Ship lifecycle electronic file construction and tracing system

Abstract

This invention provides a system for constructing and tracing electronic archives of a ship's entire lifecycle, relating to the field of ship engineering technology. It includes: an acquisition module for acquiring operational status data generated during ship operation in real time; an extension module for serializing the collected operational status data to form a status dataset and selecting a baseline status vector, extending two state change axes from the baseline status vector as the origin; and a fitting module for delineating sectors based on the directional differences between the two axes, and setting two types of reference sequences inside and outside each sector; and fitting a continuous elliptical state evolution trajectory of the ship's state change characteristics based on the temporal correlation of the two types of reference sequences. This invention achieves unified management, reliable storage, and efficient traceability of ship lifecycle data.

Description

Technical Field

[0001] This invention relates to the field of marine engineering technology, and in particular to a system for constructing and tracing electronic records of the entire life cycle of a ship. Background Technology

[0002] Although the shipbuilding industry has completed the initial transformation from paper records to digital records, data from all stages of the entire life cycle, including design, construction, operation, and maintenance, are stored in different systems. There is a lack of unified data standards and integration mechanisms, data formats are incompatible, ownership is unclear, and credibility is difficult to verify. At the same time, electronic log specifications have not been fully unified, and cross-entity data flow and traceability face many obstacles.

[0003] When a vessel in operation experienced an equipment malfunction during navigation, the parties involved attempted to trace the root cause of the failure. However, they discovered that the technical parameters from the vessel's design phase, installation records from the construction phase, operational data from the operational phase, and maintenance records from previous maintenance were scattered across multiple independent systems belonging to the design institute, shipyard, and shipowner. The data formats varied and could not be cross-referenced, and some key data was at risk of being tampered with. This caused the failure tracing work to stall, exposing the lack of a unified data management system covering the entire lifecycle of the vessel, as well as the absence of an effective, tamper-proof evidence storage and traceability mechanism. This makes it difficult to achieve complete data integration, reliable verification, and efficient traceability, failing to meet the industry's core needs for vessel safety supervision, fault diagnosis, and compliance management. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a system for constructing and tracing electronic records for the entire life cycle of ships, so as to realize unified management, reliable storage and efficient traceability of ship life cycle data.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] The first aspect is the construction and traceability system for electronic records throughout the entire lifecycle of ships, including:

[0007] The acquisition module is used to acquire real-time operational status data generated by the ship during operation.

[0008] The acquisition module is used to serialize the collected running status data to form a status dataset and select a reference status vector, extending two state change axes with the reference status vector as the origin.

[0009] The acquisition module is used to delineate sectors based on the directional differences between two axes, and to set two types of reference sequences inside and outside the sectors respectively; based on the temporal correlation of the two types of reference sequences, a continuous elliptical state evolution trajectory of the ship's state change characteristics is fitted.

[0010] The acquisition module is used to calculate the data calibration coefficient based on the elliptical state evolution trajectory, and to calibrate the original operating state data using the calibration coefficient to obtain the calibrated data;

[0011] The acquisition module is used to drive and update the digital twin ship model synchronized with the physical ship using calibrated data, and at the same time input the operating status data into the preset artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain the identification results;

[0012] The acquisition module is used to encrypt and upload the updated information, identification results and original operating status data of the digital twin ship model to the blockchain using blockchain technology, so as to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications.

[0013] The acquisition module is used to build an electronic archive of the entire life cycle of a ship using blockchain logs as the only trusted data source, and provides a traceability query interface based on blockchain logs for trusted verification and tracing of the ship's historical status, operation records and abnormal events.

[0014] Furthermore, real-time acquisition of operational status data generated by the ship during operation, including:

[0015] It receives encrypted data streams periodically sent by at least one unmanned vessel in real time via a satellite communication link.

[0016] By decrypting the encrypted data stream and parsing the communication protocol, the structured original runtime status data packets are extracted; and the integrity of the parsed data packets is verified and the source is authenticated to filter out trustworthy data packets.

[0017] Trusted data packets are collected and cached according to their corresponding ship identifiers and data types to form a raw dataset initially organized by ship and time sequence; the data includes ship position, speed and heading, power parameters, equipment status, compartment monitoring data and emergency signals;

[0018] By performing timestamp alignment, missing value marking, and standardization of the data format on the original dataset, a unified time-series runtime status data is finally obtained.

[0019] Furthermore, the collected operational status data is serialized to form a status dataset, and a reference state vector is selected. Two state change axes are extended from the reference state vector as the origin, including:

[0020] The unified time-series operational status data is arranged in chronological order and organized into a time-series dataset containing multidimensional state variables;

[0021] Based on the time series dataset, a state vector at a representative time point is selected from the dataset as the benchmark state vector for analyzing the evolution of ship state.

[0022] By using the selected baseline state vector as the origin of the coordinate system, and based on the statistical characteristics and variation patterns of each state variable in the time series dataset, two mutually orthogonal state change axes are calculated and defined.

[0023] Furthermore, sectors are defined based on the directional differences between the two axes, and two types of reference sequences are set inside and outside each sector. Based on the temporal correlation of the two types of reference sequences, a continuous elliptical state evolution trajectory of the ship's state change characteristics is fitted, including:

[0024] Each state vector in the time series dataset is projected onto a two-dimensional analysis coordinate system formed by two mutually orthogonal state change axes to form a series of time series data points.

[0025] Based on the time series data points, calculate the orientation angle of each data point relative to the origin of the reference state vector; and divide the plane of the entire two-dimensional analysis coordinate system into several sectors according to a preset angle threshold.

[0026] For each sector, the set of time-series data points within the sector is defined as the first type of reference sequence; the set of adjacent time-series data points associated with each sector but located outside the sector boundary is defined as the second type of reference sequence.

[0027] Based on the temporal correlation between the two types of reference sequences corresponding to each sector, an ellipse fitting algorithm is used to fit all time series data points projected onto the two-dimensional analysis coordinate system to obtain a continuous and closed elliptical curve as the state evolution trajectory characterizing the long-term and periodic changes in the ship's state.

[0028] Furthermore, data calibration coefficients are calculated based on the elliptical state evolution trajectory. These coefficients are then used to calibrate the original operating state data, resulting in calibrated data, including:

[0029] Based on the obtained elliptical state evolution trajectory, the principal axis length, eccentricity, and orientation parameters relative to the two-dimensional analysis coordinate system are extracted.

[0030] Using the principal axis length, eccentricity, and orientation parameters, and in accordance with predetermined calculation rules, a set of data calibration coefficients is obtained to compensate for the axial deviation between two states in the two-dimensional analysis coordinate system.

[0031] The data calibration coefficients are applied to the uniform time-series operating status data to correct the component values ​​of the data in the two state change axes defined in the two-dimensional analysis coordinate system, thus obtaining calibrated operating status data that removes long-term trend errors.

[0032] Furthermore, the calibrated data drives and updates the digital twin ship model synchronized with the physical ship, while simultaneously inputting operational status data into a pre-set artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain identification results, including:

[0033] Based on the calibrated operational status data, the digital twin ship model corresponding to the physical ship is driven in real time to update the status parameters in the twin model, so that the status parameters are synchronized with the operational status of the physical ship.

[0034] The calibrated operating status data is simultaneously input into a preset artificial intelligence anomaly detection model; the anomaly detection model performs pattern recognition and deviation analysis in real time based on the data to obtain the identification results of the equipment health status, anomaly type and probability of occurrence.

[0035] Furthermore, the updated information, identification results, and original operational status data of the digital twin ship model are encrypted and uploaded to the blockchain using blockchain technology to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications, including:

[0036] The twin model update information and anomaly detection and identification results contained in the comprehensive status report are combined and encapsulated with the original operating status data in a unified time series according to a predefined structure to obtain a complete data record block.

[0037] Based on the encapsulated data record blocks, they are converted into a standardized data structure that conforms to the requirements of the Maritime Organization's electronic log specifications;

[0038] By applying blockchain cryptography to encrypt and timestamp standardized data structures, and broadcasting the processed data as a transaction to the blockchain network for consensus verification and on-chain storage, a blockchain log with immutable and traceable characteristics is generated.

[0039] Furthermore, using blockchain logs as the sole trusted data source, an electronic archive of the entire ship's lifecycle is constructed. Based on these blockchain logs, a traceability query interface is provided for the trusted verification and tracing of the ship's historical status, operational records, and abnormal events, including:

[0040] By extracting all historical data records related to the target vessel from the blockchain log, in chronological order and by key event identifiers;

[0041] Historical data is recorded, reorganized, classified, and indexed according to the ship lifecycle management standards, a structured electronic archive of ships is constructed, and the archive summary information is stored and anchored in the blockchain;

[0042] Based on the electronic archive and the anchoring information in the blockchain, deploy a traceability query service interface with identity authentication and access control. The interface supports trusted retrieval by time range, event type and data characteristics.

[0043] The system responds to user traceability requests through a deployed query interface, retrieves relevant data records from the electronic archive, and automatically verifies the integrity and authenticity of the data by comparing it with the corresponding evidence anchoring information in the blockchain. Finally, it returns trusted and verified historical status, operation records, and abnormal event traceability reports to the user.

[0044] In a second aspect, a computing device includes:

[0045] One or more processors;

[0046] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to execute the system.

[0047] Thirdly, a computer-readable storage medium storing a program that, when executed by a processor, performs the system.

[0048] The above-described solution of the present invention has at least the following beneficial effects:

[0049] Because this system employs a combination of standardized data acquisition and processing, state evolution modeling and calibration, real-time digital twin updates, artificial intelligence anomaly detection, blockchain encrypted evidence storage, and the deployment of a full lifecycle archive and trusted traceability interface, it effectively overcomes the existing technical problems of scattered storage, incompatible formats, difficulty in ensuring the credibility of ship lifecycle data, and the lack of a unified integration and efficient traceability mechanism. This enables unified management, tamper-proof evidence storage, and accurate traceability of ship lifecycle data, providing reliable support for ship safety supervision, fault root cause investigation, compliance auditing, and cross-entity data collaboration, thereby improving the standardization, security, and efficiency of ship lifecycle management. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of a ship's full life-cycle electronic archive construction and traceability system provided by an embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram of the process of constructing and tracing electronic archives for the entire life cycle of a ship, provided by an embodiment of the present invention. The system calculates data calibration coefficients based on an elliptical state evolution trajectory, calibrates the original operating state data using the calibration coefficients, and obtains calibrated data. Detailed Implementation

[0052] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0053] like Figure 1 As shown, embodiments of the present invention propose a system for constructing and tracing electronic records throughout the entire lifecycle of a ship, including:

[0054] The acquisition module is used to acquire real-time operational status data generated by the ship during operation.

[0055] The extension module is used to serialize the collected running status data to form a status dataset and select a reference status vector, extending two status change axes with the reference status vector as the origin.

[0056] The fitting module is used to delineate sectors based on the directional differences between two axes, and to set two types of reference sequences inside and outside the sectors respectively; based on the temporal correlation of the two types of reference sequences, a continuous elliptical state evolution trajectory of the ship's state change characteristics is fitted.

[0057] The calibration module is used to calculate the data calibration coefficient based on the elliptical state evolution trajectory, and to calibrate the original operating state data using the calibration coefficient to obtain calibrated data.

[0058] The identification module is used to drive and update the digital twin ship model synchronized with the physical ship through calibrated data, and at the same time input the operating status data into the preset artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain the identification results.

[0059] The encryption module is used to encrypt and upload the updated information, identification results and original operating status data of the digital twin ship model to the blockchain using blockchain technology, so as to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications.

[0060] The processing module is used to build an electronic archive of the entire life cycle of a ship using blockchain logs as the sole trusted data source, and to provide a traceability query interface based on blockchain logs for trusted verification and tracing of the ship's historical status, operation records and abnormal events.

[0061] In this embodiment of the invention, by employing a comprehensive set of technical means—including real-time acquisition of ship operation status data, serialization of the data to set a baseline state vector and dual state change axes, construction of a state evolution trajectory through sector division and two types of reference sequences combined with an ellipse fitting algorithm, calculation of calibration coefficients to optimize the data based on the trajectory, driving the digital twin model update with calibration data and combining it with an AI model for anomaly detection, encrypting and uploading relevant data to the blockchain to form a compliant and tamper-proof log, and building a full lifecycle electronic archive based on this log and deploying a trusted traceability interface—the invention effectively overcomes the existing technical problems of lacking a unified integration mechanism for ship lifecycle data, difficulty in ensuring credibility, and low traceability efficiency. This achieves precise optimization of ship operation data, secure storage of key information, and trusted traceability of full lifecycle data, providing reliable data support for ship operation supervision, fault diagnosis, and compliance management, and improving the scientific and efficient nature of ship lifecycle management.

[0062] In a preferred embodiment of the present invention, real-time acquisition of operational status data generated by the ship during operation includes:

[0063] Through a satellite communication link, encrypted data streams periodically transmitted from at least one unmanned vessel are received in real time. Specifically, this involves: first, establishing a stable two-way communication link with the global satellite communication network to ensure stable data transmission for the unmanned vessel in various navigation areas, including open sea and coastal waters; second, each unmanned vessel is equipped with a dedicated satellite communication terminal and dedicated components with data encryption capabilities. During operation, the vessel collects various operational status data in real time, and after collection, the original data is encrypted using a dedicated symmetric encryption algorithm to form a secure encrypted data stream; subsequently, the unmanned vessel transmits the encrypted data stream to the satellite communication network at preset fixed time intervals, and then the encrypted data stream is forwarded to the shore-based data receiving center through the satellite communication network, ensuring the security and stability of data transmission throughout the entire process and preventing data from being stolen or tampered with during transmission.

[0064] By decrypting the encrypted data stream and parsing the communication protocol, structured original operational status data packets are extracted. The parsed data packets undergo integrity verification and source authentication to filter out trustworthy data packets. Specifically, after receiving the encrypted data stream, the shore-based data receiving center calls a decryption algorithm and preset decryption key that match the unmanned vessel's encryption components to decrypt the encrypted data stream segment by segment, restoring the original data transmission packets. Then, according to a preset standardized communication protocol, the decrypted original data transmission packets are parsed to extract structured original operational status data packets containing various operational status information of the vessel. During the parsing process, different types of original data transmission formats are automatically identified and converted to ensure that the extracted data packet structure is uniform and standardized. Next, a cyclic redundancy check algorithm is used to verify the integrity of each parsed data packet, checking for transmission loss, data corruption, etc. Simultaneously, the unique vessel identification information carried in the data packet is accurately compared with a pre-stored database of legitimate vessel identity information to complete the authentication of the data source. Finally, data packets that fail integrity verification or authentication are discarded, retaining only trustworthy data packets that meet security and regulatory requirements.

[0065] Trusted data packets are collected and cached according to their corresponding ship identifiers and data types, forming a preliminary raw dataset organized by ship and time sequence. The data includes ship position, speed and heading, power parameters, equipment status, cabin monitoring data, and emergency signals. Specifically, the selected trusted data packets are classified and collected. First, based on the unique ship identifier carried in each data packet, the data packets of different unmanned vessels are distinguished to ensure that the data of each vessel is stored independently and to avoid confusion between different vessel data. Then, the trusted data packets of each vessel are further subdivided according to data type, and different types of data such as ship position data, speed and heading data, power parameter data, equipment status data, cabin monitoring data, and emergency signal data are respectively assigned to their corresponding dedicated dataset directories. At the same time, a high-speed cache server is used to temporarily store the classified data packets. During the caching process, a timestamp corresponding to the reception time is added to each data packet. Finally, a preliminary raw dataset is formed, which is distinguished by ship identity, classified by data type, and sorted by reception time.

[0066] By standardizing the original dataset through timestamp alignment, missing value marking, and unified data format, a unified time-series operational status data is obtained. Specifically, this involves: first, calibrating the timestamps of all data packets in the initially organized original dataset using the standard time of the shore-based data receiving center as a benchmark, correcting timestamps of data packets with time discrepancies, ensuring consistency in timestamps across all data packets, and achieving time synchronization of data of different types and receiving periods; then, checking each data set for each data type line by line, and if any data that should have been received within a certain time period is found to be missing, adding a missing value marker at the corresponding location and indicating the specific time node of the missing data to avoid misjudgments due to missing data during data processing; finally, according to the preset standardized data format requirements, all data that has undergone timestamp alignment and missing value marking undergoes a unified format conversion, transforming all the different original data formats into a recognized standard data format, ultimately resulting in operational status data that is differentiated by ship, arranged chronologically, with a unified data format, and consistent time sequence.

[0067] In this embodiment of the invention, because the satellite communication link is used to receive encrypted data streams from unmanned vessels, and after decryption and parsing, integrity verification and source authentication are performed, and the data is collected and cached according to vessel identification and data type, and then through timestamp alignment, missing value marking and unified format standardization processing, the technical problems of insufficient security of unmanned vessel operation status data transmission, difficulty in verifying source credibility, and scattered, disordered and inconsistent data formats are effectively overcome. Thus, secure, reliable, orderly and time-series unified operation status data is obtained, providing a high-quality data foundation for vessel status evolution modeling, data calibration and full life cycle archive construction.

[0068] In a preferred embodiment of the present invention, the collected operating state data is serialized to form a state dataset, and a reference state vector is selected. Two state change axes are extended from the reference state vector as the origin, including:

[0069] The unified time-series operational status data is arranged chronologically and organized into a time-series dataset containing multi-dimensional state variables. Specifically, this involves: first, extracting the timestamp information of all data entries from the unified time-series operational status data; then, sorting all data entries in ascending order of timestamp, using the standard time of the shore-based data receiving center as a reference, to ensure data continuity and order in the time dimension; next, identifying all types of state variables included in the data, covering all collected variable types such as ship position (longitude, latitude), speed, heading, power output, propeller speed, equipment operating temperature, equipment operating voltage, cabin pressure, cabin humidity, and emergency alarm signals; then, associating and binding each sorted timestamp data with all corresponding state variable values, forming a data record containing complete multi-dimensional state variable information at each time point; finally, integrating all associated time-series data records chronologically to construct a well-structured time-series dataset, where each dataset entry corresponds to a unique time point and the ship's full-dimensional operational status data at that time point.

[0070] Based on the time series dataset, a representative time point state vector is selected from the dataset as the baseline state vector for analyzing ship state evolution. Specifically, this involves: selecting the baseline state vector using a combination of preset rules and representative time points; firstly, segmenting the time series dataset into multiple consecutive data segments at fixed time intervals, with each segment containing the same number of time points; for each data segment, calculating the coefficient of variation (COP) of all state variables, which is the ratio of the standard deviation to the mean; if the COP of all state variables in a data segment is less than 0.3, and there are no emergency signal records in that segment, then that data segment is considered a stable operating period for the ship; selecting the data segment with the longest duration from all stable operating periods; if multiple longest stable segments with the same duration exist, selecting the segment with the middle timestamp as the target segment; using the middle time point of the target segment as the representative time point, extracting all state variables corresponding to that time point, and combining the state variables according to a preset variable order to form the baseline state vector for analyzing ship state evolution.

[0071] Using the selected baseline state vector as the origin, and based on the statistical characteristics and variation patterns of each state variable in the time series dataset, two mutually orthogonal state change axes are calculated and defined. Specifically, this includes: using the selected baseline state vector as the origin, first performing statistical characteristic analysis on all state variables in the time series dataset; calculating the variance, maximum value, minimum value, and mean rate of change for each state variable across the entire dataset, where the rate of change is calculated as the ratio of the difference between variable values ​​at two adjacent time points to the time interval; selecting the state variables with the most significant variation characteristics based on the statistical results, using the top five state variables in terms of variance; and then calculating the pairwise correlation coefficients between these five state variables using the Pixley method. The Wilson correlation coefficient calculation method determines that if the absolute value of the correlation coefficient between two variables is less than 0.3, the linear correlation between the two variables is considered to be extremely low. From these five variables, a pair of variables with the smallest absolute value of the correlation coefficient is selected. The first variable is defined as the first state change axis, which reflects the main changing trend of the ship's core operating parameters. Using the baseline state vector as the origin, a second state change axis perpendicular to the first state change axis is constructed based on the principle of vector orthogonality. The second state variable must be completely orthogonal to the first state variable and reflect the changing characteristics of another key dimension of the ship. For example, if the first axis is a dynamic-related changing dimension, the second axis can be defined as a navigation attitude-related changing dimension. Finally, two mutually perpendicular orthogonal axes are formed that comprehensively cover the main state changes of the ship.

[0072] In this embodiment of the invention, because the embodiment adopts the technical means of organizing the unified time-series operational status data into a multi-dimensional time series dataset in chronological order, selecting a representative state vector as a reference state vector, and then defining two mutually orthogonal state change axes with the reference vector as the origin and based on the statistical characteristics and change laws of each state variable, it effectively overcomes the technical problems of lacking a unified analysis benchmark for ship multi-dimensional operational status data and difficulty in systematically quantifying and representing the state change trend. Thus, it provides a structured analysis framework for fitting ship state evolution trajectory and data calibration, making the quantitative analysis of ship state changes more targeted and accurate.

[0073] In a preferred embodiment of the present invention, sectors are defined based on the directional differences between two axes, and two types of reference sequences are set inside and outside the sectors respectively; based on the temporal correlation of the two types of reference sequences, a continuous elliptical state evolution trajectory of the ship's state change characteristics is fitted, including:

[0074] Each state vector in the time series dataset is projected onto a two-dimensional analysis coordinate system formed by two mutually orthogonal state change axes, forming a series of time series data points. Specifically, this involves: first, extracting each complete state vector from the time series dataset, each state vector containing multidimensional state variable information during ship operation; for each state vector, analyzing the correlation between each state variable and the two previously defined mutually orthogonal state change axes; and, based on the influence weights of each state variable on the two axes, converting the comprehensive value of the multidimensional state variables into individual projected values ​​corresponding to the two axes respectively; in this way, each state vector can find a unique corresponding coordinate point in the two-dimensional analysis coordinate system formed by the two mutually orthogonal state change axes, and simultaneously associating this coordinate point with the timestamp information corresponding to the original state vector; after all state vectors have undergone projection transformation, a series of time series data points arranged in timestamp order are formed, each data point containing both spatial location information in the two-dimensional coordinate system and retaining the corresponding time dimension information.

[0075] Based on the time series data points, calculate the orientation angle of each data point relative to the origin of the reference state vector; based on the preset angle threshold, divide the plane of the entire two-dimensional analysis coordinate system into several sectors, specifically including: taking the coordinate origin corresponding to the reference state vector as the core, calculate the orientation angle of each time series data point in the two-dimensional analysis coordinate system. During the calculation, the first state change axis is used as the starting edge. The angle between the line connecting the data point and the origin and the starting edge is measured in a counterclockwise direction to obtain the direction angle of each data point. The value range of the direction angle covers 0 degrees to 360 degrees. Then, the angle threshold is set to 30 degrees. The entire two-dimensional analysis coordinate system plane is divided into sectors according to the fixed threshold. Starting from the 0-degree position corresponding to the first state change axis, a sector boundary is defined every 30 degrees, and 12 equally sized sector regions are divided in sequence. Each sector corresponds to a fixed angle range. For example, the first sector is from 0 degrees to 30 degrees, the second sector is from 30 degrees to 60 degrees, and so on, until the sector division of the entire 360-degree plane is completed, ensuring that each time series data point can be accurately classified into the corresponding sector according to the direction angle.

[0076] For each segmented sector, the set of time-series data points within the sector is defined as the first type of reference sequence; the set of adjacent time-series data points associated with each sector but located outside the sector boundary is defined as the second type of reference sequence. Specifically, for each segmented sector, all time-series data points are traversed, and their azimuth angles are used to determine whether they fall within the current sector's angular range. All time-series data points with azimuth angles within the current sector's angular range are selected and sorted according to their corresponding timestamps to form the first type of reference sequence. This sequence centrally reflects the operational characteristics of the ship within the state change interval corresponding to the current sector. Then, the two adjacent sectors are searched, and the time-series data points within these two adjacent sectors that are closest to the current sector's boundary are selected. At the same time, adjacent data points outside the current sector's boundary that have continuous timestamps with the current sector's data points are included, and these data points are also arranged according to their timestamps to form the second type of reference sequence. The second type of reference sequence supplements the preceding and following information of the current sector's data points.

[0077] Based on the temporal correlation between the two types of reference sequences corresponding to each sector, an ellipse fitting algorithm is used to fit all time-series data points projected onto the two-dimensional analysis coordinate system, resulting in a continuous and closed elliptical curve as the state evolution trajectory characterizing the long-term, periodic changes in the ship's state. Specifically, this involves: firstly, performing temporal correlation analysis on the first and second types of reference sequences corresponding to each sector, determining the temporal order of data points in different sequences, and clarifying the logical connection between numerical changes in data points to ensure that all sequences form a complete and coherent data flow along the time axis; then, starting the ellipse fitting algorithm, using all time-series data points in the two-dimensional analysis coordinate system as the basic samples for the fitting analysis. The algorithm automatically analyzes the overall distribution characteristics, numerical fluctuation patterns, and temporal correlation patterns of these data points; during the fitting process, fully utilizing the boundary connection data and internal feature data provided by the two types of reference sequences, continuously adjusting the fitting parameters, and correcting the curve shape to ensure that the fitted curve accurately matches the distribution trend of the data points while maintaining the curve's smoothness, continuity, and closure characteristics; after multiple rounds of parameter optimization, a continuous and closed elliptical curve is finally obtained, which comprehensively and accurately characterizes the periodic characteristics of the ship's state changes during long-term operation.

[0078] In this embodiment of the invention, because the technical means of projecting the state vector in the time series dataset onto a two-dimensional analysis coordinate system, delineating sectors by direction angle and angle threshold and setting observation reference sequences inside and outside the sectors, and fitting the state evolution trajectory by combining the temporal correlation of the two types of reference sequences with an ellipse fitting algorithm, this embodiment effectively overcomes the technical problems of the chaotic nature of ship multidimensional operating state data and the difficulty in systematically capturing and intuitively representing long-term periodic change characteristics. Thus, it accurately constructs a continuous trajectory that can reflect the long-term evolution law of ship state, providing a concrete and quantifiable core basis for data calibration and ship state anomaly identification.

[0079] like Figure 2 As shown, in another preferred embodiment of the present invention, a data calibration coefficient is calculated based on the elliptical state evolution trajectory. The original operating state data is then calibrated using the calibration coefficient to obtain calibrated data, including:

[0080] Based on the obtained elliptical state evolution trajectory, the principal axis length, eccentricity, and orientation parameters relative to the two-dimensional analysis coordinate system are extracted. Specifically, this includes: first, fully traversing the fitted elliptical state evolution trajectory and recording the two-dimensional coordinate information of all points on the trajectory; second, calculating the straight-line distance between any two points on the trajectory and selecting the two points with the largest distance, the line connecting these two points being the principal axis of the ellipse, and the distance between them being the principal axis length; third, finding all pairs of points perpendicular to the principal axis direction on the elliptical trajectory and calculating the straight-line distance between these pairs, with the line corresponding to the longest distance being the principal axis length of the ellipse. Minor axis: Record the length of the minor axis. Based on the relationship between the principal axis length and the minor axis length, calculate the eccentricity of the ellipse. Specifically, subtract the square of the minor axis length from the square of the principal axis length, take the square root of the result, and then divide it by the principal axis length. This gives the eccentricity value, which reflects the flatness of the ellipse. Finally, determine the specific direction of the principal axis. Using the first state change axis of the two-dimensional analysis coordinate system as a reference, measure the angle between the center line of the principal axis and the first state change axis in a counterclockwise direction. This angle is the orientation parameter of the ellipse relative to the two-dimensional analysis coordinate system, with a value ranging from 0 degrees to 180 degrees. Record this parameter value accurately.

[0081] Using the principal axis length, eccentricity, and orientation parameters, and following predetermined calculation rules, a set of data calibration coefficients is obtained to compensate for axial deviations between two states in the two-dimensional analysis coordinate system. Specifically, this includes: pre-setting standard parameters for the elliptical trajectory under normal ship operating conditions, where the standard principal axis length is 100, the standard eccentricity is 0.5, and the standard orientation parameter is 0 degrees; according to the predetermined calculation rules, firstly, the principal axis length calibration coefficient is calculated, which is equal to the ratio of the standard principal axis length to the actual extracted principal axis length; then, the eccentricity calibration coefficient is calculated, which is equal to the ratio of the standard eccentricity to the actual extracted eccentricity; finally, the orientation parameter calibration coefficient is calculated, if the actual orientation parameter is 0 degrees... The coefficient is set to 1. For every 0-30 degree deviation of the actual orientation parameter, the coefficient is adjusted by 0.1. The adjustment range increases proportionally as the deviation angle increases. The three individual coefficients are weighted and summed with equal weights to obtain the comprehensive data calibration coefficient for the first state change axis. Using the same calculation logic, and combining the correlation characteristics between the second state change axis and the elliptical trajectory, the weight ratio of each parameter is adjusted. The principal axis length calibration coefficient accounts for 40%, the eccentricity calibration coefficient accounts for 30%, and the orientation parameter calibration coefficient accounts for 30%. The comprehensive data calibration coefficient for the second state change axis is calculated, and finally a complete set of data calibration coefficients is formed to compensate for the deviation of the two state change axes.

[0082] The data calibration coefficient is applied to the unified time-series operational status data to correct the component values ​​of the data along the two state change axes defined in the two-dimensional analysis coordinate system, resulting in calibrated operational status data free from long-term trend errors. Specifically, this involves: extracting the component values ​​of each data point along the two state change axes in the unified time-series operational status data; for the first state change axis, multiplying the component value corresponding to each data point by the comprehensive data calibration coefficient for that axis to obtain the corrected first-axis component value; for the second state change axis, using the same method, multiplying the second-axis component value of each data point by the corresponding comprehensive data calibration coefficient to obtain the corrected second-axis component value; during the correction process, each data point is checked individually to ensure that each component value is accurately calculated. For data points whose corrected values ​​exceed a reasonable range, the calibration coefficient and the original component values ​​are rechecked, and the correction results are retained after confirming that the calculations are correct. Through this correction process, the original operational status data, which originally contained long-term trend errors and deviations, is adjusted to a reasonable range that conforms to the elliptical state evolution trajectory, ultimately resulting in calibrated operational status data with improved accuracy and free from long-term trend errors.

[0083] In this embodiment of the invention, because the embodiment uses the technique of extracting the principal axis length, eccentricity and orientation parameters from the elliptical state evolution trajectory, calculating the data calibration coefficient to compensate for the axial deviation according to a predetermined rule, and applying the coefficient to the uniform time-series operating state data to correct the axial component value, it effectively overcomes the technical problem of insufficient data accuracy caused by long-term trend errors and deviations in the original ship operating state data. As a result, it obtains calibrated operating state data with higher accuracy and smaller errors, providing high-quality data support for the accurate updating of digital twin ship models and the reliable analysis of artificial intelligence anomaly detection models.

[0084] In a preferred embodiment of the present invention, the calibrated data drives and updates the digital twin ship model synchronized with the physical ship, while simultaneously inputting the operational status data into a preset artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain identification results, including:

[0085] Based on the calibrated operational status data, the digital twin ship model corresponding to the physical ship is driven in real time, updating the status parameters in the twin model to keep them synchronized with the operational status of the physical ship. Specifically, the construction of the digital twin ship model is based on fundamental data from the entire ship lifecycle, including detailed drawing dimensions from the design phase, material specifications and installation location data from the construction phase, rated performance parameters of equipment at the time of delivery, and baseline operational data from the initial operational phase. Through 3D modeling and multi-source data integration technologies, a full-dimensional virtual model is established that corresponds one-to-one with the physical ship in terms of structure, equipment, and function. The full-dimensional virtual model has a pre-defined unified parameter system, with each parameter corresponding to a specific operational status indicator of the physical ship, covering all key dimensions such as ship navigation status, power status, operational status of various equipment, and cabin environment status. A real-time mapping channel is established between the calibrated operational status data and the parameters of the digital twin ship model. The calibrated data is continuously transmitted to the model's parameter update module at fixed 1-second intervals. This module processes the received data... The system performs classification and analysis, accurately matching navigation data such as ship position (longitude, latitude, speed, and heading), power parameters such as power output and propeller speed, equipment status data such as operating temperature and voltage, and environmental data such as cabin pressure and humidity to the corresponding parameter items in the model. The model adjusts the corresponding virtual parameter values ​​in real time based on the analyzed calibration data. For example, when calibration data shows the actual ship's propeller speed is 1500 revolutions per minute, the virtual propeller speed parameter in the model is updated to 1500 revolutions per minute; when the equipment operating temperature is 75 degrees Celsius, the corresponding equipment temperature parameter in the model is adjusted to 75 degrees Celsius. Simultaneously, a parameter synchronization verification mechanism is set up, comparing the model parameters with the real-time status sampling data fed back by the actual ship via satellite communication every 5 seconds. If a parameter deviation exceeds 0.5, a secondary calibration update is immediately initiated, ensuring that all status parameters of the digital twin ship model remain highly synchronized with the actual operating status of the actual ship, providing accurate virtual mapping data for ship status monitoring and full lifecycle management.

[0086] The calibrated operational status data is simultaneously input into a pre-set artificial intelligence anomaly detection model. The anomaly detection model performs real-time pattern recognition and deviation analysis based on the data to obtain identification results regarding equipment health status, anomaly types, and their probabilities. Specifically, this includes: first, a detailed explanation of the construction process of the artificial intelligence anomaly detection model, i.e., the model's origin; the first step is to collect training data, gathering historical calibrated operational status data from multiple similar vessels under normal operating conditions, covering various navigation environments (near and far seas), load conditions (light and heavy loads), and various data under different operating durations (short and long voyages). Simultaneously, data on past equipment anomaly failures of these vessels is collected, including operational data before the failure, and data at the time of the failure. The first step involves detailed information such as abnormal data, specific fault types, causes of faults, and fault handling results, with a total data volume exceeding 100,000 records. The second step is data preprocessing, which cleans the collected historical data, removes data entries with missing values ​​exceeding 10%, and corrects extreme values ​​with excessively large fluctuations using the mean replacement method. Subsequently, the data is standardized according to the same format and standards as the real-time calibrated operational status data to ensure complete structural consistency between the training data and the real-time input data. The third step is feature extraction, which extracts key features closely related to the health status of ship equipment from the preprocessed historical data, including statistical features such as the mean, variance, rate of change, peak value, and duration of various status parameters, and screens them out. The first step involves selecting 50 core features to form the feature set for model training. The second step divides the processed feature set into training, validation, and test sets in a 7:2:1 ratio. The training set is used for core model training, the validation set is used to adjust model parameters during training, and the test set is used to ultimately verify the model's generalization ability and recognition accuracy. The third step involves model training, using the gradient boosting tree algorithm as the base algorithm. The training set is input into the algorithm with a learning rate of 0.1, 100 decision trees, and a tree depth of 5. Through multiple rounds of iterative training, the model gradually learns the feature patterns of normal operation and different fault types. The sixth step involves model validation and optimization, using the validation set to refine the trained model. The model undergoes performance testing, and the anomaly detection accuracy is calculated. If the accuracy is below 95%, parameters such as the learning rate and the number of decision trees are adjusted and the model is retrained. This iterative optimization continues until the model's anomaly detection accuracy on the validation set reaches above 95%. At the same time, the test set is used for final performance verification to ensure that the model can maintain stable recognition performance in unseen data scenarios. The seventh step is to determine the judgment threshold. Based on the fault case data in the validation and test sets, the distribution range of feature deviations under normal and various fault states is statistically analyzed. The threshold for a single feature deviation is set to 5, the cumulative threshold for multiple feature deviations is set to 20, and the anomaly probability threshold is set to 80, which serve as the judgment standard for real-time detection. Finally, a preset artificial intelligence anomaly detection model is formed.

[0087] After the model is built, the specific process of real-time analysis and monitoring is as follows: The continuously received calibrated operating status data is processed according to the feature extraction rules used during model training to extract 50 core features in real time. These features are then input into the artificial intelligence anomaly detection model. The model first performs pattern recognition on the input real-time features, comparing the current feature pattern with the normal operating status feature patterns learned during training, and calculating the deviation value between each core feature and the corresponding feature of the normal pattern. Subsequently, deviation analysis is performed, checking whether the deviation value of a single feature exceeds a set threshold of 5, and whether the cumulative deviation values ​​of multiple features exceed a set threshold of 20. If the deviation value of a single feature exceeds 5 or the cumulative deviation value of multiple features exceeds 20, the model further... The model first matches the current abnormal feature pattern with various fault feature patterns in the fault case feature library, calculates the similarity between the current abnormal pattern and each fault type, and then infers the probability of occurrence of each abnormal type. When the probability of occurrence of a certain abnormal type exceeds 80%, the model determines that the abnormal type is the current possible fault type. If the probability of occurrence of all abnormal types is less than 80%, the health status of the ship's equipment is determined to be normal. If the probability of occurrence of multiple abnormal types exceeds 80%, they are arranged in descending order of probability, and the main abnormal types and their corresponding probabilities are listed. Finally, the model outputs a complete identification result that includes whether the health status of the ship's equipment is normal or abnormal, the specific name of the abnormal type, and the corresponding probability of occurrence.

[0088] In this embodiment of the invention, because it uses high-precision calibrated operational status data as a basis to drive the digital twin ship model to update in real time to maintain synchronization with the physical ship's status, and simultaneously inputs the calibration data into a preset artificial intelligence anomaly detection model for pattern recognition and deviation analysis, it effectively overcomes the technical problems of the lack of accurate data support for the digital twin ship model's status update and the unreliability of the identification results of artificial intelligence anomaly detection due to errors in the original data. Thus, it achieves accurate synchronization between the digital twin ship model and the physical ship's operational status, as well as accurate identification of the ship's equipment health status, anomaly types, and probability of occurrence, providing timely and reliable decision-making basis for ship operation monitoring and anomaly handling.

[0089] In a preferred embodiment of the present invention, the updated information, identification results, and original operational status data of the digital twin ship model are encrypted and uploaded to the blockchain using blockchain technology to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications, including:

[0090] The twin model update information and anomaly detection and identification results contained in the comprehensive status report are combined and encapsulated with the unified time-series raw operational status data according to a predefined structure to obtain a complete data record block. Specifically, this includes: firstly, a comprehensive collection and classification of three types of core data; secondly, twin model update information covering the update details of all status parameters in the model, including the original values ​​of each parameter, the updated new values, the equipment name corresponding to the parameter, and the specific time point of the update; thirdly, anomaly detection and identification results including equipment health status judgment results, the name of the identified anomaly type, the probability value of the anomaly occurrence, the time point of anomaly detection, and the preliminary risk level given by the model; and fourthly, the unified time-series raw operational status data encompassing the ship's position... All uncorrected raw data, including latitude, speed, heading, power parameters, equipment operation data, cabin monitoring data, and emergency signals, are collected, with each data point accompanied by a complete timestamp and a unique ship identifier. The three types of data are then combined according to a predefined fixed structure. First, a general data identifier field is set, along with the ship number and data batch corresponding to the record block. Then, modules for twin model update information, anomaly detection and identification results, and raw operational status data are sequentially divided. Within each module, data is sorted by importance, with key core data placed at the top. Finally, the combined data is encapsulated using a unified encapsulation format to integrate all data content, forming a single data record block that is structurally complete, logically clear, and without any omissions.

[0091] Based on the encapsulated data record blocks, the data is converted into a standardized data structure conforming to the requirements of the Maritime Organization's electronic log specifications. This includes: first, obtaining the data fields, field order, data format, unit standards, and mandatory field requirements as specified in the Maritime Organization's electronic log specifications, which will serve as the unified basis for data structure conversion; then, parsing the encapsulated data record blocks, extracting the twin model update information, anomaly detection and identification results, and the content of each field corresponding to the original operational status data; and finally, adjusting the field names and their order according to the requirements of the Maritime Organization's electronic log specifications, uniformly modifying the internally defined field names to the standard field names specified in the specifications to ensure the accuracy of the field names. The meaning is unambiguous when circulating across entities; at the same time, the data format is uniformly standardized, for example, time data is uniformly converted to Coordinated Universal Time format, accurate to the millisecond level; all numerical data is uniformly retained to two decimal places, and the units are converted to the international standard units required by the standard; text data is converted according to the character encoding format required by the standard; in addition, the required fields required by the standard are supplemented, such as the ship's International Maritime Organization number, data acquisition equipment number, data processing node identifier, etc. If the original record block lacks relevant information, it will be automatically extracted and supplemented from the pre-stored ship basic information database; finally, a standardized data structure that fully complies with the requirements of the Maritime Organization electronic log is formed.

[0092] By applying blockchain cryptography to encrypt and timestaminate standardized data structures, and broadcasting the processed data as a transaction to the blockchain network for consensus verification and on-chain storage, a blockchain log with immutable and traceable characteristics is generated. Specifically, this involves: first, encrypting the standardized data structure using an asymmetric encryption algorithm; using a dedicated private key to perform encryption operations on the entire standardized data to generate encrypted ciphertext data; and simultaneously generating corresponding encrypted verification information using a public key for decryption and authenticity verification by the data recipient; then, adding a precise timestamp to the encrypted ciphertext data. The timestamp is based on a globally unified time base and is accurate to the millisecond level. The system records the exact moment the data encryption process is completed, and this timestamp is bound to the encrypted data and cannot be modified independently. The encrypted data, corresponding public key verification information, and precise timestamp are then combined into a complete blockchain transaction. This transaction data is broadcast to the entire blockchain network through blockchain network nodes. Upon receiving the transaction data, each node in the blockchain network initiates a consensus verification process, employing a practical Byzantine fault-tolerant consensus mechanism. Nodes mutually verify the validity of the encrypted signature, the compliance of the standardized data structure, the rationality of the timestamp, and the completeness of the data content. Only when more than two-thirds of the nodes have verified the transaction can it pass consensus verification. After verification, the transaction data is permanently stored in a new block of the blockchain. The new block records all information about the transaction and generates a unique block hash value. This block hash value is also associated with the hash value of the previous block, forming a chained storage structure. Through this series of operations, a blockchain log with immutable and traceable characteristics is generated.

[0093] In this embodiment of the invention, because the embodiment combines and encapsulates the twin model update information, anomaly detection and identification results, and unified time-series original operating status data, converts them into a standardized data structure that conforms to the Maritime Organization's electronic log specifications, and then encrypts, timestamps, and verifies the data through blockchain cryptography before storing it on the blockchain, it effectively overcomes the technical problems of inconsistent ship-related core data formats, lack of compliance support, susceptibility to tampering, and insufficient credibility of evidence storage. This generates a blockchain log that conforms to industry standards and has the characteristics of immutability and traceability, providing a solid guarantee for the credible evidence storage, compliance review, and traceability of ship data.

[0094] In a preferred embodiment of the present invention, a blockchain log is used as the sole trusted data source to construct an electronic archive of the ship's entire lifecycle. A traceability query interface is provided based on the blockchain log for trusted verification and tracing of the ship's historical status, operation records, and abnormal events, including:

[0095] By extracting all historical data records related to the target vessel from the blockchain log in chronological order and based on key event identifiers, the process involves: first, identifying the target vessel's unique identifier, such as its International Maritime Organization (IMO) number or registration number, to ensure accurate vessel location; then, initiating a blockchain log traversal process to read transaction data from each block of the blockchain, comparing the vessel identifier information within the data to filter out all transaction records related to the target vessel; and simultaneously extracting the timestamp for each record and processing the selected records in ascending order of timestamp. The data is sorted to ensure continuity and integrity over time. Simultaneously, key event identifiers are identified and marked in each record. These identifiers include ship design parameter confirmation, construction milestone completion, equipment malfunction, maintenance operation execution, major digital twin model update, and emergency response activation. Records with these identifiers are categorized separately to ensure no relevant data from any key stage of the ship's entire lifecycle is omitted. Finally, the sorted time-series data and the marked key event data are aggregated to form a complete historical data set covering the entire lifecycle of the target ship, from design and construction to operation, maintenance, and scrapping.

[0096] Historical data is recorded, reorganized, classified, and indexed according to the ship lifecycle management standards to construct a structured electronic ship archive. The archive summary information is then stored and anchored on the blockchain. Specifically, based on the ship lifecycle management standards, the extracted historical data is divided into five core categories according to stages: design stage data, construction stage data, operation stage data, maintenance stage data, and scrapping stage data. Design stage data includes parameters related to ship design drawings, material selection standards, performance design indicators, and design review records. Construction stage data includes material procurement contracts and quality inspection reports, construction and installation logs, acceptance data for each construction process, ship launching test records, and delivery and acceptance documents. Operation stage data includes daily navigation status records, real-time equipment operating parameters, fuel consumption statistics, anomaly detection results records, and digital twin model update logs. Maintenance stage data includes periodic maintenance plan documents, detailed repair operation records, replacement component models and installation information, and post-maintenance performance test reports. Scrapping stage data includes ship scrapping assessment reports, dismantling process planning and execution records, and environmental compliance documents. Within each stage category, further subdivisions are made based on data type, such as equipment parameters, event logs, and document reports. A multi-level indexing system is then established: the first-level index is categorized by stage, the second-level index is arranged chronologically, and the third-level index is classified by event type and data characteristics. Each index item is associated with the storage path of the corresponding data in the electronic archive, ensuring rapid location of the required data. Subsequently, a structured ship electronic archive is constructed, storing all categorized data in an orderly manner according to the indexing system. Simultaneously, archive summary information is generated, containing key information such as the hash value of core data for each stage, the archive's creation time, update time, the data time period covered, and a list of key events. Finally, the archive summary information is broadcast to the blockchain network via blockchain transactions. After consensus verification by more than two-thirds of the nodes in the network, the summary information is permanently stored in a newly generated block on the blockchain, completing the notarization and anchoring of the archive summary on the blockchain.

[0097] Based on the electronic archives and their anchoring information in the blockchain, a traceability query service interface with identity authentication and access control is deployed. The interface supports trusted retrieval by time range, event type, and data characteristics. Specifically, this includes: first, integrating the storage path information of the electronic archives with the evidence anchoring information of the archive summaries in the blockchain to build the underlying data association architecture of the query interface, ensuring that the interface can quickly link the electronic archives and blockchain evidence data; embedding identity authentication during interface deployment, requiring users to submit identity information before initiating a query request, including their organization name, user name, user role, and authorization certificate number. The submitted identity information is compared item by item with a pre-stored authorized user information database. Only users with an identity information matching rate of 60% or higher and whose authorization certificates are valid can pass identity authentication and enter the query stage; simultaneously, a hierarchical access control mechanism is set up, dividing user permissions into three levels: Level 1 permissions are for maritime regulatory agencies, allowing querying all categories of data throughout the ship's entire lifecycle; Level 2 permissions are for ship owners and operating units, allowing querying all data except for classified design data; Level 3 permissions are for maintenance units, allowing querying only maintenance-related stage data and corresponding equipment operation data. The interface supports multi-condition combined search functionality. Users can set time range search conditions, precisely inputting the start and end times of the query, with time precision down to the year, month, day, hour, minute, and second. They can also select event type search conditions, including design changes, construction milestone completion, equipment failure, routine maintenance, emergency response, and model updates. Furthermore, users can set data feature search conditions, inputting features such as equipment name, parameter value range, and data file type, which will then be used for precise data matching based on the user-defined combination of conditions. The interface also includes auxiliary functions such as search result preview, data export, and report generation. Additionally, a query request logging module automatically records information such as user identity, query conditions, query time, and query results for each query, ensuring traceability of query operations.

[0098] The system responds to user tracing requests via a deployed query interface, retrieves relevant data records from the electronic archive, and automatically verifies the integrity and authenticity of the data by comparing it with the corresponding evidence anchoring information in the blockchain. Finally, it returns verified historical status, operation records, and abnormal event tracing reports to the user. Specifically, this includes: submitting a tracing request through the deployed tracing query interface, requiring key parameters such as the target vessel identifier, the time range of the query, the event type, and data characteristics; and uploading relevant supplementary authorization documents if necessary. Upon receiving the user's tracing request, the system first re-verifies whether the user's identity and permissions match the category of the queried vessel data. If the user's permissions are insufficient or the identity information is inconsistent with the previous authentication results, an insufficient permissions message is returned, and the request is recorded in the abnormal request log. If the permissions verification is successful, based on the user's submitted query parameters, the system accurately locates and extracts the corresponding historical data records from the structured electronic archive using a multi-level indexing system. This includes a time series table of the vessel's operating status parameters within the query range, detailed process records of key operations, and relevant data such as the time, location, cause, and handling results of abnormal events. Subsequently, the system extracts... The hash values ​​of the archive summaries corresponding to some data are compared one by one with the corresponding stage summary information anchored in the blockchain. At the same time, the timestamps, key event identifiers, data lengths and other information of the data are checked to see if they are completely consistent with the blockchain records. If the hash values ​​are consistent and there are no differences in the key information, the integrity and authenticity of the data are verified. If the hash values ​​are inconsistent or there are deviations in the key information, the data is determined to have been tampered with or damaged. The data is automatically marked as a verification anomaly and the verification process log is recorded. After verification, all valid data that has passed verification is organized and classified according to time sequence and event logic to generate a structured traceability report. The report includes a description of the query conditions, a description of the data source and verification status, a time-series statistical chart of the ship's historical status, a detailed list of key operation records, and an in-depth analysis of the time, cause and handling results of the abnormal event. The report is presented in a combination of text and graphics to ensure that it is intuitive and easy for users to understand. Finally, the traceability report is returned to the user through the interface, and a PDF report download function is also provided for users to archive or submit to relevant review departments. This ensures that users can obtain reliable, complete and detailed traceability results, effectively supporting fault tracing and compliance review work.

[0099] In this embodiment of the invention, because it uses blockchain logs as the sole trusted data source, extracts historical data of the target vessel in chronological order and key event identifiers, constructs a structured electronic archive by reorganizing and classifying the index, and anchors the summary information to the blockchain for evidence storage, and deploys a traceability query interface with identity authentication and access control that supports multi-condition retrieval, and automatically verifies the integrity and authenticity of the data and returns a trusted traceability report when responding to user requests, it effectively overcomes the technical problems of lacking a unified trusted archive carrier for the entire life cycle of vessel data, lacking accurate retrieval capabilities for traceability queries, and difficulty in verifying the authenticity of data. Thus, it achieves efficient and trusted traceability of the vessel's historical status, operation records, and abnormal events, providing an authoritative and reliable basis for the entire life cycle management of vessels, fault tracing, liability determination, and compliance review.

[0100] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the system as described above. All implementations in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0101] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the system as described above. All implementations in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0102] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A system for constructing and tracing electronic records throughout the entire lifecycle of a ship, characterized in that: include: The acquisition module is used to acquire real-time operational status data generated by the ship during operation. The extension module is used to serialize the collected running status data to form a status dataset and select a reference status vector, extending two status change axes with the reference status vector as the origin. The fitting module is used to delineate sectors based on the directional differences between two axes, and to set two types of reference sequences inside and outside each sector. Based on the temporal correlation of the two types of reference sequences, a continuous elliptical state evolution trajectory of the ship's state change characteristics is fitted, including: Each state vector in the time series dataset is projected onto a two-dimensional analysis coordinate system formed by two mutually orthogonal state change axes to form a series of time series data points. Based on the time series data points, calculate the direction angle of each data point relative to the reference state vector as the origin; based on the preset angle threshold, divide the plane of the entire two-dimensional analysis coordinate system into several sectors; For each sector, the set of time-series data points within the sector is defined as the first type of reference sequence; the set of adjacent time-series data points associated with each sector but located outside the sector boundary is defined as the second type of reference sequence. Based on the temporal correlation between the two types of reference sequences corresponding to each sector, the ellipse fitting algorithm is used to fit all the temporal data points projected in the two-dimensional analysis coordinate system to obtain a continuous and closed elliptical curve as the state evolution trajectory characterizing the long-term and periodic changes in the ship's state. The calibration module is used to calculate data calibration coefficients based on the elliptical state evolution trajectory, and then calibrate the original operating state data using these coefficients to obtain calibrated data, including: Based on the obtained elliptical state evolution trajectory, the principal axis length, eccentricity, and orientation parameters relative to the two-dimensional analysis coordinate system are extracted. Using the principal axis length, eccentricity, and orientation parameters, and in accordance with predetermined calculation rules, a set of data calibration coefficients is obtained to compensate for the axial deviation between two states in the two-dimensional analysis coordinate system. The data calibration coefficients are applied to the uniform time series of operating status data to correct the component values ​​of the data in the two state change axes defined in the two-dimensional analysis coordinate system, so as to obtain the calibrated operating status data after removing long-term trend errors. The identification module is used to drive and update the digital twin ship model synchronized with the physical ship through calibrated data, and at the same time input the operating status data into the preset artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain the identification results. The encryption module is used to encrypt and upload the updated information, identification results and original operating status data of the digital twin ship model to the blockchain using blockchain technology, so as to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications. The processing module is used to build an electronic archive of the entire life cycle of a ship using blockchain logs as the sole trusted data source, and to provide a traceability query interface based on blockchain logs for trusted verification and tracing of the ship's historical status, operation records and abnormal events.

2. The ship lifecycle electronic record construction and traceability system according to claim 1, characterized in that, Real-time acquisition of operational status data generated by the ship during operation, including: It receives encrypted data streams periodically sent by at least one unmanned vessel in real time via a satellite communication link. By decrypting the encrypted data stream and parsing the communication protocol, the structured original runtime status data packets are extracted; and the integrity of the parsed data packets is verified and the source is authenticated to filter out trustworthy data packets. Trusted data packets are collected and cached according to their corresponding ship identifiers and data types to form a raw dataset initially organized by ship and time sequence; the data includes ship position, speed and heading, power parameters, equipment status, compartment monitoring data and emergency signals; By performing timestamp alignment, missing value marking, and standardization of the data format on the original dataset, a unified time-series runtime status data is finally obtained.

3. The ship lifecycle electronic record construction and traceability system according to claim 2, characterized in that, The collected operational status data is serialized to form a status dataset, and a baseline state vector is selected. Two state change axes are extended from the baseline state vector as the origin, including: The unified time-series operational status data is arranged in chronological order and organized into a time-series dataset containing multidimensional state variables; Based on the time series dataset, a state vector at a representative time point is selected from the dataset as the benchmark state vector for analyzing the evolution of ship state. By using the selected baseline state vector as the origin of the coordinate system, and based on the statistical characteristics and variation patterns of each state variable in the time series dataset, two mutually orthogonal state change axes are calculated and defined.

4. The ship lifecycle electronic record construction and traceability system according to claim 3, characterized in that, The calibrated data drives and updates the digital twin ship model synchronized with the physical ship, while simultaneously inputting operational status data into a pre-set artificial intelligence anomaly detection model for real-time analysis and monitoring to obtain identification results, including: Based on the calibrated operational status data, the digital twin ship model corresponding to the physical ship is driven in real time to update the status parameters in the twin model, so that the status parameters are synchronized with the operational status of the physical ship. The calibrated operating status data is simultaneously input into a preset artificial intelligence anomaly detection model; the anomaly detection model performs pattern recognition and deviation analysis in real time based on the data to obtain the identification results of the equipment health status, anomaly type and probability of occurrence.

5. The ship lifecycle electronic record construction and traceability system according to claim 4, characterized in that, The updated information, identification results, and original operational status data of the digital twin ship model are encrypted and uploaded to the blockchain using blockchain technology to obtain an immutable blockchain log that conforms to the Maritime Organization's electronic log specifications, including: The twin model update information and anomaly detection and identification results contained in the comprehensive status report are combined and encapsulated with the original operating status data in a unified time series according to a predefined structure to obtain a complete data record block. Based on the encapsulated data record blocks, they are converted into a standardized data structure that conforms to the requirements of the Maritime Organization's electronic log specifications; By applying blockchain cryptography to encrypt and timestamp standardized data structures, and broadcasting the processed data as a transaction to the blockchain network for consensus verification and on-chain storage, a blockchain log with immutable and traceable characteristics is generated.

6. The ship lifecycle electronic record construction and traceability system according to claim 5, characterized in that, Using blockchain logs as the sole trusted data source, an electronic archive of the entire ship lifecycle is constructed. Based on these blockchain logs, a traceability query interface is provided for the trusted verification and tracing of the ship's historical status, operational records, and abnormal events, including: By extracting all historical data records related to the target vessel from the blockchain log, in chronological order and by key event identifiers; Historical data is recorded, reorganized, classified, and indexed according to the ship lifecycle management standards, a structured electronic archive of ships is constructed, and the archive summary information is stored and anchored in the blockchain; Based on the electronic archive and the anchoring information in the blockchain, deploy a traceability query service interface with identity authentication and access control. The interface supports trusted retrieval by time range, event type and data characteristics. The system responds to user traceability requests through a deployed query interface, retrieves relevant data records from the electronic archive, and automatically verifies the integrity and authenticity of the data by comparing it with the corresponding evidence anchoring information in the blockchain. Finally, it returns trusted and verified historical status, operation records, and abnormal event traceability reports to the user.

7. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to perform the system as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, performs the system as described in any one of claims 1 to 6.

Images