Data fusion and integration method and system based on navigation multi-source data

Abstract

The invention relates to the field of navigation risk prediction, in particular to a data fusion and integration method and system based on navigation multi-source data, and the method comprises the steps: constructing a correlation analysis model; acquiring source data of different scene types in real time; inputting different types of source data into the correlation analysis model, and extracting risk features of the source data through the correlation analysis model; setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model, and analyzing a correlation coefficient between each risk feature and historical navigation data through the hierarchical clustering model; determining a weight coefficient of the risk feature according to the correlation coefficient, wherein the weight coefficient is used for reflecting the importance level of the risk feature in navigation risk prediction; and constructing a data framework of the navigation risk data, and integrating the risk features and the weight coefficients thereof into the data framework. And making a corresponding risk management strategy. A ship operator can process navigation multi-source data more effectively, accuracy of navigation risk prediction is improved, navigation safety is guaranteed, and a navigation plan is optimized.

Description

Technical Field

[0001] The present application relates to the technical field of navigation risk prediction, and in particular to a data fusion and integration method and system based on multi-source navigation data. Background Art

[0002] Multi-source data of navigation refers to data collected from multiple different sources and types during navigation. These data may include meteorological data, ocean current data, ship performance data, navigation log satellite data, historical accident data, traffic data, etc. These data sources include manual records, automatic sensors, communication systems, satellite remote sensing, etc. The fusion and integration of multi-source data of navigation is the key to improving navigation safety and efficiency, because they provide comprehensive information about the navigation environment, ship status and historical navigation behavior. By analyzing these data, useful risk features can be extracted for navigation risk prediction and management.

[0003] When predicting navigation risks, existing technologies usually collect a variety of data sources, including meteorological data, navigation logs, ship performance data, etc. These data are directly input into a risk identification system, and the system attempts to identify features directly related to navigation risks. This may only focus on features directly related to navigation risks, such as wind speed, wave height, etc., and may lack effective feature extraction methods, making it difficult to accurately predict navigation risks, resulting in insufficient or overreaction of preventive measures, and easily overlooking those features that are strongly related to navigation risks but are not direct causes. In navigation, there is more than just risk prediction and risk avoidance. When some risks cannot be avoided, you also need to deal with various situations where you need to face risks directly.

[0004] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0005] In order to solve one or several problems in the prior art, the main purpose of the present application is to provide a data fusion and integration method and system based on multi-source navigation data.

[0006] In order to achieve the above-mentioned invention object, the present application proposes a data fusion and integration method based on multi-source navigation data, the method comprising:

[0007] Build an association analysis model to extract risk characteristics of data;

[0008] Obtain source data of different scene types in real time;

[0009] Inputting different types of source data into the association analysis model, and extracting risk features of the source data through the association analysis model;

[0010] Setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model respectively, and analyzing the correlation coefficient between each risk feature and the historical navigation data through the hierarchical clustering model;

[0011] Determine a weight coefficient of the risk feature according to the correlation coefficient, which is used to reflect the importance level of the risk feature in the navigation risk prediction;

[0012] A data framework for navigation risk data is constructed, and the risk characteristics and their weight coefficients are integrated into the data framework.

[0013] The embodiment of the present application also provides a data fusion and integration system based on multi-source navigation data, including:

[0014] A construction module is used to build an association analysis model to extract risk features of data;

[0015] The acquisition module is used to obtain source data of different scene types in real time;

[0016] An input module, used to input different types of source data into the association analysis model, and extract risk features of the source data through the association analysis model;

[0017] An analysis module, used for setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model, and analyzing the correlation coefficient between each risk feature and the historical navigation data through the hierarchical clustering model;

[0018] A determination module, used to determine a weight coefficient of the risk feature according to the correlation coefficient, so as to reflect the importance level of the risk feature in the navigation risk prediction;

[0019] The integration module is used to construct a data framework of navigation risk data and integrate the risk characteristics and their weight coefficients into the data framework.

[0020] The present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0021] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0022] The data fusion and integration method and system based on multi-source navigation data of the embodiment of the present application can extract risk features from multiple data sources, including those features that are strongly associated with navigation risks, by constructing association analysis models and hierarchical clustering models. The weight coefficient of each risk feature is determined according to the correlation coefficient to reflect its importance level in navigation risk prediction. The determination of the weight coefficient helps to improve the prediction performance and stability of the model, enabling decision makers to formulate corresponding risk management strategies based on the importance of risk features. A data framework for navigation risk data is established to integrate risk features and their weight coefficients. The data framework can be used as a decision support tool to help ship operators and maritime administrations better understand navigation risks and formulate corresponding risk management strategies. Ship operators can process multi-source navigation data more effectively and improve the accuracy of navigation risk prediction, thereby ensuring navigation safety and optimizing navigation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of a flow chart of a data fusion and integration method based on multi-source navigation data according to an embodiment of the present application;

[0024] Figure 2 A schematic diagram of a flow chart of a data fusion and integration method based on multi-source navigation data according to an embodiment of the present application;

[0025] Figure 3 This is a schematic structural block diagram of a data fusion and integration system based on multi-source navigation data according to an embodiment of the present application;

[0026] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0027] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] Reference Figure 1 In an embodiment of the present application, a data fusion and integration method based on multi-source navigation data is provided, the method comprising:

[0030] S1. Build an association analysis model to extract risk characteristics of data;

[0031] S2, real-time acquisition of source data of different scene types;

[0032] S3, inputting different types of source data into the association analysis model, and extracting risk features of the source data through the association analysis model;

[0033] S4. Setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model respectively, and analyzing the correlation coefficient between each risk feature and the historical navigation data through the hierarchical clustering model;

[0034] S5. Determine a weight coefficient of the risk feature according to the correlation coefficient, which is used to reflect the importance level of the risk feature in the navigation risk prediction;

[0035] S6. Construct a data framework for navigation risk data, and integrate the risk characteristics and their weight coefficients into the data framework.

[0036] As described in steps S1-S2 above, the association analysis model extracts risk features by analyzing different types of source data to find the correlation between data. It may include statistical analysis, machine learning algorithms or other data mining techniques. Through association analysis, the potential relationship between different data sources can be discovered to help identify key factors that may affect navigation risks. Real-time acquisition of source data of different scenario types means that the system can continuously update data to reflect real-time environmental changes. Real-time data can provide more accurate risk features, help predict navigation risks in real time, and take corresponding preventive measures.

[0037] As described in steps S3-S6 above, the hierarchical clustering model inputs each risk feature into the model separately, and determines the importance of the risk feature by analyzing the correlation coefficient between the risk feature and the historical navigation data. The hierarchical clustering model can help determine which risk features are most important for navigation risk prediction, thereby improving the accuracy and reliability of the prediction. According to the analysis results of the hierarchical clustering model, the weight coefficient of each risk feature is determined to reflect its level of importance in navigation risk prediction. The weight coefficient can help decision makers understand which risk features require more attention and resources, so as to manage navigation risks more effectively. Construct a data framework for navigation risk data, and integrate risk features and their weight coefficients into a structured data model. The data framework can be used as a decision support tool to help ship operators and maritime management agencies better understand navigation risks and formulate corresponding risk management strategies.

[0038] As mentioned above, by constructing an association analysis model and a hierarchical clustering model, risk features can be extracted from multiple data sources, including those that are strongly associated with navigation risks. The weight coefficient of each risk feature is determined based on the correlation coefficient, reflecting its level of importance in the navigation risk prediction. The determination of the weight coefficient helps to improve the prediction performance and stability of the model, enabling decision makers to formulate corresponding risk management strategies based on the importance of the risk features. A data framework for navigation risk data is established to integrate risk features and their weight coefficients. The data framework can be used as a decision support tool to help ship operators and maritime administrations better understand navigation risks and formulate corresponding risk management strategies. Ship operators can more effectively process multi-source navigation data and improve the accuracy of navigation risk prediction, thereby ensuring navigation safety and optimizing navigation plans.

[0039] Reference Figure 2 In one embodiment, each risk feature is input into the hierarchical clustering model, and the correlation coefficient between each risk feature and the historical navigation data is analyzed by the hierarchical clustering model. The method includes:

[0040] S41, obtaining a scene classification corresponding to a risk feature, inputting the risk feature of each scene classification into the hierarchical clustering model, mapping the risk feature into the historical navigation data through the hierarchical clustering model, and determining the target feature and auxiliary feature in the navigation risk data corresponding to the risk feature;

[0041] S42, based on the target features, calculating the correlation ratio between the target features of each scene classification and the historical navigation data, wherein the scene classification includes meteorological features, ocean current features, and geographical features;

[0042] S43. Quantify the association ratio through the hierarchical clustering model to obtain the correlation coefficient between the risk characteristics and the historical navigation data.

[0043] As described in the above steps, risk features are classified into different scenarios according to their attributes, such as meteorological features, ocean current features, and geographical features. Classification helps to group similar risk features together, which facilitates further analysis of their relationship with historical navigation data. The risk features classified for each scenario are input into the hierarchical clustering model, and the model analyzes the correlation coefficient between each risk feature and the historical navigation data. The hierarchical clustering model can identify which risk features have a strong correlation with the target features and auxiliary features in the navigation risk data, thereby helping to determine which risk features are important. Based on the analysis results of the hierarchical clustering model, the target features and auxiliary features in the navigation risk data corresponding to the risk features are determined. Identifying the target features and auxiliary features helps to more accurately understand the relationship between the risk features and the navigation risk, thereby improving the accuracy of the prediction. Based on the target features, the correlation ratio between the target features classified for each scenario and the historical navigation data is calculated. The calculation of the correlation ratio helps to quantify the degree of correlation between the risk features and the navigation risk data, providing a quantitative basis for decision-making. The correlation ratio is quantified by the hierarchical clustering model to obtain the correlation coefficient between the risk features and the historical navigation data. The quantification of the correlation coefficient helps to more accurately reflect the relationship between risk characteristics and navigation risk data, and provide a quantitative basis for decision-making.

[0044] In one embodiment, the determining risk feature corresponds to the target feature and the auxiliary feature in the navigation risk data, and the method comprises:

[0045] When the scene is classified as a meteorological feature, obtaining the navigation performance parameters and the navigation delay parameters of the navigation risk data, wherein the meteorological feature includes wind speed, wind direction and air pressure parameters;

[0046] Based on the navigation performance parameters, analyzing the speed variation parameters of the ship under the meteorological characteristic conditions;

[0047] Based on the navigation delay parameters, analyzing the number and duration of navigation delays of the ship under the meteorological characteristic conditions;

[0048] Inputting the speed variation parameters and the meteorological characteristic parameters into the hierarchical clustering model, analyzing the meteorological parameters directly and indirectly affecting the navigation system through the hierarchical clustering model, and outputting the analysis results;

[0049] Inputting the number and time of the flight delays and the parameters of the meteorological characteristics into the hierarchical clustering model, and analyzing the meteorological parameters having direct and indirect effects through the hierarchical clustering model;

[0050] The directly affected meteorological parameters are used as target features, and the indirectly affected meteorological parameters are used as auxiliary parameters.

[0051] As mentioned above, navigation performance parameters and navigation delay parameters related to meteorological characteristics are obtained from navigation risk data. Meteorological characteristics are key factors affecting navigation safety. Obtaining these data helps to more accurately assess the impact of meteorological conditions on navigation risks. Based on navigation performance parameters, analyze the change in ship speed under specific meteorological characteristics. Speed ​​change is an important indicator that directly reflects the impact of meteorological conditions on ship navigation. Analyzing speed changes helps to determine the direct relationship between meteorological characteristics and navigation risks. Based on navigation delay parameters, analyze the number and time of navigation delays of ships under specific meteorological characteristics. Navigation delays reflect the impact of meteorological conditions on navigation plans and ship operations. Analyzing delays helps to determine the indirect relationship between meteorological characteristics and navigation risks. Input speed change parameters and meteorological characteristic parameters into the hierarchical clustering model to analyze meteorological parameters with direct and indirect effects. The hierarchical clustering model can help identify which meteorological parameters have a direct impact on navigation risks and which have an indirect impact, thereby more accurately assessing the impact of meteorological conditions on navigation risks. Directly affected meteorological parameters are used as target features, and indirectly affected meteorological parameters are used as auxiliary features. Distinguishing target features from auxiliary features helps to more accurately assess the impact of meteorological conditions on navigation risks and provide more targeted information for risk prediction and management.

[0052] In a possible embodiment, it is assumed that a ship operator uses a method to determine the target features and auxiliary features of meteorological features corresponding to navigation risk data. They have multiple data sources, including meteorological data, navigation logs, ship performance data, etc. These data are input into a risk prediction system in real time.

[0053] Assume that the meteorological characteristics are: wind speed (m / s), wind direction (°), air pressure (hPa)

[0054] Sailing performance parameters: speed (m / s), heading (°), engine status (normal / abnormal)

[0055] Navigation delay parameters: number of delays, delay time (minutes). Collect meteorological data in real time, including wind speed, wind direction, and air pressure. Collect navigation logs, including speed, heading, and engine status. Collect ship performance data, including the number of navigation delays and delay time. Preprocess the data, including cleaning, standardization, and normalization. Obtain navigation performance parameters and navigation delay parameters related to meteorological characteristics from navigation risk data. Based on navigation performance parameters, analyze the speed changes of ships under specific meteorological characteristic conditions. It is found that when the wind speed exceeds a certain threshold, the speed drops significantly. Based on navigation delay parameters, analyze the number and time of navigation delays of ships under specific meteorological characteristic conditions. It is found that when the wind direction changes, the number of delays increases and the delay time is extended. Input the speed change parameters and meteorological characteristic parameters into the hierarchical clustering model to analyze the meteorological parameters with direct and indirect effects. Input the number and time of navigation delays and meteorological characteristic parameters into the hierarchical clustering model to analyze the meteorological parameters with direct and indirect effects. The results of the model analysis show that wind speed is the key factor that directly affects the speed change, while the change in wind direction leads to navigation delays. The meteorological parameters that directly affect (such as wind speed) are used as target features, and the meteorological parameters that indirectly affect (such as wind direction) are used as auxiliary features. Based on the target features and auxiliary features, the navigation risk is predicted and the corresponding risk management strategy is formulated. For example, when the wind speed exceeds the safety threshold, the ship should slow down and consider adjusting the route to avoid the strong wind area.

[0056] In one embodiment, the risk feature is determined to correspond to the target feature and the auxiliary feature in the navigation risk data, and the method further comprises:

[0057] When the scenario is classified as ocean current characteristics, obtaining navigation performance parameters and navigation delay parameters of the navigation risk data, wherein the ocean current characteristics include ocean current speed, ocean current direction, ocean current density and ocean current change rate parameters;

[0058] Based on the navigation performance parameters, analyzing the speed variation parameters of the ship under the characteristic conditions of the ocean current;

[0059] Based on the navigation delay parameters, analyzing the number and duration of navigation delays of the ship under the ocean current characteristic conditions;

[0060] According to the dynamic changes of the ocean current characteristics, the speed change parameters and the parameters of the ocean current characteristics are input into the hierarchical clustering model, and the ocean current parameters directly and indirectly affected are analyzed by the hierarchical clustering model;

[0061] According to the real-time impact of ocean current characteristics on navigation performance and delay, the number and time of navigation delays and the parameters of ocean current characteristics are input into the hierarchical clustering model, the ocean current parameters with direct and indirect impact are analyzed through the hierarchical clustering model, and the analysis results are output;

[0062] The directly affected ocean current parameters are used as target features, and the indirectly affected ocean current parameters are used as auxiliary parameters.

[0063] As mentioned above, the navigation performance parameters and navigation delay parameters related to the current characteristics are obtained from the navigation risk data. The current characteristics are an important factor affecting navigation safety. Obtaining these data helps to more accurately assess the impact of current conditions on navigation risks. Based on the navigation performance parameters, the speed changes of the ship under specific current characteristics are analyzed. The speed change is an important indicator that directly reflects the impact of current conditions on ship navigation. Analyzing the speed change helps to determine the direct relationship between current characteristics and navigation risks. Based on the navigation delay parameters, the number and time of navigation delays of the ship under specific current characteristics are analyzed. Navigation delays reflect the impact of current conditions on navigation plans and ship operations. Analyzing delays helps to determine the indirect relationship between current characteristics and navigation risks. The speed change parameters and current characteristic parameters are input into the hierarchical clustering model to analyze the current parameters with direct and indirect effects. The hierarchical clustering model can help identify which current parameters have a direct impact on navigation risks and which have an indirect impact, thereby more accurately assessing the impact of current conditions on navigation risks. The current parameters with direct impact are taken as the target features, and the current parameters with indirect impact are taken as the auxiliary features. Distinguishing target features from auxiliary features helps to more accurately assess the impact of current conditions on navigation risks and provide more targeted information for risk prediction and management.

[0064] In one embodiment, the risk feature is determined to correspond to the target feature and the auxiliary feature in the navigation risk data, and the method further comprises:

[0065] When the scenario is classified as a geographical feature, obtaining the navigation performance parameters and the navigation delay parameters of the navigation risk data, wherein the geographical features include geographical location, port distance, ocean depth and seabed topography parameters;

[0066] Based on the navigation performance parameters, analyzing the speed variation parameters of the ship under the geographical characteristics;

[0067] Based on the navigation delay parameters, analyzing the number and duration of navigation delays of the ship under geographical characteristics;

[0068] Inputting the speed variation parameters and the parameters of the geographical features into the hierarchical clustering model, and analyzing the geographical parameters directly and indirectly affected by the hierarchical clustering model;

[0069] Inputting the number and time of the flight delays and the parameters of the geographical features into the hierarchical clustering model, analyzing the geographical parameters directly and indirectly affected by the hierarchical clustering model, and outputting the analysis results;

[0070] The directly affected geographical parameters are used as target features, and the indirectly affected geographical parameters are used as auxiliary parameters.

[0071] As mentioned above, navigation performance parameters and navigation delay parameters related to geographical features are obtained from navigation risk data. Geographical features are important factors affecting navigation safety. Obtaining these data helps to more accurately assess the impact of geographical conditions on navigation risks. Based on navigation performance parameters, analyze the speed changes of ships under specific geographical feature conditions. Speed ​​changes are an important indicator that directly reflects the impact of geographical conditions on ship navigation. Analyzing speed changes helps to determine the direct relationship between geographical features and navigation risks. Based on navigation delay parameters, analyze the number and time of navigation delays of ships under specific geographical feature conditions. Navigation delays reflect the impact of geographical conditions on navigation plans and ship operations. Analyzing delays helps to determine the indirect relationship between geographical features and navigation risks. Input speed change parameters and geographical feature parameters into the hierarchical clustering model to analyze the geographical parameters with direct and indirect effects. The hierarchical clustering model can help identify which geographical parameters have a direct impact on navigation risks and which have an indirect impact, thereby more accurately assessing the impact of geographical conditions on navigation risks. Directly affected geographical parameters are used as target features, and indirectly affected geographical parameters are used as auxiliary features. Distinguishing target features from auxiliary features helps to more accurately assess the impact of geographical conditions on navigation risks and provide more targeted information for risk prediction and management.

[0072] In one embodiment, the method further comprises:

[0073] Obtain historical target features and auxiliary features of the same scene, and analyze whether the historical target features and auxiliary features of the same scene have time sequence attributes;

[0074] When the historical target features and auxiliary features of the same scene have a time sequence attribute, the historical target features and auxiliary features are extracted;

[0075] Construct a probability analysis model, input the historical target features and auxiliary features into the prediction model, analyze the probability of the auxiliary features triggering the target features through the prediction model, and output the probability analysis results;

[0076] When the probability analysis result shows that the probability that the auxiliary feature triggers the target feature is greater than a preset probability threshold, the auxiliary feature is marked as a warning feature.

[0077] As described above, data on historical target features and auxiliary features of the same scenario are collected. By analyzing historical data, patterns and trends related to navigation risks can be identified. Analyze whether there is a temporal relationship between historical target features and auxiliary features. Determining the temporal relationship helps to understand the causal relationship between risk features and provide a basis for prediction. When it is found that historical target features and auxiliary features have temporal attributes, these features are extracted. Extracting these features helps to further analyze the relationship between them and navigation risks. Construct a probability analysis model and input historical target features and auxiliary features into the model. The probability analysis model can quantify the probability that auxiliary features trigger target features and provide quantitative risk assessment. When the probability analysis results show that the probability that auxiliary features trigger target features reaches a preset probability threshold, the auxiliary feature is marked as a warning feature. The marking of warning features helps to detect potential risks in a timely manner and provide warning information for decision-making so that preventive measures can be taken.

[0078] In one embodiment, the navigation risk prediction includes navigation accident prediction, ship damage prediction, navigation delay prediction and personnel risk prediction. Analyze historical navigation accident data, identify accident patterns and risk factors, and establish a prediction model. Predicting navigation accidents helps to take preventive measures in advance and reduce the risk of accidents. Analyze ship maintenance records and performance data, identify potential factors that may cause damage, and establish a prediction model. Predicting ship damage helps to detect potential problems in a timely manner, take maintenance measures, and avoid major damage. Analyze weather forecasts, navigation logs and traffic data, predict possible navigation delays, and establish a prediction model. Predicting navigation delays helps to adjust navigation plans and reduce unnecessary delays and costs. Analyze crew work records, health status and psychological state, identify factors that may cause personnel risks, and establish a prediction model. Predicting personnel risks helps to formulate corresponding safety management measures to ensure the safety and health of crew members.

[0079] In one embodiment, after the step of determining the weight coefficient of the risk feature according to the correlation coefficient to reflect the importance level of the risk feature in the navigation risk prediction, the method further includes:

[0080] Analyzing the risk type of the risk feature based on the importance level of the risk feature in the navigation risk prediction, wherein the risk type includes avoidable risk and unavoidable risk;

[0081] When the risk type is an unavoidable risk, obtaining current ship performance data and historical response strategies for the risk characteristics;

[0082] According to the current ship performance data and the historical response strategy of the risk characteristics, the current ship performance data is simulated and tested using the historical response strategy, and the historical response strategy is adjusted according to the test result as the current response strategy, wherein the response strategy includes adjusting the speed, adjusting the load and adjusting the route;

[0083] When the risk type is an avoidable risk, determining the target feature and auxiliary feature of the risk feature, and formulating an avoidance strategy according to the target feature and the auxiliary feature;

[0084] The avoidance strategy and / or the response strategy of the risk characteristic is input into the data framework.

[0085] As mentioned above, the existing risk prediction system can usually only identify and predict some common avoidable risks, such as meteorological conditions, current speed, etc. When a ship encounters sudden bad weather, such as a storm, the existing risk prediction system may not be able to accurately predict the arrival of the storm, resulting in the ship being unable to take avoidance measures in time. It is necessary to face the risk and reduce the danger. This application uses the correlation coefficient to determine the weight coefficient of the risk feature to reflect its importance level in the navigation risk prediction. Help identify the most important features for navigation risk prediction and provide a basis for subsequent analysis. Based on the weight coefficient of the risk feature, analyze the risk type of the risk feature, including avoidable risks and unavoidable risks. Provide a basis for formulating targeted response strategies. Obtain the historical response strategy of the current ship performance data and risk features, simulate and test the current ship performance data using the historical response strategy, and adjust the historical response strategy as the current response strategy based on the test results. Improve the ability to respond to unavoidable risks and reduce losses. Determine the target features and auxiliary features of the risk feature, and formulate an avoidance strategy based on the target features and auxiliary features. Through the avoidance strategy, reduce the probability of risk and improve the safety of navigation. Input the avoidance strategies and / or response strategies of risk characteristics into the data framework. Integrate risk characteristics and their response strategies into a structured data model for easy management and application.

[0086] In one feasible embodiment, when a ship encounters a storm, the impact of the storm may not be avoided, but the risk can be reduced by adjusting the speed, adjusting the load and adjusting the route. Analyze the risk index of the crew in combination with the current ship performance data and the adjusted coping strategy. Determine the factors that affect the crew risk index, such as workload, psychological pressure, fatigue level, etc. Establish a crew risk index assessment model, taking into account the crew's working conditions, physiological state and psychological state. Evaluate the risk index of the crew under the current coping strategy. If the crew's risk index is greater than the preset index threshold, it is judged that the coping strategy does not meet the output conditions. Readjust the coping strategy according to the risk index to reduce the crew's risk index.

[0087] In a specific embodiment, after the step of adjusting the response strategy according to the test results, the method further includes:

[0088] Obtain the crew's current physiological parameters, and combine the physiological parameters with current ship performance data to assess the crew's acceptable risk index range;

[0089] Inputting the acceptable risk index range of the crew, the current ship performance data and the response strategy into a preset evaluation model respectively, and analyzing whether the response strategy is within the acceptable risk index range of the crew through the evaluation model;

[0090] If the response strategy is within the risk index range acceptable to the crew, it is judged that the response strategy meets the output conditions;

[0091] Based on the conditions for satisfying the output, the adjusted response strategy is used as the response strategy for the current unavoidable risks.

[0092] As mentioned above, sensors or other monitoring devices can be used to obtain physiological parameters of the crew in real time, such as heart rate, blood pressure, body temperature, etc. Providing instant health status information of the crew can help quickly identify possible health problems. For example, when encountering severe weather, real-time monitoring of the crew's heart rate and blood pressure can timely understand the physiological response of the crew under stress, so as to take measures to prevent overwork or health crisis. Combine the crew's physiological parameters with the ship's performance data (such as stability, speed, etc.) and calculate the risk index through an algorithm. Determine the level of risk that the crew can withstand under specific operating conditions. If the ship is sailing in a storm, combining the crew's physiological data with the ship's stability data can help decide whether it is necessary to adjust the route or take other measures to reduce the risk. Analyze the input data using a preset evaluation model to determine the effectiveness of the response strategy. Ensure that the response strategy matches the current risk situation and is safe for the crew. If the response strategy includes having the crew perform emergency repairs, the evaluation model will analyze whether such repairs are feasible and safe under the current ship performance and crew physiological status. The evaluation model will compare the risks that the response strategy may bring with the acceptable risk index range for the crew. Prevent the adoption of response measures that may cause damage to the health or safety of the crew. If the response strategy requires the crew to work under high-risk conditions, the assessment model can identify this situation and recommend safer alternatives. After confirming that the risk level of the response strategy is within the acceptable range, it is considered a feasible option. Ensure that the output response strategy is both effective and safe. After confirming that the physiological parameters of all crew members are within the normal range and the ship's performance is stable, the response strategy adopted is more likely to succeed and ensure the safety of the crew. Adjust and implement the response strategy based on the assessment results. Provide a verified response plan for the current risk. After adjusting the response strategy, such as reducing the crew's working hours or providing additional safety equipment, it can ensure that the safety and health of the crew are maximally protected when facing unavoidable risks.

[0093] In another embodiment, suppose we have a cargo ship crossing a sea area that has multiple risk features in past navigation records, such as encountering a large number of fragmented ice blocks and reefs during navigation and being unable to bypass.

[0094] The specific method is: through data analysis, determine the weight coefficients of the following risk characteristics:

[0095] Broken ice: weight factor 0.8

[0096] Reef: weight factor 0.9

[0097] Analysis shows that both fragmented ice and reefs are unavoidable risks. Current ship performance data: ship's icebreaking capability, structural strength, speed, etc.

[0098] Historical response strategy: In similar situations, ships usually slow down and choose the safest path to go straight through the obstacle. Simulation tests were conducted using current ship performance data and historical response strategies. The test results showed that it is feasible to choose a smaller and relatively safe ice block to hit it head-on without exceeding a speed of 5 knots.

[0099] According to the test results, the response strategy was adjusted: it was decided to reduce the speed to 5 knots and choose a relatively safe path to directly pass through the fragmented ice. The current physiological parameters of the crew: the average heart rate is 80 beats / minute, and the average blood pressure is 110 / 70mmHg. Combining physiological parameters and ship performance data, the acceptable risk index range for the crew is evaluated to be 0.4-0.8. The acceptable risk index range for the crew (0.4-0.8), current ship performance data (5 knots speed, icebreaker structural strength) and response strategy (direct impact with smaller ice) were input into the evaluation model. The analysis results of the evaluation model showed that the risk index of the response strategy for direct impact with smaller ice was 0.6, which was within the acceptable risk index range for the crew. Since the response strategy meets the output conditions, the adjusted response strategy (decelerating to 5 knots and choosing smaller ice to face the impact) is used as the current response strategy for unavoidable risks. The response strategy for direct impact is input into the data framework of navigation risk data and is ready for implementation. From the above analysis, the ship must face the unavoidable risks and reduce the damage caused by the impact by slowing down and choosing a relatively safe path. This approach demonstrates how to ensure the safety of crew and vessel in the face of extreme situations.

[0100] The data fusion and integration method based on multi-source navigation data of the present application can extract risk features from multiple data sources, including those features that are strongly associated with navigation risks, by constructing association analysis models and hierarchical clustering models. The weight coefficient of each risk feature is determined according to the correlation coefficient to reflect its level of importance in navigation risk prediction. The determination of the weight coefficient helps to improve the prediction performance and stability of the model, enabling decision makers to formulate corresponding risk management strategies based on the importance of risk features. A data framework for navigation risk data is established to integrate risk features and their weight coefficients. The data framework can be used as a decision support tool to help ship operators and maritime administrations better understand navigation risks and formulate corresponding risk management strategies. Ship operators can process multi-source navigation data more effectively and improve the accuracy of navigation risk prediction, thereby ensuring navigation safety and optimizing navigation plans.

[0101] Reference Figure 3 In an embodiment of the present application, a data fusion and integration system based on multi-source navigation data is also provided, including:

[0102] The first acquisition module 1 is used to acquire the marine environment data in real time and identify the sea ice density parameter in the marine environment data;

[0103] A second acquisition module 2 is used to acquire the current preset channel position data, and mark the target sea ice data within the preset range of the channel position data according to the channel position data and the sea ice density parameter;

[0104] A judgment module 3 is used to judge whether the target sea ice data meets the conditions for replacing the channel position data;

[0105] A determination module 4 is used to determine that the target sea ice data meets the condition for replacing the channel position data when the sea ice concentration parameter constitutes a passing obstacle on the channel position data;

[0106] A route avoidance module 5, configured to determine route avoidance data according to the target sea ice data based on the result of the determination;

[0107] The modification module 6 is used to modify the channel position data based on the route avoidance data.

[0108] As described above, it can be understood that the various components of the data fusion and integration system based on navigation multi-source data proposed in this application can realize the functions of any of the data fusion and integration methods based on navigation multi-source data described above, and the specific structure will not be repeated.

[0109] Reference Figure 4 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as monitoring data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a data fusion and integration method based on multi-source navigation data is implemented.

[0110] The above-mentioned processor executes the above-mentioned data fusion and integration method based on navigation multi-source data, including: constructing an association analysis model to extract the risk characteristics of data; acquiring source data of different scenario types in real time; inputting different types of source data into the association analysis model, and extracting the risk characteristics of the source data through the association analysis model; setting a hierarchical clustering model, inputting each risk characteristic into the hierarchical clustering model respectively, and analyzing the correlation coefficient between each risk characteristic and the historical navigation data through the hierarchical clustering model; determining the weight coefficient of the risk characteristic according to the correlation coefficient to reflect the importance level of the risk characteristic in navigation risk prediction; constructing a data framework of navigation risk data, and integrating the risk characteristics and their weight coefficients into the data framework.

[0111] The above-mentioned data fusion and integration method based on multi-source navigation data can extract risk features from multiple data sources, including those with strong correlation with navigation risks, by constructing association analysis models and hierarchical clustering models. The weight coefficient of each risk feature is determined according to the correlation coefficient, reflecting its importance level in navigation risk prediction. The determination of the weight coefficient helps to improve the prediction performance and stability of the model, enabling decision makers to formulate corresponding risk management strategies based on the importance of risk features. Construct a data framework for navigation risk data to integrate risk features and their weight coefficients. The data framework can be used as a decision support tool to help ship operators and maritime administration agencies better understand navigation risks and formulate corresponding risk management strategies. Ship operators can process multi-source navigation data more effectively and improve the accuracy of navigation risk prediction, thereby ensuring navigation safety and optimizing navigation plans.

[0112] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a data fusion and integration method based on multi-source navigation data is implemented, including the following steps: constructing an association analysis model for extracting risk characteristics of data; acquiring source data of different scenario types in real time; inputting different types of source data into the association analysis model, and extracting risk characteristics of the source data through the association analysis model; setting a hierarchical clustering model, inputting each risk characteristic into the hierarchical clustering model respectively, and analyzing the correlation coefficient between each risk characteristic and historical navigation data through the hierarchical clustering model; determining a weight coefficient of the risk characteristic based on the correlation coefficient to reflect the importance level of the risk characteristic in navigation risk prediction; and constructing a data framework for navigation risk data, and integrating the risk characteristics and their weight coefficients into the data framework.

[0113] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0114] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0115] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A data fusion and integration method based on multi-source navigation data for navigation risk prediction. It is characterized in that The method comprises: Build an association analysis model to extract risk characteristics of data; Obtain source data of different scene types in real time; Inputting different types of source data into the association analysis model, and extracting risk features of the source data through the association analysis model; Setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model respectively, and analyzing the correlation coefficient between each risk feature and the historical navigation data through the hierarchical clustering model; Determine a weight coefficient of the risk feature according to the correlation coefficient, which is used to reflect the importance level of the risk feature in the navigation risk prediction; A data framework for navigation risk data is constructed, and the risk characteristics and their weight coefficients are integrated into the data framework.

2. The data fusion and integration method based on multi-source navigation data according to claim 1, It is characterized in that The method of inputting each risk feature into the hierarchical clustering model respectively and analyzing the correlation coefficient between each risk feature and the historical navigation data by the hierarchical clustering model comprises: Obtaining a scenario classification corresponding to a risk feature, inputting the risk feature of each scenario classification into the hierarchical clustering model, mapping the risk feature into the historical navigation data through the hierarchical clustering model, and determining the target feature and auxiliary feature corresponding to the risk feature in the navigation risk data; Based on the target features, the correlation ratio between the target features and the historical navigation data of each scenario classification is calculated, where the scenario classification includes meteorological features, ocean current features and geographical features; The association ratio is quantified by the hierarchical clustering model to obtain the correlation coefficient between the risk characteristics and the historical navigation data.

3. The data fusion and integration method based on multi-source navigation data according to claim 2, It is characterized in that The method of determining that the risk feature corresponds to the target feature and the auxiliary feature in the navigation risk data comprises: When the scene is classified as a meteorological feature, obtaining the navigation performance parameters and the navigation delay parameters of the navigation risk data, wherein the meteorological feature includes wind speed, wind direction and air pressure parameters; Based on the navigation performance parameters, analyzing the speed variation parameters of the ship under the meteorological characteristic conditions; Based on the navigation delay parameters, analyzing the number and duration of navigation delays of the ship under the meteorological characteristic conditions; Inputting the speed variation parameters and the meteorological characteristic parameters into the hierarchical clustering model, analyzing the meteorological parameters directly and indirectly affecting the navigation system through the hierarchical clustering model, and outputting the analysis results; Inputting the number and time of the flight delays and the parameters of the meteorological characteristics into the hierarchical clustering model, and analyzing the meteorological parameters having direct and indirect effects through the hierarchical clustering model; The directly affected meteorological parameters are used as target features, and the indirectly affected meteorological parameters are used as auxiliary parameters.

4. The data fusion and integration method based on multi-source navigation data according to claim 3, It is characterized in that The determining risk feature corresponds to the target feature and the auxiliary feature in the navigation risk data, and the method further comprises: When the scenario is classified as ocean current characteristics, obtaining navigation performance parameters and navigation delay parameters of the navigation risk data, wherein the ocean current characteristics include ocean current speed, ocean current direction, ocean current density and ocean current change rate parameters; Based on the navigation performance parameters, analyzing the speed variation parameters of the ship under the characteristic conditions of the ocean current; Based on the navigation delay parameters, analyzing the number and duration of navigation delays of the ship under the ocean current characteristic conditions; According to the dynamic changes of the ocean current characteristics, the speed change parameters and the parameters of the ocean current characteristics are input into the hierarchical clustering model, and the ocean current parameters directly and indirectly affected are analyzed by the hierarchical clustering model; According to the real-time impact of ocean current characteristics on navigation performance and delay, the number and time of navigation delays and the parameters of ocean current characteristics are input into the hierarchical clustering model, the ocean current parameters with direct and indirect impact are analyzed through the hierarchical clustering model, and the analysis results are output; The directly affected ocean current parameters are used as target features, and the indirectly affected ocean current parameters are used as auxiliary parameters.

5. The data fusion and integration method based on multi-source navigation data according to claim 2, It is characterized in that The method further comprises: Obtain historical target features and auxiliary features of the same scene, and analyze whether the historical target features and auxiliary features of the same scene have time sequence attributes; When the historical target features and auxiliary features of the same scene have a time sequence attribute, the historical target features and auxiliary features are extracted; Construct a probability analysis model, input the historical target features and auxiliary features into the prediction model, analyze the probability of the auxiliary features triggering the target features through the prediction model, and output the probability analysis results; When the probability analysis result shows that the probability that the auxiliary feature triggers the target feature is greater than a preset probability threshold, the auxiliary feature is marked as a warning feature.

6. The data fusion and integration method based on multi-source navigation data according to claim 2, It is characterized in that After the step of determining the weight coefficient of the risk feature according to the correlation coefficient to reflect the importance level of the risk feature in the navigation risk prediction, the method further includes: Analyzing the risk type of the risk feature based on the importance level of the risk feature in the navigation risk prediction, wherein the risk type includes avoidable risk and unavoidable risk; When the risk type is an unavoidable risk, obtaining current ship performance data and historical response strategies for the risk characteristics; According to the current ship performance data and the historical response strategy of the risk characteristics, the current ship performance data is simulated and tested using the historical response strategy, and the historical response strategy is adjusted according to the test result as the current response strategy, wherein the response strategy includes adjusting the speed, adjusting the load and adjusting the route; When the risk type is an avoidable risk, determining the target feature and auxiliary feature of the risk feature, and formulating an avoidance strategy according to the target feature and the auxiliary feature; The avoidance strategy and / or the response strategy of the risk characteristic is input into the data framework.

7. The data fusion and integration method based on multi-source navigation data according to claim 6, It is characterized in that After the step of adjusting the response strategy according to the test results, the method further includes: Obtain the crew's current physiological parameters, and combine the physiological parameters with current ship performance data to assess the crew's acceptable risk index range; Inputting the acceptable risk index range of the crew, the current ship performance data and the response strategy into a preset evaluation model respectively, and analyzing whether the response strategy is within the acceptable risk index range of the crew through the evaluation model; If the response strategy is within the risk index range acceptable to the crew, it is judged that the response strategy meets the output conditions; Based on the conditions for satisfying the output, the adjusted response strategy is used as the response strategy for the current unavoidable risks.

8. A data fusion and integration system based on multi-source navigation data. It is characterized in that include: A construction module is used to build an association analysis model to extract risk features of data; The acquisition module is used to obtain source data of different scene types in real time; An input module, used to input different types of source data into the association analysis model, and extract risk features of the source data through the association analysis model; An analysis module, used for setting a hierarchical clustering model, inputting each risk feature into the hierarchical clustering model, and analyzing the correlation coefficient between each risk feature and the historical navigation data through the hierarchical clustering model; A determination module, used to determine a weight coefficient of the risk feature according to the correlation coefficient, so as to reflect the importance level of the risk feature in the navigation risk prediction; The integration module is used to construct a data framework of navigation risk data and integrate the risk characteristics and their weight coefficients into the data framework.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images