A remote real-time early warning system for ships based on cloud computing

Abstract

The present invention provides a ship remote real-time early warning system based on cloud computing, which relates to the technical field of data processing, and comprises: a data acquisition module, which is used for acquiring ship operation data, environmental status data and external interaction data; a data preprocessing module, which is used for generating standard processing data; a data grouping module, which is used for dividing into a priority data group and a common data group; a data transmission module, which is used for sending to a cloud analysis module; the cloud analysis module, which is used for performing multi-source data aggregation processing on the priority data group and the common data group respectively, performing real-time evaluation on the ship operation status, identifying abnormal behavior recognition and emergency event prediction, and generating corresponding final early warning instructions, including avoidance operation instructions and route adjustment suggestions; an early warning feedback module, which is used for sending the early warning instructions to a ship terminal; and a local response module, which is used for performing an immediate trigger operation on the instruction classification data. The present invention improves the autonomy and accuracy of the ship remote real-time early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a cloud computing-based ship remote real-time early warning system. Background Art

[0002] With the rapid development of the shipping industry, the improvement of navigation safety and operational efficiency has become the focus of the industry. In the existing technology, the ship monitoring and early warning system usually collects the ship's operating status and environmental data through shipboard sensors, and uses wireless communication technology or satellite networks to transmit these data to the shore-based control center for analysis and processing. This type of technology can realize basic functions such as ship positioning and equipment operating status monitoring. However, the traditional shore-based monitoring mode has great limitations. For example, when processing complex data in the ocean environment, real-time performance is difficult to guarantee due to communication delays and bandwidth limitations. In addition, the sole reliance on shore-based analysis also makes the ship lack autonomy and unable to respond quickly to emergencies. This method is difficult to adapt to the high requirements of the modern shipping industry for real-time warnings.

[0003] Specifically, when a ship crosses a dense sea route, the ship's position, heading, surrounding environment, and other dynamic information of other ships need to be collected and processed in real time. Traditional systems rely too much on shore-based analysis centers, which may cause the data analysis results to become invalid due to delays, making it impossible to generate risk avoidance instructions in a timely manner. This delay will significantly increase the risk of collision in areas with dense ships. Therefore, there is an urgent need for a remote real-time early warning system based on cloud computing that can make full use of the powerful computing power and low-latency communication of cloud resources to achieve a fast and accurate early warning mechanism that collaborates between ships and the cloud to address the limitations of existing technologies. Summary of the invention

[0004] The purpose of the present invention is to provide a cloud computing-based ship remote real-time early warning system, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A cloud computing-based ship remote real-time early warning system, the system comprising:

[0007] The data acquisition module is used to collect ship operation data, environmental status data and external interaction data through various types of sensors. The operation data includes ship speed, heading and equipment operation parameters. The environmental status data includes real-time meteorological information and sea condition parameters. The external interaction data includes the dynamic position and navigation information of surrounding ships.

[0008] The data preprocessing module is used to perform denoising, formatting and normalization on the operation data, environmental status data and external interaction data to generate standard processed data;

[0009] A data grouping module is used to determine data priority according to data type, data time sensitivity, data source, and data collection scenario, and divide standard processing data into priority data groups and common data groups according to data priority;

[0010] A data transmission module, used to send the priority data group and the common data group to the cloud analysis module through the wireless communication network according to different transmission protocols and transmission channels according to the data priority;

[0011] The cloud analysis module is used to generate warning instructions according to the priority data group and the common data group through the following steps:

[0012] Perform multi-source data aggregation processing on the priority data group and the common data group respectively to generate final aggregated data;

[0013] Based on the aggregated data, the ship's operating status is evaluated in real time, abnormal behavior recognition and emergency event prediction are identified, and risk level determination results are generated;

[0014] Generate corresponding final warning instructions based on the risk level determination results, including avoidance operation instructions and route adjustment suggestions;

[0015] The warning feedback module is used to assign warning priorities to warning instructions according to the risk level determination results, generate instruction classification data, and send the instruction classification data to the ship terminal in sequence through a low-latency communication link;

[0016] The local response module is used to perform immediate trigger operations on command classification data to adjust ship navigation and regulate the status of mechanical equipment.

[0017] Preferably, the data acquisition module comprises:

[0018] An environmental collection unit is used to collect meteorological information and sea condition parameters through meteorological sensors and sea condition monitoring equipment respectively, and generate environmental status data;

[0019] The operation collection unit is used to collect the speed, heading and equipment operation parameters through the ship-borne navigation system and mechanical equipment sensors respectively;

[0020] The external interactive collection unit is used to receive the dynamic position and navigation information of surrounding ships in real time through the ship automatic identification system.

[0021] Preferably, the data preprocessing module includes:

[0022] A data filtering unit is used to clean the operation data, environmental status data and external interaction data to generate denoised data;

[0023] A data conversion unit, used for encoding the denoised data according to different types to generate preliminary structured data;

[0024] The data normalization unit is used to normalize the preliminary structured data to generate standard processed data.

[0025] Preferably, the data grouping module comprises:

[0026] The data characteristic analysis unit is used to calculate the priority score of standard processing data and determine the data priority according to the data type, data time sensitivity, data source, and data collection scenario; the calculation formula of the priority score is:

[0027] ,

[0028] in, Score data priority, is the data update interval, is the reliability score of the data source, and the calculation formula is ,in is the historical error rate of the data, is the data transmission delay, is the weight coefficient of the data collection scenario, is the number of sensors used in the data collection scenario, is the maximum number of sensors supported by the system, is the number of data types involved in the data collection scenario, is the maximum number of data types supported by the system, is the weight coefficient of the data type, , , , and is the adjustment coefficient;

[0029] A data priority allocation unit, used to allocate a priority label to the standard processing data according to the data priority and a preset data priority threshold, wherein the priority label includes a high priority and a normal priority;

[0030] A data grouping unit, used for dividing the standard processing data into a priority data group and a common data group according to the priority label;

[0031] The data dynamic adjustment unit is used to dynamically update the data priority according to the real-time changes in the ship's operating status.

[0032] Preferably, the data transmission module comprises:

[0033] The transmission path selection unit is used to select a transmission channel and a transmission protocol according to the priority label, wherein:

[0034] The priority data group selects a low-latency transmission channel and combines it with the UDP protocol for transmission, while the common data group selects a high-bandwidth transmission channel and combines it with the TCP protocol for transmission;

[0035] Switch the transmission protocol in real time according to the network status. When the low-latency channel is unstable, switch to the backup channel and use the QUIC protocol first.

[0036] A data compression unit, used for compressing a common data group to generate a compressed data packet;

[0037] The data confirmation unit is used to receive a confirmation signal from the cloud after each data packet is sent. If no confirmation signal is received, the unconfirmed data packet is automatically resent through the backup channel, wherein:

[0038] The confirmation mechanism for the priority data group is a fast retransmission mechanism, and the confirmation mechanism for the normal data group is a batch confirmation mechanism.

[0039] Preferably, multi-source data aggregation processing is performed on the priority data group and the common data group respectively to generate final aggregated data, including:

[0040] The multi-source data in the priority data group and the general data group are processed in layers according to the data type, and the data of the same type are classified and integrated to generate a layered data set;

[0041] Align the data in the hierarchical data set according to the timestamp to eliminate the time difference between different data sources and generate time-aligned data;

[0042] Conduct correlation analysis on time-aligned data, identify the coupling relationship between ship operation status data and environmental status data, and generate associated aggregated data;

[0043] The associated aggregated data is calibrated based on historical navigation data to eliminate existing sensor bias and generate final aggregated data.

[0044] Preferably, based on the final aggregated data, the ship's operating status is evaluated in real time, abnormal behavior recognition and emergency event prediction are identified, and risk level determination results are generated, including:

[0045] Based on the final aggregated data, the behavioral characteristics of the ship's operation are extracted, including speed fluctuation characteristics, heading deviation characteristics, and equipment operation status characteristics;

[0046] Perform abnormal detection on the behavior characteristics of ship operation according to preset behavior characteristic thresholds to generate abnormal behavior data;

[0047] Based on the abnormal behavior data and the environmental status data in the final aggregated data, a dynamic prediction algorithm is used to predict possible emergencies in the future and generate emergency data. The emergencies include collision risks, equipment failures, and severe sea conditions risks.

[0048] Based on the abnormal behavior data and emergency event data, the current ship status is risk assessed, the risk score is calculated, and the risk level determination result is generated according to the preset risk threshold range. The risk levels include low risk, medium risk and high risk. The calculation formula for the risk score is:

[0049] ,

[0050] in, Score the risk, is the total number of operational characteristics monitored, For the current The real-time value of the operating characteristics, For the current The expected value of a running characteristic, For the current The direction value of the running characteristic, For the current The expected direction value of the running feature, is the total number of emergency event types, is the total number of emergencies, For the The weight coefficient of the emergency event, For the The historical data weight coefficient of the emergency event, For the Kind The historical risk value of an emergency event is expressed as the historical total data. The ratio of the number of emergencies to the total number of emergencies. For the The real-time data weight coefficient of the emergency event, For the Kind The real-time risk value of each emergency event, is the current wind speed, is the safety threshold of wind speed, is the current wave height, is the safety threshold of wave height, , and is the weight coefficient;

[0051] , For the The total number of risk features associated with each incident, For the Kind The first The real-time value of each risk feature, For the Kind The first The safety threshold of each risk feature is For the Kind The first The weight coefficient of each risk feature.

[0052] Preferably, a corresponding final warning instruction is generated according to the risk level determination result, and the final warning instruction includes an avoidance operation instruction and a route adjustment suggestion, including:

[0053] Generate preliminary warning instructions based on the risk level determination results, including equipment operation suggestions and route adjustment suggestions;

[0054] The preliminary warning instructions are divided into global instructions and local instructions according to the scope of influence. The global instructions include route adjustment plans, and the local instructions include equipment operation plans.

[0055] According to the ship's operating constraints and navigation mission requirements, global and local instructions are optimized to generate final warning instructions.

[0056] Preferably, the early warning feedback module includes:

[0057] The warning priority determination module is used to determine the warning priority of the warning instruction according to the risk level determination result and generate instruction classification data;

[0058] The data distribution unit is used to send the command classification data to different modules of the ship terminal in sequence through a low-latency communication link according to the warning priority.

[0059] Preferably, the local response module includes:

[0060] A navigation adjustment unit, used to adjust the ship's course and speed in real time according to the global instructions in the instruction classification data, and generate route adjustment data;

[0061] An equipment adjustment unit, used for adjusting the equipment operation state and equipment adjustment data according to the local instructions in the instruction classification data;

[0062] The status feedback unit is used to feed back route adjustment data and equipment adjustment data to the cloud to optimize early warning instructions.

[0063] The above solution of the present invention includes at least the following beneficial effects:

[0064] The present invention provides a remote real-time early warning system for ships based on cloud computing, which effectively overcomes the limitations of traditional shore-based monitoring modes in terms of real-time and autonomy, and significantly improves the navigation safety and operation efficiency of ships in complex marine environments. In traditional technologies, due to communication delays and bandwidth limitations, data may lag in the process of remote transmission to the shore-based control center, and the analysis results lack real-time performance due to reliance on shore-based resources, making it difficult to meet the rapid early warning needs in high-risk scenarios such as densely populated ship routes. The present invention transmits the operating data, environmental status data, and interactive data collected by the ship to the cloud in real time for processing and analysis, and uses the powerful computing power of cloud computing to achieve rapid processing and analysis of multi-source data, thereby improving the efficiency and accuracy of early warning response.

[0065] First, the data acquisition module of the present invention can collect dynamic information about the ship's operation, environmental status and surrounding ships in real time through multiple types of sensors. It not only covers key data reflecting the safety status of the ship itself, such as speed, heading, and equipment operation status, but also introduces external interactive information such as real-time meteorological information and the dynamic position of surrounding ships. This multi-dimensional data acquisition significantly improves the system's ability to perceive the overall operation status of the ship, and provides accurate data support for subsequent risk assessment and instruction generation.

[0066] Secondly, the present invention uses the data preprocessing module to denoise, format and normalize the collected multi-source data, solves the problem of data heterogeneity, and ensures the accuracy and availability of the data. Combined with the dynamic priority division mechanism of the data grouping module, the system can divide the data into priority data groups and ordinary data groups according to the time sensitivity, importance and source of the data when facing a large amount of complex data, ensuring that key data can be processed first on a low-latency transmission channel, while auxiliary data is transmitted to the cloud through a high-bandwidth channel. This not only optimizes the utilization of communication resources, but also effectively reduces the delay of data analysis.

[0067] The cloud analysis module is one of the core innovations of the system. Through multi-source aggregation processing of priority data groups and ordinary data groups, it can quickly complete the real-time evaluation of the ship's operating status, and identify potential risks and emergencies in combination with dynamic prediction algorithms. Traditional shore-based monitoring systems rely on fixed computing resources for single data analysis, while the present invention utilizes the elastic expansion capabilities of cloud computing resources to complete the integration and analysis of large amounts of data in a short period of time, achieving rapid response in high-risk scenarios. For example, when a ship travels through dense sea routes, the cloud analysis module can identify the dynamic position changes of surrounding ships in real time, and combine heading and speed data to accurately predict potential collision risks, thereby generating avoidance operation instructions and route adjustment suggestions, significantly reducing the risk of collision.

[0068] In addition, the present invention realizes a closed-loop control mechanism from the cloud to the ship through the collaborative work of the early warning feedback module and the local response module. The early warning feedback module can quickly transmit the instructions to the ship terminal through a low-latency communication link according to the priority of the early warning instructions generated by the cloud analysis results; and the local response module can adjust the navigation parameters and equipment operating status of the ship according to the early warning instructions, ensuring the real-time execution capability of the early warning. This setting gives full play to the synergistic advantages of cloud computing and local execution of ships, significantly improving the early warning response speed while reducing the cost of manual decision-making in emergencies.

[0069] In summary, the present invention provides a real-time, efficient, and intelligent ship warning solution in the entire process from data collection, transmission, analysis to execution by introducing cloud computing technology and a multi-layer module collaborative architecture. Especially in high-risk scenarios such as densely populated ship routes or severe sea conditions, the system can quickly respond to emergencies, generate accurate warning instructions, effectively reduce the risks caused by delays or insufficient analysis, and ensure the safety and efficiency of ship navigation. Compared with the traditional shore-based monitoring mode, the present invention not only solves the problems of communication delays and bandwidth limitations, but also realizes the rapid processing and accurate warning of complex data through the full utilization of cloud computing resources, and has significant technical advantages and application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is an architecture diagram of a cloud computing-based ship remote real-time early warning system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0072] like Figure 1 As shown, an embodiment of the present invention provides a cloud computing-based ship remote real-time warning system, the system comprising:

[0073] The data acquisition module is used to collect ship operation data, environmental status data and external interaction data through various types of sensors. The operation data includes ship speed, heading and equipment operation parameters. The environmental status data includes real-time meteorological information and sea condition parameters. The external interaction data includes the dynamic position and navigation information of surrounding ships.

[0074] The data preprocessing module is used to perform denoising, formatting and normalization on the operation data, environmental status data and external interaction data to generate standard processed data;

[0075] A data grouping module is used to determine data priority according to data type, data time sensitivity, data source, and data collection scenario, and divide standard processing data into priority data groups and common data groups according to data priority;

[0076] A data transmission module, used to send the priority data group and the common data group to the cloud analysis module through the wireless communication network according to different transmission protocols and transmission channels according to the data priority;

[0077] The cloud analysis module is used to generate warning instructions according to the priority data group and the common data group through the following steps:

[0078] Perform multi-source data aggregation processing on the priority data group and the common data group respectively to generate final aggregated data;

[0079] Based on the aggregated data, the ship's operating status is evaluated in real time, abnormal behavior recognition and emergency event prediction are identified, and risk level determination results are generated;

[0080] Generate corresponding final warning instructions based on the risk level determination results, including avoidance operation instructions and route adjustment suggestions;

[0081] The warning feedback module is used to assign warning priorities to warning instructions according to the risk level determination results, generate instruction classification data, and send the instruction classification data to the ship terminal in sequence through a low-latency communication link;

[0082] The local response module is used to perform immediate trigger operations on command classification data to adjust ship navigation and regulate the status of mechanical equipment.

[0083] In the embodiment of the present invention, the system provides technical support for the safe navigation and efficient operation of ships in complex marine environments through the collaborative work of multiple functional modules. The data acquisition module can collect key data of ship operation, real-time environmental status information and external interaction information through high-precision sensors to form a full range of operational situation awareness. Operational data such as speed, heading and equipment operating parameters can directly reflect the operating status and potential problems of the ship; environmental data such as real-time meteorological information and sea condition parameters can timely capture changes in severe weather or extreme sea conditions; external interaction data obtains the dynamic position and navigation direction information of surrounding ships through the ship automatic identification system, providing a more comprehensive analysis basis for the early warning system.

[0084] The data preprocessing module effectively improves the quality and consistency of the data by denoising, formatting and normalizing the raw data, ensuring the reliability of subsequent analysis. Combined with the data grouping module, the system divides the data into priority data groups and common data groups according to their importance and real-time nature. The priority data group is quickly transmitted to the cloud through a low-latency transmission channel, ensuring real-time processing of key data; the common data group is transmitted using a high-bandwidth protocol, which not only optimizes resource allocation, but also provides auxiliary information support for the analysis system.

[0085] In the cloud analysis module, the system generates high-precision risk level determination results through multi-source data aggregation, real-time evaluation, and emergency prediction technologies, and outputs avoidance operation instructions and route adjustment suggestions. These instructions are sent to the ship terminal in a low-latency manner through the early warning feedback module. The ship can respond quickly and adjust the course, speed or equipment operating status through the local response module to achieve safe active avoidance. The core technical effect of the system is that through the combination of distributed data processing and cloud computing capabilities, it not only improves the real-time and accuracy of early warnings, but also realizes the automated guarantee of safe navigation of ships.

[0086] In a preferred embodiment of the present invention, the data acquisition module includes:

[0087] An environmental collection unit is used to collect meteorological information and sea condition parameters through meteorological sensors and sea condition monitoring equipment respectively, and generate environmental status data;

[0088] The operation collection unit is used to collect the speed, heading and equipment operation parameters through the ship-borne navigation system and mechanical equipment sensors respectively;

[0089] The external interactive collection unit is used to receive the dynamic position and navigation information of surrounding ships in real time through the ship automatic identification system.

[0090] In the embodiment of the present invention, the data acquisition module uses multiple types of sensors, which are characterized by high precision and real-time performance, and provides a full range of input data support for the system. The environmental acquisition unit captures real-time meteorological information such as wind speed, wind direction, temperature and humidity through meteorological sensors, as well as wave height and flow rate data collected by sea condition monitoring equipment. This information can provide timely basis for early warning of severe weather and complex sea conditions. For example, when the wind speed exceeds the safe navigation threshold, the module can quickly transmit environmental change information, prompting the system to activate the early warning mechanism.

[0091] The operation collection unit relies on the ship's navigation system and equipment sensors to collect speed, heading and equipment operation status parameters. For example, when the equipment sensor detects an abnormal vibration frequency of mechanical equipment, it can report its data to the system in a timely manner, and further determine whether emergency maintenance is required through the cloud analysis module. The external interactive collection unit receives dynamic data of surrounding ships through the automatic identification system (AIS), including their position, speed, heading, etc. These interactive data are particularly important in areas with dense routes and can significantly reduce the probability of collision risks.

[0092] The setting of this module realizes the precision and real-time collection of multi-dimensional data, providing strong support for subsequent data analysis, risk assessment and decision-making.

[0093] In a preferred embodiment of the present invention, the data preprocessing module includes:

[0094] A data filtering unit is used to clean the operation data, environmental status data and external interaction data to generate denoised data;

[0095] A data conversion unit, used for encoding the denoised data according to different types to generate preliminary structured data;

[0096] The data normalization unit is used to normalize the preliminary structured data to generate standard processed data.

[0097] In the embodiment of the present invention, the data preprocessing module significantly improves the processing quality of the raw data and the reliability of subsequent analysis. First, the data filtering unit performs denoising on the collected raw data, which can effectively eliminate abnormal values ​​caused by external interference or equipment aging during the sensor acquisition process. For example, for speed data, if there is a sudden change in a short period of time, the module can classify it as abnormal data and eliminate it through a filtering algorithm, thereby avoiding interference with subsequent risk assessment.

[0098] Secondly, the data conversion unit encodes data from different sources in a unified standard format so that it can be uniformly recognized and processed by subsequent modules of the system. This function is particularly important in multi-source data fusion scenarios, such as the formatting of environmental data, operating data, and interactive data to avoid data incompatibility issues.

[0099] Finally, the data normalization unit normalizes multidimensional data and converts data of different dimensions into the same range, ensuring that the magnitude differences of data from different sources will not bias the analysis results. For example, the magnitudes of wave height and wind speed are different, but after normalization, they can be compared and analyzed under a unified standard. This module effectively improves the data consistency and processing efficiency of the system.

[0100] More specifically, the data filtering unit includes:

[0101] Use the sliding average method: take the average of continuous data over a period of time to smooth out data fluctuations. For example, if there are short-term mutations in the continuous data collected by a ship speed sensor, such as a sudden change from 20 knots to 60 knots, the sliding window can be used to remove outliers.

[0102] Or use median filtering: take the middle value in a data window to replace the outlier value to process single-point abnormal interference data.

[0103] Threshold detection-based: Set a reasonable range, such as a speed range of 0-50 knots, and automatically remove data outside this range.

[0104] Detection based on statistical distribution: Assuming that the data follows a normal distribution, detect data that exceeds two standard deviations of the mean and mark them as outliers for elimination.

[0105] More specifically, the data conversion unit includes:

[0106] Code the data according to the source type. For example, operation data is marked as type code "01", environmental status data is marked as "02", and external interaction data is marked as "03".

[0107] Use key-value pairs to mark and structure data.

[0108] Convert data units: for example, convert wind speed to "m / s" and wave height to "m".

[0109] Timestamp standardization: The timestamps collected by different sensors are unified into UTC format to facilitate subsequent data synchronization and time series analysis.

[0110] More specifically, the data normalization unit can be processed by minimum-maximum normalization, Z-Score normalization, or quantile normalization.

[0111] In a preferred embodiment of the present invention, the data grouping module includes:

[0112] The data characteristic analysis unit is used to calculate the priority score of standard processing data and determine the data priority according to the data type, data time sensitivity, data source, and data collection scenario; the calculation formula of the priority score is:

[0113] ,

[0114] in, Score data priority, is the data update interval, is the reliability score of the data source, and the calculation formula is ,in is the historical error rate of the data, is the data transmission delay, is the weight coefficient of the data collection scenario, is the number of sensors used in the data collection scenario, is the maximum number of sensors supported by the system, is the number of data types involved in the data collection scenario, is the maximum number of data types supported by the system, is the weight coefficient of the data type, , , , and is the adjustment coefficient;

[0115] A data priority allocation unit, used to allocate a priority label to the standard processing data according to the data priority and a preset data priority threshold, wherein the priority label includes a high priority and a normal priority;

[0116] A data grouping unit, used for dividing the standard processing data into a priority data group and a common data group according to the priority label;

[0117] The data dynamic adjustment unit is used to dynamically update the data priority according to the real-time changes in the ship's operating status.

[0118] In an embodiment of the present invention, the data grouping module can dynamically determine the data priority based on factors such as data type, time sensitivity, data source, and collection scenario. The calculation of the priority score integrates multiple key parameters, such as data update time interval, source reliability, transmission delay, and scenario weight. Through reasonable weighted calculation, the module can quickly determine the importance of data. For example, in typhoon weather, environmental data is given a higher priority due to its direct impact on navigation safety; in areas with dense routes, the priority of interactive data is significantly improved due to its predictive value for collision risk.

[0119] Priority data groups are quickly transmitted to the cloud analysis module through low-latency channels, ensuring real-time processing of key data; ordinary data groups are transmitted in batches through high-bandwidth channels, effectively reducing the occupation of communication resources. At the same time, the module supports dynamic adjustment of priorities. For example, when the ship's operating status changes significantly, the priority of environmental data can be automatically increased. This dynamic adjustment mechanism enables the system to flexibly adapt to different navigation scenarios, significantly enhancing the accuracy and real-time nature of early warnings.

[0120] More specifically, the weight coefficients of the data collection scenarios include:

[0121] Predefine weight values ​​according to different collection scenarios. For example:

[0122] Typhoon area: the weight of environmental data is set to 5; in areas with dense air routes: the weight of interactive data is set to 4; in normal scenarios: the weight of operational data is set to 3.

[0123] Real-time monitoring of environmental conditions, such as wind speed and surrounding ship density, triggers weight adjustment based on specific thresholds. For example, when the wind speed exceeds 20m / s, the weight of environmental data is automatically increased to high priority.

[0124] More specifically, the weight coefficients of data types include:

[0125] Preset weight rules: the default weight of operation data is 0.4; the default weight of environmental data is 0.35; the default weight of interaction data is 0.25.

[0126] According to the scene monitoring results, the weight of the data type is adjusted in real time. For example, in a scene with a high risk of collision, the weight of the interaction data can be increased to 5.

[0127] More specifically, the data dynamic adjustment unit includes:

[0128] The priority score is recalculated by combining the weight coefficient of the data collection scenario and the weight coefficient of the data type.

[0129] In a preferred embodiment of the present invention, the data transmission module includes:

[0130] The transmission path selection unit is used to select a transmission channel and a transmission protocol according to the priority label, wherein:

[0131] The priority data group selects a low-latency transmission channel and combines it with the UDP protocol for transmission, while the common data group selects a high-bandwidth transmission channel and combines it with the TCP protocol for transmission;

[0132] Switch the transmission protocol in real time according to the network status. When the low-latency channel is unstable, switch to the backup channel and use the QUIC protocol first.

[0133] A data compression unit, used for compressing a common data group to generate a compressed data packet;

[0134] The data confirmation unit is used to receive a confirmation signal from the cloud after each data packet is sent. If no confirmation signal is received, the unconfirmed data packet is automatically resent through the backup channel, wherein:

[0135] The confirmation mechanism for the priority data group is a fast retransmission mechanism, and the confirmation mechanism for the normal data group is a batch confirmation mechanism.

[0136] In an embodiment of the present invention, the data transmission module greatly improves the efficiency and reliability of data transmission by intelligently selecting different transmission protocols and transmission channels. The priority data group selects a low-latency transmission channel and combines it with the UDP protocol for real-time transmission. The UDP protocol can ensure the rapid transmission of higher priority data while reducing communication delays. For example, environmental status data such as wind speed and wave height can be transmitted to the cloud analysis module with a millisecond response time in severe weather, thereby achieving more timely warnings and avoidance suggestions. The ordinary data group is transmitted through a high-bandwidth transmission channel combined with the TCP protocol. The reliability of the TCP protocol can ensure the integrity of low-priority data, and efficient data transmission can still be achieved even in the case of unstable network quality.

[0137] The dynamic switching function within the module can quickly switch to the backup channel when the transmission is interrupted according to the real-time network status. In particular, when the network delay exceeds the allowable threshold, the system can automatically switch to the QUIC protocol, which combines the low latency of UDP and the reliability of TCP to ensure the stability and continuity of data transmission. In addition, the data compression unit significantly reduces the bandwidth resources required for data transmission by efficiently compressing ordinary data groups, enabling the system to maintain high transmission performance with limited network resources. The data confirmation unit provides a confirmation mechanism for each data packet transmission, and ensures the loss-free transmission of critical data through a fast retransmission strategy, while ordinary data groups are confirmed in batches, further reducing the occupation of network resources.

[0138] This module significantly improves the reliability and efficiency of data transmission through flexible selection of transmission protocols, intelligent switching of channels, and optimization of data confirmation mechanisms, providing technical support for the data flow of the entire early warning system.

[0139] More specifically, the fast retransmission mechanism detects and retransmits lost data packets in a timely manner during the data transmission process to ensure that critical data can reach the receiving end on time. The traditional timeout retransmission method requires waiting for a certain time window before resending the lost data packets. This method may cause large delays in time-sensitive data transmission scenarios.

[0140] The fast retransmit mechanism detects packet loss by monitoring the acknowledgment signal (ACK) from the receiving end. When the sending end receives the same acknowledgment signal three times in a row, the system will immediately determine that the data packet may be lost, skip the waiting timeout process, and directly retransmit the lost data packet. This setting can significantly shorten the time interval from packet loss detection to retransmission, ensuring the real-time nature of key data. For example, when a ship transmits high-priority data such as wind speed and speed in real time, the fast retransmit mechanism can ensure that this data can still be delivered to the cloud on time in the event of network fluctuations, supporting real-time evaluation and command generation of the system.

[0141] More specifically, the batch confirmation mechanism is to solve the problem of too frequent confirmation signals during the transmission of ordinary data groups. In the prior art, if each data packet requires a separate confirmation signal, it will lead to a waste of bandwidth resources and a decrease in communication efficiency. The batch confirmation mechanism groups data packets, such as 10 data packets as a batch, and the receiving end sends a confirmation signal uniformly after receiving the entire batch of data, thereby reducing the frequency of confirmation signals.

[0142] This mechanism is suitable for ordinary data with low real-time requirements, such as equipment operation logs or historical navigation records. In actual applications, if the receiving end finds that a batch of data packets is lost, the system will only retransmit the lost data in that batch without affecting the data transmission of other batches. The batch confirmation mechanism reduces communication overhead, optimizes the use of network resources, and ensures the integrity of ordinary data.

[0143] If there are lost packets in a batch of data, only the lost packets in the batch are retransmitted without retransmitting the entire batch.

[0144] If the receiving end does not receive a complete batch of data within the specified time, it will request the sending end to retransmit the entire batch of data.

[0145] In a preferred embodiment of the present invention, multi-source data aggregation processing is performed on the priority data group and the common data group respectively to generate final aggregated data, including:

[0146] The multi-source data in the priority data group and the general data group are processed in layers according to the data type, and the data of the same type are classified and integrated to generate a layered data set;

[0147] Align the data in the hierarchical data set according to the timestamp to eliminate the time difference between different data sources and generate time-aligned data;

[0148] Conduct correlation analysis on time-aligned data, identify the coupling relationship between ship operation status data and environmental status data, and generate associated aggregated data;

[0149] The associated aggregated data is calibrated based on historical navigation data to eliminate existing sensor bias and generate final aggregated data.

[0150] In the embodiment of the present invention, the multi-source data aggregation processing module provides the system with efficient multi-dimensional data integration capabilities. The module first hierarchically processes the multi-source data in the priority data group and the general data group according to the data type, and classifies and integrates the data of the same type. For example, environmental data, such as wind speed and wave height, and operational data, such as speed and heading, are respectively integrated into a unified hierarchical data set, which can significantly improve the structuring degree of the data and facilitate subsequent analysis and processing.

[0151] Through time alignment technology, the module accurately aligns multi-source data in the layered data set according to the timestamp, eliminating the errors caused by the difference in sensor acquisition time. For example, in the real-time speed and wind speed data collected by the sensor, the timestamp difference may lead to evaluation errors. The time alignment algorithm can dynamically adjust these data to ensure the accuracy of data analysis.

[0152] In addition, the module can accurately identify the coupling relationship between the ship's operating status data and environmental status data through the correlation analysis function. For example, by analyzing the correlation between the speed fluctuation and the wave height, the impact of waves on the stability of the ship's speed can be determined, which is of great significance for navigation optimization under complex sea conditions. The historical data calibration function corrects the current aggregated data through historical navigation data, eliminating the long-term accumulated errors caused by sensor deviations, and providing more reliable data input for subsequent analysis. Ultimately, the generated aggregated data provides a high-quality data foundation for the real-time evaluation of the system and the prediction of emergencies.

[0153] More specifically, correlation analysis is performed on time-aligned data to identify the coupling relationship between ship operation status data and environmental status data, and to generate associated aggregated data. The Pearson correlation coefficient method or cross-correlation method can be used to analyze the lag relationship between time series data. For example, the lag relationship between speed fluctuation and wave height is calculated through cross-correlation to identify the time delay of the impact of waves on speed. Through correlation analysis, the coupling strength between different data can be quantified, providing a scientific basis for subsequent risk assessment and decision-making. For example, in the case of high correlation between wind speed and heading changes, the direct impact of bad weather on ship control can be determined.

[0154] More specifically, the associated aggregated data is calibrated based on historical navigation data to eliminate existing sensor biases and generate final aggregated data. The deviation can be calculated by comparing the real-time collected data with the historical mean. The sliding window technology is used to calculate the dynamic mean and standard deviation of historical data and eliminate outliers in historical data. The real-time data is corrected based on historical deviations. Historical calibration can significantly improve the reliability of sensor data, eliminate errors caused by the decline in sensor accuracy during long-term use, and provide higher accuracy guarantees for the final aggregated data.

[0155] In a preferred embodiment of the present invention, the ship operation status is evaluated in real time based on the final aggregated data, abnormal behavior recognition and emergency event prediction are identified, and risk level determination results are generated, including:

[0156] Based on the final aggregated data, the behavioral characteristics of the ship's operation are extracted, including speed fluctuation characteristics, heading deviation characteristics, and equipment operation status characteristics;

[0157] Perform abnormal detection on the behavior characteristics of ship operation according to preset behavior characteristic thresholds to generate abnormal behavior data;

[0158] Based on the abnormal behavior data and the environmental status data in the final aggregated data, a dynamic prediction algorithm is used to predict possible emergencies in the future and generate emergency data. The emergencies include collision risks, equipment failures, and severe sea conditions risks.

[0159] Based on the abnormal behavior data and emergency event data, the current ship status is risk assessed, the risk score is calculated, and the risk level determination result is generated according to the preset risk threshold range. The risk levels include low risk, medium risk and high risk. The calculation formula for the risk score is:

[0160] ,

[0161] in, Score the risk, is the total number of operational characteristics monitored, For the current The real-time value of the operating characteristics, For the current The expected value of a running characteristic, For the current The direction value of the running characteristic, For the current The expected direction value of the running feature, is the total number of emergency event types, is the total number of emergencies, For the The weight coefficient of the emergency event, For the The historical data weight coefficient of the emergency event, For the Kind The historical risk value of an emergency event is expressed as the historical total data. The ratio of the number of emergencies to the total number of emergencies. For the The real-time data weight coefficient of the emergency event, For the Kind The real-time risk value of each emergency event, is the current wind speed, is the safety threshold of wind speed, is the current wave height, is the safety threshold of wave height, , and is the weight coefficient;

[0162] , For the The total number of risk features associated with each incident, For the Kind The first The real-time value of each risk feature, For the Kind The first The safety threshold of each risk feature is For the Kind The first The weight coefficient of each risk feature.

[0163] In an embodiment of the present invention, the risk assessment and level determination module can evaluate the ship's operating status in real time, predict emergencies, and provide accurate risk level classification through in-depth analysis of the final aggregated data. First, the module extracts the behavioral characteristics of the ship's operation, including key data such as speed fluctuations, heading deviations, and equipment operating status. The extraction of behavioral characteristics provides a core basis for identifying abnormal operating conditions. For example, the system can quickly detect drastic fluctuations in speed and determine whether there is a collision risk by analyzing heading deviations.

[0164] Combined with dynamic prediction algorithms, this module can predict possible future emergencies, such as collision risks, equipment failures, and severe sea conditions, based on extracted behavioral characteristics and environmental status data. For example, by continuously monitoring equipment operating parameters such as temperature and vibration, the system can issue early warnings when the equipment is close to a critical state, thereby avoiding safety hazards caused by equipment failures. Through the risk scoring calculation formula, the module can integrate operating data, emergencies, and environmental data to generate a quantitative risk score value, and divide the risk level into low risk, medium risk, and high risk according to the preset risk threshold range. The determination of the risk level provides a reliable basis for the generation of subsequent warning instructions, ensuring the system's ability to respond quickly to sudden risks.

[0165] More specifically, based on the abnormal behavior data and the environmental status data in the final aggregated data, a dynamic prediction algorithm is used to predict possible emergencies in the future and generate emergency data. The emergencies include collision risks, equipment failures, and severe sea conditions. By analyzing the changing trends of abnormal behavior data and environmental status data, emergencies that may occur in the future can be effectively predicted. Abnormal behavior data include ship speed fluctuations, heading deviations, abnormal equipment operating parameters, etc. Environmental status data include external factors such as wind speed and wave height. By combining these data for dynamic analysis, the system can identify potential signs of risks.

[0166] For example, when the speed fluctuation of a ship exceeds the normal range and the wave height continues to increase, this may be a signal that the ship is becoming more difficult to operate, and the system can predict the possibility of collision risk. For another example, when the equipment sensor continues to detect an abnormal increase in the frequency of mechanical vibration, this may indicate an increased risk of equipment failure. The system can issue an early warning based on these characteristics to remind the crew to check or maintain the equipment.

[0167] The prediction of emergencies is usually combined with dynamic algorithms, such as time series-based data analysis technology, which can monitor the changes of risk factors in real time. For example, in the prediction of equipment failure, by continuously monitoring the operating status of the equipment, such as temperature and vibration frequency, the system can discover long-term accumulated trend changes. This trend change may indicate that the equipment is gradually entering a critical state of failure, and the system can issue maintenance suggestions in advance to avoid safety hazards caused by the failure.

[0168] For the prediction of collision risk, the system can determine the proximity between two ships by real-time monitoring of the ship's speed, heading and the dynamic position data of surrounding ships. When the approach distance is lower than the safety threshold, the system can predict the possibility of future collision based on the dynamic trajectory of surrounding ships and generate avoidance instructions.

[0169] For risk prediction of severe sea conditions, the system can determine whether there is extreme weather such as typhoons and huge waves based on the changing trends of meteorological data, such as wind speed and wave height, and determine whether the route needs to be adjusted to avoid danger based on the ship's current route.

[0170] It can be based on a time series prediction model, such as the ARIMA model, which inputs the feature values ​​of the past R time points and then outputs the predicted probability of an emergency. It can also be based on a machine learning classification model, such as the random forest, which inputs abnormal behavior characteristics and environmental characteristics and then outputs the category of an emergency, such as collision risk, equipment failure, and severe sea conditions.

[0171] In a preferred embodiment of the present invention, a corresponding final warning instruction is generated according to the risk level determination result, and the final warning instruction includes an avoidance operation instruction and a route adjustment suggestion, including:

[0172] Generate preliminary warning instructions based on the risk level determination results, including equipment operation suggestions and route adjustment suggestions;

[0173] The preliminary warning instructions are divided into global instructions and local instructions according to the scope of influence. The global instructions include route adjustment plans, and the local instructions include equipment operation plans.

[0174] According to the ship's operating constraints and navigation mission requirements, global and local instructions are optimized to generate final warning instructions.

[0175] In an embodiment of the present invention, the warning instruction generation module can provide specific avoidance operations and route adjustment suggestions for the ship by analyzing the risk level determination results. The initially generated instructions are subdivided into global instructions and local instructions. Global instructions, such as route adjustment plans, can optimize the ship's navigation path as a whole to avoid entering high-risk areas; local instructions, such as equipment operation plans, optimize and adjust specific equipment or parameters. For example, when sailing in a typhoon area, global instructions can plan a safe route that avoids the typhoon path, while local instructions can adjust the operating status of mechanical equipment to reduce the impact on the hull.

[0176] The module also optimizes the global and local instructions initially generated based on the ship's operating constraints and current mission requirements. For example, the system can generate a safe and efficient route based on the ship's fuel consumption status and time constraints. The final generated warning instructions are transmitted to the ship terminal through the warning feedback module, providing accurate operating guidance for crew members and automated equipment. The core technical effect of this module is that through multi-dimensional optimization analysis, it provides a solution that is both global and local, thereby ensuring the safety and efficiency of ship operations.

[0177] More specifically, a preliminary warning instruction is generated based on the risk level determination result. The preliminary warning instruction includes equipment operation suggestions and route adjustment suggestions, including:

[0178] In the event of abnormal equipment operation, the system can generate operation suggestions for the equipment. For example, when the vibration frequency of the host exceeds the normal range, the system will automatically generate an operation suggestion of "load reduction operation" to prevent further damage to the equipment. When the equipment temperature rises abnormally or the pressure value exceeds the safe range, the system can provide suggestions for reducing the equipment output power or performing a short-term shutdown inspection.

[0179] In addition, in response to long-term high-load operation detected by sensors, the system can generate equipment operation adjustment reminders or regular maintenance recommendations to ensure the continued stable operation of the equipment.

[0180] In terms of route planning, the system can generate adjustment suggestions based on the risk information of the route ahead. For example, when it detects that there is a dense area of ​​ships ahead and the risk of collision is high, the system can suggest changing the course to avoid the high-risk area.

[0181] In severe weather conditions, if high wind speeds or huge waves are detected on the route, the system can provide detour route recommendations to help ships avoid dangerous waters.

[0182] The route adjustment proposal will also take into account the current speed and expected arrival time, achieving a balance between safety and efficiency by optimizing the speed.

[0183] More specifically, the preliminary warning instructions are divided into global instructions and local instructions according to the scope of influence. The global instructions include route adjustment plans, and the local instructions include equipment operation plans, including:

[0184] The core of the global command is the route adjustment plan, which optimizes the overall route planning. For example, if there is bad weather or dense routes in the current sea area, the system will generate a global detour adjustment suggestion to reduce the probability of entering the high-risk area. The route adjustment plan will take into account factors such as the length of the current route, the size of the risk area, and fuel consumption to ensure the safety and economy of navigation.

[0185] Based on the risk assessment results, the global directive may also include speed optimization strategies. For example, when passing through areas with severe sea conditions, it is recommended to reduce speed to reduce wave impact; and after entering a safe route, resume normal speed to ensure on-time arrival at the port.

[0186] Local instructions generate solutions for specific equipment problems. For example, when the host temperature rises abnormally, the system will generate an operation suggestion to reduce the host power; if the operating parameters of the auxiliary equipment exceed the safe range, the system will recommend shutting down for maintenance.

[0187] Local instructions provide immediate solutions to short-term operational needs of the ship, such as adjusting the course or changing the operating status of equipment. For example, when the ship's course deviates from the target route, the system will generate instructions to fine-tune the course; when the wind speed increases suddenly in a short period of time, causing the hull's stability to decrease, the system will recommend adjusting the equipment to balance the ship's center of gravity.

[0188] More specifically, the global and local instructions are optimized according to the ship's operating constraints and navigation mission requirements to generate the final warning instructions, including:

[0189] Combined with the operational constraints: fuel consumption limit, the system will consider the fuel consumption of different routes in the optimization of route adjustment plans. For example, if the detour distance of the alternative route is too long and exceeds the fuel budget, the system will re-evaluate the passability of the risk area and find a balance between fuel consumption and risk avoidance.

[0190] Equipment load limit: For equipment operation plans, the system will optimize the equipment based on the current load conditions. For example, if the equipment is close to full load, the system will recommend reducing the equipment operating time or reducing the output power to avoid further losses.

[0191] Combined with the navigation mission requirements:

[0192] When generating the final speed optimization plan, the system will take into account the arrival time limit of the mission. For example, after detouring a bad weather area, if time permits, it is recommended to reduce the speed to save fuel; if time is tight, it is recommended to increase the speed to make up for the loss of detour time.

[0193] If there are special requirements in the ship's mission, such as sailing to a designated location to pick up cargo as planned, the system will try to meet the mission requirements while avoiding risks. For example, when adjusting the route, it will give priority to alternative routes close to the designated docking point.

[0194] The optimized final warning instructions are output in priority order to ensure that high-priority instructions, such as global bypass instructions, are executed at the first time.

[0195] Local instructions serve as a supplement to ensure that the ship’s equipment and operating status are in the best condition when executing global instructions.

[0196] In a preferred embodiment of the present invention, the early warning feedback module includes:

[0197] The warning priority determination module is used to determine the warning priority of the warning instruction according to the risk level determination result and generate instruction classification data;

[0198] The data distribution unit is used to send the command classification data to different modules of the ship terminal in sequence through a low-latency communication link according to the warning priority.

[0199] In an embodiment of the present invention, the early warning feedback module significantly improves the response speed and execution efficiency of the system by assigning priorities and efficiently transmitting the generated early warning instructions. The module first prioritizes the early warning instructions according to the risk level, for example, marking the avoidance instructions of high-risk areas as high priority to ensure that they are transmitted to the terminal in a very short time. The data distribution unit uses a low-latency communication link to quickly send the classified instruction data to different terminal modules. For example, when a ship enters a dangerous sea area, the system can send high-priority avoidance instructions to the navigation system within a few milliseconds, and transmit equipment adjustment instructions to the mechanical control unit to ensure the linkage response of the entire ship system.

[0200] Through the hierarchical command transmission mechanism, the module can also dynamically adjust the order of command transmission, prioritize the execution of key commands, and reduce the transmission delay and bandwidth usage of non-essential data. The technical effect of this module is reflected in ensuring the efficient transmission and distributed execution of early warning commands, enabling ships to achieve rapid response and precise control in complex environments.

[0201] In a preferred embodiment of the present invention, the local response module includes:

[0202] A navigation adjustment unit, used to adjust the ship's course and speed in real time according to the global instructions in the instruction classification data, and generate route adjustment data;

[0203] An equipment adjustment unit, used for adjusting the equipment operation state and equipment adjustment data according to the local instructions in the instruction classification data;

[0204] The status feedback unit is used to feed back route adjustment data and equipment adjustment data to the cloud to optimize early warning instructions.

[0205] In the embodiment of the present invention, the local response module directly adjusts the ship's operating status and equipment parameters according to the received warning instructions to ensure that the ship can respond to sudden risks in a timely manner. The navigation adjustment function of this module can modify the ship's heading and speed in real time. For example, when a collision risk is identified in the route ahead, the system can quickly adjust the ship's heading to avoid obstacles and optimize the avoidance path by controlling the speed.

[0206] The equipment adjustment function adjusts the operating status of mechanical equipment according to local instructions. For example, when sailing in high-wave areas, the system can adjust the main engine power output according to local instructions to reduce the possibility of equipment failure due to overload. The status feedback function provides accurate input data for the subsequent early warning instruction generation by feeding back the status data after navigation adjustment and equipment adjustment to the cloud analysis module in real time. This module effectively ensures the dynamic adaptability of the ship in complex environments by achieving local rapid response and closed-loop feedback.

[0207] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cloud computing-based ship remote real-time early warning system, characterized in that: The system comprises: The data acquisition module is used to collect ship operation data, environmental status data and external interaction data through various types of sensors. The operation data includes ship speed, heading and equipment operation parameters. The environmental status data includes real-time meteorological information and sea condition parameters. The external interaction data includes the dynamic position and navigation information of surrounding ships. The data preprocessing module is used to perform denoising, formatting and normalization on the operation data, environmental status data and external interaction data to generate standard processed data; A data grouping module is used to determine data priority according to data type, data time sensitivity, data source, and data collection scenario, and divide standard processing data into priority data groups and common data groups according to data priority; A data transmission module, used to send the priority data group and the common data group to the cloud analysis module through the wireless communication network according to different transmission protocols and transmission channels according to the data priority; The cloud analysis module is used to generate warning instructions according to the priority data group and the common data group through the following steps: Perform multi-source data aggregation processing on the priority data group and the common data group respectively to generate final aggregated data; Based on the aggregated data, the ship's operating status is evaluated in real time, abnormal behavior recognition and emergency event prediction are identified, and risk level determination results are generated; Generate corresponding final warning instructions based on the risk level determination results, including avoidance operation instructions and route adjustment suggestions; The warning feedback module is used to assign warning priorities to warning instructions according to the risk level determination results, generate instruction classification data, and send the instruction classification data to the ship terminal in sequence through a low-latency communication link; Local response module, used to perform instant trigger operations on command classification data, adjust ship navigation and adjust the status of mechanical equipment; The data grouping module comprises: The data characteristic analysis unit is used to calculate the priority score of standard processing data and determine the data priority according to the data type, data time sensitivity, data source, and data collection scenario; the calculation formula of the priority score is: ， in, Score data priority, is the data update interval, is the reliability score of the data source, and the calculation formula is ,in is the historical error rate of the data, is the data transmission delay, is the weight coefficient of the data collection scenario, is the number of sensors used in the data collection scenario, is the maximum number of sensors supported by the system, is the number of data types involved in the data collection scenario, is the maximum number of data types supported by the system, is the weight coefficient of the data type, , , , and is the adjustment coefficient; A data priority allocation unit, used to allocate a priority label to the standard processing data according to the data priority and a preset data priority threshold, wherein the priority label includes a high priority and a normal priority; A data grouping unit, used for dividing the standard processing data into a priority data group and a common data group according to the priority label; The data dynamic adjustment unit is used to dynamically update the data priority according to the real-time changes in the ship's operating status.

2. According to the cloud computing-based ship remote real-time early warning system of claim 1, it is characterized in that: The data acquisition module comprises: An environmental collection unit is used to collect meteorological information and sea condition parameters through meteorological sensors and sea condition monitoring equipment respectively, and generate environmental status data; The operation collection unit is used to collect the speed, heading and equipment operation parameters through the ship-borne navigation system and mechanical equipment sensors respectively; The external interactive collection unit is used to receive the dynamic position and navigation information of surrounding ships in real time through the ship automatic identification system.

3. According to the cloud computing-based ship remote real-time early warning system of claim 2, it is characterized in that: The data preprocessing module comprises: A data filtering unit is used to clean the operation data, environmental status data and external interaction data to generate denoised data; A data conversion unit, used for encoding the denoised data according to different types to generate preliminary structured data; The data normalization unit is used to normalize the preliminary structured data to generate standard processed data.

4. According to the cloud computing-based ship remote real-time early warning system of claim 3, it is characterized in that: The data transmission module comprises: The transmission path selection unit is used to select a transmission channel and a transmission protocol according to the priority label, wherein: The priority data group selects a low-latency transmission channel and combines it with the UDP protocol for transmission, while the common data group selects a high-bandwidth transmission channel and combines it with the TCP protocol for transmission; Switch the transmission protocol in real time according to the network status. When the low-latency channel is unstable, switch to the backup channel and use the QUIC protocol first. A data compression unit, used for compressing a common data group to generate a compressed data packet; The data confirmation unit is used to receive a confirmation signal from the cloud after each data packet is sent. If no confirmation signal is received, the unconfirmed data packet is automatically resent through the backup channel, wherein: The confirmation mechanism for the priority data group is a fast retransmission mechanism, and the confirmation mechanism for the normal data group is a batch confirmation mechanism.

5. A cloud computing-based ship remote real-time early warning system according to claim 4, characterized in that: Multi-source data aggregation is performed on the priority data group and the common data group to generate the final aggregated data, including: The multi-source data in the priority data group and the general data group are processed in layers according to the data type, and the data of the same type are classified and integrated to generate a layered data set; Align the data in the hierarchical data set according to the timestamp to eliminate the time difference between different data sources and generate time-aligned data; Conduct correlation analysis on time-aligned data, identify the coupling relationship between ship operation status data and environmental status data, and generate associated aggregated data; The associated aggregated data is calibrated based on historical navigation data to eliminate existing sensor biases and generate final aggregated data.

6. The cloud computing-based ship remote real-time early warning system according to claim 5, characterized in that: Based on the final aggregated data, the ship's operating status is evaluated in real time, abnormal behavior identification and emergency event prediction are identified, and risk level determination results are generated, including: Based on the final aggregated data, the behavioral characteristics of the ship's operation are extracted, including speed fluctuation characteristics, heading deviation characteristics, and equipment operation status characteristics; Perform abnormal detection on the behavior characteristics of ship operation according to preset behavior characteristic thresholds to generate abnormal behavior data; Based on the abnormal behavior data and the environmental status data in the final aggregated data, a dynamic prediction algorithm is used to predict possible emergencies in the future and generate emergency data. The emergencies include collision risks, equipment failures, and severe sea conditions risks. Based on the abnormal behavior data and emergency event data, the current ship status is risk assessed, the risk score is calculated, and the risk level determination result is generated according to the preset risk threshold range. The risk levels include low risk, medium risk and high risk. The calculation formula for the risk score is: ， in, Score the risk, is the total number of operational characteristics monitored, For the current The real-time value of the operating characteristics, For the current The expected value of a running characteristic, For the current The direction value of the running characteristic, For the current The expected direction value of the running feature, is the total number of emergency event types, is the total number of emergencies, For the The weight coefficient of the emergency event, For the The historical data weight coefficient of the emergency event, For the Kind The historical risk value of an emergency event is expressed as the historical total data. The ratio of the number of emergencies to the total number of emergencies. For the The real-time data weight coefficient of the emergency event, For the Kind The real-time risk value of each emergency event, is the current wind speed, is the safety threshold of wind speed, is the current wave height, is the safety threshold of wave height, , and is the weight coefficient; , For the The total number of risk features associated with each incident, For the Kind The first The real-time value of each risk feature, For the Kind The first The safety threshold of each risk feature is For the Kind The first The weight coefficient of each risk feature.

7. The cloud computing-based ship remote real-time early warning system according to claim 6, characterized in that: The corresponding final warning instructions are generated according to the risk level determination results. The final warning instructions include avoidance operation instructions and route adjustment suggestions, including: Generate preliminary warning instructions based on the risk level determination results, including equipment operation suggestions and route adjustment suggestions; The preliminary warning instructions are divided into global instructions and local instructions according to the scope of influence. The global instructions include route adjustment plans, and the local instructions include equipment operation plans. According to the ship's operating constraints and navigation mission requirements, global and local instructions are optimized to generate final warning instructions.

8. The cloud computing-based ship remote real-time early warning system according to claim 7, characterized in that: The early warning feedback module includes: The warning priority determination module is used to determine the warning priority of the warning instruction according to the risk level determination result and generate instruction classification data; The data distribution unit is used to send the command classification data to different modules of the ship terminal in sequence through a low-latency communication link according to the warning priority.

9. The cloud computing-based ship remote real-time early warning system according to claim 8, characterized in that: The local response module includes: A navigation adjustment unit, used to adjust the ship's course and speed in real time according to the global instructions in the instruction classification data, and generate route adjustment data; An equipment adjustment unit, used for adjusting the equipment operation state and equipment adjustment data according to the local instructions in the instruction classification data; The status feedback unit is used to feed back route adjustment data and equipment adjustment data to the cloud to optimize early warning instructions.

Images