Intelligent document paperless system based on ship environment self-adaptation and blockchain storage

Abstract

The application discloses an intelligent document paperless system based on ship environment self-adaptation and blockchain storage, relates to the technical field of ship data processing, and comprises a ship motion perception and correction module, a multi-source sensor data dynamic compensation based on a ship type physical model, and corrected sensor data generation; a cause and effect reasoning and evidence chain module for building a cause and effect graph model between ship motion state and document parameters, performing cause and effect inference based on current motion state and sensor data, and generating a cause and effect evidence chain; a document generation and verification module for dynamically selecting a document template according to the ship motion state; a blockchain storage and feedback module for packing document content, motion context and cause and effect evidence chain abstract to generate composite storage information, and implementing a hierarchical storage strategy according to document credibility; and the application ensures the accuracy and reliability of sensor data.

Description

Technical Field

[0001] This invention relates to the field of ship data processing technology, specifically to an intelligent paperless document system based on ship environment adaptation and blockchain evidence storage. Background Technology

[0002] Paperless ship documentation is an important direction for the digital transformation of the shipping industry. Currently, the paper documents used by tugboats include various types such as nautical logs, engine logs, oil record books, garbage record books, radio communication logs, night navigation command books, ozone depleting substance record books, and ship safety inspection records. Traditional paper documents are filled out by hand by the crew, which has problems such as inconsistent standards, easy damage, and difficulty in handwriting recognition.

[0003] Existing paperless ship documentation systems primarily use electronic document formats to replace paper documents, with data entry via computer terminals. Some systems have implemented automatic data collection, using PLCs (Programmable Logic Controllers) deployed on ship equipment to collect data such as engine speed, ship speed, and oil and water levels. For tamper protection, existing technologies mostly employ hash algorithms to digitally sign electronic documents and submit the hash values ​​to a third-party evidence storage institution.

[0004] However, existing ships experience multiple motions such as rolling, pitching, and vibration during navigation. These motions not only cause distortion in the readings of individual sensors (such as oil tank level errors), but also cause timing discrepancies in the data acquisition of different sensors (such as PLC sampling clock drift). Current blockchain evidence storage mechanisms can only verify the bit integrity of the document content and cannot identify and distinguish these motion artifacts caused by the physical environment and inherent in the data generation stage. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent paperless document system based on ship environment adaptation and blockchain evidence storage.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides an intelligent paperless document system based on ship environment adaptation and blockchain evidence storage, comprising:

[0008] The ship motion sensing and correction module is used to monitor the six degrees of freedom motion of the ship in real time, and to dynamically compensate multi-source sensor data based on the ship's physical model to generate corrected sensor data.

[0009] The causal reasoning and evidence chain module is used to construct a causal graph model between the ship's motion state and document parameters, perform causal inference based on the current motion state and sensor data, and generate a causal evidence chain.

[0010] The document generation and verification module is used to dynamically select document templates based on the ship's motion status, implement three-layer closed-loop verification during the document generation process, and adaptively adjust the document content based on the verification results.

[0011] The blockchain-based evidence storage and feedback module is used to package document content, motion context, and causal evidence chain summaries to generate composite evidence storage information, and implement a hierarchical evidence storage strategy based on the credibility of the document.

[0012] The beneficial effects of this invention are:

[0013] 1. This invention utilizes a sensor calibration engine to generate personalized compensation parameters for multi-source sensors and performs compensation calculations in a trusted execution environment. For example, it can perform liquid surface tilt compensation and sensor installation position offset compensation for liquid level sensors. This effectively corrects the sensor reading distortion caused by ship motion and ensures the accuracy and reliability of sensor data.

[0014] 2. This invention considers the impact of ship vibration on clock signals and uses the ship's characteristic motion pattern as the data acquisition trigger signal through a motion trigger acquisition unit. Combined with a data watermark embedding unit, digital watermarks are embedded in the sensor data packets, thereby effectively solving the problem of data timing inaccuracy caused by vibration and ensuring the time logic consistency of multi-source sensor data acquisition.

[0015] 3. This invention packages document content, motion context, and causal evidence chain summary to generate composite evidence information. It implements a hierarchical evidence storage strategy based on the document's credibility, enabling blockchain evidence storage to not only verify the integrity of document content but also provide explanations for motion artifacts caused by the physical environment, thereby comprehensively improving the overall credibility of electronic documents. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0017] Figure 1 This is a schematic diagram of the overall structure of the intelligent paperless document system of the present invention;

[0018] Figure 2 This is a schematic diagram of the feedback loop optimization structure in the intelligent paperless document system of the present invention. Detailed Implementation

[0019] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] like Figures 1-2As shown, the intelligent paperless document system based on ship environment adaptation and blockchain evidence storage includes:

[0021] The ship motion sensing and correction module is used to monitor the six degrees of freedom motion of the ship in real time, and to dynamically compensate multi-source sensor data based on the ship's physical model to generate corrected sensor data.

[0022] The causal reasoning and evidence chain module is used to construct a causal graph model between the ship's motion state and document parameters, perform causal inference based on the current motion state and sensor data, and generate a causal evidence chain.

[0023] The document generation and verification module is used to dynamically select document templates based on the ship's motion status, implement three-layer closed-loop verification during the document generation process, and adaptively adjust the document content based on the verification results.

[0024] The blockchain-based evidence storage and feedback module is used to package document content, motion context, and causal evidence chain summaries to generate composite evidence storage information, and implement a hierarchical evidence storage strategy based on the credibility of the document.

[0025] When the system is working, it first collects and corrects sensor data through the ship motion sensing and correction module, then analyzes the cause of data anomalies and generates an evidence chain through the causal reasoning and evidence chain module, then generates verified document content through the document generation and verification module, and finally completes credible evidence storage through the blockchain evidence storage and feedback module. The whole process forms a complete closed-loop processing flow to ensure the authenticity and credibility of document data.

[0026] Furthermore, the ship motion sensing and correction module includes:

[0027] The six-degree-of-freedom motion monitoring unit consists of an inertial measurement unit (including a three-axis gyroscope and a three-axis accelerometer) and a global navigation satellite system receiver (BeiDou). The sampling frequency is set to 100Hz to ensure that the high-frequency vibration components of the ship can be captured. It is used to collect the ship's roll, pitch, heel, sway, sway and heel motion data.

[0028] The ship hull physical model library stores physical deformation models built for different ship hull types. The physical deformation models include hull structural parameters, elastic deformation characteristics, and sensor installation location information.

[0029] Each physical deformation model mainly includes three types of key parameters: hull structural parameters, including basic geometric features such as length, beam, depth, and displacement; elastic deformation characteristics, including mechanical property parameters such as elastic modulus, yield strength, and fatigue strength of the hull material, which describe the deformation characteristics of the hull material under stress; and sensor installation location information, including the three-dimensional coordinates and installation direction of each sensor in the hull coordinate system. These parameters can be obtained through ship design drawings, structural calculations, and actual ship measurements, and are classified and stored according to ship type.

[0030] The physical deformation model is constructed based on the principles of ship structural mechanics. By discretizing the hull into multiple structural units, a deformation calculation model of the hull under external forces such as wave loads and inertial loads is established. This model can calculate the deformation of the hull at various positions under different motion states, providing a theoretical basis for sensor data correction.

[0031] The sensor calibration engine generates personalized compensation parameters for each sensor based on real-time six-degree-of-freedom motion state and ship hull physical model. The calculation of these compensation parameters is performed in a trusted execution environment, which employs hardware-level security isolation technology to ensure the integrity and immutability of the calculation process.

[0032] The calculation process for compensation parameters mainly includes two steps: first, calculating the theoretical error based on the ship's motion state; and then generating the corresponding compensation parameters based on the sensor characteristics. For example, for a liquid level sensor, the calibration engine will simultaneously calculate the compensation values ​​for two components: compensation for liquid surface tilt caused by motion and compensation for sensor installation position offset caused by hull elastic deformation.

[0033] The sensors include a liquid level sensor, an attitude sensor, and a speed sensor. The sensor calibration engine performs compensation calculations for the liquid level sensor, including compensation for liquid surface tilt caused by motion and compensation for sensor installation position offset caused by hull elastic deformation. The two compensation components are vector-superimposed in a trusted execution environment to generate the final liquid level compensation parameters.

[0034] In practical applications, when a ship encounters severe sea conditions and rolls at a large angle, the sensor correction engine can calculate and apply compensation parameters in real time, effectively eliminating liquid level measurement errors caused by ship motion and ensuring the accuracy and reliability of liquid level data. For example, under severe conditions with a roll angle of 15°, the system can correct 8%-12% of the measurement error caused by liquid surface tilt, providing an accurate data basis for ship stability calculation and fuel management.

[0035] Furthermore, the causal reasoning and evidence chain module includes:

[0036] The causal graph model construction unit is used to establish a three-layer causal graph model between ship motion state and document parameters. The three-layer causal graph model includes a physical layer, a parameter layer, and a logic layer. The physical layer describes the physical relationship between ship motion and sensor readings, including specific causal paths such as ship roll angle and liquid level reading error, and hull vibration and sensor timing deviation. The parameter layer describes the conversion relationship between sensor readings and document parameters, including parameter mapping rules such as liquid level reading to fuel quantity and speed reading to power. The logic layer describes the logical relationship between document parameters, including logical constraints such as the theoretical correspondence between main engine power and fuel consumption, and the matching relationship between ship speed and main engine speed.

[0037] This model can achieve uncertain reasoning through Bayesian networks, where each node contains a state probability distribution and edges represent conditional dependencies.

[0038] The causal inference engine unit performs causal inference based on the current motion state and corrected sensor data, predicting the theoretical value range of each document parameter. The calculation of the theoretical value range takes into account the uncertainty of the ship's motion state and is usually represented by the interval probability method. It identifies the deviation between the actual value and the theoretical value. Deviation detection adopts the hypothesis testing method, sets the statistical significance threshold, and determines the cause of the anomaly through causal tracing. Causal tracing adopts the counterfactual inference method to analyze the contribution of different factors to the abnormal result. For example, when an abnormal oil tank level is detected, the inference engine will analyze possible causes such as roll angle, hull deformation, and sensor failure in sequence, and determine the most likely root cause through evidence weight. The statistical significance threshold is set to an absolute Z-score of greater than 2.58, corresponding to a 99% confidence level.

[0039] The evidence chain generation unit transforms the causal reasoning process into a verifiable evidence chain, which includes the cause of the anomaly, the impact path, the theoretical error, and the nodes of the compensation measures.

[0040] The cause node records the root cause of the problem identified through causal tracing; the influence path node describes the complete propagation path from cause to effect, including intermediate variables and transmission relationships; the theoretical error node records the theoretical deviation range calculated based on the physical model; and the compensation measure node records the specific correction methods and parameters adopted by the system.

[0041] The evidence chain generation process uses a workflow engine to build evidence nodes according to a fixed logical order of reasoning, ensuring the integrity and logical consistency of the evidence chain. Each evidence node contains metadata such as timestamp, data source, calculation method, and credibility score, providing a complete information foundation for subsequent verification.

[0042] The evidence chain compressor compresses a complete causal evidence chain into a natural language interpretation suitable for document format, while preserving causal nodes and numerical relationships during the compression process;

[0043] The compression process employs a rule-based natural language generation method. First, a node importance analysis algorithm identifies key causal nodes in the evidence chain, with importance assessment based on the node's centrality in the causal path and its contribution to the final result. Then, a template filling method converts the key node information into natural language descriptions, with a template library containing preset language expression patterns for different types of anomalies. Finally, a text optimization algorithm ensures the accuracy and readability of the generated explanation. The compression process strictly preserves all key causal nodes and numerical relationships, ensuring that the compressed explanation does not lose important technical information.

[0044] For example, in the event of an abnormal liquid level, the compressor will generate a concise explanation such as "The liquid surface tilted due to a 15° roll, with a theoretical error of 8%, and corresponding compensation has been applied."

[0045] In particular, when an abnormal situation occurs during berthing operations where the main engine speed and rudder angle do not match, the causal reasoning and evidence chain module can analyze the rationality of the operation sequence through a causal graph model, identify temporal logic problems through a causal reasoning engine, construct complete abnormal analysis evidence through an evidence chain generation unit, and finally generate a natural language description suitable for nautical log recording through an evidence chain compressor. This systematic causal analysis method effectively solves the problem of the lack of physical explanation for abnormal records in traditional document systems, and provides a reliable technical analysis basis for ship operations.

[0046] Furthermore, the document generation and verification module includes:

[0047] A dynamic document template library contains preset document templates for different motion states, including templates for calm waters, templates for high motion states, and templates for berthing operations.

[0048] The calm waters template is suitable for ships in normal navigation conditions in calm seas. The template adopts a standard document format and a regular data recording frequency, focusing on recording the ship's basic operating parameters and navigation status. The preferred applicable conditions are roll angle less than 5° and pitch angle less than 3°.

[0049] The high motion state template is specifically designed for the special state of ships when they experience large roll and pitch in severe sea conditions. This template increases the recording density of ship motion state parameters, adjusts the allowable error range of key parameters, and reserves a dedicated field for motion impact description. The preferred applicable conditions are roll angle greater than or equal to 5° or pitch angle greater than or equal to 3°.

[0050] The berthing operation template is suitable for special scenarios where ships are berthing and unberthing in port. This template improves the time resolution of operation parameters such as rudder angle and main engine speed, and increases the requirements for detailed recording of operation sequences. The preferred conditions for application are a speed of less than 2 knots and a distance of less than 100 meters from the dock.

[0051] By setting up multiple templates, we can ensure that the most suitable document recording solution can be provided under different movement conditions.

[0052] A three-layer closed-loop verification engine is used to implement single-parameter rationality checks, inter-parameter logical consistency verification, and motion state matching degree evaluation. The single-parameter rationality check is based on the dynamically calculated reasonable range of the parameter, which is calculated using the mean plus or minus three standard deviations. The inter-parameter logical consistency verification is based on the ship dynamics model (a mathematical model based on the principles of ship motion, used to describe the physical relationships between various ship parameters. This ship dynamics model includes sub-models such as propulsion system model, resistance model, and maneuverability model, and can calculate the theoretical correspondence of each parameter under different operating conditions). The motion state matching degree evaluation calculates the degree of consistency between the document records and the current motion state, verifying whether the physical relationships between key parameters conform to the laws of ship motion, such as verifying the correspondence between main engine power and speed, and the matching relationship between fuel consumption and travel distance. During the verification process, physical constraints are used to satisfy algorithms to detect logical contradictions. For example, the threshold for single-parameter rationality checks is an absolute Z-score greater than 3; the threshold for inter-parameter logical consistency verification is a correlation coefficient less than 0.8; and the threshold for motion state matching degree evaluation is a matching score less than 0.7.

[0053] An explanatory content generator automatically generates explanatory annotations and explanations based on verification results and causal inference results. The explanatory annotations are embedded in the main text of the document, and the explanations are included in the footnotes of the document.

[0054] Explanatory annotations are generated based on a pre-set annotation template library. For different types of verification anomalies and causal analysis results, the appropriate explanation template is selected and specific parameter values ​​are filled in. The generated explanatory annotations are directly embedded in the relevant data record positions in the main text of the document, providing immediate explanations for abnormal data or special processing.

[0055] The adaptive adjustment unit dynamically adjusts the document template and content based on the verification results. When a preset motion condition is detected, it automatically switches to the corresponding document template. The template switching control is based on the detection results of the preset motion condition. When the system detects that the roll angle continuously exceeds a threshold, the ship enters a specific navigation state, or encounters special sea conditions, it automatically triggers the document template switching operation to ensure that the document format matches the current ship motion state. In this embodiment, the template switching conditions include: switching to the high motion state template when the roll angle exceeds 5° for 30 seconds; and switching to the berthing operation template when the speed is below 2 knots and the distance to the dock is less than 100 meters.

[0056] The template switching control is based on the detection results of preset motion conditions. When the system detects that the roll angle continuously exceeds the threshold, the ship enters a specific navigation state, or encounters special sea conditions, it automatically triggers the document template switching operation to ensure that the document format matches the current ship motion state.

[0057] Specifically, when a vessel enters rough seas from calm waters, the document generation and verification module automatically switches from the calm water template to the high-motion state template, adjusting the recording frequency and allowable error range accordingly. A three-layer closed-loop verification process ensures the accuracy of document recordings. Any anomalies detected during verification are automatically annotated using an explanatory content generator, providing comprehensive background information for subsequent data use and analysis. This adaptive document generation and verification mechanism effectively solves the problems of rigid formatting and lack of state adaptability in traditional document systems, providing reliable technical support for document recording by vessels under various navigation conditions.

[0058] Furthermore, the blockchain-based evidence storage and feedback module includes:

[0059] A composite evidence generator is used to combine document content hash value, motion context digest and causal evidence chain digest to generate composite evidence information. The motion context digest includes motion parameter and compensation parameter digests, and the causal evidence chain digest includes causal inference nodes.

[0060] The document content hash value is calculated using the SHA-256 algorithm to ensure the integrity verification of the document content. The motion context summary is extracted from the key motion parameters and compensation parameters output by the ship motion perception and correction module, including motion characteristic parameters such as roll angle, pitch angle, and vibration intensity, as well as specific compensation values ​​for each sensor. These parameters are standardized to form a summary information in a fixed format. The causal evidence chain summary is derived from the key nodes of the evidence chain generated by the causal reasoning and evidence chain module, including core information such as the cause of the anomaly, the impact path, and the theoretical error. The most representative reasoning nodes are retained through feature extraction algorithms.

[0061] The composite process uses a cascaded hashing method, which first calculates the hash value of each of the three components, and then combines these hash values ​​to generate the final composite evidence information, ensuring that the evidence content can both verify the integrity of the document and provide complete background information support.

[0062] The hierarchical evidence storage controller dynamically adjusts the evidence storage strategy based on the document credibility score. When the document credibility is higher than the first threshold, only the document hash and basic metadata are stored. When the document credibility is between the first and second thresholds, a summary of the causal evidence chain is added. When the document credibility is lower than the second threshold, additional original data fragments from the sensor are stored.

[0063] When the document's credibility is higher than the first threshold, the system determines that the document is of high quality and only stores the document hash and basic metadata. This basic evidence storage mode saves storage resources to the maximum extent while ensuring verifiability. When the document's credibility is between the first and second thresholds, the system determines that there is a certain quality risk and adds a summary of the causal evidence chain to provide reasoning support for possible disputes. When the document's credibility is lower than the second threshold, the system determines that there is a significant quality problem and additionally stores original sensor data fragments to retain the most original data evidence for in-depth analysis. This hierarchical evidence storage strategy optimizes the system's resource utilization efficiency while ensuring the sufficiency of evidence through intelligent resource allocation.

[0064] Specifically, the threshold values ​​for the evidence storage strategy in the tiered evidence storage controller can be set as follows:

[0065] The first threshold is set to 0.8. When the document credibility is higher than 0.8, the system determines that the document quality is high and only stores the document hash and basic metadata. This basic evidence storage mode saves storage resources to the maximum extent while ensuring verifiability.

[0066] The second threshold is set to 0.6. When the credibility of a document is between 0.6 and 0.8, the system determines that there is a certain quality risk and adds a summary of the causal evidence chain to provide reasoning support for possible disputes.

[0067] When the document's credibility score is below 0.6, the system determines that there is a significant quality problem and additionally stores the original sensor data fragments to preserve the most original data evidence for in-depth analysis.

[0068] In addition, the blockchain evidence storage and feedback module also includes: a smart contract verification interface, which provides functions for document integrity verification, causal evidence chain verification, and motion state replay verification.

[0069] Document integrity verification confirms whether the document content has been tampered with by comparing the hash value of the document to be verified with the original hash value stored on the blockchain. The verification process is executed entirely on-chain, ensuring the immutability of the verification result. Causal evidence chain verification verifies whether the explanations in the document conform to the logical reasoning process by replaying the causal evidence chain summary stored on the blockchain. This verification combines the reasoning rules stored on-chain and the complete evidence chain data maintained off-chain.

[0070] The feedback optimization unit collects document verification results and user feedback, analyzes verification failure modes, and regularly optimizes physical model parameters and verification rules.

[0071] Furthermore, the intelligent paperless document system also includes a time synchronization and data acquisition module, which includes:

[0072] A precision clock synchronization unit provides time synchronization for distributed programmable logic controllers based on a precision time protocol.

[0073] The precision clock synchronization unit provides a unified time reference for the ship's distributed programmable logic controllers (PLCs) based on the Precision Time Protocol (PTP).

[0074] The motion-triggered acquisition unit uses the ship's characteristic motion pattern as the data acquisition trigger signal. When the rudder operation angle is detected to be greater than the threshold, the generated characteristic vibration will trigger the synchronous acquisition of all data acquisition nodes on the ship. In this embodiment, the trigger threshold is set as follows: the rudder operation angle threshold is set to 10°, and the roll acceleration threshold is set to 5° / s². When these characteristic motions are detected, the system automatically triggers the synchronous acquisition of all data acquisition nodes on the ship, and the acquisition window is set to 30 seconds before and after the trigger.

[0075] The precision clock synchronization unit adopts a master-slave architecture. In the ship network, a node with a stable clock source is selected as the master clock, and all other PLC devices are slave clocks. The synchronization process is achieved through a timestamp exchange mechanism. The master clock periodically sends synchronization messages to the slave clocks, and the slave clocks record the message reception time. The network transmission delay and clock offset are calculated through a delay request-response mechanism. This unit takes into account the impact of ship vibration on the clock signal and introduces a vibration compensation factor in the clock offset calculation. This factor is dynamically adjusted according to the real-time monitored vibration intensity to ensure that the ship's data acquisition system can maintain accurate time consistency even in the ship's vibration environment.

[0076] The data watermark embedding unit embeds a digital watermark generated by a precision time protocol timestamp and a trigger event identifier into each sensor data packet. The digital watermark provides the function of verifying the timing correctness of the data packet.

[0077] The data watermark embedding unit receives a precision time protocol timestamp from the precision clock synchronization unit and a trigger event identifier from the motion trigger acquisition unit. It converts this information into a digital watermark using a watermark generation algorithm. The watermark generation process employs a lightweight encryption algorithm to ensure the authenticity and integrity of the watermark information. The generated watermark is embedded into a specified field of the sensor data packet, forming an inseparable whole with the acquired data. The digital watermark provides the function of verifying the timing correctness of the data packet. In subsequent data processing stages, the system can extract and verify the watermark information to confirm whether the acquisition time, trigger event, and timing relationship of the data packet meet expectations.

[0078] Furthermore, the intelligent paperless document system also includes a device authentication and data integrity module, which includes:

[0079] The vibration fingerprint extraction unit analyzes the vibration spectrum characteristics of each sensor installation location and generates a vibration fingerprint for that location.

[0080] The vibration fingerprint extraction unit collects vibration signals of the ship during operation through an accelerometer installed near the sensor. The sampling frequency is not less than 1kHz to ensure that the complete vibration spectrum is captured. The collected raw vibration signal is first preprocessed, including noise reduction, filtering and signal enhancement. Then, the time domain signal is converted into frequency domain features through fast Fourier transform. The feature extraction algorithm selects distinguishable feature parameters from the spectrum, including the main resonant frequency, spectral envelope features, harmonic distribution mode, etc. These parameters together constitute the vibration fingerprint of that location.

[0081] The device identity binding unit binds the digital certificate of the sensor device to the vibration fingerprint of its installation location, establishing a correspondence between the device and the location;

[0082] The digital certificate adopts the X.509 standard format and contains elements such as the sensor's unique identifier, public key information, and issuing authority signature. The binding process first verifies the sensor's digital certificate to ensure its authenticity and validity. Then, it extracts the vibration fingerprint feature vector of the installation location. The binding algorithm combines the hash value of the digital certificate with the vibration fingerprint feature vector to generate a device-location binding credential. This credential is digitally signed to ensure its immutability. The binding information is stored in a secure credential database, and the hash value of the binding credential is uploaded to the blockchain for notarization, establishing a traceable binding record.

[0083] The in-situ verification unit periodically requires the sensor to simultaneously acquire a vibration signal while collecting data. It verifies the sensor's presence by comparing the current vibration signal with a pre-stored vibration fingerprint. The in-situ verification unit triggers the verification process according to a preset time interval or a specific event. During verification, the system requires the sensor to simultaneously acquire a vibration signal through its built-in vibration sensor while collecting data. The acquired vibration signal undergoes the same processing procedure as the vibration fingerprint extraction unit to extract the current feature parameters. The comparison algorithm calculates the similarity between the current feature parameters and the pre-stored vibration fingerprint. A threshold is set to determine the sensor's presence. In this embodiment, the similarity threshold is set to 0.85, and the verification frequency is set to once per hour. The verification result, along with a timestamp, is recorded in the verification log, providing a complete audit trail for device status monitoring.

[0084] The in-situ verification unit triggers the verification process according to a preset time interval or a specific event. During verification, the system requires the sensor to collect a vibration signal simultaneously through the built-in vibration sensor while collecting data. The collected current vibration signal undergoes the same processing process as the vibration fingerprint extraction unit to extract the current feature parameters. The comparison algorithm calculates the similarity between the current feature parameters and the pre-stored vibration fingerprint. A threshold is set to determine whether the sensor is in place. The verification result, along with the timestamp, is recorded in the verification log, providing a complete audit trail for device status monitoring.

[0085] The data integrity signature unit requires the sensor to use a private key to digitally sign the data content and vibration fingerprint fragments, providing the function of verifying the source of the data.

[0086] This unit is responsible for ensuring the integrity and authenticity of sensor data. The data integrity signing unit requires the sensor to digitally sign the data content and vibration fingerprint fragments using its private key before sending data. The signing process first combines the data content collected by the sensor with the feature fragments of the current vibration fingerprint to generate a data packet to be signed. Then, the sensor's private key is used to digitally sign the data packet to generate a digital signature. The signed data packet contains three parts: the original data, the vibration fingerprint fragments, and the digital signature. The function of verifying the source of data is achieved by verifying the digital signature. The receiver uses the sensor's public key to verify the validity of the signature, thereby confirming that the data does indeed come from the designated sensor and has not been tampered with during transmission.

[0087] When the system detects that the vibration characteristics of a certain oil tank level sensor do not match the pre-stored fingerprint, it will immediately trigger an alarm and record the abnormal event to prevent data tampering caused by malicious replacement of the sensor. At the same time, through the digital signature mechanism provided by the data integrity signature unit, the authenticity and integrity of all sensor data are ensured, providing a reliable data foundation for subsequent document generation and blockchain evidence storage.

[0088] Furthermore, the intelligent paperless document system also includes an external data import and verification module, which includes:

[0089] The oracle interface unit introduces external environmental data corresponding to the ship's position and time published by an authoritative institution through a blockchain oracle. The external environmental data includes ocean currents, wind speed and direction, and water temperature data.

[0090] All external environmental data is timestamped and digitally signed by authoritative institutions to ensure its timeliness and reliability. The data format adopts the internationally standardized marine data exchange format, which facilitates unified processing and analysis by the system. The data update frequency is set to once per hour to ensure the real-time nature of the data.

[0091] The environmental consistency verification unit uses external environmental data combined with a ship dynamics model to calculate the expected range of sensor readings and verify whether its own sensor data is within the expected range.

[0092] The environmental consistency verification unit first calculates the theoretical expected range of sensor readings under the current environmental conditions based on external environmental data and the ship's basic parameters using a ship motion mathematical model. The calculation process considers multiple factors such as ship type, loading status, and sailing speed, and employs a multiphysics coupling analysis method to ensure the accuracy of the expected range. In the verification phase, the actual sensor readings are compared with the expected range, and statistical hypothesis testing methods are used to determine the consistency of the data. For data that exceeds the expected range, the system records the magnitude and direction of the deviation and analyzes possible influencing factors. The verification results generate an environmental consistency report, providing important reference for subsequent data processing and document generation.

[0093] The external evidence association unit associates external environmental data with internal document data. When the internal data deviates significantly from the expected range due to motion distortion, it generates environmental consistency proof. Specifically, it is automatically triggered when the deviation between the actual data and the expected range exceeds 20%.

[0094] The external evidence association unit establishes a mapping relationship between external environmental data and internal document parameters. It identifies the influence patterns of environmental factors on document data through data association algorithms. When it detects that internal data deviates significantly from the expected range due to ship motion distortion, the system automatically generates an environmental consistency certificate. This certificate records in detail the external environmental conditions, theoretical expected range, actual measurement data, and deviation analysis results, forming a complete chain of evidence. The environmental consistency certificate adopts a standardized electronic evidence format, which includes timestamps and data source metadata.

[0095] For example, in typhoon-affected sea areas, the system obtains officially released typhoon path and intensity data through the oracle interface, and calculates the reasonable range of sensor readings under the current wind and wave conditions by combining the ship dynamics model. When the ship's movement causes abnormal sensor readings, the environmental consistency verification unit can identify whether these abnormalities are within the explainable range of environmental factors, and the external evidence association unit generates corresponding environmental consistency proofs to provide external evidence for the abnormal records in the documents.

[0096] Furthermore, the intelligent paperless document system also includes an offline processing and consensus module, which includes:

[0097] The micro-block consensus unit runs a lightweight consensus algorithm among multiple data acquisition gateways within the ship's local area network to reach consensus on micro-blocks of data during specific motion events;

[0098] The micro-block consensus unit employs an optimized practical Byzantine fault-tolerant algorithm, which is specifically optimized for the characteristics of the ship network environment. The consensus process begins with the master node packaging the data collected during a specific motion event into data micro-blocks. Each data micro-block contains the data collected by all relevant sensors during the event and its metadata. Then, the master node broadcasts the data micro-blocks to all slave nodes. The slave nodes verify the received data micro-blocks, including data integrity, timestamp continuity, and logical consistency. When more than two-thirds of the nodes reach a consensus, the consensus is complete. This lightweight consensus algorithm is suitable for the limited network resources of ships.

[0099] The motion event detection unit detects violent motion events of the ship in real time, and uses large-angle turns and emergency collision avoidance operations as trigger signals for micro-block generation.

[0100] The motion event detection unit continuously monitors six-degree-of-freedom motion data from the ship's motion perception and correction module. It identifies severe motion events through a multi-threshold detection algorithm. For large-angle turning events, the detection algorithm makes a comprehensive judgment based on the rudder angle change rate and roll acceleration. For emergency collision avoidance operations, the detection algorithm combines radar data, AIS data, and ship motion status for multi-source information fusion judgment. When a severe motion event that meets the conditions is detected (in this embodiment, the rudder angle threshold for large-angle turning events is set to 30°, and the rudder angle change rate threshold is set to 15° / s; the target distance threshold for emergency collision avoidance operations is set to 500 meters, and the approach speed threshold is set to 10 knots), the motion event detection unit immediately sends a trigger signal to the micro-block consensus unit to start the generation and consensus process of data micro-blocks.

[0101] The offline buffer management unit encrypts and caches the micro-blocks that have reached a consensus on distributed nodes during network interruption, and puts the aggregate hash of the entire micro-block sequence on the chain after the network is restored;

[0102] The offline buffer management unit adopts a distributed storage architecture, establishing encrypted cache areas on multiple data acquisition gateway nodes of the ship's local area network. The cache areas employ a hierarchical storage strategy, storing recently acquired micro-blocks in memory to improve access speed, while historical micro-blocks are transferred to persistent storage. Each micro-block is encrypted before caching to ensure data security during offline periods. When the network connection is restored, the offline buffer management unit performs aggregate hash calculations on all cached micro-block sequences in chronological order to generate an aggregate hash value representing the integrity of the data throughout the offline period. This aggregate hash value is submitted to the blockchain network for notarization, forming a trusted proof of the data during the offline period.

[0103] The consensus credible proof unit generates a credible proof of the consensus process for each micro-block, wherein the credible proof includes the signatures of the nodes participating in the consensus and the consensus result;

[0104] The proof generation process uses a Merkle tree structure to organize signature data, ensuring the integrity and verifiability of the proof. The generated trusted proof is bound to the corresponding data micro-block and stored together in the offline buffer management unit. In the subsequent data verification process, the node signature in the trusted proof can be verified to confirm that the data micro-block has indeed been verified by the consensus process.

[0105] When a ship encounters a network outage while sailing in the open sea, the offline processing and consensus module can ensure the reliable recording and storage of critical motion event data. For example, during an emergency collision avoidance operation, even if the connection between the ship and the shore-based network is interrupted, the system can still reach a consensus on the data micro-blocks during the operation within the ship's local area network and store them securely through the offline buffer management unit. After the network is restored, the sequence of micro-blocks during the entire operation is aggregated, hashed, and uploaded to the blockchain, providing complete and reliable data evidence for accident analysis.

[0106] Furthermore, the intelligent paperless document system also includes an adaptive learning and optimization module. This module continuously learns and optimizes to improve the system's adaptability to multi-source data acquisition systems under complex ship motion conditions, thereby ensuring the authenticity and credibility of the generated electronic documents. The adaptive learning and optimization module includes:

[0107] The online learning unit uses online machine learning algorithms to continuously learn the error characteristics and correction parameters of each sensor of a specific ship under different motion modes, so as to realize the self-evolution of the model;

[0108] The online learning unit is based on an incremental learning framework, such as using random forest or neural network algorithms. It takes real-time ship motion data (such as roll and pitch angles) and raw sensor data as input and outputs correction parameters for sensor errors. The unit dynamically adjusts the correction model parameters by analyzing the correlation between motion patterns and sensor reading deviations in historical data. For example, when a ship encounters a large roll angle, the reading of the liquid level sensor will be incorrect due to the tilt of the liquid surface. The online learning unit learns and updates the tilt compensation parameters of the liquid level sensor by comparing the sensor data under multiple roll events with the expected values.

[0109] The parameter locking unit locks the hash value of the parameters on the blockchain after the parameters of the calibration model have been verified to be stable, providing a verification function for the credibility of the model evolution.

[0110] The parameter locking unit monitors the parameter change rate output by the online learning unit. When the parameter changes less than a preset threshold in multiple consecutive motion events, it determines that the parameter is stabilizing. Then, it generates a hash value of the parameter (such as using the SHA-256 algorithm) and stores the hash value on the blockchain through the blockchain notarization and feedback module.

[0111] The version management unit assigns a version identifier to each calibration model, associates the model version identifier with document storage, and provides a function to trace the document generation process.

[0112] After the parameter locking unit locks the parameters, the version management unit automatically generates a unique version identifier (such as a timestamp or serial number) and binds this identifier to the calibration model parameters and document storage records. For example, when the liquid level sensor calibration model is updated, the version management unit assigns a new version number and embeds this version number into the document metadata when generating the document.

[0113] Among them, the adaptive learning and optimization module, through the collaborative work of the above units, enables the system to continuously adapt to the ship's motion environment. In practical application scenarios, such as when the ship is sailing in severe sea conditions, this module can learn historical motion data, optimize sensor correction parameters, and ensure the credibility of model evolution through version management and parameter locking, ultimately generating highly realistic electronic documents.

[0114] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart paperless document system based on ship environment adaptation and blockchain evidence storage, characterized in that: include: The ship motion sensing and correction module is used to monitor the six degrees of freedom motion of the ship in real time, and to dynamically compensate multi-source sensor data based on the ship's physical model to generate corrected sensor data. The causal reasoning and evidence chain module is used to construct a causal graph model between the ship's motion state and document parameters, perform causal inference based on the current motion state and sensor data, and generate a causal evidence chain. The document generation and verification module is used to dynamically select document templates based on the ship's motion status, implement three-layer closed-loop verification during the document generation process, and adaptively adjust the document content based on the verification results. The blockchain evidence storage and feedback module is used to package document content, motion context and causal evidence chain summary to generate composite evidence storage information, and implement a hierarchical evidence storage strategy based on the credibility of the document; The causal reasoning and evidence chain module includes: The cause-effect graph model building unit is used to establish a three-layer cause-effect graph model between ship motion state and document parameters. The three-layer cause-effect graph model includes a physical layer, a parameter layer and a logic layer. The physical layer is used to describe the physical relationship between ship motion and sensor readings. The parameter layer is used to describe the conversion relationship between sensor readings and document parameters. The logic layer is used to describe the logical relationship between document parameters. The causal reasoning engine unit performs causal inference based on the current motion state and corrected sensor data, predicts the theoretical value range of each document parameter, identifies the deviation between the actual value and the theoretical value, and determines the cause of the anomaly through causal tracing. The evidence chain generation unit transforms the causal reasoning process into a verifiable evidence chain, which includes the cause of the anomaly, the impact path, the theoretical error, and the nodes of the compensation measures. Evidence chain compressor compresses complete causal evidence chains into natural language interpretations suitable for document format.

2. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage as described in claim 1, characterized in that, The ship motion sensing and correction module includes: The six-degree-of-freedom motion monitoring unit is used to collect data on the ship's roll, pitch, heel, sway, sway, and heave motions. The ship hull physical model library stores physical deformation models built for different ship hull types. The physical deformation models include hull structural parameters, elastic deformation characteristics, and sensor installation location information. The sensor calibration engine generates personalized compensation parameters for each sensor based on real-time six-degree-of-freedom motion state and ship hull physical model. The calculation of these compensation parameters is performed in a trusted execution environment. The sensors include a level sensor, an attitude sensor, and a speed sensor.

3. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage as described in claim 2, characterized in that, The document generation and verification module includes: A dynamic document template library contains preset document templates for different motion states, including templates for calm waters, templates for high motion states, and templates for berthing operations. A three-layer closed-loop verification engine is used to implement single-parameter rationality checks, inter-parameter logical consistency verification, and motion state matching degree evaluation. The single-parameter rationality checks are based on the dynamically calculated reasonable range of parameters, the inter-parameter logical consistency verification is based on the ship dynamics model, and the motion state matching degree evaluation calculates the degree of consistency between the document records and the current motion state. An explanatory content generator generates explanatory annotations and explanations based on verification results and causal inference results. The explanatory annotations are embedded in the main text of the document, and the explanations are included in the footnotes of the document. The adaptive adjustment unit dynamically adjusts the document template and content based on the verification results, and automatically switches to the corresponding document template when a preset motion condition is detected.

4. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage as described in claim 3, characterized in that, The blockchain-based evidence storage and feedback module includes: A composite evidence generator is used to combine document content hash value, motion context digest and causal evidence chain digest to generate composite evidence information. The motion context digest includes motion parameter and compensation parameter digests, and the causal evidence chain digest includes causal inference nodes. The hierarchical evidence storage controller dynamically adjusts the evidence storage strategy based on the document credibility score. When the document credibility is higher than the first threshold, only the document hash and basic metadata are stored. When the document credibility is between the first and second thresholds, a summary of the causal evidence chain is added. When the document credibility is lower than the second threshold, additional original data fragments from the sensors are stored.

5. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage as described in claim 1, characterized in that, The intelligent paperless document system also includes a time synchronization and data acquisition module, which includes: A precision clock synchronization unit provides time synchronization for distributed programmable logic controllers based on a precision time protocol. The motion-triggered acquisition unit uses the ship's characteristic motion pattern as the data acquisition trigger signal. When the rudder operation angle is detected to be greater than the threshold, the resulting characteristic vibration will trigger the synchronous acquisition of data from all data acquisition nodes on the ship. The data watermark embedding unit embeds a digital watermark generated by a timestamp and a trigger event identifier into each sensor data packet.

6. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage according to claim 1, characterized in that, The intelligent paperless document system also includes a device authentication and data integrity module, which includes: The vibration fingerprint extraction unit analyzes the vibration spectrum characteristics of each sensor installation location and generates a vibration fingerprint for that location. The device identity binding unit binds the digital certificate of the sensor device to the vibration fingerprint of its installation location, establishing a correspondence between the device and the location; The in-situ verification unit verifies whether the sensor is in place by comparing the current vibration signal with the pre-stored vibration fingerprint. The data integrity signature unit requires the sensor to use a private key to digitally sign the data content and vibration fingerprint fragments, providing the function of verifying the source of the data.

7. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage according to claim 1, characterized in that, The intelligent paperless document system also includes an external data import and verification module, which includes: The oracle interface unit introduces external environmental data corresponding to the ship's position and time from authoritative institutions through a blockchain oracle; The environmental consistency verification unit uses external environmental data combined with a ship dynamics model to calculate the expected range of sensor readings and verify whether its own sensor data is within the expected range. The external evidence association unit links external environmental data with internal document data.

8. The intelligent paperless document system based on ship environment adaptation and blockchain evidence storage according to claim 1, characterized in that, The intelligent paperless document system also includes an offline processing and consensus module, which includes: Micro-block consensus units run lightweight consensus algorithms among multiple data acquisition gateways within the ship's local area network; The motion event detection unit detects violent motion events of the ship in real time, and uses large-angle turns and emergency collision avoidance operations as trigger signals for micro-block generation. The offline buffer management unit encrypts and caches the micro-blocks that have reached a consensus on distributed nodes during network interruption, and puts the aggregate hash of the entire micro-block sequence on the chain after the network is restored; The consensus credible proof unit generates a credible proof of the consensus process for each micro-block. The credible proof includes the signatures of the nodes participating in the consensus and the consensus result.

Images