the number of clusters required by combining the elbow rule with the search area risk index ; As an example, the number of clusters It depends on the size of the search area. The value range is usually 1~10.

[0036] For each Value, Execution Mean clustering, calculate the risk index for each grid point The corresponding cluster center risk index The sum of squares of the difference As the cost function for clustering; The sum of the squares of the calculated differences With the corresponding The values ​​are plotted as a curve with the horizontal axis being Value, vertical axis is value; As an example, in the drawn graph, find “ The value decreases at a significantly slower rate, which usually takes on an "elbow" shape. The value is the optimal number of clusters.

[0037] Select the K value where the curve in the curve graph has an inflection point to determine the number of clusters; S242: Using random sampling, select data points as the initial cluster centers; As an example, the random sampling method ensures that each individual has an equal probability of being selected, and the selection of each individual in the sample is independent of each other and is not affected by other individuals. The random number generator is used to ensure the randomness of the sample selection process.

[0038] S243: Based on each data point The risk index of ) and the risk index of each cluster center ( ) , assign each data point to the nearest cluster center; S244: For each cluster , calculate all points The average risk index , update the cluster center value;

[0039]

[0040] S245: Repeat the process of step S243 and step S244 until the cluster center value Terminate the iteration unchanged; Based on clusters, the search area is divided into sub-areas with different risk levels; S246: In accordance with the highest risk level principle, the highest risk level in each sub-sea area shall be used as the sea area risk level of the search area.

[0041] As an embodiment, based on the marine meteorological environmental data such as sea surface wind speed, wave height, sea visibility, current speed, etc., the comprehensive hierarchical analysis method is used to comprehensively calculate the sea area risk index, and the search sea area is divided into different sub-sea areas according to the risk level based on the K-means clustering method. Among them, the comprehensive hierarchical analysis method is a multiple hierarchical analysis model for different types of marine ships, taking into account the comprehensive influence of single or multiple factors of marine meteorological environmental data such as sea surface wind speed, wave height, sea visibility, current speed, etc. When calculating the sea area risk index, multiple hierarchical analysis models are used to calculate the search sea area risk index respectively, and the search sea area risk level is determined according to the principle of the highest risk level; among them, the K-means clustering method clusters each point in the search sea area according to the risk index, determines the center of each cluster, and divides the sea areas with similar risk levels into the same category, thereby realizing the hierarchical assessment of the sea area risk level.

[0042] Step S3 includes: S31: Determine search resources; S311: Obtain the anti-risk levels of all available search aircraft and search ships around the search area; S312: Screening search aircraft and search ships based on the sea area risk level and anti-risk level; S313: Calculate the round trip time between the search aircraft base and the search area as follows:

[0043] in To search for the round trip time of the aircraft, is the distance between the flight base and the search area. The maximum speed of the search aircraft; is the impact coefficient of risk level, Indicates the risk level of strong wind warning during the round trip of the aircraft; Get the maximum flight time of the search aircraft; Eliminate search aircraft whose maximum endurance time is less than the round-trip time, conduct a secondary screening of search aircraft, and retain the final available search aircraft; S314: The search aircraft and search ships that are finally retained are the search resources finally determined; S32: Establishing a search resource decision model; obtaining a comprehensive index of the decision plan by combining the search resource decision model with the search resources finally determined; The search resource decision model is a multi-objective integer nonlinear programming model with the goals of minimizing the search action duration, minimizing the search resource cost, and maximizing the search action safety, which can be expressed as

[0044] in represents the comprehensive index of decision options, To search for time, is the search cost, In search of safety, , , They are respectively , , The corresponding weight is calculated based on the urgency of the search operation; , ,as well as Determined by the following function:

[0045] in, Indicates the total available amount of searched vessels after initial screening, Indicates the total available number of search aircraft after initial screening, Indicates Search ships, , Indicates Search plane, ; Indicates Search status of search vessels, , When Search ships participated in the search operation. When Search vessels do not participate in the search operation. Indicates The search status of the search aircraft, , When Search aircraft participated in the search operation. When The search aircraft did not participate in the search operation. Indicates The time required for a ship to sail at full speed to the sea area to be searched, Indicates The search capability of the ship Indicates The search capability of the aircraft, Indicates the area of ​​the search area. Indicates The time required for an aircraft to travel between the flight base and the sea area to be searched. Indicates The maximum flight time of an aircraft, Indicates The number of sorties dispatched by search aircraft during the entire search operation. Indicates The dispatching cost of a ship, Indicates The cost of dispatching an aircraft, Indicates Fuel consumption rate of a ship, Indicates The fuel consumption rate of the aircraft, Indicates The unit price of the fuel used by the ship, Indicates The unit price of fuel used by an aircraft, Indicates The risk resistance level of the ship, For the The risk level of the aircraft, To determine the risk level of the search area, No. The distance between the aircraft and the sea area to be searched, Indicates the risk level of the round trip. Indicates The maximum speed of the aircraft, is the impact coefficient of risk level, The upper limit of the number of search vessels that can be accommodated in the sea area to be searched. To search for an upper limit on the number of aircraft; , , The allocation method is determined by the urgency of the search operation and is as follows: When the search operation is extremely urgent, , , They are 0.9, 0.05, and 0.05 respectively; When the search operation is of high urgency, , , They are 0.8, 0.1, and 0.1 respectively; When the search operation is of medium urgency, , , They are 0.7, 0.15, and 0.15 respectively; When the search operation is of low urgency, , , They are 0.6, 0.2, and 0.2 respectively; S33: The search resource planning scheme with the smallest comprehensive index of the decision schemes output by the search resource decision model is determined as the best search resource planning scheme.

[0046] As an embodiment, the search resource planning scheme derived from the search resource decision model specifically includes the number of search aircraft and the number of search ships. The comprehensive index of the search resource planning scheme is calculated based on the search resource decision model, and the search resource planning scheme with the smallest comprehensive index of the decision scheme is determined as the optimal search resource planning scheme.

[0047] Step S4 includes: S41: Determine basic information of the search force based on the optimal search resource planning scheme and the divided sub-areas; the basic information of the search force includes: risk level of the search sub-area, area of ​​the search sub-area, risk resistance level of the search force, and maximum workload of the search force; As an embodiment, the areas of sub-areas of different sea risk levels are calculated; based on the optimal search resource planning scheme, the maximum workload that the search force can complete within the search time is calculated, which is specifically expressed as the product of the search capability of the search force and the working time.

[0048] S42: allocating the work layout of the search force based on the greedy algorithm, including: the greedy algorithm allocates the search force in descending order according to the matching priority of the anti-risk level of the search force and the risk level of the search sub-area; S421: Set input data sub-area set ,in Indicates Search sub-area risk level, Indicates Search sub-area area; input data search force set ,in Indicates Search force risk resistance level, Indicates Maximum workload of a search force; S422: Search by sub-area risk level Sort descending ; According to the risk level of search force Sort descending ; S423: Initialize search resource allocation result set , traverse the search force set , for the current search force , find the first Sub-area ; like , then the allocated area to the sub-region and update ; like , then the allocated area Give the search power and update ; By traversing the search force set , output the search resource allocation result set , i.e. the assigned area of ​​each search force and the corresponding area, i.e. the working sea area of ​​the search force; S43: Based on the search resource allocation result set , perform optimal global path planning for the search force, and obtain the optimal global path planning for the search force.

[0049] As an embodiment, the global search method mainly includes fan-shaped search, extended square search, and parallel line search, which are specifically selected according to the following rules: The sector search method is mainly applicable when the search target is located accurately or the search area is small. For searching ships, the sector search radius is usually 2 to 5 nautical miles. If the target is not found after completing a sector search, rotate the sector and move down half of the first search radius for a second sector search.

[0050] The extended square search is suitable for situations when the search target location is in a relatively close area. In this method, the search starting point is always the reference position, and the search action expands outward in concentric squares, thereby basically evenly covering the area centered on the reference point. However, if the reference is not a point but a short line, it should be changed to an outward-expanding rectangle. The length of the two search routes is equal to the search line spacing, and the length of each two search segments will increase by one search line spacing on the original basis. If continuous searches are conducted in the same area, the search line is generally turned 45° to continue the search.

[0051] Parallel line search is suitable for situations where the search target location is very uncertain and requires uniform coverage of a wide area. First, set a search area and determine the search line spacing based on the on-site situation. The search facility uses a corner of the search area as the search starting point. The search starting point is usually located within the search rectangle at a distance of 1 / 2 of the search line spacing from each of the two right-angled sides. Then, the search is continued back and forth along the long side of the rectangle while maintaining the spacing.

[0052] The present application also discloses an electronic device. Figure 2 , Figure 2 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0053] The communication bus 502 is used to realize the connection and communication between these components.

[0054] The user interface 503 may include a display screen, and the optional user interface 503 may also include a standard wired interface or a wireless interface.

[0055] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0056] The present application also discloses a computer-readable storage medium storing a plurality of instructions suitable for loading by a processor to execute the above-mentioned method, device and medium for auxiliary decision-making for searching for distressed targets at sea.

[0057] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure.

[0058] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method, device and medium for assisting decision-making in searching for a target in distress at sea, characterized in that: The method comprises the following steps: S1: Obtain the marine meteorological environment forecast information of the distress area and nearby sea areas, and determine the search area; S2: Calculate the sea risk level of the search area and the urgency of the search operation; S3: Acquire search resources; establish a search resource decision model, combine the urgency of the search operation, the risk level of the sea area and the search resources to obtain the best search resource planning plan; S4: Divide the search sea area into sub-areas according to the sea risk level, combine the best search resource planning scheme, obtain the working sea area of ​​the search force and determine the optimal global path planning of the search force.

2. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 1, characterized in that: Step S1 includes: S11: Establish a wind-wave-current coupling numerical prediction model to obtain the marine meteorological environment forecast information of the distress area and nearby sea areas through the wind-wave-current coupling numerical prediction model; The wind-wave-current coupled numerical prediction model includes: atmosphere module, ocean wave module and ocean current module; S12: Build a deep learning correction model; obtain historical actual marine meteorological environment observation data and historical marine meteorological environment forecast information; The deep learning correction model is trained through historical marine meteorological environment forecast information and historical actual marine meteorological environment observation data; Correct the marine meteorological environment forecast information through the trained deep learning correction model; S13: Based on the prior information of the distress target at sea and the corrected marine meteorological environment forecast information, a deep learning model of the initial distribution of the distress target at sea based on the convolutional long short-term memory network algorithm is constructed and trained; the prior information of the distress includes: the planned navigation route of the distressed ship at sea, the basic information of the distressed ship at sea, the basic information of the people who fell into the water at sea, and the feedback information of the passing ships; Obtain real-time prior information on distress; predict the real-time prior information on distress through the deep learning model of initial distribution of distress targets at sea, and obtain the initial distribution set of targets; S14: Based on wind-induced drift, flow-induced drift and wave-induced drift, a drift trajectory prediction model for targets in distress at sea is constructed, and based on the initial distribution set of the target, the drift trajectory is predicted to obtain the final distribution set; S15: Based on the final distribution set and in combination with the convex hull algorithm, determine the search area.

3. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 1, characterized in that: Step S2 includes: S21: Obtaining distress information of the target in distress at sea; the distress information includes: the last reported position, the last reported time, the type of distress accident, the number of people in distress and the sea temperature; S22: Based on the distress information, combined with the empirical evaluation method and the analytic hierarchy process, the urgency of the search operation is quantitatively evaluated to obtain the urgency of the search operation; S23: Obtain marine meteorological environmental data, and calculate the marine risk index by combining the empirical evaluation method and the analytic hierarchy process; the marine meteorological environmental data include: sea surface wind speed, wave height, sea visibility and current speed; S24: Based on the K-means clustering method and combined with the sea area risk index, the search sea area is divided into different sub-sea areas according to the risk level to obtain the sea area risk level of the search sea area; Risk levels include: extreme risk, high risk, medium risk, low risk and no risk.

4. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 3, characterized in that: Step S22 includes: S221: The type of distress incident, number of people in distress, and sea temperature of the distress information are set as evaluation indicators, that is, ; Set the evaluation level set to ={extreme urgency, high urgency, moderate urgency, slight urgency}; S222: Use the empirical evaluation method to construct an urgency evaluation matrix; Set evaluation indicators The degree of membership of each element in the evaluation set V , Indicates the evaluation index The evaluation result of the kth level is The proportion of the number of people who have been evaluated to the total number of people who have been evaluated, that is, the evaluation index The kth level has the comment the extent of; Depend on Construct a single evaluation index evaluation set : in Indicates The parameters are divided into m levels, Characterization The probability of the kth urgency level; ; ; S223: Use the analytic hierarchy process to determine the weight set of evaluation indicators ; Evaluate indicators for urgency , construct the judgment matrix as follows: Among them, the judgment matrix has , , The nature of; elements Indicates the impact factor on the urgency of the search operation and the relative importance of Calculate the product of each row of the judgment matrix , ; calculate The cube root of , ; Vector Normalized to , ; Calculate the largest eigenvalue, ; Use consistency indicators , compare CI with the average random consistency index RI to test whether the judgment matrix has satisfactory consistency and obtain the consistency test result; According to the results of the consistency test, determine the weight set of evaluation indicators , represents the importance ranking weight of the three evaluation indicators of sea temperature, type of distress, and number of people in distress to the urgency of the search operation; S224: Perform comprehensive evaluation based on the weight set of evaluation indicators to obtain the search urgency probability; Search Urgency Probability Matrix ,in In planning this search operation, The probability of occurrence of each urgency level; Represents a single evaluation metric evaluation set; S225: Calculate the urgency of the search operation based on the search urgency probability; Set a risk level ratio threshold L. If the ratio of the largest element to the second largest element in the search urgency probability matrix A is greater than the threshold L, the urgency corresponding to the largest element in the search urgency probability matrix A is used as the final search action urgency. If the ratio of the largest element to the second largest element in the search urgency probability matrix A is less than the threshold L, the final search action urgency is determined by the median of each urgency in A.

5. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 4, characterized in that: Step S23 includes: S231: Set the marine meteorological environment data as the evaluation index, that is, ; Set the risk assessment level set to ; S232: Repeat steps S222 to S225 to obtain the marine risk index, as follows: Adopting the empirical evaluation method, construct the urgency evaluation matrix; The analytic hierarchy process is used, combined with the urgency evaluation matrix, to determine the weight set of evaluation indicators; A comprehensive evaluation is performed based on the weight set of evaluation indicators to obtain the search risk level probability of all grid points in the search area; By searching the risk level probability, the risk index of all grid points in the search sea area is calculated, that is, the sea area risk index.

6. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 3, characterized in that: Step S24 includes: S241: Determine the number of clusters required by combining the elbow rule with the search area risk index ; For each Value, Execution Mean clustering, calculate the risk index for each grid point The corresponding cluster center risk index The sum of squares of the difference As the cost function for clustering; The sum of the squares of the calculated differences With the corresponding The values ​​are plotted as a curve with the horizontal axis being Value, vertical axis is value; Select the K value where the curve in the curve graph has an inflection point to determine the number of clusters; S242: Using random sampling, select data points as the initial cluster centers; S243: Based on each data point Risk Index The risk index of each cluster center The difference , assign each data point to the nearest cluster center; S244: For each cluster , calculate all points The average risk index , update the cluster center value; S245: Repeat the process of step S243 and step S244 until the cluster center value Terminate the iteration unchanged; Based on clusters, the search area is divided into sub-areas with different risk levels; S246: In accordance with the highest risk level principle, the highest risk level in each sub-sea area shall be used as the sea area risk level of the search area.

7. A method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 1, characterized in that: Step S3 includes: S31: Determine search resources; S311: Obtain the anti-risk levels of all available search aircraft and search ships around the search area; S312: Screening search aircraft and search ships based on the sea area risk level and anti-risk level; S313: Calculate the round trip time of the search aircraft between the aircraft base and the search sea area; Get the maximum flight time of the search aircraft; Eliminate search aircraft whose maximum endurance time is less than the round-trip time, conduct a secondary screening of search aircraft, and retain the final available search aircraft; S314: The search aircraft and search ships that are finally retained are the search resources finally determined; S32: Establishing a search resource decision model; obtaining a comprehensive index of the decision plan by combining the search resource decision model with the search resources finally determined; The search resource decision model is a multi-objective integer nonlinear programming model with the goals of minimizing the search action duration, minimizing the search resource cost, and maximizing the search action safety, which can be expressed as in represents the comprehensive index of decision options, To search for time, is the search cost, In search of safety, , , They are respectively , , The corresponding weight is calculated based on the urgency of the search operation; S33: The search resource planning scheme with the smallest comprehensive index of the decision schemes output by the search resource decision model is determined as the best search resource planning scheme.

8. The method, device and medium for assisting decision-making in searching for a target in distress at sea as claimed in claim 1, characterized in that: Step S4 includes: S41: Determine basic information of the search force based on the optimal search resource planning scheme and the divided sub-areas; the basic information of the search force includes: risk level of the search sub-area, area of ​​the search sub-area, risk resistance level of the search force, and maximum workload of the search force; S42: allocating the work layout of the search force based on the greedy algorithm, including: the greedy algorithm allocates the search force in descending order according to the matching priority of the anti-risk level of the search force and the risk level of the search sub-area; S421: Set input data sub-area set ,in Indicates Search sub-area risk level, Indicates Search sub-area area; input data search force set ,in Indicates Search force risk resistance level, Indicates Maximum workload of a search force; S422: Search by sub-area risk level Sort descending ; According to the search force risk level Sort descending ; S423: Initialize search resource allocation result set , traverse the search force set , for the current search force , find the first Sub-area ; like , then the allocated area to the sub-region and update ; like , then the allocated area Give the search power and update ; By traversing the search force set , output the search resource allocation result set , i.e. the assigned area of ​​each search force and the corresponding area, i.e. the working sea area of ​​the search force; S43: Based on the search resource allocation result set , perform optimal global path planning for the search force, and obtain the optimal global path planning for the search force.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed by a computer, the method according to any one of claims 1 to 8 is executed.

Images