A method, device, equipment and readable storage medium for analyzing ocean vortex evolution

Abstract

The present invention provides an ocean vortex evolution analysis method, device, equipment and readable storage medium, which relates to the field of marine information technology technology, including constructing an ocean vortex process model; extracting key events in the ocean vortex evolution process, identifying and tracking the ocean vortex according to the vortex detection method and tracking method; determining all key events in the ocean vortex evolution cycle according to the identified ocean vortex and the key events, and generating key analysis results of the ocean vortex evolution using a model combining a recurrent neural network and a long short-term memory artificial neural network. The beneficial effects of the present invention are to use in-situ visualization to process simulation analysis in large-scale flow fields; to generate high-precision vortex streamline structures in real time, and to support vortex statistical analysis and tracking analysis in different ocean regions and at different scales; to prove the efficiency and effectiveness of the system; and to help experts track and analyze the evolution of ocean vortices.

Description

Technical Field

[0001] The present invention relates to the field of marine information technology technology, and in particular to an ocean vortex evolution analysis method, device, equipment and a readable storage medium. Background Art

[0002] Eddies are one of the important research objects in ocean simulation. The diameter of mesoscale eddies in the ocean is usually between tens and hundreds of kilometers. The transmission of ocean eddies often lasts for months and plays an important role in the transportation of heat, salt, biological materials and energy. Among them, eddy analysis is a key step in the analysis of ocean flow field data and one of the commonly used visualization methods.

[0003] At present, the time series of large-scale flow field data and vortices result in a huge amount of vortex data that needs to be analyzed. Experts need to obtain structured small-scale information from large-scale vortex data, such as the structural characteristics of specific vortices. The gap between them makes it difficult for users to obtain the required vortex tracking information even if they use high-performance GPUs for calculations. In addition, existing visualization solutions cannot convert large-scale vectorized data into usable information sets, which requires compression of large-scale data. In general, the analysis of large-scale ocean vortex evolution faces the following challenges: there is a huge gap between the temporal nature of ocean vortex data and small-scale information; and there is an end-to-end visualization analysis requirement from vectorized data to structured information. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device, equipment and readable storage medium for analyzing the evolution of ocean vortices to improve the above problems. In order to achieve the above objectives, the technical solutions adopted by the present invention are as follows:

[0005] In a first aspect, the present application provides a method for analyzing ocean vortex evolution, comprising:

[0006] Extract process information from the ocean eddy historical sample library and build an ocean eddy process model;

[0007] Using a heuristic workflow algorithm to mine the ocean vortex process model, extract key events in the evolution of the ocean vortex, wherein the key events include the merging, splitting, generation and extinction of the ocean vortex;

[0008] The acquired ocean flow data is input into the preset high-performance computing cluster for analysis, and the ocean vortex is identified and tracked according to the vortex detection and tracking methods;

[0009] Determine all key events in the evolution cycle of the ocean vortex according to the identified ocean vortex and the key events, and store the time points of all the key events and high-precision streamline images before and after the time points;

[0010] Based on the high-precision streamline images and the types of all the key events, a model combining a recurrent neural network and a long short-term memory artificial neural network is used to generate key analysis results of the evolution of ocean vortices.

[0011] In a second aspect, the present application also provides an ocean vortex evolution analysis device, including an extraction module, a mining module, an identification module, a determination module and a generation module, wherein:

[0012] Extraction module: used to extract process information from the ocean vortex historical sample library and build an ocean vortex process model;

[0013] A mining module is used to mine the ocean vortex process model using a heuristic workflow algorithm to extract key events in the evolution of the ocean vortex, wherein the key events include the merging, splitting, generation and extinction of the ocean vortex;

[0014] Identification module: used to input the acquired ocean flow field data into a preset high-performance computing cluster for analysis, identify and track ocean vortices based on vortex detection and tracking methods;

[0015] A determination module is used to determine all key events in the evolution cycle of the ocean vortex according to the identified ocean vortex and the key events, and store the time points of all the key events and the high-precision streamline images before and after the time points;

[0016] Generation module: used to generate key analysis results of ocean vortex evolution based on the high-precision streamline image and the types of all the key events, using a model combining a recurrent neural network and a long short-term memory artificial neural network.

[0017] In a third aspect, the present application also provides an ocean vortex evolution analysis device, comprising:

[0018] Memory for storing computer programs;

[0019] A processor is used to implement the steps of the ocean eddy evolution analysis method when executing the computer program.

[0020] In a fourth aspect, the present application further provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned ocean vortex evolution analysis method are implemented.

[0021] Currently, there are two common methods for analyzing ocean eddy evolution. The first is to use event detection methods to detect ocean eddy evolution events by simply visualizing all events and providing them to experts for analysis. The other method is to reduce the size of the original data as much as possible through in-situ visualization. Experts usually use compressed data for research through traditional ocean simulation analysis methods. However, these two methods are relatively isolated and cannot completely solve the above challenges.

[0022] The beneficial effects of the present invention are: using in-situ visualization to process simulation analysis in large-scale flow fields; combining narrative visualization analysis, designing a large-scale ocean flow field vortex evolution analysis system based on in-situ visualization; being able to generate high-precision vortex streamline structures in real time, and supporting vortex statistical analysis and tracking analysis in different ocean regions and at different scales; proving the efficiency and effectiveness of the system through analysis of ocean simulation case data; based on the in-situ visualization analysis system of complex multivariate and multi-center feature evolution, it can help experts track and analyze the evolution of ocean vortices, as well as the statistics and analysis of regional vortices.

[0023] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 Schematic diagram of the process of the ocean vortex evolution analysis method described in an embodiment of the present invention;

[0026] Figure 2 Schematic diagram of the structure of the ocean vortex evolution analysis device described in an embodiment of the present invention;

[0027] Figure 3 It is a schematic diagram of the structure of the ocean vortex evolution analysis equipment described in an embodiment of the present invention.

[0028] In the figure: 701, extraction module; 702, mining module; 7021, selection unit; 7022, first acquisition unit; 7023, processing unit; 7024, merging unit; 7025, construction unit; 703, recognition module; 7031, first generation unit; 7032, filtering unit; 7033, visualization unit; 7034, display unit; 7035, recording unit; 7036, recognition unit; 7037, acquisition of second information unit; 7038, acquisition of third information unit; 7039, acquisition of fourth information unit Information unit; 7040, data information unit; 704, determination module; 705, generation module; 7051, determination unit; 7052, extraction unit; 7053, second generation unit; 7054, conversion unit; 7055, third generation unit; 706, data set building module; 707, calculation module; 708, optimization module; 800, ocean vortex evolution analysis equipment; 801, processor; 802, memory; 803, multimedia component; 804, input / output (I / O) interface; 805, communication component. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0031] Embodiment 1:

[0032] This embodiment provides a method for analyzing the evolution of ocean vortices.

[0033] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400 and step S500.

[0034] S100, extracting process information from an ocean vortex historical sample library and constructing an ocean vortex process model.

[0035] It can be understood that the process before step S100 includes S101, S102 and S103, wherein:

[0036] S101, using a narrative visualization method to identify global ocean vortices and establish an ocean vortex dataset, wherein the ocean vortex dataset includes flow field vector information, temperature information, and salinity information of the ocean vortex;

[0037] S102, calculating an ocean vortex process model according to the Heuristics Miner algorithm and the ocean vortex data set;

[0038] S103, analyzing the ocean vortex process model to obtain an optimized ocean vortex process model.

[0039] Among them, the narrative visualization method and the Heuristics Miner algorithm are both existing technologies. Simply put, they use visual presentation to tell stories, and the Heuristics Miner algorithm is an inductive mining algorithm.

[0040] It should be noted that narrative visualization methods and interactive visualization analysis methods based on storytelling have been applied to the field of information visualization and have also provided new ideas for scientific visualization. Hullman et al. outlined how to use automatic sorting (the order in which visualizations are presented) when designing systems to help non-designers navigate structured decisions when creating narrative visualizations. Fisher et al. used narrative as a new visualization technology based on topic detection, tracking, and time visualization. At present, narrative visualization methods are often used in information visualization, but rarely in the field of scientific visualization. However, it still has high prospects in improving the interactive and analytical effects of scientific visualization.

[0041] Therefore, it is necessary to describe key feature events based on feature detection and image generation to visualize event descriptions. These descriptions can be completed with a small amount of text and event sequences. Therefore, it is necessary to generate description information of key feature events; the main idea of ​​workflow mining is to extract information about the process from the transaction log, obtain the process model and support analysis and optimization. The Heuristics Miner algorithm is used to generate event descriptions for processes with a low degree of structure.

[0042] S200, mining the ocean vortex process model using a heuristic workflow algorithm to extract key events in the evolution of the ocean vortex, wherein the key events include the merging, splitting, generation and extinction of the ocean vortex.

[0043] It can be understood that step S200 includes S201, S202, S203, S204 and S205, wherein:

[0044] S201, selecting a first object as a case object;

[0045] S202, obtaining first information, where the first information includes obtaining the radius, energy and type information of the ocean vortex of the first object;

[0046] S203, processing the first information by using a heuristic workflow mining algorithm;

[0047] S204, deriving a keyword of the key event according to the processed first information and the merging and disappearance information of the ocean vortex in the first object;

[0048] S205: Construct a description table of the key events according to the keywords.

[0049] It should be noted that the merging and splitting events of the vortex evolution process are: Users can choose to explore in depth and enter the vortex evolution tracking interface of the local area. In vortex evolution tracking, experts usually focus on important events in the vortex evolution process.

[0050] Based on the specific sea area selected by the user, streamline generation and vortex tracking are performed on a high-performance computer, and all vortices that split and merge during the vortex evolution are filtered out as visualization objects in the main view. The main view displays the events of each vortex changing over time in the tracking graph, and displays the merging and splitting of vortices through branch merging and splitting. The evolution direction of the tracking graph is the time continuation direction. Each node can be selected to obtain detailed information. The width of each node represents the radius of the vortex. The user can select the vortex structure represented by a specific node, visualize the local streamline image at the corresponding time, and describe the event on the right.

[0051] Among them, in this embodiment, the data set of the first information includes the public data set AVISO and the Red Sea simulation data set. AVISO contains global ocean data from 1993 to 2014, with a grid scale of 1440×720 and a time granularity of 1 day. It contains information such as velocity field, longitude and latitude, temperature, etc. of the global ocean, and the overall scale is about 2TB. The Red Sea simulation data set consists of the bathymetry and regular grid (500×500×50, 60 time steps) of the Red Sea covering a simulation time of one month. In each member, it contains five physical attributes: temperature, salinity, and velocity field components in three directions, with 64-bit floating point data accuracy. In the visualization, we mainly use the data of temperature and velocity attributes. The size of each member is about 22.35GB (500*500*50*60*8KB*4 (number of variables) = 22.35GB$), and the total amount is about 1.09TB.

[0052] S300: input the acquired ocean flow field data into a preset high-performance computing cluster for analysis, and identify and track ocean vortices based on vortex detection methods and tracking methods.

[0053] It can be understood that step S300 includes S301, S302, S303, S304, S305 and S306, wherein:

[0054] S301, generating streamlines and vortex tracking according to the specific sea area to be selected;

[0055] S302, filtering out all ocean vortices that split and merge during the evolution of the ocean vortex;

[0056] S303, using the filtered ocean vortex as a visualization object;

[0057] S304, displaying the events of each ocean vortex in the visualization object changing over time;

[0058] S305, recording each node data in the event as the ocean flow field data;

[0059] S306: Analyze and identify the ocean flow field data through the high performance computing cluster to obtain the identified ocean vortex data.

[0060] It should be noted that in the vortex detection and tracking methods, the vortex is studied as the main body of the event, and its detection method is particularly critical. Commonly used vortex detection methods are divided into two categories, mainly physics-based methods and geometry-based methods. The physics-based method involves the calculation of the dynamic flow field and requires a closed contour vortex definition based on a threshold. Commonly used methods include the SSH method, the vorticity method, the velocity gradient tensor method, or the Okubo-Weiss parameter method. The geometry-based method mainly uses the curvature or shape of the instantaneous flow field to detect vortices. They are mainly based on winding angle, vector geometry, wavelet and Lagrangian pseudo-level structure detection. Among them, the OW method is simple and efficient and is the most widely used.

[0061] Vortex tracking is an important part of vortex evolution analysis. Corinne B. Trott uses the difference in vortex radius, amplitude and kinetic energy to judge the continuity of vortices based on the characteristic differences between two vortices at adjacent moments. The two vortices with the smallest difference are continuous. The research of Penven et al. shows that the smaller the distance between the vortex cores, the greater the correlation between the two vortices.

[0062] The vortex detection method is based on the physical and geometric characteristics of the flow field. The method obtains the Okubo-Weiss (OW) parameter, sea level height and velocity vector through numerical simulation. The OW parameter describes the relative advantage of deformation over flow rotation, and its calculation method is as follows:

[0063]

[0064] Among them, s n 、s s and ω are the normal component of tension, the shear component, and the relative vorticity of the flow field, respectively. They are defined as follows:

[0065]

[0066] s n =ux-vy,s s =vx+vy,ω=vx-vy

[0067] Among them, u and v represent the velocity components, and x and y represent the directions.

[0068] Vortex tracking is achieved by tracking distance (ΔD), radius (ΔR), shear deformation rate (ΔS s ), tensile deformation rate (ΔS n ) and amplitude (ΔA) through the vortex core, etc. For example, if the identified vortices e1 and e2 intersect at t1 and t2, then vortex e2 is the continuation of e1 from t1 to t2 when the difference in characteristics between e1 and e2 is minimal.

[0069]

[0070] Among them, Δf represents the above-mentioned specific feature differences, and σ is the number of features.

[0071] It should be mentioned that in this embodiment, the high-performance computer CPU model used is Intel (R) Xeon (R) Gold 6248CPU @ 2.50GHz, and the GPU is Tesla V100. The GPU-accelerated scientific visualization algorithm toolkit VTK-m is mainly used to support fine-grained concurrency of data analysis and visualization algorithms by providing abstract models for data and execution. The VTK-m tool supporting GPU acceleration is deployed to perform in-situ real-time streamline tracking and streamline image output according to the flow field area selected by the user.

[0072] Step S300 also includes S307, S308, S309 and S310, wherein:

[0073] S307, obtaining second information, wherein the second information includes information on a normal component and a shear component of the ocean velocity field and a relative vorticity of the flow field;

[0074] S308. Obtain third information by numerical simulation according to the second information, where the third information is an Okubo-Weiss parameter;

[0075] S309, obtaining fourth information by calculation according to the third information, wherein the fourth information is ocean vortex parameters, sea level height and velocity vector information;

[0076] S310. Calculate and obtain ocean flow field data information based on the fourth information and the tracking method.

[0077] It should be noted that vortex tracking is the distance (ΔD), radius (ΔR), shear deformation rate (ΔS s ), tensile deformation rate (ΔS n ) and the amplitude (ΔA). The formula is used to determine that when the characteristic difference is the smallest, the vortices have a sequential relationship (that is, vortex tracking).

[0078] S400, determining all key events in the evolution cycle of the ocean vortex based on the identified ocean vortex and the key events, and storing the time points of all the key events and high-precision streamline images before and after the time points.

[0079] It can be understood that this step includes storing high-precision streamline images of key event time points of vortex evolution and the moments before and after them under in-situ simulation; wherein, in-situ means running the program on a high-performance computing cluster and only saving intermediate or result images or data features, but not the original data set or intermediate data set.

[0080] S500, based on the high-precision streamline image and the types of all the key events, a model combining a recurrent neural network and a long short-term memory artificial neural network is used to generate key analysis results of the evolution of the ocean vortex.

[0081] It can be understood that step S500 includes steps S501, S502, S503, S504, and S505, wherein:

[0082] S501, determining the event type to which the key event belongs according to the key event;

[0083] S502, extracting keyword information corresponding to the key event according to the event type and preset feature information;

[0084] S503, generating a keyword table according to the keyword information and the high-precision streamline image;

[0085] S504, converting the keyword list into a narrative text description result through a model combining a recurrent neural network and a long short-term memory artificial neural network;

[0086] S505, generating key analysis results of ocean vortex evolution from the narrative text description results.

[0087] It should be noted that workflow mining uses the Heuristics Miner algorithm to generate event descriptions for processes with a low degree of structure. The basic idea of ​​this method is to mine dependency graphs, split and connect relationships. By setting case patterns and setting objects, key monitoring information such as the radius, energy and type of the vortex is processed by the heuristic workflow mining algorithm. Using the dimensional information of the event, such as the merger and extinction of the vortex, a description of the event can be obtained.

[0088] Based on workflow mining, keywords of events can be derived, and then an event description table can be constructed. The table usually consists of fields and values.

[0089] Generally speaking, generating natural language from tables is now a common method in machine learning. The machine learning model of Cao et al. is mainly used to achieve the work of generating natural language from tables. The specific model is mainly composed of a two-stage LSTM-RNN (Long Short-Term Memory, Recurrent Neural Network) convolutional neural network (CNN) and a long short-term memory artificial neural network (LSTM) encoder.

[0090] The entire system is controlled based on user interaction. The in-situ calculation module is responsible for vortex identification and tracking, and outputs statistical data and high-precision streamlines for the corresponding area according to the user's selection. The workflow module extracts keyword information of the corresponding event based on the corresponding event and related features, forms a keyword table, and provides it to the machine learning processing module. The machine learning processing module mainly uses the RNN+LSTM model to convert the keyword table into a narrative text description. The visual interaction module provides users with images, texts, features and other information from other modules for visual analysis.

[0091] In this example, the AVISO dataset and our system were used to identify and track eddies in the North Pacific, and the spatiotemporal characteristics of regional eddy distribution and evolution were analyzed.

[0092] Based on relevant laws in the ocean field, experts divide the North Pacific region into seven areas, including the Oyashio Current (OC), Kuroshio Extension (KE), Subtropical Countercurrent (STCC), North Equatorial Current (NEC), Alaska Current (AC), California Current (CC) and Northeast Tropical Pacific (NETP).

[0093] Experts tracked and analyzed all vortices generated in the North Pacific from 1993 to 2014. Based on the analysis results of the histogram, it was concluded that more than half of the vortices have a lifespan of less than 10 weeks. For the average radius of the vortex, the cyclonic vortex is 5 km smaller than the anticyclonic vortex, and the average radius of all vortices is about 100 km. Experts analyzed the generation of vortices from a seasonal perspective and found that there are obvious seasonal changes in the generation of vortices, with more vortices in spring and fewer vortices in autumn. In addition, through the identification and tracking of vortices in different regions of the North Pacific, it was found that the STCC region generated the most vortices. Based on this result, combined with on-site analysis, experts believe that it is caused by the interaction between wind stress and ocean currents in the region.

[0094] In this case study, experts tracked the evolution of the Red Sea vortex life cycle by analyzing the Red Sea ensemble data, focusing on the vortices that split and merged during the evolution process. In addition, by comparing the vortex merger and splitting events, experts found that the vortex's radius, vorticity and other physical information changed significantly when the corresponding events occurred. We selected the evolution of two vortices to demonstrate the visualization of our system for vortex tracking evolution.

[0095] Vortex No. 17 on day 0, an anticyclonic vortex, after 2 days of evolution, merged with the new vortex that appeared on the 2nd day. The merged vortex split into two vortices on the 5th day, and the two vortices merged into the same vortex one day later. After that, the vortex continued to evolve, and no vortex split or merged with other vortices. By the 39th day, the vortex disappeared. Before the first merger, the scale (radius) of the vortex showed small-scale fluctuations. On the 3rd day, the vortex increased significantly after merging because the two vortices were of the same type. Subsequently, the vortex radius continued to increase. On the 6th day, the average radius decreased significantly due to the splitting of the vortex. After the split and the second merger, the radius was relatively stable, decreased on the 21st day, and then slowly disappeared. In addition, by selecting the corresponding nodes of the tracking graph, observing the local streamline characteristics at the corresponding time, the streamline morphology analysis was performed. On the 4th day, the streamlines in the green area had not yet formed a vortex structure, but some streamlines had already shown different flow trends. On the 5th day, the vortex basically began to split, and the central core of the new vortex began to form. By the sixth day, the split was essentially complete.

[0096] On day 0, the 28th and 30th cyclonic vortices each evolved for a period of time and merged on day 7. The merged vortex split into two vortices one day later, and then merged into the same vortex one day later. After that, the vortex continued to evolve, split into two vortices on day 40, and then merged quickly. On day 55, a similar situation occurred again.

[0097] Based on the visualization results of multiple vortices, experts in this field believe that due to the narrowness of the Red Sea, it is difficult for vortices to split again after merging. Even if they split again, the split vortices tend to merge quickly and are difficult to evolve and dissipate. In addition, experts also found that each time the evolving vortices merge, they usually merge and split in a short period of time. Based on this, experts believe that vortex mergers are usually a process, during which factors such as energy collisions can cause vortices to become unstable in a short period of time.

[0098] The technical effects produced by this embodiment are:

[0099] 1. IO performance: In the in-situ visualization framework, for one time step, only 498KB is needed to store an image of the North Pacific region with a resolution of 4160×1600. At the same time, the corresponding pipeline data is saved for comparison, and 381MB (using VTK binary format) is required to save the pipeline data.

[0100] 2. Time performance: The time required to store the corresponding image and the corresponding streamline file is 0.974s and 9.533s respectively.

[0101] 3. Interactive performance: The interactive time of switching flow charts at different times is about 1 second, while if the original storage method is used, the interactive time is at least 10 seconds.

[0102] Embodiment 2:

[0103] like Figure 2 As shown, this embodiment provides an ocean vortex evolution analysis device, see Figure 2 The device includes an extraction module 701, a mining module 702, an identification module 703, a determination module 704 and a generation module 705, wherein:

[0104] Extraction module 701: used to extract process information from the ocean vortex historical sample library and construct an ocean vortex process model;

[0105] Mining module 702: used to mine the ocean vortex process model using a heuristic workflow algorithm to extract key events in the evolution of the ocean vortex, including the merging, splitting, generation and extinction of the ocean vortex;

[0106] Identification module 703: used to input the acquired ocean flow field data into a preset high-performance computing cluster for analysis, identify and track ocean vortices according to vortex detection methods and tracking methods;

[0107] Determination module 704: used to determine all key events in the evolution cycle of the ocean vortex according to the identified ocean vortex and key events, and store the time points of all key events and the high-precision streamline images before and after the time points;

[0108] Generation module 705: used to generate key analysis results of ocean vortex evolution based on high-precision streamline images and the types of all key events using a model that combines a recurrent neural network and a long short-term memory artificial neural network.

[0109] Preferably, the extraction module 701 includes a data set building module 706, a calculation module 707 and an optimization module 708, wherein:

[0110] A data set building module 706 is used to identify global ocean vortices using a narrative visualization method and build an ocean vortex data set, where the ocean vortex data set includes flow field vector information, temperature information, and salinity information of the ocean vortex;

[0111] Calculation module 707: used to calculate the ocean vortex process model according to the Heuristics Miner algorithm and the ocean vortex data set;

[0112] Optimization module 708: used to analyze the ocean vortex process model to obtain an optimized ocean vortex process model.

[0113] Preferably, the mining module 702 includes a selection unit 7021, a first acquisition unit 7022, a processing unit 7023, a merging unit 7024 and a construction unit 7025, wherein:

[0114] The selection unit 7021 is used to select the first object as a case object;

[0115] A first acquisition unit 7022 is used to acquire first information, where the first information includes the radius, energy and type information of the ocean vortex of the first object;

[0116] Processing unit 7023: used to process the first information through a heuristic workflow mining algorithm;

[0117] A merging unit 7024 is used to derive a keyword of a key event according to the processed first information and the merging and disappearance information of the ocean vortex in the first object;

[0118] Construction unit 7025: used to construct a description table of key events based on keywords.

[0119] Preferably, the identification module 703 includes a first generating unit 7031, a filtering unit 7032, a visualization unit 7033, a display unit 7034, a recording unit 7035 and an identification unit 7036, wherein:

[0120] The first generating unit 7031 is used to generate streamlines and vortex tracking according to the specific sea area to be selected;

[0121] Filtering unit 7032: used to filter out all ocean vortices that split and merge during the evolution of the ocean vortex;

[0122] Visualization unit 7033: used to use the filtered ocean vortex as a visualization object;

[0123] Display unit 7034: used to display the events of each ocean vortex in the visualization object changing over time;

[0124] Recording unit 7035: used to record each node data in the event as ocean flow field data;

[0125] Identification unit 7036: used to analyze and identify ocean flow field data through a high-performance computing cluster to obtain identified ocean vortex data.

[0126] Preferably, the identification module 703 further includes a second information acquisition unit 7037, a third information acquisition unit 7038, a fourth information acquisition unit 7039 and a data information unit 7040, wherein:

[0127] The second information acquisition unit 7037 is used to acquire the second information, where the second information includes the normal component and shear component of the ocean velocity field and the relative vorticity of the flow field;

[0128] The third information acquisition unit 7038 is used to acquire the third information by numerical simulation method according to the second information, where the third information is the Okubo-Weiss parameter;

[0129] The fourth information acquisition unit 7039 is used to obtain the fourth information by calculation according to the third information, where the fourth information is the ocean vortex parameter, the sea level height and the velocity vector information;

[0130] Data information unit 7040: used to calculate the ocean flow field data information according to the fourth information and the tracking method.

[0131] Preferably, the generating module 705 includes a determining unit 7051, an extracting unit 7052, a second generating unit 7053, a transforming unit 7054 and a third generating unit 7055, wherein:

[0132] Determining unit 7051: used to determine the event type to which the key event belongs according to the key event;

[0133] Extraction unit 7052: used to extract keyword information corresponding to the key event according to the event type and preset feature information;

[0134] The second generating unit 7053 is used to generate a keyword table according to the keyword information and the high-precision streamline image;

[0135] Conversion unit 7054: used to convert the keyword list into a narrative text description result through a model combining a recurrent neural network and a long short-term memory artificial neural network;

[0136] The third generation unit 7055 is used to generate key analysis results of ocean vortex evolution from narrative text description results.

[0137] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0138] Embodiment 3:

[0139] Corresponding to the above method embodiment, an ocean vortex evolution analysis device is also provided in this embodiment. The ocean vortex evolution analysis device described below and the ocean vortex evolution analysis method described above can refer to each other.

[0140] Figure 3 FIG. 8 is a block diagram of an ocean vortex evolution analysis device 800 according to an exemplary embodiment. Figure 3As shown, the ocean vortex evolution analysis device 800 may include: a processor 801 and a memory 802. The ocean vortex evolution analysis device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0141] The processor 801 is used to control the overall operation of the ocean vortex evolution analysis device 800 to complete all or part of the steps in the above-mentioned ocean vortex evolution analysis method. The memory 802 is used to store various types of data to support the operation of the ocean vortex evolution analysis device 800, and these data may include, for example, instructions for any application or method operating on the ocean vortex evolution analysis device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the ocean vortex evolution analysis device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0142] In an exemplary embodiment, the ocean vortex evolution analysis device 800 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned ocean vortex evolution analysis method.

[0143] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned ocean vortex evolution analysis method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the ocean vortex evolution analysis device 800 to complete the above-mentioned ocean vortex evolution analysis method.

[0144] Embodiment 4:

[0145] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the ocean vortex evolution analysis method described above can refer to each other.

[0146] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the ocean vortex evolution analysis method of the above method embodiment.

[0147] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0148] In summary, the beneficial effects of the present invention are the use of in-situ visualization to process simulation analysis in large-scale flow fields; the ability to generate high-precision vortex streamline structures in real time, and support vortex statistical analysis and tracking analysis in different ocean regions and at different scales; the efficiency and effectiveness of the system are demonstrated; and it can help experts track and analyze the evolution of ocean vortices.

[0149] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0150] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art 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 analyzing ocean eddy evolution, characterized in that: include: Extract process information from the ocean eddy historical sample library and build an ocean eddy process model; Using a heuristic workflow algorithm to mine the ocean vortex process model, extract key events in the evolution of the ocean vortex, wherein the key events include the merging, splitting, generation and extinction of the ocean vortex; The acquired ocean flow data is input into the preset high-performance computing cluster for analysis, and the ocean vortex is identified and tracked according to the vortex detection and tracking methods; Determine all key events in the evolution cycle of the ocean vortex according to the identified ocean vortex and the key events, and store the time points of all the key events and high-precision streamline images before and after the time points; Based on the high-precision streamline images and the types of all the key events, a model combining a recurrent neural network and a long short-term memory artificial neural network is used to generate key analysis results of the evolution of ocean vortices; The method of mining the ocean vortex process model using a heuristic workflow algorithm to extract key events in the evolution of the ocean vortex includes: Select the first object as the case object; Acquiring first information, wherein the first information includes acquiring information on the radius, energy, and type of the ocean vortex of the first object; Processing the first information by using a heuristic workflow mining algorithm; Derived keywords of the key event according to the processed first information and the merging and disappearance information of the ocean vortex in the first object; Constructing a description table of the key events according to the keywords; The acquired ocean flow field data is input into a preset high-performance computing cluster for analysis, and the ocean vortex is identified and tracked according to the vortex detection method and tracking method, including: Generate streamlines and vortex tracking according to the specific sea area to be selected; filtering out all ocean eddies that split and merge during the evolution of the ocean eddies; Using the filtered ocean vortex as a visualization object; Displaying the events of each ocean eddy in the visualization object over time; Recording each node data in the event as the ocean flow field data; The ocean flow field data is analyzed and identified by the high performance computing cluster to obtain the identified ocean vortex data.

2. The method for analyzing ocean vortex evolution according to claim 1, characterized in that According to the high-precision streamline image and the types of all the key events, a model combining a recurrent neural network and a long short-term memory artificial neural network is used to generate key analysis results of the evolution of ocean vortices, including: Determine the event type to which the key event belongs according to the key event; Extracting keyword information corresponding to the key event according to the event type and preset feature information; Generate a keyword table according to the keyword information and the high-precision streamline image; The keyword list is converted into a narrative text description result through a model combining a recurrent neural network and a long short-term memory artificial neural network; The narrative text description results are used to generate key analytical results of ocean eddy evolution.

3. An ocean vortex evolution analysis device, characterized in that: include: Extraction module: used to extract process information from the ocean vortex historical sample library and build an ocean vortex process model; A mining module is used to mine the ocean vortex process model using a heuristic workflow algorithm to extract key events in the evolution of the ocean vortex, wherein the key events include the merging, splitting, generation and extinction of the ocean vortex; Identification module: used to input the acquired ocean flow field data into a preset high-performance computing cluster for analysis, identify and track ocean vortices based on vortex detection and tracking methods; A determination module is used to determine all key events in the evolution cycle of the ocean vortex according to the identified ocean vortex and the key events, and store the time points of all the key events and the high-precision streamline images before and after the time points; A generation module is used to generate key analysis results of ocean vortex evolution based on the high-precision streamline image and the types of all the key events using a model combining a recurrent neural network and a long short-term memory artificial neural network; Wherein, the mining module includes: A selection unit: used for selecting the first object as a case object; A first acquisition unit is used to acquire first information, wherein the first information includes obtaining the radius, energy and type information of the ocean vortex of the first object; A processing unit: used for processing the first information by using a heuristic workflow mining algorithm; A merging unit: used to derive a keyword of the key event according to the processed first information and the merging and disappearance information of the ocean vortex in the first object; A construction unit: used to construct a description table of the key events according to the keywords; Wherein, the identification module includes: The first generating unit is used to generate streamlines and vortex tracking according to the specific sea area to be selected; A filtering unit is used to filter out all ocean vortices that split and merge during the evolution of the ocean vortex; A visualization unit: used for taking the filtered ocean vortex as a visualization object; Display unit: used for displaying the events of each ocean vortex in the visualization object changing with time; Recording unit: used for recording each node data in the event as the ocean flow field data; Identification unit: used for analyzing and identifying the ocean flow field data through the high-performance computing cluster to obtain the identified ocean vortex data.

4. The ocean vortex evolution analysis device according to claim 3, characterized in that: The generating module comprises: a determining unit: used for determining the event type to which the key event belongs according to the key event; Extraction unit: used for extracting keyword information corresponding to the key event according to the event type and preset feature information; A second generating unit is used to generate a keyword table according to the keyword information and the high-precision streamline image; A conversion unit is used to convert the keyword list into a narrative text description result through a model combining a recurrent neural network and a long short-term memory artificial neural network; The third generating unit is used to generate key analysis results of ocean vortex evolution from the narrative text description results.

5. An ocean vortex evolution analysis device, characterized in that: include: Memory for storing computer programs; A processor, used to implement the steps of the ocean vortex evolution analysis method as described in any one of claims 1 to 2 when executing the computer program.

6. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the ocean vortex evolution analysis method as described in any one of claims 1 to 2 are implemented.

Images