Method for establishing ship climate route database

Abstract

This application provides a method for establishing a ship climate route database, which includes: acquiring historical ocean climate record files, historical ship route record files, and historical ship voyage record files; parsing the historical ocean climate record files, historical ship route record files, and historical ship voyage record files respectively to construct an ocean climate description ontology, a historical ship route description ontology, and a historical ship voyage description ontology; generating ship climate route data standard data based on unified data description according to the ocean climate description ontology, historical ship route description ontology, and historical ship voyage description ontology; and establishing a mapping relationship between the ship climate route data standard data and ship integrated codes to establish a ship climate route database, which at least improves the correlation and consistency between data and the efficiency of data processing.

Description

Technical Field

[0001] The present application relates to the field of safety technology, and in particular to a method for establishing a ship climate route database. Background Art

[0002] In the shipping industry, the choice of ship climate routes plays a vital role in ensuring navigation safety, improving navigation efficiency and reducing operating costs. However, there are still many challenges in the processing and use of ship climate route data.

[0003] First, key data such as historical ocean climate records, historical ship route records, and historical ship voyage records are usually stored in different formats and standards, resulting in poor relevance and consistency between data. This inconsistency of data not only increases the difficulty of data processing, but also affects the accuracy and reliability of data analysis.

[0004] Secondly, the existing ship climate route database often lacks a unified data description standard, making it difficult to effectively integrate and share data from different sources. This data island phenomenon limits the mining and utilization of data value and cannot meet the shipping industry's demand for comprehensive and comprehensive climate route data. Summary of the invention

[0005] The purpose of this application is to propose a method for establishing a ship climate route database to solve or alleviate the technical problems existing in the prior art.

[0006] The present application embodiment provides a method for establishing a ship climate route database, which includes:

[0007] Obtain historical ocean climate record files, historical ship route record files, and historical ship voyage record files;

[0008] Respectively parse the historical ocean climate record files, the ship historical route record files, and the ship historical voyage record files to construct an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology;

[0009] Generate ship climate route data standard data based on unified data description according to the ocean climate description ontology, ship historical route description ontology, and ship historical voyage description ontology;

[0010] A mapping relationship between the ship climate route data standard data and the ship comprehensive code is established to establish a ship climate route database.

[0011] In the embodiment of the present application, historical ocean climate record files, ship historical route record files and ship historical voyage record files are obtained, and these data are the key basis for climate route selection in the shipping field. By parsing these data files from different sources and in different formats, the ocean climate description ontology, the ship historical route description ontology and the ship historical voyage description ontology are constructed. This step is actually to standardize the data, convert the originally scattered and heterogeneous data into data entities with unified description standards, thereby improving the relevance and consistency between the data. In addition, after constructing each description ontology, the scheme further generates ship climate route data standard data based on unified data description according to these ontologies. This step is actually to achieve data integration, integrating the originally scattered and different description standards into a data set with a unified description standard, thereby breaking the data island phenomenon and enabling data from different sources to be effectively shared and utilized. Finally, the scheme establishes a mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database. This step is actually to achieve refined management and efficient use of data. Through the unique identification of the ship comprehensive code, data can be easily tracked and managed, improving the efficiency of data processing. At the same time, since the data has been standardized and integrated, its utilization value has been significantly improved, which can better meet the shipping industry's demand for comprehensive and comprehensive climate route data. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Hereinafter, some specific embodiments of the present application will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0013] Figure 1 The present invention is a flowchart of a method for establishing a ship climate route database according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.

[0015] Figure 1 The present invention is a flowchart of a method for establishing a ship climate route database according to an embodiment of the present application.

[0016] like Figure 1 As shown, it includes:

[0017] Obtain historical ocean climate record files, historical ship route record files, and historical ship voyage record files;

[0018] Respectively parse the historical ocean climate record files, the ship historical route record files, and the ship historical voyage record files to construct an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology;

[0019] Generate ship climate route data standard data based on unified data description according to the ocean climate description ontology, ship historical route description ontology, and ship historical voyage description ontology;

[0020] A mapping relationship between the ship climate route data standard data and the ship comprehensive code is established to establish a ship climate route database.

[0021] To this end, in the embodiment of the present application, historical ocean climate record files, ship historical route record files and ship historical voyage record files are obtained, and these data are the key basis for climate route selection in the shipping field. By parsing these data files from different sources and in different formats, the ocean climate description ontology, the ship historical route description ontology and the ship historical voyage description ontology are constructed. This step is actually to standardize the data, convert the originally scattered and heterogeneous data into data entities with unified description standards, thereby improving the relevance and consistency between the data. In addition, after constructing each description ontology, the scheme further generates ship climate route data standard data based on unified data description according to these ontologies. This step is actually to achieve data integration, integrating the originally scattered and different description standards into a data set with a unified description standard, thereby breaking the data island phenomenon and enabling data from different sources to be effectively shared and utilized. Finally, the scheme establishes a mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database. This step is actually to achieve refined management and efficient use of data. Through the unique identification of the ship comprehensive code, data can be easily tracked and managed, improving the efficiency of data processing. At the same time, since the data has been standardized and integrated, its utilization value has been significantly improved, which can better meet the shipping industry's demand for comprehensive and comprehensive climate route data.

[0022] Optionally, respectively parsing the historical ocean climate record files, the ship historical route record files, and the ship historical voyage record files to construct an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology includes:

[0023] Respectively analyzing the historical ocean climate record files, the historical ship route record files, and the historical ship voyage record files to generate historical ocean climate data item sets, historical ship route data item sets, and historical ship voyage data item sets;

[0024] According to the historical ocean climate data item set, the ship historical route data item set, and the ship historical voyage data item set, an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology are constructed.

[0025] To this end, the above technical solution has the following technical advantages:

[0026] (1) By parsing the historical ocean climate record files, historical ship route record files, and historical ship voyage record files, historical ocean climate data item sets, historical ship route data item sets, and historical ship voyage data item sets were generated. This step achieves the refined processing of the original data and splits the data into units that are easier to manage and use. The generation of data item sets makes the originally scattered and disordered data structured, which facilitates the subsequent data integration, analysis, and utilization.

[0027] (2) Based on the generated data item set, the ocean climate description ontology, ship historical route description ontology, and ship historical voyage description ontology were constructed. The construction of ontology is actually a higher-level knowledge representation of data, which transforms data into entities and concepts with clear meanings and relationships. The introduction of ontology has significantly improved the relevance and consistency between data, which helps to better understand and use data.

[0028] (3) By constructing a descriptive ontology, the historical ocean climate data, historical ship route data, and historical ship voyage data are integrated. This step breaks the data island phenomenon and enables data from different sources and formats to be effectively shared and utilized. The improvement of data integration and sharing capabilities will help the shipping industry to more comprehensively grasp climate and route information and make more accurate decisions.

[0029] (4) The construction of structured data item sets and description ontology reduces the difficulty and complexity of subsequent data processing. Data processing personnel can query, analyze and mine data more conveniently, improving data processing efficiency. At the same time, since the data has been refined and organized, its utilization value has also been significantly improved. The shipping industry can better use this data to make decisions in terms of climate route selection, navigation efficiency improvement and operating cost reduction.

[0030] The following is a simplified example code to illustrate how to parse the historical ocean climate record file and generate a data item set:

[0031]

[0032]

[0033] In the above example, regular expressions are used to extract data items from the historical marine climate record file and store them in the data_items dictionary. Each data item (such as location, date, and temperature) is added to a list for subsequent processing and analysis.

[0034] Optionally, the historical ocean climate record file, the ship historical route record file, and the ship historical voyage record file are parsed respectively to generate a historical ocean climate data item set, a ship historical route data item set, and a ship historical voyage data item set, including:

[0035] Performing lexical analysis on the historical ocean climate record files, the historical ship route record files, and the historical ship voyage record files respectively to obtain a plurality of words or phrases, and performing part-of-speech tagging on the plurality of words or phrases;

[0036] Performing syntactic analysis on the multiple words or phrases that have been tagged with parts of speech, including dependency syntactic analysis and generative syntactic analysis, to generate a structure tree of dependency relationships between the words and / or phrases;

[0037] Based on the structure tree of dependency relationships between the words and / or phrases, context analysis and role labeling are performed on the multiple words or phrases to generate historical ocean climate data item sets, historical ship route data item sets, and historical ship voyage data item sets.

[0038] To this end, in combination with the technical solution, it has the following technical benefits:

[0039] (1) Through lexical analysis, the original text data is broken down into smaller units (words or phrases), and these units are tagged with part of speech. This step helps to understand the meaning and structure of the data more accurately, because different words may have different meanings and usages, and part of speech tagging can help distinguish them. For example, in marine climate records, "strong" and "wind" are two different words. Through lexical analysis and part of speech tagging, it can be made clear that "strong" is an adjective that means the strength of the wind, while "wind" is a noun that means one of the climate elements.

[0040] (2) Syntactic analysis (including dependency syntactic analysis and generative syntactic analysis) can generate a structure tree of dependency relationships between words and / or phrases. This step clearly shows the associations and hierarchical relationships between data, making the data structure more intuitive and easy to understand. For example, in a ship route record, "from Shanghai to New York" can be analyzed through dependency syntactic analysis to make it clear that "Shanghai" is the starting point and "New York" is the end point. The dependency relationship between them and "from" and "to" is clear at a glance, making it easier to understand the structure of the entire route.

[0041] (3) Context analysis and role labeling based on the structure tree can generate structured data item sets. These data item sets organize data into a more structured and easy-to-use form, improving the usability and value of the data. For example, in the historical voyage records, "On March 15, 2023, Ship A completed the voyage from Shanghai to New York" can generate a structured data item set containing information such as date, ship name, starting point, and end point through context analysis and role labeling, making subsequent data processing and analysis more efficient.

[0042] Improved data processing efficiency and cost-effectiveness:

[0043] Through the above technical solutions, key data is effectively parsed and organized into structured data item sets, which reduces the difficulty and complexity of subsequent data processing, improves the efficiency of data processing, and reduces manpower and time costs.

[0044] The following is an example code to illustrate the implementation of the above process:

[0045]

[0046] Optionally, constructing an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology according to the historical ocean climate data item set, the ship historical route data item set, and the ship historical voyage data item set includes:

[0047] Extracting historical ocean climate entities, historical ocean climate attributes, and historical ocean climate relationships from the historical ocean climate data item set to construct an ocean climate description ontology;

[0048] Extracting ship historical route entities, ship historical route attributes, and ship historical route relationships from the ship historical route data item set to construct a ship historical route description ontology;

[0049] The ship historical voyage data item set is subjected to ship historical voyage entity, ship historical voyage attribute, and ship historical voyage relationship extraction to construct a ship historical voyage description ontology.

[0050] To this end, the above process of constructing the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology has the following technical benefits:

[0051] (1) By extracting entities, attributes, and relationships from data item sets, more accurate and rich knowledge representation models can be constructed. These models can more accurately describe the actual conditions of ocean climate, ship routes, and voyages, providing a solid foundation for subsequent data analysis and application. For example, in the ocean climate description ontology, the associations and mutual influences between different climate elements (such as temperature, humidity, wind speed, etc.) can be clearly expressed, which is crucial for understanding climate change patterns and predicting future climate trends.

[0052] (2) The construction of ontology helps to break down the barriers between different data sources and realize data integration and sharing. By defining common entities and attributes, data from different sources can be aligned and merged to form a unified data view. For example, in the ontology of ship historical route description, the route data provided by different shipping companies can be integrated to form a comprehensive route network, providing strong support for planning and decision-making in the shipping industry.

[0053] (3) Ontology provides rich semantic information, enabling more intelligent and accurate query and reasoning. By utilizing the entities, attributes, and relationships in the ontology, complex query statements can be constructed to achieve information retrieval and integration across data sources. For example, in the ontology of ship historical voyage descriptions, all voyage information of a specific ship within a specific time period can be queried, or the average sailing speed and most commonly used routes of a certain ship can be inferred.

[0054] (4) The ontology construction process includes operations such as data cleaning, verification, and standardization, which helps to ensure the quality and consistency of the data. By defining clear entities and attributes, the data can be effectively constrained and verified, reducing the occurrence of errors and outliers. For example, in the marine climate description ontology, the range and unit of temperature can be defined to ensure that all climate data meet this standard, thereby improving the accuracy and reliability of the data.

[0055] The following is an example code to illustrate the implementation process of the above solution:

[0056]

[0057]

[0058] Optionally, generating ship climate route data standard data based on unified data description according to the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology includes:

[0059] Multi-layer adaptive aggregation and multi-dimensional feature analysis are performed on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate standard data of ship climate route data based on unified data description.

[0060] To this end, the above process of generating standard data of ship climate route data based on unified data description has the following technical benefits:

[0061] (1) Through multi-layer adaptive aggregation and multi-dimensional feature analysis, data from different ontologies (ocean climate, ship routes, ship voyages) can be unified and standardized. This ensures the consistency and comparability of the data, providing a solid foundation for subsequent data analysis and application. For example, different units or formats may be used in different ontologies to represent the same climate element (such as temperature). Through standardization, these data can be unified into the same unit and format, thus avoiding problems caused by inconsistent data.

[0062] (2) The ontology aggregation and feature analysis process can integrate information from different data sources to generate a richer and more complete data set. This helps to reveal hidden relationships and patterns between data, providing the possibility for deeper data mining and analysis. For example, by integrating ocean climate data and ship route data, the navigation patterns and efficiency of ships under specific climate conditions can be analyzed, providing strong support for the optimization of the shipping industry.

[0063] (3) The generated standard data has better interpretability and comprehensibility. Through multi-layer adaptive aggregation and multi-dimensional feature analysis, the data can be converted into a more intuitive and easy-to-understand form, so that non-professionals can also easily understand and use the data. For example, by visualizing the navigation of ships under different climatic conditions, the impact of climate on navigation can be intuitively seen without the need to deeply understand complex climate and navigation data.

[0064] (4) Standardized data can be more easily used by various data analysis and decision support systems. Whether it is data mining, machine learning or visualization, standard data can provide better support and results. For example, in the decision support system of the shipping industry, standardized ship climate route data can be directly used for tasks such as route planning and ship scheduling, improving the accuracy and efficiency of decision-making.

[0065] (5) Standard data formats and descriptions facilitate data sharing and interaction between different systems. By following unified data standards, different systems can easily exchange and share data, achieving information interoperability and collaboration. For example, sharing standardized ship climate route data among multiple related systems in the shipping industry can achieve more efficient information transmission and collaborative work.

[0066] Another exemplary code is provided below to illustrate the implementation of the above-mentioned data processing process for generating the ship climate route data standard based on the unified data description:

[0067]

[0068]

[0069] Optionally, the performing of multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate ship climate route data standard data based on unified data description includes:

[0070] Performing spatial feature extraction and temporal feature extraction on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology respectively;

[0071] The extracted spatial features and temporal features are aligned to perform multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate standard ship climate route data based on unified data description.

[0072] To this end, the above process of multi-layer adaptive aggregation and multi-dimensional feature analysis of the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology has the following technical benefits:

[0073] (1) It can provide a deeper understanding of the distribution and variation of data in spatial and temporal dimensions. This helps to reveal hidden spatial and temporal patterns and provide rich feature information for subsequent data analysis and application. For example, spatial feature extraction can help discover the climate characteristics of a specific sea area, such as the spatial distribution of temperature and wind speed; while temporal feature extraction can reveal the changing trends of these climate characteristics over time, such as seasonal changes and long-term trends.

[0074] (2) Aligning the extracted spatial and temporal features can ensure that the data between different entities are consistent in time and space, thereby facilitating multi-layer adaptive aggregation and multi-dimensional feature analysis. This alignment and integration process helps to eliminate inconsistencies and redundancies between data and improve the overall quality of the data. For example, through feature alignment, ship route data can be matched with ocean climate data in time and space, thereby analyzing the navigation of ships under specific climate conditions.

[0075] (3) Multidimensional feature analysis can comprehensively consider space, time and other possible dimensions (such as climate type, ship type, etc.), thereby revealing the complex relationship between data in a more comprehensive way. This comprehensive analysis helps to gain a deeper understanding of the inherent laws and patterns of ship climate route data. For example, through multidimensional feature analysis, multiple dimensions such as climate type, sea area location, and ship type can be considered simultaneously to analyze the navigation efficiency and safety of ships under different conditions.

[0076] (4) Generating standard data of ship climate route data based on unified data description will help achieve data standardization and uniformity. Such standardized and uniform data can be more easily used by different systems and tools, improving the shareability and reusability of data. For example, standardized ship climate route data can be adopted and used by different shipping industry systems, thus achieving data consistency and comparability.

[0077] (5) Data that has undergone multi-layer adaptive aggregation and multi-dimensional feature analysis can be more easily applied to various data analysis and decision support systems. Whether it is data mining, machine learning or visualization, these data can provide better support and results. For example, in the decision support system of the shipping industry, ship climate route data that has been deeply analyzed and processed can be used in tasks such as route planning, ship scheduling, and risk assessment to improve the accuracy and efficiency of decision-making.

[0078] The following is an example code to illustrate the above process of feature extraction and multi-layer adaptive aggregation of ocean climate, ship routes and voyages:

[0079]

[0080]

[0081]

[0082] Optionally, the extracted spatial features and temporal features are aligned to perform multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate ship climate route data standard data based on unified data description, including:

[0083] Based on the extracted spatial features and temporal features, the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology are aligned, and then the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology are vectorized to obtain an ocean climate representation vector, a ship historical route representation vector, and a ship historical voyage representation vector;

[0084] Based on the trained multi-level feature aggregation network, multi-layer adaptive aggregation and multi-dimensional feature analysis are performed on the ocean climate characterization vector, the ship historical route characterization vector, and the ship historical voyage characterization vector to generate standard ship climate route data based on a unified data description.

[0085] To this end, the above process of performing multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate standard data of ship climate route data based on unified data description has the following technical benefits:

[0086] (1) By vectorizing the ontology of ocean climate, ship historical routes and voyage descriptions, the unstructured ontology data can be converted into a structured vector form, which is convenient for subsequent computer processing and analysis. This vectorized representation method can more accurately capture and represent the key information in the data and improve the data representation ability.

[0087] (2) Using the trained multi-level feature aggregation network, the vectorized data can be adaptively aggregated in multiple layers. This aggregation method can fully consider the characteristics of the data at different levels, thereby achieving more comprehensive and in-depth feature extraction. Compared with single-level feature extraction, multi-level feature aggregation can more accurately reveal the complex relationship between data.

[0088] (3) When conducting multi-dimensional feature analysis, space, time and other possible dimensions are considered at the same time, so as to more comprehensively reveal the inherent laws and patterns between the data. This comprehensive analysis method helps to more deeply understand the essential characteristics of ship climate route data and provide strong support for subsequent data applications.

[0089] (4) By generating standard data of ship climate route data based on unified data description, data standardization and uniformity can be achieved. Such standardized and uniform data can be more easily used by different systems and tools, improving the shareability and reusability of data. At the same time, this also facilitates subsequent data analysis and application.

[0090] (5) Data that has undergone multi-layer adaptive aggregation and multi-dimensional feature analysis can be more easily applied to various data analysis and decision support systems. Whether it is data mining, machine learning or visualization, these data can provide better support and results. Especially in the decision support system of the shipping industry, the application of these data can significantly improve the accuracy and efficiency of decision-making.

[0091] The following is an exemplary code to illustrate the above implementation process of vectorizing ocean climate, ship historical routes and voyage description ontology, and using a multi-level feature aggregation network to perform multi-layer adaptive aggregation and multi-dimensional feature analysis:

[0092]

[0093]

[0094] Optionally, the multi-level feature aggregation network includes a fusion module, an aggregation module, and a stacked multi-layer spatiotemporal convolutional neural network module;

[0095] The multi-level feature aggregation network based on the training is used to perform multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate characterization vector, the ship historical route characterization vector, and the ship historical voyage characterization vector to generate standard data of ship climate route data based on a unified data description, including:

[0096] Based on the fusion module, the ocean climate representation vector, the ship historical route representation vector, and the ship historical voyage representation vector are fused to obtain a comprehensive fusion vector;

[0097] Based on the aggregation module, context aggregation is performed on the comprehensive fusion vector to obtain a multi-level feature pyramid;

[0098] Based on the stacked multi-layer spatiotemporal convolutional neural network module, multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies are performed on the multi-level feature pyramid to generate standard data of ship climate route data based on unified data description.

[0099] To this end, a multi-level feature aggregation network is used, including a fusion module, an aggregation module, and a stacked multi-layer spatiotemporal convolutional neural network module. Multi-layer adaptive aggregation and multi-dimensional feature analysis are performed on the ocean climate representation vector, the ship historical route representation vector, and the ship historical voyage representation vector, thereby generating a process of generating standard data of ship climate route data based on a unified data description. This process has the following technical benefits:

[0100] (1) The fusion module can effectively fuse different representation vectors of ocean climate, ship historical routes and voyages to form a comprehensive fusion vector. This fusion method can fully consider the complementarity between different data sources, thereby improving the comprehensive representation ability of the data. The aggregation module can generate a multi-level feature pyramid by contextually aggregating the comprehensive fusion vector. This multi-level feature representation method can better capture the hierarchical structure in the data, thereby enhancing the ability to understand and analyze the data.

[0101] (2) The stacked multi-layer spatiotemporal convolutional neural network module can fully consider the spatial and temporal dependencies in the data and perform deep multi-layer adaptive aggregation and multi-dimensional feature analysis. This deep spatiotemporal analysis method can more accurately reveal the complex patterns and laws in the data.

[0102] (3) Through the processing of multi-level feature aggregation networks, standard data of ship climate route data based on unified data description can be generated. This standardized data is easier to be used by different systems and tools, further improving the shareability and reusability of the data.

[0103] (4) The data processed by the multi-level feature aggregation network has higher quality and consistency, which can significantly improve the efficiency of subsequent data analysis and application. Whether in the decision support system of the shipping industry or in other related data analysis and mining tasks, these data can provide better support and results.

[0104] The exemplary code implemented using a multi-level feature aggregation network including a fusion module, an aggregation module, and a stacked multi-layer spatiotemporal convolutional neural network module is as follows:

[0105]

[0106]

[0107] Optionally, the stacked multi-layer spatiotemporal convolutional neural network module performs multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies on the multi-level feature pyramid to generate standard data of ship climate route data based on unified data description, including:

[0108] Based on the stacked multi-layer spatiotemporal convolutional neural network module, the multi-level feature pyramid is upsampled to obtain a multi-scale feature map, and multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies are performed based on the multi-scale feature map to generate standard data of ship climate route data based on unified data description.

[0109] To this end, the stacked multi-layer spatiotemporal convolutional neural network module upsamples the multi-level feature pyramid to obtain multi-scale feature maps, and performs multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies based on these feature maps, which has the following technical benefits:

[0110] (1) The multi-scale feature map obtained by upsampling can capture spatial and temporal features at different scales. This multi-scale feature extraction method is crucial for understanding complex and variable ship climate route data because it can consider both local and global information at the same time.

[0111] (2) Multi-layer adaptive aggregation based on spatial and temporal dependencies on multi-scale feature maps can more deeply explore the spatiotemporal patterns in the data. This deep spatiotemporal analysis helps to reveal hidden laws and trends in ship climate route data.

[0112] (3) Multi-level and multi-dimensional feature analysis can enhance the robustness of feature representation. By comprehensively analyzing features at different scales and levels, a more stable and reliable feature representation can be generated, which is crucial for subsequent data analysis and application.

[0113] (4) Through the processing of stacked multi-layer spatiotemporal convolutional neural network modules, data standardization can be further deepened. The generated ship climate route data standard based on unified data description has higher quality and consistency, which is convenient for sharing and reuse among different systems and tools.

[0114] (5) Data that has undergone multi-level and multi-dimensional feature analysis and standardization can provide higher quality input for subsequent data analysis, machine learning or deep learning tasks. This will directly improve the performance and effectiveness of these tasks, thereby promoting the development of related fields such as the shipping industry.

[0115] The following is an example code for implementing the stacked multi-layer spatiotemporal convolutional neural network module to process a multi-level feature pyramid scheme.

[0116]

[0117]

[0118] Optionally, the multi-level feature aggregation network further includes an attention module;

[0119] The stacked multi-layer spatiotemporal convolutional neural network module upsamples the multi-level feature pyramid to obtain a multi-scale feature map, and performs multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies based on the multi-scale feature map to generate standard data of ship climate route data based on a unified data description. The attention weights between the multi-scale feature maps are calculated, and based on the attention weights, the attention strength between the multi-scale feature maps is calculated to perform multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies.

[0120] To this end, in particular, the newly added attention module in the multi-level feature aggregation network, as well as the process of calculating attention weights for multi-scale feature maps in the stacked multi-layer spatiotemporal convolutional neural network module and performing multi-layer adaptive aggregation and multi-dimensional feature analysis based on these weights, bring the following technical benefits:

[0121] (1) The attention module can calculate the attention weights between multi-scale feature maps, which helps the network automatically select features that are more important to the task. Through this feature selection mechanism, the quality of features can be further improved and the impact of noise and redundant information can be reduced.

[0122] (2) Multi-layer adaptive aggregation based on attention weights can capture spatial and temporal dependencies more accurately. The attention mechanism ensures that more important features are given greater weights during the aggregation process, thereby improving the accuracy and reliability of the analysis.

[0123] (3) By introducing the attention module, a more robust feature representation can be generated. This is because the attention mechanism can adaptively adjust the contribution between different features and scales, making the feature representation more stable and less sensitive to noise and outliers.

[0124] (4) The attention module can accelerate the data standardization process. By focusing on more important features, the goal of data standardization can be achieved faster, thereby reducing the consumption of computing resources and time.

[0125] (5) The data enhanced by the attention module can provide higher quality input for subsequent data analysis, machine learning or deep learning tasks. This will directly improve the performance and effect of these tasks, especially in applications that require high accuracy and reliability.

[0126] The following is an example code for processing ship climate route data using a multi-level feature aggregation network (including an attention module)

[0127]

[0128]

[0129] Optionally, based on the following formula, the multi-level feature pyramid is upsampled to obtain a multi-scale feature map:

[0130] Among them, The kth multi-scale feature map is obtained after upsampling. is the i-th input feature map of the multi-level feature pyramid, W i and b are the weights and biases of the transposed convolutional layer, * represents the convolution operation, and σ is the activation function;

[0131] The attention weights between the multi-scale feature maps are calculated based on the following formula:

[0132] Among them, A ij is the attention weight of the i-th multi-scale feature map to the j-th multi-scale feature map, E ij is the calculated energy value of the i-th multi-scale feature map relative to the j-th multi-scale feature map, and m is the number of multi-scale feature maps;

[0133] The attention strength between the multi-scale feature maps is calculated based on the following formula:

[0134] Among them, s j represents the attention strength of the j-th multi-scale feature map, is the position of the k-th multi-scale feature map relative to the j-th multi-scale feature map after upsampling, and ⊙ represents element-wise multiplication;

[0135] Based on the following formula, multi-layer adaptive aggregation is performed:

[0136] F agg =AGG({S j}), where F agg is the feature representation after aggregation, AGG is the aggregation function, which is at least one of average pooling and maximum pooling;

[0137] Based on the following formula, multi-dimensional feature analysis is performed:

[0138] F final =W f *F agg +b f , where F final is the final feature representation, W f and b f are the weights and biases of the fully connected layers.

[0139] To this end, the upsampling operation can restore the spatial resolution from the deep features, which helps to retain more detailed information, which is crucial for accurate climate route prediction. The attention mechanism allows the model to dynamically focus on the most important parts when processing the input data. In ship climate route data, some features (such as climate data for a specific time period, navigation conditions in key sea areas, etc.) may be more important than other features. The attention mechanism can automatically learn and emphasize these key features, thereby improving the prediction performance of the model. Multi-layer adaptive aggregation operations can effectively merge the information of multiple feature maps into a single feature representation, which helps the model to be more efficient and accurate when processing complex data. Multi-dimensional feature analysis further allows for deeper processing and transformation of the aggregated features to capture more useful information in the data.

[0140] Optionally, before establishing the mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database, it also includes:

[0141] Obtain ship type data, size data, navigation area data, and climate adaptation data, and encode them to generate basic ship codes;

[0142] Adding a timestamp and a serial number to the basic ship code to generate an extended ship code;

[0143] A hash operation is performed on the ship extended code to generate the ship comprehensive code.

[0144] Therefore, before establishing the mapping relationship between the ship climate route data standard data and the ship comprehensive code, the generation process of ship basic code, extended code and comprehensive code is carried out, which has the following technical benefits:

[0145] (1) By acquiring data such as ship type, size, navigation area, and climate adaptation, and encoding them to generate a basic ship code, the standardization of ship data is achieved. This helps to eliminate redundancy and inconsistency in the data and improve the quality and availability of the data. Adding a timestamp and serial number to the basic code to generate a ship extended code further enhances the uniqueness and traceability of the data.

[0146] (2) The ship comprehensive code is a key identifier in the ship climate route database, which is generated by hash operation and has a fixed length and uniqueness. This makes it more efficient and faster to retrieve data for a specific ship in the database.

[0147] (3) Hash operation is a one-way function that can convert input of any length into output of fixed length, and this process is irreversible. Therefore, using hash operation to generate ship comprehensive code can protect the original data from being easily leaked, thus enhancing the security of data.

[0148] (4) Standardized and integrated ship data is easier to use for big data analysis and machine learning. As the unique identifier of the data, the ship's comprehensive code can be easily associated and integrated with other data sets, thus supporting more complex data analysis and mining tasks.

[0149] An exemplary code is provided below to illustrate the process of generating the ship comprehensive code:

[0150]

[0151]

[0152] Optionally, the establishing of a mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database includes:

[0153] The ship's comprehensive code is used as a primary key, and a composite primary key is generated with the ship's unique code, wherein the ship's unique code includes at least one of a ship registration code, an International Maritime Organization code, and a ship's radio call sign code;

[0154] Based on the object mapping framework, the composite primary key is mapped to an object in a database table, and the standard data attributes of the ship climate route data are mapped to predicates in the database table, so as to establish a mapping relationship between the standard data of the ship climate route data and the ship comprehensive code;

[0155] According to the mapping relationship, a ship climate route database is established.

[0156] To this end, by establishing a mapping relationship between the ship climate route data standard data and the ship comprehensive code, and establishing a ship climate route database based on this mapping relationship, the following technical benefits are brought about:

[0157] (1) By using the ship's comprehensive code and the ship's unique code (such as ship registration code, International Maritime Organization code, ship radio call sign code) to generate a composite primary key, the uniqueness and consistency of each record in the database is ensured. This helps avoid data duplication and conflict and improves data accuracy and reliability.

[0158] (2) The composite primary key, as a unique identifier in the database table, makes data retrieval and management more efficient. The composite primary key can be used to quickly locate a specific ship record and perform query, update or delete operations without traversing the entire database table.

[0159] (3) Using the object mapping framework to map the composite primary key to an object in the database table and to map the standard data attributes of the ship climate route data to predicates in the database table, complex data relationships can be easily represented and managed. This helps to achieve hierarchical and structured storage of data and improves data readability and maintainability.

[0160] (4) A unified ship climate route database has been established, making it easier to share and exchange ship data between different systems and platforms. This helps break down information silos, promote information sharing and cooperation within the shipping industry, and improve the efficiency and competitiveness of the entire industry.

[0161] (5) Structured and standardized ship climate route data is stored in the database, which is convenient for data analysis and mining. Through the analysis of the data, potential patterns, trends and associations can be discovered, providing strong support for decision-making in the shipping industry.

[0162] The exemplary code of the process of establishing a mapping relationship between the ship climate route data standard data and the ship comprehensive code, and establishing a ship climate route database based on the mapping relationship is as follows:

[0163]

[0164]

[0165] Optionally, the method further includes:

[0166] The ship climate route data in the ship climate route database is converted into embedded data, and the ocean risk prediction pre-model is trained to obtain an ocean risk prediction model, which is used to predict ocean risks based on ocean meteorological sea conditions data and real-time ship route data.

[0167] The expressions "first", "second", "the first" or "the second" used in various embodiments of the present disclosure may modify various components regardless of order and / or importance, but these expressions do not limit the corresponding components. The above expressions are only configured for the purpose of distinguishing an element from other elements. For example, a first user device and a second user device represent different user devices, although both are user devices. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of the present disclosure.

[0168] When one element (e.g., a first element) is referred to as being "(operably or communicatively) coupled" or "(operably or communicatively) coupled to" or "connected to" another element (e.g., a second element), it is understood that the one element is directly connected to the other element or the one element is indirectly connected to the other element via yet another element (e.g., a third element). Conversely, it is understood that when an element (e.g., a first element) is referred to as being "directly connected" or "directly coupled" to another element (the second element), no element (e.g., a third element) is interposed between the two.

[0169] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for establishing a ship climate route database, characterized in that: include: Obtain historical ocean climate record files, historical ship route record files, and historical ship voyage record files; Respectively parse the historical ocean climate record files, the ship historical route record files, and the ship historical voyage record files to construct an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology; Generate ship climate route data standard data based on unified data description according to the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology, including: perform multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate ship climate route data standard data based on unified data description; Establishing a mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database; Among them, the multi-layer adaptive aggregation and multi-dimensional feature analysis of the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology are performed to generate ship climate route data standard data based on unified data description, including: Performing spatial feature extraction and temporal feature extraction on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology respectively; The extracted spatial features and temporal features are aligned to perform multi-layer adaptive aggregation and multi-dimensional feature analysis on the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology to generate standard data of ship climate route data based on unified data description, including: Based on the extracted spatial features and temporal features, the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology are aligned, and then the ocean climate description ontology, the ship historical route description ontology, and the ship historical voyage description ontology are vectorized to obtain an ocean climate representation vector, a ship historical route representation vector, and a ship historical voyage representation vector; Based on the trained multi-level feature aggregation network, including a fusion module, an aggregation module, and a stacked multi-layer spatiotemporal convolutional neural network module, multi-layer adaptive aggregation and multi-dimensional feature analysis are performed on the ocean climate characterization vector, the ship historical route characterization vector, and the ship historical voyage characterization vector to generate standard data of ship climate route data based on a unified data description, including: Based on the fusion module, the ocean climate representation vector, the ship historical route representation vector, and the ship historical voyage representation vector are fused to obtain a comprehensive fusion vector; Based on the aggregation module, context aggregation is performed on the comprehensive fusion vector to obtain a multi-level feature pyramid; Based on the stacked multi-layer spatiotemporal convolutional neural network module, multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies are performed on the multi-level feature pyramid to generate standard data of ship climate route data based on unified data description.

2. The method for establishing a ship climate route database according to claim 1, characterized in that: The historical ocean climate record files, the historical ship route record files, and the historical ship voyage record files are parsed respectively to construct an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology, including: Respectively analyzing the historical ocean climate record files, the historical ship route record files, and the historical ship voyage record files to generate historical ocean climate data item sets, historical ship route data item sets, and historical ship voyage data item sets; According to the historical ocean climate data item set, the ship historical route data item set, and the ship historical voyage data item set, an ocean climate description ontology, a ship historical route description ontology, and a ship historical voyage description ontology are constructed.

3. The method for establishing a ship climate route database according to claim 2, characterized in that: The historical ocean climate record file, the ship historical route record file, and the ship historical voyage record file are respectively parsed to generate a historical ocean climate data item set, a ship historical route data item set, and a ship historical voyage data item set, including: Performing lexical analysis on the historical ocean climate record files, the historical ship route record files, and the historical ship voyage record files respectively to obtain a plurality of words or phrases, and performing part-of-speech tagging on the plurality of words or phrases; Performing syntactic analysis on the multiple words or phrases that have been tagged with parts of speech, including dependency syntactic analysis and generative syntactic analysis, to generate a structure tree of dependency relationships between the words and / or phrases; Based on the structure tree of dependency relationships between the words and / or phrases, context analysis and role labeling are performed on the multiple words or phrases to generate historical ocean climate data item sets, historical ship route data item sets, and historical ship voyage data item sets.

4. The method for establishing a ship climate route database according to claim 1, characterized in that: The stacked multi-layer spatiotemporal convolutional neural network module performs multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies on the multi-level feature pyramid to generate standard data of ship climate route data based on unified data description, including: Based on the stacked multi-layer spatiotemporal convolutional neural network module, the multi-level feature pyramid is upsampled to obtain a multi-scale feature map, and multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies are performed based on the multi-scale feature map to generate standard data of ship climate route data based on unified data description.

5. The method for establishing a ship climate route database according to claim 4, characterized in that: The multi-level feature aggregation network also includes an attention module; The stacked multi-layer spatiotemporal convolutional neural network module upsamples the multi-level feature pyramid to obtain a multi-scale feature map, and performs multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies based on the multi-scale feature map to generate standard data of ship climate route data based on a unified data description. The attention weights between the multi-scale feature maps are calculated, and based on the attention weights, the attention strength between the multi-scale feature maps is calculated to perform multi-layer adaptive aggregation and multi-dimensional feature analysis based on spatial and temporal dependencies.

6. The method for establishing a ship climate route database according to claim 1, characterized in that: Before establishing the mapping relationship between the ship climate route data standard data and the ship comprehensive code to establish a ship climate route database, it also includes: Obtain ship type data, size data, navigation area data, and climate adaptation data, and encode them to generate basic ship codes; Adding a timestamp and a serial number to the basic ship code to generate an extended ship code; A hash operation is performed on the ship extended code to generate the ship comprehensive code.

Images