Fishing boat dynamic drift prediction method and system based on South China Sea water area distress analysis

Abstract

The invention relates to the technical field of fishing boat navigation positioning, in particular to a fishing boat dynamic drift prediction method and system based on South China Sea water area distress analysis. Constructing a dynamic drift prediction model of the target fishing boat and a future meteorological environment field model of the South China Sea water area, stripping a first future meteorological environment sub-simulation field of the current target fishing boat in the first static distress environment area through the future meteorological environment field model, and performing dynamic prediction in the dynamic drift prediction model, obtaining an initial dynamic drift route; and planning a second static distress environment area formed by dynamic drift of the current target fishing boat through the initial dynamic drift route to analyze a vector disturbance index of the meteorological environment, adjusting a second future meteorological environment sub-simulation field according to the vector disturbance index, and performing dynamic prediction again to obtain a final dynamic drift route. According to the method, the dynamic drifting route and coordinates of the fishing boat under the influence of the meteorological environment when the fishing boat is in danger in the South China Sea can be accurately predicted, so that the search and rescue efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fishing vessel navigation positioning in distress, and in particular to a method and system for predicting the dynamic drift of a fishing vessel based on distress analysis in the South China Sea. Background Art

[0002] The South China Sea waters usually refer to the South China Sea in southern China, which is an important marine area in my country. The geographical location of the South China Sea waters is very important, but the vast area of ​​the South China Sea, complex marine conditions and frequent maritime traffic make fishing boats often affected by extreme weather such as typhoons, storms, and turbulent ocean currents, which makes fishing boats fishing in the South China Sea waters suffer from distress such as ship failure, loss of control, distress or sinking, which is prone to casualties. The complex meteorological environment changes will cause fishing boats to drift dynamically. Therefore, in order to ensure timely rescue, it is necessary to accurately obtain the dynamic drift trajectory of fishing boats. However, the dynamic drift prediction method of traditional fishing boats in distress can generally only meet the scene of stable meteorological environment, and the performance of analyzing meteorological environment data that changes all the time is poor. Therefore, the results of dynamic drift prediction have large errors, and the traditional dynamic drift prediction method is difficult to further predict and analyze the impact of the geological and topographic distribution of the South China Sea waters on the distress of fishing boats, which reduces the accuracy of dynamic drift prediction, reduces the quality of rescue plan formulation, and reduces the rescue efficiency of fishing boats in distress. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method and system for predicting the dynamic drift of a fishing vessel based on distress analysis in the South China Sea.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention provides a method for predicting the dynamic drift of a fishing vessel based on distress analysis in the South China Sea, comprising the following steps:

[0006] Construct a historical dynamic drift model of the target fishing boat when it is in distress in the waters of the South China Sea, and analyze the drift vector of the target fishing boat on a multidimensional geographic coordinate map based on the historical dynamic drift model to obtain a set of historical drift vectors of the target fishing boat in each sub-dynamic drift area;

[0007] Acquire several historical meteorological environment data of the target fishing boat, introduce the FCM algorithm to cluster the several historical meteorological environment data into a historical drift vector set, generate a fuzzy clustering result, adjust the state transition model in the historical dynamic drift model based on the fuzzy clustering result, and obtain a dynamic drift prediction model of the target fishing boat;

[0008] Constructing a future meteorological environment simulation field for the South China Sea waters area, planning a first dynamic distress environment area according to the distress coordinate position of the current target fishing boat, stripping a first future meteorological environment sub-simulation field in the first dynamic distress environment area through the future meteorological environment field model, and performing dynamic prediction in a dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift trajectory;

[0009] The second dynamic distress environment area formed by the dynamic drift of the current target fishing boat is planned through the initial dynamic drift route, the vector disturbance index of the meteorological environment is analyzed through the geological and topographic characteristic vector distribution map of the second dynamic distress environment area, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift trajectory and the corresponding coordinate values.

[0010] Furthermore, in a preferred embodiment of the present invention, the historical dynamic drift model of the target fishing boat when it is in distress in the waters of the South China Sea is constructed, and the drift vector of the target fishing boat is analyzed on a multidimensional geographic coordinate map according to the historical dynamic drift model to obtain a historical drift vector set of the target fishing boat in each sub-dynamic drift area, specifically comprising the following steps:

[0011] Obtaining a navigation log of the target fishing vessel in the waters of the South China Sea, and extracting a number of historical scheduled route coordinates and a number of actual route coordinates of the target fishing vessel in distress sequence through the navigation log;

[0012] A particle filter algorithm is introduced, and a state transition model and an observation model of the drift of the target fishing boat are constructed in the particle filter algorithm based on a number of historical scheduled route coordinates and a number of actual route coordinates, and a group of particles are initialized according to the state transition model and the observation model;

[0013] The weight of each particle is updated according to the coordinate deviation between the historical scheduled route coordinates and the actual route coordinates and the observation model, and new particles are resampled with the new weights. The above steps are repeated to continuously recursively update the weights and states of the particles until all particles are updated, thus generating a historical dynamic drift model of the target fishing vessel.

[0014] Obtain a multidimensional geographic coordinate map of the South China Sea waters area, construct a multidimensional space coordinate domain based on the multidimensional geographic coordinate map, and dynamically fit the historical scheduled route and the actual route in the multidimensional space coordinate domain based on the historical dynamic drift model to extract the dynamic drift area formed between the historical scheduled route and the actual route;

[0015] The scheduled sailing time of the target fishing vessel is obtained through the sailing log, the dynamic drift area is divided into N sub-dynamic drift areas based on the scheduled sailing time, and all drift coordinates in each sub-dynamic drift area and the distribution pattern of each drift coordinate are obtained;

[0016] The drift changes of the target fishing boat in the dynamic drift area are analyzed by the changes and distribution patterns of the drift coordinates, and the historical drift vector sets of the target fishing boat in each sub-dynamic drift area are obtained.

[0017] Furthermore, in a preferred embodiment of the present invention, the method of obtaining a plurality of historical meteorological environment data of the target fishing boat, introducing the FCM algorithm to cluster the plurality of historical meteorological environment data into a historical drift vector set, generating a fuzzy clustering result, and adjusting the state transition model in the historical dynamic drift model based on the fuzzy clustering result to obtain a dynamic drift prediction model of the target fishing boat specifically includes the following steps:

[0018] Obtain a number of historical meteorological environment data corresponding to the target fishing vessel sailing in each sub-drift area through the navigation log, and divide the scheduled sailing time into a number of sailing time sequence nodes;

[0019] Obtain multidimensional specification parameters of the target fishing boat, and obtain physical action vectors of the multidimensional specification parameters under different combinations of meteorological environmental conditions based on a big data network; wherein the multidimensional specification parameters include the volume, weight, height and load of the target fishing boat;

[0020] Based on the historical drift vector set, the historical drift vector corresponding to each navigation time series node is extracted, the FCM algorithm is introduced to construct a fuzzy membership matrix between each historical drift vector and different physical action vectors, and multiple clusters are set according to a number of historical meteorological environment data;

[0021] A convergence threshold is preset, the cluster center of each cluster is calculated based on the fuzzy membership matrix, the fuzzy membership of each cluster center is obtained, the fuzzy membership matrix is ​​iteratively updated by the fuzzy membership until the convergence threshold is reached, and the fuzzy clustering result of the historical drift vector generated by the historical meteorological environment data is obtained;

[0022] The state transition model of the target fishing boat drift in the historical dynamic drift model is adjusted based on the fuzzy clustering result to obtain an adjusted state transition model, and the dynamic drift prediction model of the target fishing boat is obtained by resampling the states and weights of new particles to train and verify the adjusted state transition model.

[0023] Furthermore, in a preferred embodiment of the present invention, the future meteorological environment simulation field of the South China Sea water area is constructed, a first dynamic distress environment area is planned according to the distress coordinate position of the current target fishing boat, a first future meteorological environment sub-simulation field in the first dynamic distress environment area is stripped through the future meteorological environment field model, and a dynamic prediction is performed in a dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift route, specifically including the following steps:

[0024] Acquire a number of meteorological forecast data for the South China Sea waters in a future time period, generate sub-models of different meteorological environment types according to the meteorological forecast data, and perform spatial coupling and temporal coupling on each of the sub-models to construct a future meteorological environment simulation field for the South China Sea waters;

[0025] Constructing a three-dimensional model of the current fishing boat, querying the distress coordinate position of the current fishing boat in the multi-dimensional geographic coordinate map through the GPS positioning function, and presetting a distress environment planning threshold based on the three-dimensional model of the current fishing boat;

[0026] Planning the distress environment range of the current fishing vessel in the multi-dimensional geographic coordinate map until the distress environment planning threshold is reached, and obtaining the static distress environment range of the current fishing vessel at the distress coordinate position;

[0027] The preset ideal navigation route of the current target fishing boat is obtained, and the static distress environment range of the current target fishing boat is planned in the multi-dimensional geographic coordinate map according to the ideal navigation route, and the area formed after the current target fishing boat travels along the ideal navigation route is obtained to obtain the first dynamic distress environment area;

[0028] In the future meteorological environment field model, multiple future meteorological environment sub-simulation fields in the future time series in which the current target fishing boat is in the first dynamic distress environment area are separated and defined as the first future meteorological environment sub-simulation field. The dynamic drift prediction model of the target fishing boat is used to perform dynamic drift prediction analysis on the multiple first future meteorological environment sub-simulation fields to obtain an initial dynamic drift trajectory.

[0029] Furthermore, in a preferred embodiment of the present invention, the method of obtaining a plurality of meteorological forecast data in a future time period of the South China Sea water area, generating sub-models of different meteorological environment types according to the plurality of meteorological forecast data, and spatially coupling and temporally coupling each of the sub-models to construct a future meteorological environment simulation field of the South China Sea water area specifically includes the following steps:

[0030] Acquire a number of meteorological forecast data for the South China Sea waters in a future time period, parse and reconstruct the meteorological forecast data using meteorological environment simulation software, and generate sub-state models of different meteorological environment types; wherein the climate environment types include atmosphere, ocean currents, and land;

[0031] Extracting meteorological environment variable equations expressed by each meteorological environment type based on the sub-state model, introducing a finite difference algorithm to discretize each meteorological environment variable equation to generate a number of discretized variable equations, and constructing a coupled spatial grid and a time step of a sub-state model of each meteorological environment type according to the several discretized variable equations;

[0032] The coupling sequence of each sub-state model is set according to the time step, and the coupling space grids of the sub-state model of each meteorological environment type are matched one by one based on the coupling sequence, and the meteorological environment variables of each sub-state model are exchanged during the grid matching process to obtain a meteorological environment simulation field;

[0033] Based on the coupling time series, a plurality of historical meteorological environment data are coupled and tested in the meteorological environment simulation field in sequence to obtain a meteorological environment test simulation field, and a meteorological environment benchmark simulation field after a plurality of historical meteorological environment data standard couplings are created;

[0034] The degree of fit between the meteorological environment test simulation field and the meteorological environment benchmark simulation field is calculated. If the degree of fit is less than the preset degree of fit, the spatial coupling model and coupling time series of each meteorological environment type are coupled and calibrated according to the degree of fit to obtain the future meteorological environment simulation field in the South China Sea water area.

[0035] Furthermore, in a preferred embodiment of the present invention, the second dynamic distress environment area formed by the dynamic drift of the current target fishing boat is planned through the initial dynamic drift route, the vector disturbance index of the meteorological environment is analyzed through the geological and topographical characteristic vector distribution map of the second dynamic distress environment area, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift route and the corresponding coordinate values, which specifically includes the following steps:

[0036] Based on the multi-dimensional geographic coordinate map, the static distress environment range of the current target fishing boat is replanned according to the area formed after the initial dynamic drift route, to obtain a second dynamic distress environment area;

[0037] Extracting geographic coordinate parameters of the second dynamic distress environment area through a multi-dimensional geographic coordinate map, and transmitting the geographic coordinate parameters to a GPS satellite positioning system;

[0038] Remotely locate and photograph the second dynamic distress environment area through a GPS satellite positioning system to obtain satellite terrain image data of the second dynamic distress environment area, and introduce an edge detection algorithm to analyze and extract the satellite terrain image data to obtain a geological terrain feature vector of the second dynamic distress environment area;

[0039] Based on geographic coordinate parameters, all geological and topographical feature vectors are marked in the multidimensional geographic coordinate map to generate a distribution map of geological and topographical feature vectors in the second dynamic distress environment area. At the same time, the geographical and geomorphic knowledge map of the South China Sea waters is obtained based on the big data network.

[0040] The geological terrain characteristic vector distribution map and the future meteorological environment simulation field are simultaneously imported into the geographical and geomorphic knowledge map of the South China Sea waters for identification, and the vector disturbance index of the geological terrain distribution of the second dynamic distress environment area for the future meteorological environment is obtained;

[0041] A future meteorological environment sub-simulation field in which the current target fishing boat is in a second dynamic distress environment area is separated from the future meteorological environment field model, which is defined as a second future meteorological environment sub-simulation field; and the meteorological environment vector of the second future meteorological environment sub-simulation field is adjusted and updated according to the vector disturbance index to obtain the adjusted and updated second future meteorological environment sub-simulation field;

[0042] The dynamic drift prediction model is used to re-analyze the adjusted and updated second future meteorological environment sub-simulation field to obtain the final dynamic drift trajectory of the current target fishing boat and the coordinate value of the final dynamic drift trajectory, and the final dynamic drift trajectory and coordinate value are uploaded to the distress terminal of the target fishing boat.

[0043] A second aspect of the present invention provides a fishing vessel dynamic drift prediction system based on distress analysis in South China Sea waters, the fishing vessel dynamic drift prediction system comprises a memory and a processor, the memory stores a fishing vessel dynamic drift prediction method program based on distress analysis in South China Sea waters, when the fishing vessel dynamic drift prediction method program based on distress analysis in South China Sea waters is executed by the processor, any one of the steps of the fishing vessel dynamic drift prediction method based on distress analysis in South China Sea waters is implemented.

[0044] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0045] A historical dynamic drift model of the target fishing boat in the waters of the South China Sea is constructed, and the drift vector of the target fishing boat is analyzed on a multidimensional geographic coordinate map according to the historical dynamic drift model to obtain a historical drift vector set of the target fishing boat in each sub-dynamic drift area; a number of historical meteorological environmental data of the target fishing boat are obtained, and the FCM algorithm is introduced to cluster the several historical meteorological environmental data into a historical drift vector set to generate a fuzzy clustering result, and the state transition model in the historical dynamic drift model is adjusted based on the fuzzy clustering result to obtain a dynamic drift prediction model for the target fishing boat; a future meteorological environment field model of the South China Sea waters area is constructed, and according to the current distress coordinates of the target fishing boat, a prediction model of the dynamic drift of the target fishing boat is obtained. The first static distress environment area is planned in the position, the first future meteorological environment sub-simulation field in the first static distress environment area is stripped by the future meteorological environment field model, and dynamic prediction is performed in the dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift route; the second static distress environment area formed by the dynamic drift of the current target fishing boat is planned by the initial dynamic drift route and the corresponding geological terrain characteristic vector distribution map is obtained to analyze the vector disturbance index of the meteorological environment, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift route and the corresponding coordinate values. The present invention can predict the route and coordinates of the dynamic drift of the fishing boat under the influence of the meteorological environment when the fishing boat is in distress in the waters of the South China Sea, and can enable the search and rescue personnel to accurately locate the dynamic distress position of the current fishing boat, thereby improving the search and rescue efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A first method flow chart of a method for predicting the dynamic drift of a fishing vessel based on distress analysis in the South China Sea is shown;

[0048] Figure 2 A second method flow chart of the method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea is shown;

[0049] Figure 3 A third method flow chart of the method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea is shown;

[0050] Figure 4 The system framework diagram of the dynamic drift prediction system of fishing vessels based on distress analysis in the South China Sea waters is shown. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0053] The first aspect of the present invention provides a method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea waters, such as Figure 1 As shown, the following steps are included:

[0054] S102: constructing a historical dynamic drift model of the target fishing boat when it is in distress in the waters of the South China Sea, and analyzing the drift vector of the target fishing boat on a multi-dimensional geographic coordinate map according to the historical dynamic drift model to obtain a historical drift vector set of the target fishing boat in each sub-dynamic drift area;

[0055] S104: obtaining a number of historical meteorological environment data of the target fishing boat, introducing the FCM algorithm to cluster the several historical meteorological environment data into a historical drift vector set, generating a fuzzy clustering result, and adjusting the state transition model in the historical dynamic drift model based on the fuzzy clustering result to obtain a dynamic drift prediction model of the target fishing boat;

[0056] S106: construct a future meteorological environment simulation field for the South China Sea waters area, plan a first dynamic distress environment area according to the distress coordinate position of the current target fishing boat, separate the first future meteorological environment sub-simulation field in the first dynamic distress environment area through the future meteorological environment field model, and perform dynamic prediction in the dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift trajectory;

[0057] S108: Planning the second dynamic distress environment area formed by the dynamic drift of the current target fishing boat through the initial dynamic drift route, analyzing the vector disturbance index of the meteorological environment through the geological and topographic characteristic vector distribution map of the second dynamic distress environment area, adjusting the second future meteorological environment sub-simulation field according to the vector disturbance index and re-dynamically predicting to obtain the final dynamic drift trajectory and the corresponding coordinate values.

[0058] Furthermore, in a preferred embodiment of the present invention, the historical dynamic drift model of the target fishing boat when it is in distress in the waters of the South China Sea is constructed, and the drift vector of the target fishing boat is analyzed on a multidimensional geographic coordinate map according to the historical dynamic drift model to obtain a historical drift vector set of the target fishing boat in each sub-dynamic drift area, specifically comprising the following steps:

[0059] Obtaining a navigation log of the target fishing vessel in the waters of the South China Sea, and extracting a number of historical scheduled route coordinates and a number of actual route coordinates of the target fishing vessel in distress sequence through the navigation log;

[0060] A particle filter algorithm is introduced, and a state transition model and an observation model of the drift of the target fishing boat are constructed in the particle filter algorithm based on a number of historical scheduled route coordinates and a number of actual route coordinates, and a group of particles are initialized according to the state transition model and the observation model;

[0061] The weight of each particle is updated according to the coordinate deviation between the historical scheduled route coordinates and the actual route coordinates and the observation model, and new particles are resampled with the new weights. The above steps are repeated to continuously recursively update the weights and states of the particles until all particles are updated, thus generating a historical dynamic drift model of the target fishing vessel.

[0062] Obtain a multidimensional geographic coordinate map of the South China Sea waters area, construct a multidimensional space coordinate domain based on the multidimensional geographic coordinate map, and dynamically fit the historical scheduled route and the actual route in the multidimensional space coordinate domain based on the historical dynamic drift model to extract the dynamic drift area formed between the historical scheduled route and the actual route;

[0063] The scheduled sailing time of the target fishing vessel is obtained through the sailing log, the dynamic drift area is divided into N sub-dynamic drift areas based on the scheduled sailing time, and all drift coordinates in each sub-dynamic drift area and the distribution pattern of each drift coordinate are obtained;

[0064] The drift changes of the target fishing boat in the dynamic drift area are analyzed by the changes and distribution patterns of the drift coordinates, and the historical drift vector sets of the target fishing boat in each sub-dynamic drift area are obtained.

[0065] It should be noted that the environment in the South China Sea is complex and changeable, and the possibility of fishing boats getting into distress when operating in the South China Sea is greatly increased. However, the transmission prediction method for the dynamic prediction of fishing boats in distress in the South China Sea has not been able to further analyze the historical dynamic drift data, so the final dynamic drift prediction is not suitable and the prediction result is low in accuracy, which greatly reduces the search and rescue efficiency of fishing boats in distress in the South China Sea. Therefore, in order to ensure accurate dynamic prediction of the drift trajectory of fishing boats when they are in distress, this method conducts in-depth and reliable analysis of the historical dynamic drift behavior of fishing boats in the waters of the South China Sea. First, the particle filter algorithm is used to calculate and analyze several historical scheduled route coordinates and several actual route coordinates of the target fishing boat in distress time sequence, so as to generate a dynamic drift model that describes the target fishing boat when it is in distress in historical time; when the actual route of the target fishing boat deviates from the scheduled route, a dynamic drift area will be formed between the two, and the drift vector of the target fishing boat in the dynamic drift area is not consistent during the scheduled sailing time, which is caused by different changes in the meteorological environment in the waters of the South China Sea. Therefore, the historical drift vector set generated by the target fishing boat in the dynamic drift area due to different meteorological environmental changes can be obtained through the historical dynamic drift model. These historical drift vector sets play a crucial training and analysis effect on dynamic drift prediction based on meteorological environmental changes; wherein, the drift vector is the direction, angle and drift travel of the drift. This method can accurately analyze the dynamic drift behavior of the target fishing boat when it is in distress in the South China Sea based on historical data, thereby providing a reliable and accurate training and analysis basis for the subsequent dynamic drift prediction of fishing boats in distress, greatly improving the accuracy of dynamic prediction of fishing boats in distress.

[0066] Furthermore, in a preferred embodiment of the present invention, the historical meteorological environment data of the target fishing boat are obtained, and the FCM algorithm is introduced to cluster the historical meteorological environment data into a historical drift vector set to generate a fuzzy clustering result. The state transition model in the historical dynamic drift model is adjusted based on the fuzzy clustering result to obtain a dynamic drift prediction model of the target fishing boat, such as Figure 2 As shown, the specific steps include:

[0067] S202: obtaining a number of historical meteorological environment data corresponding to the target fishing vessel sailing in each sub-drifting area through the navigation log, and dividing the scheduled sailing time into a number of sailing time sequence nodes;

[0068] S204: Acquire multidimensional specification parameters of the target fishing boat, and acquire physical action vectors of the multidimensional specification parameters under different combinations of meteorological environmental conditions based on the big data network; wherein the multidimensional specification parameters include the volume, weight, height and load of the target fishing boat;

[0069] S206: extracting the historical drift vector corresponding to each navigation time series node based on the historical drift vector set, introducing the FCM algorithm to construct a fuzzy membership matrix between each historical drift vector and different physical action vectors, and setting multiple clusters according to a number of historical meteorological environment data;

[0070] S208: Preset a convergence threshold, calculate the cluster center of each cluster based on the fuzzy membership matrix, obtain the fuzzy membership of each cluster center, iteratively update the fuzzy membership matrix by the fuzzy membership until the convergence threshold is reached, and obtain the fuzzy clustering result of the historical meteorological environment data to generate the historical drift vector;

[0071] S210: adjusting the state transition model of the target fishing boat drift in the historical dynamic drift model based on the fuzzy clustering result to obtain an adjusted state transition model, and obtaining a dynamic drift prediction model of the target fishing boat by resampling the states and weights of new particles and training and verifying the adjusted state transition model.

[0072] It should be noted that the physical action vector is the angle, direction and mechanical magnitude of the physical action of the meteorological environment on the target fishing boat itself. The historical meteorological environment data include wind direction, ocean current velocity, rainfall, wave intensity and fog concentration. The meteorological environment changes in the South China Sea waters will cause the target fishing boat to drift significantly or slightly. For example, the wind force in the distress area is strong and the ocean current is turbulent, which causes the target fishing boat to drift significantly, thus forming a historical drift vector set generated by the target fishing boat in each sub-drift area in the dynamic drift area; however, due to the complex and changeable meteorological environment in the South China Sea waters, it may be strong or weak at times, and meteorological environment changes of different intensities lead to the generation of different drift vectors for the target fishing boat. If the historical drift vectors corresponding to each historical meteorological environment data cannot be matched and aligned, the accuracy of the target fishing boat dynamic drift prediction result may have a large error, affecting the accuracy of search and rescue positioning of fishing boats in distress. Therefore, this method can accurately fuzzy cluster the historical meteorological environment data corresponding to the target fishing boat when it sails in each sub-drift area with the historical drift vector set of the target fishing boat in each sub-dynamic drift area, so that the time series changes of different meteorological environment data and the corresponding historical drift vectors can be clustered and matched with each other when predicting the dynamic drift of the fishing boat in distress, thereby improving the reliability and accuracy of the prediction results of the dynamic drift of the fishing boat in distress caused by the influence of meteorological environment changes, and greatly reducing the dynamic prediction error rate compared with the traditional prediction method. In addition, this method can directly adjust the state transition model of the historical dynamic drift model constructed in the previous step according to its fuzzy clustering results, so as to predict the next moment state of each particle based on the fuzzy clustering results, which saves the tedious steps of constructing another prediction model, saves computing resources and efficiency, and can improve the prediction performance based on the improved fuzzy clustering results, and further optimize the prediction accuracy of the dynamic drift prediction model.

[0073] It should be noted that the FCM algorithm is called fuzzy clustering algorithm in Chinese. It is a clustering algorithm that allows data points to belong to multiple classes and assigns different membership degrees to each data point. It is very effective when processing data with fuzzy boundaries and can better reflect the uncertainty and ambiguity in actual situations. Due to the fuzzy characteristics of FCM, it has less impact on noise data points and can provide smoother clustering results, further improving clustering accuracy and performance.

[0074] Furthermore, in a preferred embodiment of the present invention, the construction of the future meteorological environment simulation field of the South China Sea water area plans a first dynamic distress environment area according to the distress coordinate position of the current target fishing boat, and the first future meteorological environment sub-simulation field in the first dynamic distress environment area is stripped by the future meteorological environment field model, and a dynamic prediction is performed in the dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift trajectory, which specifically includes the following steps:

[0075] Acquire a number of meteorological forecast data for the South China Sea waters in a future time period, generate sub-models of different meteorological environment types according to the meteorological forecast data, and perform spatial coupling and temporal coupling on each of the sub-models to construct a future meteorological environment simulation field for the South China Sea waters;

[0076] Constructing a three-dimensional model of the current fishing boat, querying the distress coordinate position of the current fishing boat in the multi-dimensional geographic coordinate map through the GPS positioning function, and presetting a distress environment planning threshold based on the three-dimensional model of the current fishing boat;

[0077] Planning the distress environment range of the current fishing vessel in the multi-dimensional geographic coordinate map until the distress environment planning threshold is reached, and obtaining the static distress environment range of the current fishing vessel at the distress coordinate position;

[0078] The preset ideal navigation route of the current target fishing boat is obtained, and according to the ideal navigation route, a static distress environment range of the current target fishing boat is planned in a multi-dimensional geographic coordinate map, which is an area formed after the boat travels along the ideal navigation route, to obtain a first static distress environment area;

[0079] In the future meteorological environment field model, multiple future meteorological environment sub-simulation fields in the future time series in which the current target fishing boat is in the first static distress environment area are separated and defined as the first future meteorological environment sub-simulation field. The dynamic drift prediction model of the target fishing boat is used to perform dynamic drift prediction analysis on the multiple first future meteorological environment sub-simulation fields to obtain an initial dynamic drift trajectory.

[0080] It should be noted that after the dynamic drift prediction model is generated, the drift trajectory of the current target fishing boat when it is in distress can be dynamically predicted. However, since the meteorological environment in the South China Sea waters is unevenly distributed and changes all the time, the target fishing boat is sailing towards the ideal sailing route. The ever-changing meteorological environment on its ideal sailing route directly affects the accuracy of the dynamic drift prediction results. Therefore, in order to improve the accuracy of the dynamic drift prediction, it is necessary to dynamically obtain the future meteorological environment field in a certain area of ​​the target fishing boat while it is sailing along the ideal sailing route. Therefore, the method can first construct a future meteorological environment field model of the South China Sea water area based on several meteorological forecast data of the South China Sea water area; because the multi-dimensional specifications of the fishing boat itself when it is in a static state will be affected by the physical effects of the meteorological environment within a certain range, it is necessary to plan the static distress environment range when the target fishing boat is at the distress coordinate position; then the dynamic distress environment area formed during the static distress environment range of the target fishing boat along the ideal navigation route is analyzed to obtain the first dynamic distress environment area, and this first dynamic distress environment area can reflect the meteorological environment changes that are predicted to cause the dynamic drift of the current target fishing boat when it is in distress. These meteorological environment changes are necessary data conditions for the dynamic drift prediction model to predict the drift of the current target fishing boat. Therefore, the first future meteorological environment sub-simulation field of the current target fishing boat in the first dynamic distress environment area is finally separated from the constructed future meteorological environment field model, and the dynamic drift prediction model is used to perform dynamic drift prediction analysis on the first future meteorological environment sub-simulation field, so as to obtain the initial dynamic drift trajectory. This method can be used to dynamically predict the impact of changes in the meteorological environment field under normal driving conditions on the current distressed fishing boat, thereby accurately predicting the final drift trajectory of the current target fishing boat after being affected by the meteorological environment, thereby improving the accuracy of dynamic drift prediction, facilitating the formulation of rescue plans, and ensuring rescue efficiency.

[0081] Furthermore, in a preferred embodiment of the present invention, a plurality of meteorological forecast data in a future time period of the South China Sea water area are obtained, sub-models of different meteorological environment types are generated according to the plurality of meteorological forecast data, and each of the sub-models is spatially coupled and temporally coupled to construct a future meteorological environment simulation field of the South China Sea water area, such as Figure 3 As shown, the specific steps include:

[0082] S302: Acquire a number of meteorological forecast data for the South China Sea water area in a future time period, parse and reconstruct the meteorological forecast data using meteorological environment simulation software, and generate sub-state models of different meteorological environment types; wherein the climate environment types include atmosphere, ocean currents, and land;

[0083] S304: extracting meteorological environment variable equations expressed by each meteorological environment type based on the sub-state model, introducing a finite difference algorithm to discretize each meteorological environment variable equation to generate a number of discretized variable equations, and constructing a coupled spatial grid and a time step of a sub-state model of each meteorological environment type according to the number of discretized variable equations;

[0084] S306: setting a coupling sequence for each sub-state model according to the time step, performing grid docking and matching on the coupling spatial grids of the sub-state model of each meteorological environment type one by one based on the coupling sequence, and exchanging the meteorological environment variables of each sub-state model during the grid docking and matching process to obtain a meteorological environment simulation field;

[0085] S308: Based on the coupling time sequence, a plurality of historical meteorological environment data are sequentially coupled and tested in the meteorological environment simulation field to obtain a meteorological environment test simulation field, and a plurality of meteorological environment benchmark simulation fields are created after standard coupling of the historical meteorological environment data;

[0086] S310: Calculate the degree of fit between the meteorological environment test simulation field and the meteorological environment benchmark simulation field. If the degree of fit is less than a preset degree of fit, perform coupling calibration on the spatial coupling model and coupling time series of each meteorological environment type according to the degree of fit to obtain the future meteorological environment simulation field for the South China Sea water area.

[0087] It should be noted that, due to the rich and ever-changing meteorological environment types in the South China Sea waters, and the dynamic drift prediction model of the target fishing boat is in the form of dynamic prediction, if the meteorological environment data is directly imported into this dynamic drift prediction model for prediction, the dynamic drift prediction model will reduce its ability to analyze the time series dynamics of the meteorological environment data, increase the amount of data calculation of the dynamic drift prediction model, and thus may cause errors or large deviations in the final predicted results, affecting the quality and efficiency of the rescue plan for the fishing boat in distress. Therefore, this method can generate sub-models of different meteorological environment types according to several meteorological forecast data in the future time period of the South China Sea waters, and perform grid docking and matching of the coupled spatial grids of the sub-models of each meteorological environment type one by one according to the time step, such as grid docking and matching between the atmosphere and ocean current models. Among them, the meteorological environment variables include temperature, humidity, wind speed, sea surface temperature and flow rate; exchanging the meteorological environment variables of each sub-state model during the grid docking and matching process can ensure the actuality and influence of the interaction between each meteorological environment type during the coupling interaction, and further improve the reliability of the meteorological environment simulation of the meteorological environment simulation field. For the degree of fit between the meteorological environment test simulation field and the meteorological environment benchmark simulation field, if the degree of fit is less than the preset degree of fit, it means that the meteorological environment data simulation of the meteorological environment simulation field has deviations, so it is inaccurate. Therefore, it is necessary to perform corresponding coupling calibration on the network docking and coupling timing of its spatial coupling model, so as to continuously update and improve the simulation performance of the future meteorological environment simulation field. Through this method, a future meteorological environment simulation field that can realize dynamic simulation based on meteorological forecast data can be constructed, thereby effectively improving the meteorological environment analysis efficiency and accuracy of the dynamic drift prediction model, reducing the amount of calculation, improving the stability and accuracy of the prediction results of the dynamic drift prediction model, and avoiding the generation of dynamic drift prediction errors.

[0088] Furthermore, in a preferred embodiment of the present invention, the second dynamic distress environment area formed by the dynamic drift of the current target fishing boat is planned through the initial dynamic drift route, the vector disturbance index of the meteorological environment is analyzed through the geological terrain characteristic vector distribution map of the second dynamic distress environment area, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift trajectory and the corresponding coordinate values, which specifically includes the following steps:

[0089] Based on the multi-dimensional geographic coordinate map, the static distress environment range of the current target fishing boat is replanned according to the area formed after the initial dynamic drift route, to obtain a second dynamic distress environment area;

[0090] Extracting geographic coordinate parameters of the second dynamic distress environment area through a multi-dimensional geographic coordinate map, and transmitting the geographic coordinate parameters to a GPS satellite positioning system;

[0091] Remotely locate and photograph the second dynamic distress environment area through a GPS satellite positioning system to obtain satellite terrain image data of the second dynamic distress environment area, and introduce an edge detection algorithm to analyze and extract the satellite terrain image data to obtain a geological terrain feature vector of the second dynamic distress environment area;

[0092] Based on geographic coordinate parameters, all geological and topographical feature vectors are marked in the multidimensional geographic coordinate map to generate a distribution map of geological and topographical feature vectors in the second dynamic distress environment area. At the same time, the geographical and geomorphic knowledge map of the South China Sea waters is obtained based on the big data network.

[0093] The geological terrain characteristic vector distribution map and the future meteorological environment simulation field are simultaneously imported into the geographical and geomorphic knowledge map of the South China Sea waters for identification, and the vector disturbance index of the geological terrain distribution of the second dynamic distress environment area for the future meteorological environment is obtained;

[0094] A future meteorological environment sub-simulation field in which the current target fishing boat is in a second dynamic distress environment area is separated from the future meteorological environment simulation field, which is defined as a second future meteorological environment sub-simulation field; and the meteorological environment vector of the second future meteorological environment sub-simulation field is adjusted and updated according to the vector disturbance index to obtain the adjusted and updated second future meteorological environment sub-simulation field;

[0095] The dynamic drift prediction model is used to perform a dynamic drift prediction analysis on the adjusted and updated second future meteorological environment sub-simulation field to obtain the final dynamic drift route of the current target fishing boat and the coordinate value of the final dynamic drift route, and the final dynamic drift route and coordinate value are uploaded to the distress terminal of the target fishing boat.

[0096] It should be noted that the geological topography in the South China Sea waters is complex, with dense shoals, coral reefs and reefs. These geological topography may cause the meteorological environment to change or deflect, thereby further affecting the dynamic drift trajectory of the current target fishing boat. Therefore, after predicting the initial dynamic drift trajectory of the current target fishing boat, it is necessary to further analyze the impact of the geological topography on the initial dynamic drift trajectory on the dynamic drift behavior of the current target fishing boat. The method replans the static distress environment range of the current target fishing boat to form a second dynamic distress environment area after traveling along the initial dynamic drift route; then analyzes the satellite image data of the second dynamic distress environment area based on the GPS satellite positioning system to extract all geological and topographical feature vectors existing in the second dynamic distress environment area; because the different distributions of different geological and topographical features will cause vector disturbance changes such as energy weakening, physical action direction shift or direct absorption disappearance in the future meteorological environment, the vector disturbance index of the future meteorological environment can be determined according to the geological and topographical distribution of the second dynamic distress environment area; when the vector disturbance index occurs, the original second future meteorological environment sub-simulation field must change, so it is necessary to adjust and update the meteorological environment simulation change and state of the second future meteorological environment sub-simulation field according to the vector disturbance index, so as to ensure the timing, reality and rationality of the future meteorological environment sub-simulation field used for prediction when affected by geological and topographical features, thereby greatly improving the accuracy of dynamic drift prediction. This method can further analyze the impact of geological topography in the South China Sea on the dynamic drift prediction results of the current target fishing vessel, so that the prediction results are more in line with the actual distress in the South China Sea, improve the prediction accuracy of the final dynamic drift of the fishing vessel, and effectively ensure the quality of the rescue plan for fishing boats in distress and the actual rescue efficiency, with high economic benefits.

[0097] In addition, the method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea waters further includes the following steps:

[0098] Acquire a plurality of historical dynamic drift prediction trajectories of different target fishing vessels in the waters of the South China Sea, and determine the final dynamic drift waters of the different target fishing vessels according to the plurality of historical dynamic drift prediction trajectories;

[0099] Obtain the historical meteorological environment simulation field of each final dynamic drift water area, obtain several energy gradient evaluation indicators of different meteorological environment types based on big data network retrieval, introduce deep learning algorithm to train and verify the several energy gradient evaluation indicators, and obtain a meteorological environment energy gradient evaluation model;

[0100] Evaluate the historical meteorological environment simulation field using a meteorological environment energy gradient evaluation model to obtain an energy gradient evaluation value for each historical meteorological environment simulation field;

[0101] Introducing a Kriging interpolation algorithm to calculate the semivariogram of the energy gradient evaluation value of each historical meteorological environment simulation field, performing Kriging interpolation on each energy gradient evaluation value according to the semivariogram to obtain a historical meteorological environment energy hotspot map;

[0102] Preset the meteorological environment energy anomaly threshold, remove the energy hotspots whose meteorological environment energy values ​​are higher than the meteorological environment energy anomaly threshold in the historical meteorological environment energy hotspot map, and integrate the remaining energy hotspots to obtain multiple resettlement areas;

[0103] The placement layout points of the search and rescue drone are rationally planned according to the multiple placement areas, a placement layout diagram of the search and rescue drone is generated, and the placement layout diagram of the search and rescue drone is uploaded to the fishing boat terminal.

[0104] It should be noted that fishing boats are prone to encounter bad weather and get into trouble when fishing in the waters of the South China Sea. Search and rescue drones can implement efficient and reasonable rescue tasks based on the dynamic drifting behavior of fishing boats in distress, and can replace manual rescue when human rescue fails to reach the distress location in the first time. However, if search and rescue drones are easily affected by the meteorological environment during the search and rescue mission in the waters of the South China Sea, it is easy to cause search and rescue trajectory deviation, drone damage and failure, etc., so a reasonable search and rescue drone placement layout is crucial. Through this method, the placement area of ​​the search and rescue drone can be screened and planned, so that the placement layout of the search and rescue drone is more reasonable, so that the search and rescue drone can avoid the influence of bad meteorological environment when performing search and rescue missions, improve the efficiency and quality of search and rescue, and further protect the lives of people in distress.

[0105] The second aspect of the present invention provides a fishing boat dynamic drift prediction system based on South China Sea waters distress analysis, such as Figure 4 As shown, the fishing vessel dynamic drift prediction system includes a memory 41 and a processor 42. The memory 41 stores a fishing vessel dynamic drift prediction method program based on the distress analysis in the South China Sea waters. When the fishing vessel dynamic drift prediction method program based on the distress analysis in the South China Sea waters is executed by the processor 42, any one of the steps of the fishing vessel dynamic drift prediction method based on the distress analysis in the South China Sea waters is implemented.

[0106] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea, characterized in that: The following steps are involved: Construct a historical dynamic drift model of the target fishing boat when it is in distress in the waters of the South China Sea, and analyze the drift vector of the target fishing boat on a multidimensional geographic coordinate map based on the historical dynamic drift model to obtain a set of historical drift vectors of the target fishing boat in each sub-dynamic drift area; Acquire several historical meteorological environment data of the target fishing boat, introduce the FCM algorithm to cluster the several historical meteorological environment data into a historical drift vector set, generate a fuzzy clustering result, adjust the state transition model in the historical dynamic drift model based on the fuzzy clustering result, and obtain a dynamic drift prediction model of the target fishing boat; Constructing a future meteorological environment simulation field for the South China Sea waters area, planning a first dynamic distress environment area according to the distress coordinate position of the current target fishing boat, stripping a first future meteorological environment sub-simulation field in the first dynamic distress environment area through the future meteorological environment field model, and performing dynamic prediction in a dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift trajectory; The second dynamic distress environment area formed by the dynamic drift of the current target fishing boat is planned through the initial dynamic drift route, the vector disturbance index of the meteorological environment is analyzed through the geological and topographic characteristic vector distribution map of the second dynamic distress environment area, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift trajectory and the corresponding coordinate values.

2. The method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea according to claim 1 is characterized in that: The method of constructing a historical dynamic drift model of a target fishing boat when it is in distress in the waters of the South China Sea, and analyzing the drift vector of the target fishing boat on a multidimensional geographic coordinate map according to the historical dynamic drift model to obtain a historical drift vector set of the target fishing boat in each sub-dynamic drift area, specifically includes the following steps: Obtaining a navigation log of the target fishing vessel in the waters of the South China Sea, and extracting a number of historical scheduled route coordinates and a number of actual route coordinates of the target fishing vessel in distress sequence through the navigation log; A particle filter algorithm is introduced, and a state transition model and an observation model of the drift of the target fishing boat are constructed in the particle filter algorithm based on a number of historical scheduled route coordinates and a number of actual route coordinates, and a group of particles are initialized according to the state transition model and the observation model; The weight of each particle is updated according to the coordinate deviation between the historical scheduled route coordinates and the actual route coordinates and the observation model, and new particles are resampled with the new weights. The above steps are repeated to continuously recursively update the weights and states of the particles until all particles are updated, thus generating a historical dynamic drift model of the target fishing vessel. Obtain a multidimensional geographic coordinate map of the South China Sea waters area, construct a multidimensional space coordinate domain based on the multidimensional geographic coordinate map, and dynamically fit the historical scheduled route and the actual route in the multidimensional space coordinate domain based on the historical dynamic drift model to extract the dynamic drift area formed between the historical scheduled route and the actual route; The scheduled sailing time of the target fishing vessel is obtained through the sailing log, the dynamic drift area is divided into N sub-dynamic drift areas based on the scheduled sailing time, and all drift coordinates in each sub-dynamic drift area and the distribution pattern of each drift coordinate are obtained; The drift changes of the target fishing boat in the dynamic drift area are analyzed by the changes and distribution patterns of the drift coordinates, and the historical drift vector sets of the target fishing boat in each sub-dynamic drift area are obtained.

3. The method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea according to claim 1 is characterized in that: The method of obtaining a plurality of historical meteorological environment data of the target fishing boat, introducing the FCM algorithm to cluster the plurality of historical meteorological environment data into a historical drift vector set, generating a fuzzy clustering result, and adjusting the state transition model in the historical dynamic drift model based on the fuzzy clustering result to obtain a dynamic drift prediction model of the target fishing boat specifically includes the following steps: Obtain a number of historical meteorological environment data corresponding to the target fishing vessel sailing in each sub-drift area through the navigation log, and divide the scheduled sailing time into a number of sailing time sequence nodes; Obtain multidimensional specification parameters of the target fishing boat, and obtain physical action vectors of the multidimensional specification parameters under different combinations of meteorological environmental conditions based on a big data network; wherein the multidimensional specification parameters include the volume, weight, height and load of the target fishing boat; Based on the historical drift vector set, the historical drift vector corresponding to each navigation time series node is extracted, the FCM algorithm is introduced to construct a fuzzy membership matrix between each historical drift vector and different physical action vectors, and multiple clusters are set according to a number of historical meteorological environment data; A convergence threshold is preset, the cluster center of each cluster is calculated based on the fuzzy membership matrix, the fuzzy membership of each cluster center is obtained, the fuzzy membership matrix is ​​iteratively updated by the fuzzy membership until the convergence threshold is reached, and the fuzzy clustering result of the historical drift vector generated by the historical meteorological environment data is obtained; The state transition model of the target fishing boat drift in the historical dynamic drift model is adjusted based on the fuzzy clustering result to obtain an adjusted state transition model, and the dynamic drift prediction model of the target fishing boat is obtained by resampling the states and weights of new particles to train and verify the adjusted state transition model.

4. The method for predicting the dynamic drift of fishing vessels based on distress analysis in South China Sea waters according to claim 1 is characterized in that: The method of constructing a future meteorological environment simulation field for the South China Sea waters area, planning a first dynamic distress environment area according to the distress coordinate position of the current target fishing boat, separating a first future meteorological environment sub-simulation field in the first dynamic distress environment area through the future meteorological environment field model, and performing dynamic prediction in a dynamic drift prediction model based on the first future meteorological environment sub-simulation field to obtain an initial dynamic drift route, specifically includes the following steps: Acquire a number of meteorological forecast data for the South China Sea waters in a future time period, generate sub-models of different meteorological environment types according to the meteorological forecast data, and perform spatial coupling and temporal coupling on each of the sub-models to construct a future meteorological environment simulation field for the South China Sea waters; Constructing a three-dimensional model of the current fishing boat, querying the distress coordinate position of the current fishing boat in the multi-dimensional geographic coordinate map through the GPS positioning function, and presetting a distress environment planning threshold based on the three-dimensional model of the current fishing boat; Planning the distress environment range of the current fishing vessel in the multi-dimensional geographic coordinate map until the distress environment planning threshold is reached, and obtaining the static distress environment range of the current fishing vessel at the distress coordinate position; The preset ideal navigation route of the current target fishing boat is obtained, and the static distress environment range of the current target fishing boat is planned in the multi-dimensional geographic coordinate map according to the ideal navigation route, and the area formed after the current target fishing boat travels along the ideal navigation route is obtained to obtain the first dynamic distress environment area; In the future meteorological environment field model, multiple future meteorological environment sub-simulation fields in the future time series in which the current target fishing boat is in the first dynamic distress environment area are separated and defined as the first future meteorological environment sub-simulation field. The dynamic drift prediction model of the target fishing boat is used to perform dynamic drift prediction analysis on the multiple first future meteorological environment sub-simulation fields to obtain an initial dynamic drift trajectory.

5. The method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea according to claim 4 is characterized in that: The method of obtaining a plurality of meteorological forecast data in a future time period of the South China Sea water area, generating sub-models of different meteorological environment types according to the plurality of meteorological forecast data, and performing spatial coupling and temporal coupling on each of the sub-models to construct a future meteorological environment simulation field of the South China Sea water area specifically includes the following steps: Acquire a number of meteorological forecast data for the South China Sea waters in a future time period, parse and reconstruct the meteorological forecast data using meteorological environment simulation software, and generate sub-state models of different meteorological environment types; wherein the climate environment types include atmosphere, ocean currents, and land; Extracting meteorological environment variable equations expressed by each meteorological environment type based on the sub-state model, introducing a finite difference algorithm to discretize each meteorological environment variable equation to generate a number of discretized variable equations, and constructing a coupled spatial grid and a time step of a sub-state model of each meteorological environment type according to the several discretized variable equations; The coupling sequence of each sub-state model is set according to the time step, and the coupling space grids of the sub-state model of each meteorological environment type are matched one by one based on the coupling sequence, and the meteorological environment variables of each sub-state model are exchanged during the grid matching process to obtain a meteorological environment simulation field; Based on the coupling time series, a plurality of historical meteorological environment data are coupled and tested in the meteorological environment simulation field in sequence to obtain a meteorological environment test simulation field, and a meteorological environment benchmark simulation field after a plurality of historical meteorological environment data standard couplings are created; The degree of fit between the meteorological environment test simulation field and the meteorological environment benchmark simulation field is calculated. If the degree of fit is less than the preset degree of fit, the spatial coupling model and coupling time series of each meteorological environment type are coupled and calibrated according to the degree of fit to obtain the future meteorological environment simulation field in the South China Sea water area.

6. The method for predicting the dynamic drift of fishing vessels based on distress analysis in the South China Sea according to claim 1 is characterized in that: The second dynamic distress environment area formed by the dynamic drift of the current target fishing boat is planned by the initial dynamic drift route, the vector disturbance index of the meteorological environment is analyzed by the geological and topographic characteristic vector distribution map of the second dynamic distress environment area, and the second future meteorological environment sub-simulation field is adjusted according to the vector disturbance index and dynamically predicted again to obtain the final dynamic drift route and the corresponding coordinate values, which specifically includes the following steps: Based on the multi-dimensional geographic coordinate map, the static distress environment range of the current target fishing boat is replanned according to the area formed after the initial dynamic drift route, to obtain a second dynamic distress environment area; Extracting geographic coordinate parameters of the second dynamic distress environment area through a multi-dimensional geographic coordinate map, and transmitting the geographic coordinate parameters to a GPS satellite positioning system; Remotely locate and photograph the second dynamic distress environment area through a GPS satellite positioning system to obtain satellite terrain image data of the second dynamic distress environment area, and introduce an edge detection algorithm to analyze and extract the satellite terrain image data to obtain a geological terrain feature vector of the second dynamic distress environment area; Based on geographic coordinate parameters, all geological and topographical feature vectors are marked in the multidimensional geographic coordinate map to generate a distribution map of geological and topographical feature vectors in the second dynamic distress environment area. At the same time, the geographical and geomorphic knowledge map of the South China Sea waters is obtained based on the big data network. The geological terrain characteristic vector distribution map and the future meteorological environment simulation field are simultaneously imported into the geographical and geomorphic knowledge map of the South China Sea waters for identification, and the vector disturbance index of the geological terrain distribution of the second dynamic distress environment area for the future meteorological environment is obtained; A future meteorological environment sub-simulation field in which the current target fishing boat is in a second dynamic distress environment area is separated from the future meteorological environment field model, which is defined as a second future meteorological environment sub-simulation field; and the meteorological environment vector of the second future meteorological environment sub-simulation field is adjusted and updated according to the vector disturbance index to obtain the adjusted and updated second future meteorological environment sub-simulation field; The dynamic drift prediction model is used to re-analyze the adjusted and updated second future meteorological environment sub-simulation field to obtain the final dynamic drift trajectory of the current target fishing boat and the coordinate value of the final dynamic drift trajectory, and the final dynamic drift trajectory and coordinate value are uploaded to the distress terminal of the target fishing boat.

7. The dynamic drift prediction system of fishing vessels based on distress analysis in South China Sea waters is characterized by: The fishing vessel dynamic drift prediction system includes a memory and a processor, wherein the memory stores a fishing vessel dynamic drift prediction method program based on distress analysis in South China Sea waters, and when the fishing vessel dynamic drift prediction method program based on distress analysis in South China Sea waters is executed by the processor, the steps of the fishing vessel dynamic drift prediction method based on distress analysis in South China Sea waters as described in any one of claims 1-6 are implemented.

Images