Meteorological disaster risk classification and early warning method for inland river ship formations

Abstract

The present application provides a meteorological disaster risk classification warning method for inland ship formations, which includes generating a ship navigation dynamic risk description file; parsing the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints based on this; acquiring ship dynamic navigation data, and establishing correlation constraints between the ship dynamic navigation data and the risk semantic constraints; establishing a disaster derivative relationship map between various factors in meteorological disaster risks; and according to the disaster derivative relationship map and the correlation constraints, performing a graded warning for ship meteorological disaster risks, thereby improving the accuracy and timeliness of the warning, thereby enhancing the pertinence and effectiveness of the warning, improving the real-time and accuracy of the warning, and more effectively guiding the navigation decision of the ship formation.

Description

Technical Field

[0001] The present application relates to the field of ship technology, and in particular to a meteorological disaster risk classification early warning method for inland river ship formations. Background Art

[0002] During the navigation of inland river fleets, meteorological disasters are an important factor affecting navigation safety. At present, the early warning methods for ship meteorological disasters mainly rely on weather forecasts and navigation experience, but these methods have obvious limitations. Specifically, the existing technologies mainly have the following problems:

[0003] Lack of dynamic risk description: Most of the existing early warning methods are based on static meteorological data and historical navigation experience, and are unable to generate and analyze ship navigation dynamic risk description files in real time, making it difficult to accurately reflect the potential meteorological disaster risks during the current navigation process.

[0004] Insufficient risk semantic constraints: Due to the lack of an effective risk semantic constraint mechanism, existing technologies make it difficult to extract key meteorological hazard factors from complex meteorological data and establish a clear relationship between these factors and navigation risks, resulting in insufficient accuracy and pertinence of warnings.

[0005] Lack of correlation constraints: Existing technologies have failed to fully establish the correlation constraints between ship dynamic navigation data and meteorological disaster risks, making it impossible for the early warning system to dynamically adjust the warning level according to real-time navigation data, reducing the real-time and effectiveness of the warning.

[0006] Disaster derivative relationships are unclear: In meteorological disaster risk warning, the disaster derivative relationships between various factors are complex and changeable. Existing technologies lack an effective mechanism to describe and analyze these relationships, making it difficult for the warning system to make accurate judgments when facing complex meteorological conditions. Summary of the invention

[0007] The purpose of this application is to propose a meteorological disaster risk classification warning method for inland ship formations, which is used to solve or alleviate the technical problems existing in the prior art.

[0008] The present application provides a meteorological disaster risk classification warning method for an inland river vessel formation, which includes:

[0009] Generate ship navigation dynamic risk description file;

[0010] Parsing the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints accordingly;

[0011] Acquiring dynamic navigation data of the ship, and establishing a correlation constraint between the dynamic navigation data of the ship and the risk semantic constraint;

[0012] Establish a disaster derivative relationship map between various factors in meteorological disaster risks;

[0013] According to the disaster derivative relationship map and the correlation constraints, graded warnings are given to the ship meteorological disaster risks.

[0014] To this end, the meteorological disaster risk classification warning method for inland river ship formations provided in the embodiment of the present application has at least the following technical benefits:

[0015] (1) Solve the problem of missing dynamic risk description:

[0016] By generating a dynamic risk description file for ship navigation, this method can capture and describe the potential risks during the ship's navigation in real time, ensuring that the early warning system can perform risk assessment based on the latest navigation status, thereby improving the accuracy and timeliness of the early warning.

[0017] (2) Strengthening risk semantic constraints:

[0018] The ship navigation dynamic risk description file is parsed to construct potential meteorological hazard factors, and risk semantic constraints are established accordingly. This step enables the early warning system to extract key information from complex meteorological data and clarify the relationship between meteorological hazard factors and navigation risks, thereby enhancing the pertinence and effectiveness of the early warning.

[0019] (3) Establish associativity constraints:

[0020] By acquiring the dynamic navigation data of the ship and establishing the correlation constraints between these data and risk semantic constraints, this method realizes the close integration of the early warning system and real-time navigation data. This enables the early warning system to dynamically adjust the warning level according to the actual navigation situation of the ship, further improving the real-time and accuracy of the early warning.

[0021] (4) Clearly describe the disaster derivative relationship:

[0022] The method provides an intuitive and systematic way to describe and analyze the complexity and variability of meteorological disaster risks by establishing a disaster derivative relationship map between various factors in meteorological disaster risks. This enables the early warning system to make more accurate judgments when facing complex meteorological conditions, thereby more effectively guiding the navigation decisions of ship formations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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:

[0024] Figure 1 This is a schematic flow chart of a meteorological disaster risk classification warning method for inland river vessel formations according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] 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.

[0026] Figure 1 This is a flow chart of a meteorological disaster risk classification warning method for inland river ship formations according to an embodiment of the present application. Figure 1 As shown, it includes:

[0027] Generate ship navigation dynamic risk description file;

[0028] Parsing the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints accordingly;

[0029] Acquiring dynamic navigation data of the ship, and establishing a correlation constraint between the dynamic navigation data of the ship and the risk semantic constraint;

[0030] Establish a disaster derivative relationship map between various factors in meteorological disaster risks;

[0031] According to the disaster derivative relationship map and the correlation constraints, graded warnings are given to the ship meteorological disaster risks.

[0032] Optionally, generating a ship navigation dynamic risk description file includes:

[0033] Acquiring ship navigation and operation data, wherein the ship navigation and operation data includes at least one of navigation log data, meteorological data, accident report data, and ship status monitoring data;

[0034] Constructing a data item set based on the ship navigation and operation data;

[0035] The data item set is encoded to generate a ship navigation dynamic risk description file.

[0036] To this end, the meteorological disaster risk classification warning method for inland river ship formations provided in the embodiment of the present application has at least the following technical benefits:

[0037] (1) Solve the problem of missing dynamic risk description:

[0038] By generating a dynamic risk description file for ship navigation, this method can capture and describe the potential risks during the ship's navigation in real time, ensuring that the early warning system can perform risk assessment based on the latest navigation status, thereby improving the accuracy and timeliness of the early warning.

[0039] (2) Strengthening risk semantic constraints:

[0040] The ship navigation dynamic risk description file is parsed to construct potential meteorological hazard factors, and risk semantic constraints are established accordingly. This step enables the early warning system to extract key information from complex meteorological data and clarify the relationship between meteorological hazard factors and navigation risks, thereby enhancing the pertinence and effectiveness of the early warning.

[0041] (3) Establish associativity constraints:

[0042] By acquiring the dynamic navigation data of the ship and establishing the correlation constraints between these data and risk semantic constraints, this method realizes the close integration of the early warning system and real-time navigation data. This enables the early warning system to dynamically adjust the warning level according to the actual navigation situation of the ship, further improving the real-time and accuracy of the early warning.

[0043] (4) Clearly describe the disaster derivative relationship:

[0044] The method provides an intuitive and systematic way to describe and analyze the complexity and variability of meteorological disaster risks by establishing a disaster derivative relationship map between various factors in meteorological disaster risks. This enables the early warning system to make more accurate judgments when facing complex meteorological conditions, thereby more effectively guiding the navigation decisions of ship formations.

[0045] Optionally, constructing a data item set based on the ship navigation and operation data includes:

[0046] Extracting features from the ship's navigation and operation data to obtain dynamic risk characterization features of the ship's navigation;

[0047] Vectorizing the ship navigation dynamic risk characterization features to obtain a structured feature vector;

[0048] Performing linear discriminant analysis on the structured feature vector to obtain a risk discriminant matrix;

[0049] Based on the gradient boosting tree, the risk discrimination matrix is ​​feature clustered and a data item set is constructed accordingly.

[0050] To this end, the above scheme further enhances the effectiveness of the meteorological disaster risk classification warning method for inland river ship formations by constructing a data item set, which has the following technical benefits:

[0051] (1) By extracting features from ship navigation and operation data, the dynamic risk characterization features of ship navigation are obtained. This step can accurately capture the key factors affecting navigation safety and provide a solid foundation for subsequent risk assessment.

[0052] (2) The dynamic risk characterization features of ship navigation are vectorized to obtain structured feature vectors. This transformation makes the originally complex and disordered navigation and operation data orderly and easy to process, which is convenient for subsequent data analysis and model training.

[0053] (3) Performing linear discriminant analysis on the structured feature vectors to obtain the risk discrimination matrix. This step can deeply mine the risk information in the data and improve the early warning system's ability to identify and judge potential risks.

[0054] (4) Based on the gradient boosting tree, the risk discrimination matrix is ​​clustered and the data item set is constructed accordingly. This step can further optimize the clustering effect of the features, so that the early warning system can more accurately classify the risk level and provide more targeted early warning information when facing complex and changeable meteorological conditions.

[0055] The following is a sample code that shows how to perform feature extraction and vectorization processing on ship navigation and operation data:

[0056]

[0057] In this example, DictVectorizer is used to convert the ship navigation and operation data in dictionary form into structured feature vectors. The converted feature vectors can be used in subsequent steps such as risk discriminant matrix calculation and feature clustering.

[0058] Optionally, encoding the data item set to generate a ship navigation dynamic risk description file includes:

[0059] Converting the data item set into a binary bit string, and vertically grouping the binary bit string to obtain a frequent pattern of ship dynamic risk;

[0060] Based on the frequent patterns of ship dynamic risks, rules describing feature combinations at different risk levels are generated to generate a ship navigation dynamic risk description file.

[0061] To this end, by encoding the data item set to generate a dynamic risk description file for ship navigation, the practicality and effectiveness of the meteorological disaster risk classification warning method are further improved, which has the following technical benefits:

[0062] (1) Convert the data item set into a binary bit string. This step realizes the binarization of the data, making subsequent data processing more efficient. Binary data is usually stored and processed faster in computers than text or other formats of data, thereby improving the response speed of the entire early warning system.

[0063] (2) Vertically group the binary bit strings to obtain frequent patterns of ship dynamic risk. This step can identify frequently occurring feature combinations in the data, which are often associated with specific meteorological disaster risks. By mining these frequent patterns, the early warning system can more accurately identify potential risks.

[0064] (3) Based on the frequent patterns of ship dynamic risks, rules describing the combination of features at different risk levels are generated to generate a ship navigation dynamic risk description file. This step saves the mined risk information in a structured form to facilitate subsequent risk assessment and early warning. At the same time, this structured description method also makes risk information easier to understand and disseminate.

[0065] (4) By using binary bit strings and frequent patterns to describe risks, the warning system can more easily add or modify risk rules to adapt to changing meteorological conditions and navigation environments. This flexibility and scalability enables the warning system to remain effective and accurate over the long term.

[0066] The following is a code example to illustrate how to convert a data item set into a binary bit string and perform vertical grouping to obtain the frequent patterns of ship dynamic risk:

[0067]

[0068]

[0069] In this example, MultiLabelBinarizer is used to convert data itemsets into binary matrices, and TransactionEncoder is used to vertically group the binary matrices and mine frequent patterns. These frequent patterns can be used to generate rules in the risk description file of ship navigation dynamics.

[0070] Optionally, the parsing of the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints accordingly includes:

[0071] Parsing the ship navigation dynamic risk description file to obtain meteorological characteristic representation;

[0072] Converting the meteorological characteristic representation into a transaction data set;

[0073] Based on the transaction data set, determine the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies;

[0074] According to the disaster-prone environment, disaster-causing factors and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly.

[0075] To this end, by parsing the ship navigation dynamic risk description file, constructing potential meteorological disaster factors and establishing risk semantic constraints, a more accurate and comprehensive basis is provided for meteorological disaster risk classification warning, which has the following technical benefits:

[0076] (1) By parsing the ship navigation dynamic risk description file, the meteorological characteristics are obtained. This step can accurately extract the key meteorological information contained in the description file and provide an accurate data basis for subsequent risk assessment.

[0077] (2) The meteorological feature representation is converted into a transaction data set. This conversion makes the meteorological data, which may have been stored in text or non-standard formats, structured, facilitating subsequent data analysis and processing.

[0078] (3) Based on the transaction data set, the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies are determined. These three elements are the basic framework of meteorological disaster risk. By clarifying these three elements, the early warning system can more clearly identify and assess potential meteorological disaster risks.

[0079] (4) Based on the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly. This step can ensure that the early warning system can accurately understand and apply relevant meteorological disaster factors and risk semantics when identifying and assessing risks, thereby improving the accuracy and reliability of early warnings.

[0080] The following is a code example that illustrates how to convert meteorological characteristics into a transaction dataset and determine the disaster-prone environment, disaster-causing factors, and disaster-affected bodies:

[0081]

[0082] In this example, the meteorological characteristics are first converted into a transaction data set, and then the disaster-prone environment, hazard factors and disaster-susceptible bodies are simply determined based on the meteorological characteristics.

[0083] Optionally, the parsing of the ship navigation dynamic risk description file to obtain a meteorological characteristic representation includes:

[0084] Based on the knowledge graph related to ship navigation and meteorology, a subspace model for discovering risk characteristics is constructed;

[0085] Based on the risk feature subspace model, generating risk process behavior semantic trajectory;

[0086] Based on the risk process behavior semantic trajectory, the ship navigation dynamic risk description file is parsed to obtain meteorological feature representation.

[0087] To this end, by combining the knowledge graph related to ship navigation and meteorology, a risk feature subspace model is constructed, and the semantic trajectory of risk process behavior is generated based on it. Finally, the ship navigation dynamic risk description file is parsed to obtain the meteorological feature representation, which has the following technical benefits:

[0088] (1) Based on the constructed knowledge graph related to ship navigation and meteorology, this technology can make full use of professional knowledge in the field to ensure that key information related to meteorological characteristics can be accurately captured and identified during the analysis process, thereby improving the accuracy and comprehensiveness of the analysis.

[0089] (2) By constructing a risk characteristic subspace model, this technology can deeply explore the potential risk characteristics related to meteorological disasters during the navigation process of ships. These characteristics may be hidden in a large amount of navigation and meteorological data, but through the construction and analysis of the model, they can be effectively extracted and utilized.

[0090] (3) Generate semantic trajectories of risk process behavior, so that the early warning system can better understand and simulate the development process of meteorological disaster risks, which is crucial for accurately assessing the evolution trend and possible impact of risks.

[0091] (3) The entire technical solution realizes an automated process from knowledge graph construction to risk description file parsing, reduces manual intervention, improves the intelligence level of the early warning system, and enables the system to respond to potential meteorological disaster risks more quickly and accurately.

[0092] The following is a code example to illustrate how to generate risk process behavior semantic trajectories based on the risk feature subspace model and parse the ship navigation dynamic risk description file to obtain meteorological feature representations:

[0093]

[0094]

[0095] In this example, a risk feature subspace model was first created, then the risk process behavior semantic trajectory was generated by matching keywords, and finally the meteorological feature representation was obtained.

[0096] Optionally, converting the meteorological characteristic representation into a transaction data set includes:

[0097] The meteorological characteristic representation is used as an item in a transaction, and a corresponding navigation risk event is indexed as a transaction, so as to convert the meteorological characteristic representation into a transaction data set.

[0098] Optionally, determining the disaster-prone environment, disaster-causing factors, and disaster-affected bodies based on the transaction data set includes:

[0099] Clustering meteorological parameters, geographic location information, time information, and disaster-prone body types from the transaction data set;

[0100] Analyzing spatial distribution characteristics based on the geographical location information to determine the disaster-prone environment;

[0101] Performing navigation risk relationship mining based on the meteorological parameters and the time information to determine the disaster-causing factors;

[0102] Based on the type of the disaster-prone body, the disaster-prone body is classified and the vulnerability level is analyzed to determine the disaster-prone body.

[0103] To this end, by clustering key information from the transaction data set, including meteorological parameters, geographic location information, time information and disaster-prone body type, and then analyzing and determining the disaster-prone environment, disaster-causing factors and disaster-prone bodies, the following technical benefits are achieved:

[0104] (1) By clustering and analyzing the key information in the transaction data set, various factors related to meteorological disaster risks can be more accurately identified, including the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies. This data-based analysis method reduces the subjectivity of human judgment and improves the accuracy of risk identification.

[0105] (2) This technical solution takes into account multiple factors such as meteorological parameters, geographic location information, time information and disaster-prone body type, ensuring that all possible risk factors can be fully considered in the risk identification process and avoiding the omission or neglect of certain important factors.

[0106] (3) By analyzing the spatial distribution characteristics of geographic location information, mining the relationship between meteorological parameters and time information and navigation risks, and classifying the types of disaster-prone objects and analyzing their vulnerability, this technical solution can provide a deeper understanding of the nature and characteristics of risks and provide stronger support for subsequent early warning and response.

[0107] (4) The entire technical solution realizes an automated process from data clustering to risk factor determination, reduces manual intervention, and improves the intelligence and automation level of the early warning system. This enables the early warning system to respond to potential meteorological disaster risks more quickly and accurately.

[0108] The following is a Python code example to illustrate how to cluster key information from a transaction dataset and identify disaster-prone environments, hazard factors, and hazard-affected bodies:

[0109]

[0110]

[0111] In this example, the transaction data set is first traversed to cluster meteorological parameters, geographic location information, time information, and hazard-prone body types. Then, the disaster-prone environment, hazard factors, and hazard-prone bodies are simply determined based on this information.

[0112] Optionally, constructing potential meteorological disaster factors according to the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies and establishing risk semantic constraints accordingly includes:

[0113] Performing association rule mining on the disaster-prone environment, disaster-causing factors, and disaster-affected bodies to analyze the semantic association strength between the disaster-prone environment, disaster-causing factors, and disaster-affected bodies;

[0114] Based on the semantic correlation strength between the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly.

[0115] To this end, by mining association rules among disaster-prone environments, disaster-causing factors, and disaster-bearing bodies, and analyzing the semantic association strength between them, potential meteorological disaster factors are constructed and risk semantic constraints are established, which has the following technical benefits:

[0116] (1) Through association rule mining, we can reveal the hidden and complex internal connections between the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, which may be difficult to discover with traditional analysis methods. This in-depth mining helps to understand the formation and evolution mechanism of meteorological disaster risks more comprehensively and accurately.

[0117] (2) Based on the mined semantic association strength, potential meteorological disaster factors are constructed and risk semantic constraints are established, which enables the early warning system to more accurately identify potential meteorological disaster risks and issue warnings in time before the risks occur, thereby improving the accuracy and timeliness of risk warnings.

[0118] (3) The risk semantic constraints established through association rule mining can provide a clear explanation and basis for the output results of the early warning system, making the early warning results easier to understand and accept. This interpretability enhances the credibility of the early warning system and improves the user's satisfaction and trust in the early warning system.

[0119] (4) The entire technical solution realizes an automated process from association rule mining to risk semantic constraint establishment, reduces manual intervention, and improves the intelligence and automation level of the early warning system. This enables the early warning system to respond to potential meteorological disaster risks more quickly and accurately, providing stronger support for disaster prevention and mitigation work.

[0120] The following is a code example to illustrate how to mine association rules for disaster-prone environments, disaster-causing factors, and disaster-affected bodies, and to construct potential meteorological disaster factors and risk semantic constraints:

[0121]

[0122]

[0123] Optionally, the performing association rule mining on the disaster-prone environment, disaster-causing factors, and disaster-affected bodies to analyze the semantic association strength between the disaster-prone environment, disaster-causing factors, and disaster-affected bodies includes:

[0124] Based on the information entropy weight method, association rules are mined for the disaster-prone environment, disaster-causing factors, and disaster-affected bodies to generate knowledge representations and association paths between the disaster-prone environment, disaster-causing factors, and disaster-affected bodies;

[0125] The knowledge representation and association paths among the disaster-prone environment, disaster-causing factors, and disaster-affected bodies are used to analyze the semantic association strength among the disaster-prone environment, disaster-causing factors, and disaster-affected bodies.

[0126] To this end, the information entropy weight method is used to mine association rules for disaster-prone environments, disaster-causing factors, and disaster-bearing bodies to generate knowledge representations and association paths between them, and then analyze the semantic association strength between them. This has the following technical benefits:

[0127] (1) The weights of disaster-prone environments, disaster-causing factors, and disaster-affected bodies are determined according to their frequency of occurrence and distribution characteristics in the data set, avoiding the bias that may be caused by subjective weighting and improving the accuracy of association rule mining.

[0128] (2) Through the information entropy weight method, we can more deeply explore the association rules between disaster-prone environments, disaster-causing factors, and disaster-sustaining bodies, discover the hidden and complex semantic associations between them, and provide more comprehensive information support for the identification and early warning of meteorological disaster risks.

[0129] (3) The knowledge representation and association paths generated by association rule mining can provide a clear explanation and basis for the output results of the early warning system, making the early warning results easier to understand and accept, thereby enhancing the interpretability and credibility of the early warning system.

[0130] The following is a code example to illustrate how to mine association rules for disaster-prone environments, disaster-causing factors, and disaster-affected bodies based on the information entropy weight method, and generate knowledge representations and association paths between them:

[0131] #The disaster-prone environment, disaster-causing factors, and disaster-bearing bodies have been given in the form of a list, and the transaction data set has been constructed

[0132]

[0133]

[0134] Optionally, based on the semantic association strength between the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, constructing potential meteorological disaster factors and establishing risk semantic constraints accordingly includes:

[0135] Determine the attribute characteristics of the disaster-prone environment, disaster-causing factors, and disaster-affected objects, and calculate the information entropy of the disaster-prone environment, disaster-causing factors, and disaster-affected objects;

[0136] Calculating characteristic weights of the disaster-prone environment, disaster-causing factors, and disaster-affected objects according to the information entropy of the disaster-prone environment, disaster-causing factors, and disaster-affected objects;

[0137] The characteristic weights of the disaster-prone environment, disaster-causing factors, and disaster-prone bodies and the semantic association strengths among the disaster-prone environment, disaster-causing factors, and disaster-prone bodies are used to construct potential meteorological disaster factors and establish risk semantic constraints accordingly.

[0138] To this end, by determining the attribute characteristics of the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, their information entropy is calculated, and then the feature weight is calculated based on the information entropy. In combination with the semantic association strength between them, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly. This has the following technical benefits:

[0139] (1) By comprehensively considering the disaster-prone environment, disaster-causing factors, and the attributes of the disaster-bearing body, this scheme can comprehensively and systematically analyze the causes and influencing factors of meteorological disasters, avoiding the one-sidedness of single-factor analysis.

[0140] (2) The information entropy is used to calculate the characteristic weights of the disaster-prone environment, disaster-causing factors, and disaster-prone bodies, which achieves objective weighting, avoids the bias that may be caused by subjective weighting, and improves the accuracy and reliability of the analysis.

[0141] (3) By combining the characteristic weights of the disaster-prone environment, disaster-causing factors, and disaster-prone bodies and the semantic association strength between them, this scheme can deeply explore potential meteorological disaster factors and provide a new perspective and basis for the identification and early warning of meteorological disaster risks.

[0142] (4) After constructing the potential meteorological disaster factors, risk semantic constraints are further established based on them, which can provide a clear explanation and basis for the graded warning of meteorological disaster risks, making the warning results easier to understand and accept.

[0143] The following is a code example to illustrate how to calculate the information entropy and feature weights of the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, and to construct potential meteorological disaster factors and risk semantic constraints:

[0144]

[0145]

[0146] Optionally, the acquiring of the ship dynamic navigation data and establishing a correlation constraint between the ship dynamic navigation data and the risk semantic constraint includes:

[0147] Obtain ship dynamic navigation data through the AIG system;

[0148] Extracting time series features of the ship's dynamic navigation data to determine the continuity and trend of the ship's behavior;

[0149] Converting the risk semantic constraints into a time series logic expression;

[0150] The continuity and trend of the ship's behavior are matched to the time series logical expression to establish an association constraint between the ship's dynamic navigation data and the risk semantic constraint.

[0151] To this end, by obtaining the ship's dynamic navigation data and establishing the association constraints between the data and the risk semantic constraints, the following technical benefits are brought about:

[0152] (1) The AIG system obtains ship dynamic navigation data in real time, ensuring the timeliness and dynamism of the data. This enables the monitoring and early warning of ship behavior to keep up with the actual situation and respond to changes in the ship's status in a timely manner.

[0153] (2) Extracting the time series characteristics of ship dynamic navigation data helps to identify the continuity and trend of ship behavior. This in-depth understanding of ship behavior patterns provides strong support for subsequent early warning and decision-making.

[0154] (3) The risk semantic constraints are converted into time series logical expressions, realizing the quantitative application of risk semantics. This makes risk warning more specific and measurable, and improves the accuracy and practicality of warning.

[0155] (4) By matching the continuity and trend of ship behavior to the time series logic expression, the correlation constraint between the ship dynamic navigation data and the risk semantic constraint is established. The establishment of this correlation provides a direct basis for the risk assessment and early warning of ship behavior.

[0156] The following is a code example to illustrate how to extract the time series features of ship dynamic navigation data and match them with the time series logical expression of risk semantic constraints:

[0157]

[0158]

[0159] Optionally, the establishing of a disaster derivative relationship map between various factors in meteorological disaster risk includes:

[0160] Obtaining a meteorological disaster risk description, and performing spatial topological and temporal topological relationship analysis on the meteorological disaster risk description to extract disaster factors therein and semantic associations between disaster factors;

[0161] Based on the semantic association relationship between the disaster factors, entities and association relationships between entities are extracted to establish a disaster derivative relationship map between various factors in meteorological disaster risks.

[0162] To this end, by combining the disaster derivative relationship map and correlation constraints, a graded warning of ship meteorological disaster risks is carried out, which brings the following technical benefits:

[0163] (1) By establishing a risk assessment model and determining the disaster risk entity link, a refined assessment of ship meteorological disaster risks can be achieved. This assessment method takes into account multiple factors and the complex relationships between factors, improving the accuracy and comprehensiveness of risk assessment.

[0164] (2) Based on the disaster derivative relationship map and correlation constraints, graded warning can be carried out to establish a dynamic warning mechanism. When the disaster factors or correlation relationships change, the warning level will be adjusted accordingly to ensure the timeliness and effectiveness of the warning.

[0165] (3) The hierarchical warning mechanism can provide more targeted guidance for responding to ship meteorological disasters. Different levels of warnings correspond to different response measures and resource allocation plans, which helps to improve the efficiency and effectiveness of response.

[0166] (4) Through graded warnings, response resources and forces can be allocated more reasonably. For high-risk warnings, more resources and forces can be allocated to respond first to ensure the effective use of resources.

[0167] The following is a code example to illustrate how to perform graded warnings for ship meteorological disaster risks based on the disaster derivative relationship graph and correlation constraints:

[0168]

[0169]

[0170] In this example, the disaster derived relationship graph G and the association constraint data constraints_df are established. Then, a grade_warning function is defined, which traverses the association constraint data and checks whether there is a corresponding association relationship in the disaster derived relationship graph. If so, a warning is issued according to the risk level and the warning result is stored in the warnings dictionary. Finally, the result of the graded warning is output.

[0171] Optionally, the step of providing a graded warning for ship meteorological disaster risks according to the disaster derivative relationship graph and the correlation constraint includes:

[0172] Identify disaster risk entity links to build risk assessment models;

[0173] Based on the risk assessment model, a graded warning is given to the ship meteorological disaster risk according to the disaster derivative relationship map and the correlation.

[0174] To this end, by combining the disaster derivative relationship map and correlation constraints, a graded warning of ship meteorological disaster risks is carried out, which has the following technical benefits:

[0175] (1) By establishing a risk assessment model and determining the disaster risk entity link, various disaster factors and their complex relationships can be fully considered, thereby making a more accurate and comprehensive assessment of ship meteorological disaster risks. This assessment method avoids the limitations of a single factor or simple statistical method and improves the scientificity and practicality of risk assessment.

[0176] (2) Based on the disaster derivative relationship map and correlation constraints, hierarchical warning can be carried out to establish a dynamic warning mechanism. When the disaster factors change or new correlations emerge, the warning system can update the warning level in real time to ensure the timeliness and effectiveness of the warning. This dynamic warning mechanism is of great significance for responding to rapidly changing meteorological disaster risks.

[0177] (3) The graded warning mechanism can provide more targeted guidance for the response to ship meteorological disasters. Different levels of warnings correspond to different response measures and resource allocation plans, which helps to optimize resource allocation and improve response efficiency and effectiveness. At the same time, changes in warning levels can also guide the adjustment of response strategies in a timely manner to ensure the scientificity and effectiveness of response work.

[0178] (4) Through the graded warning mechanism, decision makers can more clearly understand the level of ship meteorological disaster risk and the possible impact range, so as to make more scientific and reasonable decisions. At the same time, the emergency response team can also respond quickly according to the warning level and response measures, effectively reducing disaster losses.

[0179] The following is a simplified Python code example to illustrate how to perform graded warnings for ship meteorological disaster risks based on the disaster derivative relationship graph and correlation constraints:

[0180]

[0181]

[0182] In this example, a simple disaster derivative relationship graph G is first created, and an example node "strong wind" and an example edge "strong wind" -> "ship deviation" are added. Then, a correlation constraint data constraints_df is created, which contains the disaster factor chain and the corresponding risk level. Next, a grade_warning function is defined, which traverses the correlation constraint data, checks whether there is a corresponding correlation relationship in the disaster derivative relationship graph, and issues a warning based on the risk level. Finally, the warning result is output.

[0183] Optionally, determining the disaster risk entity links to establish a risk assessment model includes:

[0184] Determine the disaster risk entity link based on the disaster-causing factors, the disaster-bearing body classification catalog and the disaster chain relationship table between them;

[0185] Based on the disaster risk entity link, the probability of disaster occurrence and the degree of loss under different scenarios are simulated according to the Monte Carlo simulation method to establish a risk assessment model.

[0186] To this end, a risk assessment model is established by determining the disaster risk entity link, which brings the following technical benefits:

[0187] (1) Based on the classification catalog of hazard factors and hazard-affected objects and the disaster chain relationship table between them, it is possible to comprehensively and systematically identify and analyze disaster risk entities and their relationships. This systematic approach helps avoid missing important risk factors and ensures the comprehensiveness and accuracy of risk assessment.

[0188] (2) Monte Carlo simulation can be used to simulate the probability of disasters and the extent of losses under different scenarios, thus upgrading risk assessment from qualitative analysis to quantitative analysis. This quantitative method makes risk assessment results more specific and comparable, helping decision makers to understand the risk level more clearly.

[0189] (3) The establishment of disaster risk entity links and risk assessment models is based on updateable data and relationship tables, which means that when new disaster information, changes in disaster-prone bodies, or disaster chain relationships appear, the model and assessment results can be flexibly updated. This flexibility and adaptability enables risk assessment to keep pace with the times and maintain its accuracy and effectiveness.

[0190] (4) By establishing a risk assessment model, decision makers can obtain detailed information about disaster risks in different scenarios, including the probability of disaster occurrence, the extent of losses, etc. This information provides a scientific basis for decision makers and helps them formulate more reasonable and effective disaster prevention and response measures.

[0191] The following is a code example to illustrate how to build a risk assessment model based on disaster risk entity linking and Monte Carlo simulation:

[0192]

[0193]

[0194] In this example, the disaster factors, disaster-prone body classification catalog, and disaster chain relationship table between them are first defined and represented by dictionaries. Then, a monte_carlo_simulation function is defined, which uses the Monte Carlo simulation method to simulate the probability of disaster occurrence and the extent of loss under different scenarios, and stores the results in a PandasDataFrame.

[0195] Optionally, based on the risk assessment model, graded warnings are given for ship meteorological disaster risks according to the disaster derivative relationship map and the correlation constraints, including: based on the risk assessment model, feature extraction and feature fusion of the disaster derivative relationship map and the correlation constraints are performed to predict the shipping meteorological index, and graded warnings are given for ship meteorological disaster risks accordingly.

[0196] To this end, based on the risk assessment model, feature extraction and feature fusion are performed on the disaster derivative relationship map and correlation constraints to predict the shipping meteorological index, and graded warning of ship meteorological disaster risks is carried out accordingly, which has the following technical benefits:

[0197] (1) By extracting and fusing features of disaster-derived relationship graphs and correlation constraints, we can more comprehensively capture the complexity and correlation of disaster occurrence and development. This method, which comprehensively considers multiple factors and correlations, can significantly improve the accuracy of early warnings compared to single factors or simple statistical methods.

[0198] (2) The feature extraction and fusion process is dynamic and can be updated as new data is added and the disaster situation changes. This means that the early warning system can be adjusted in real time to reflect the latest disaster risk conditions and provide ships with more timely and accurate early warning information.

[0199] (3) The hierarchical warning mechanism enables different levels of warning to correspond to different response measures and resource allocation plans. Through accurate forecasts of the shipping weather index, emergency resources can be allocated more reasonably to ensure that a quick and effective response can be made when a disaster occurs.

[0200] The following is a code example to illustrate how to perform feature extraction, feature fusion, and shipping weather index prediction:

[0201]

[0202]

[0203] In this example, a DataFrame containing wind speed, precipitation, temperature, disaster chain relationship characteristics, and vulnerability characteristics of the disaster-bearing body, as well as the corresponding shipping weather index label data, is first created. Then, the training set and test set are divided, and the shipping weather index is predicted using a random forest regression model. Finally, the prediction error is calculated, and a simplified processing of graded warnings is performed based on the prediction results.

[0204] 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.

[0205] 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.

[0206] 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 meteorological disaster risk classification early warning method for inland river ship formations, characterized in that: include: Generate ship navigation dynamic risk description file; Parsing the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints accordingly; Acquiring dynamic navigation data of the ship, and establishing a correlation constraint between the dynamic navigation data of the ship and the risk semantic constraint; Establish a disaster derivative relationship map between various factors in meteorological disaster risks; According to the disaster derivative relationship map and the correlation constraints, a graded warning is provided for the ship meteorological disaster risk; The parsing of the ship navigation dynamic risk description file to construct potential meteorological disaster factors and establish risk semantic constraints accordingly includes: Parsing the ship navigation dynamic risk description file to obtain meteorological characteristic representation; Converting the meteorological characteristic representation into a transaction data set; Based on the transaction data set, determine the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies; According to the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly; The process of parsing the ship navigation dynamic risk description file to obtain meteorological characteristic representations includes: Based on the knowledge graph related to ship navigation and meteorology, a subspace model for discovering risk characteristics is constructed; Based on the risk feature subspace model, generating risk process behavior semantic trajectory; Based on the risk process behavior semantic trajectory, the ship navigation dynamic risk description file is parsed to obtain meteorological feature representation.

2. A meteorological disaster risk classification early warning method for inland river ship formations according to claim 1, characterized in that: The generating of the ship navigation dynamic risk description file includes: Acquiring ship navigation and operation data, wherein the ship navigation and operation data includes at least one of navigation log data, meteorological data, accident report data, and ship status monitoring data; Constructing a data item set based on the ship navigation and operation data; The data item set is encoded to generate a ship navigation dynamic risk description file.

3. A meteorological disaster risk classification early warning method for inland river ship formations according to claim 2, characterized in that: The data item set is constructed based on the ship navigation and operation data, including: Extracting features from the ship's navigation and operation data to obtain dynamic risk characterization features of the ship's navigation; Vectorizing the ship navigation dynamic risk characterization features to obtain a structured feature vector; Performing linear discriminant analysis on the structured feature vector to obtain a risk discriminant matrix; Based on the gradient boosting tree, the risk discrimination matrix is ​​feature clustered and a data item set is constructed accordingly.

4. A meteorological disaster risk classification early warning method for inland river ship formations according to claim 2, characterized in that: The encoding of the data item set to generate a ship navigation dynamic risk description file includes: Converting the data item set into a binary bit string, and vertically grouping the binary bit string to obtain a frequent pattern of ship dynamic risk; Based on the frequent patterns of ship dynamic risks, rules describing feature combinations at different risk levels are generated to generate a ship navigation dynamic risk description file.

5. The meteorological disaster risk classification early warning method for inland river ship formations according to claim 1 is characterized in that: The step of converting the meteorological characteristic representation into a transaction data set comprises: The meteorological characteristic representation is used as an item in a transaction, and a corresponding navigation risk event is indexed as a transaction, so as to convert the meteorological characteristic representation into a transaction data set.

6. The meteorological disaster risk classification early warning method for inland river ship formations according to claim 1 is characterized in that: Determining the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies based on the transaction data set includes: Clustering meteorological parameters, geographic location information, time information, and disaster-prone body types from the transaction data set; Analyzing spatial distribution characteristics based on the geographical location information to determine the disaster-prone environment; Performing navigation risk relationship mining based on the meteorological parameters and the time information to determine the disaster-causing factors; Based on the type of the disaster-prone body, the disaster-prone body is classified and the vulnerability level is analyzed to determine the disaster-prone body.

7. The meteorological disaster risk classification early warning method for inland river ship formations according to claim 1 is characterized in that: According to the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly, including: Performing association rule mining on the disaster-prone environment, disaster-causing factors, and disaster-affected bodies to analyze the semantic association strength between the disaster-prone environment, disaster-causing factors, and disaster-affected bodies; Based on the semantic correlation strength between the disaster-prone environment, disaster-causing factors, and disaster-bearing bodies, potential meteorological disaster factors are constructed and risk semantic constraints are established accordingly.

8. The meteorological disaster risk classification early warning method for inland river ship formations according to claim 1 is characterized in that: The establishment of a disaster derivative relationship map between various factors in meteorological disaster risk includes: Obtaining a meteorological disaster risk description, and performing spatial topological and temporal topological relationship analysis on the meteorological disaster risk description to extract disaster factors therein and semantic associations between disaster factors; Based on the semantic association relationship between the disaster factors, entities and association relationships between entities are extracted to establish a disaster derivative relationship map between various factors in meteorological disaster risks.

Images