A method and system for identifying illegal declaration of a ship passing through a lock based on big data
By analyzing ship lock passage behavior using big data, combined with positioning data and start/stop logs, the problem of low accuracy and efficiency in violation identification in existing technologies has been solved. This has enabled fully automated monitoring of ship violations, improving identification accuracy and verification efficiency.
Patent Information
- Application Number
- CN202611323788.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies, the identification of illegal declarations by vessels passing through locks relies on manual reporting and verification, lacking proactive and fully automated screening capabilities. It cannot distinguish between objective signal interruptions and manual shutdowns, resulting in low identification accuracy and low verification efficiency. Furthermore, it cannot achieve real-time automated supervision of the entire process from vessel declaration to waiting for lock passage.
By employing a big data-based approach, the system acquires ship lock passage declaration information, positioning data, and positioning terminal power-on/off logs. It then performs segmented temporal neighborhood weighted smoothing and noise reduction, time-series gap hierarchical interpolation filling, and speed-geographic linkage dual-layer anomaly screening. Combined with multi-dimensional judgment of continuous points and the Haversine formula, it achieves multi-dimensional analysis of ship trajectories and identification of illegal power-off.
It has improved the automatic identification rate of violations, reduced the verification time per vessel, lowered the false judgment rate, enhanced the level of automation, and improved the fairness and efficiency of navigation scheduling.
Smart Images

Figure CN123116076A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of lock vessel scheduling and control, and in particular to a method and system for identifying illegal declarations of lock vessels based on big data. Background Technology
[0002] The Three Gorges-Gezhouba section of the Yangtze River, located between Miaohe and Zhongshuimen, is a crucial section of the Yangtze River and an essential passage for ships entering and leaving the Sichuan section. The region experiences complex and changeable climate, unique hydrological and water quality conditions, and frequent ship traffic, resulting in continuously increasing navigation pressure.
[0003] Currently, the Three Gorges Dam lock passage scheduling is mainly based on a comprehensive ranking system of vessel declaration time, actual anchoring order, and credit records, implementing a queuing rule of first-come, first-served, and based on creditworthiness. However, some ship owners, in order to obtain a better lock passage sequence, have resorted to various illegal means to disrupt the fair order. Typical behaviors include: falsely reporting the estimated anchoring time (declaring in advance to grab a spot), turning off the AIS / BeiDou positioning terminal outside the waiting area and speeding up the voyage (artificially creating a trajectory blind spot to shorten the actual voyage), and falsely claiming equipment failure to avoid penalties for delayed anchoring.
[0004] Currently, screening for the aforementioned violations mainly relies on reports from vessels, which has significant limitations: First, evidence collection depends on external reports, lacking proactive and fully automated screening capabilities, making it difficult to detect many hidden violations in a timely manner. Second, the manual evidence collection process is cumbersome, requiring cross-system retrieval of various data such as declaration forms, location trajectories, terminal power-on / off logs, and video surveillance data. Verification of violations for a single vessel takes an average of over 2.5 hours, which cannot support large-scale, real-time, and routine supervision. Third, existing anomaly analysis technologies based on location data generally rely on the continuous and stable uploading of data from vessel positioning terminals. When encountering objective situations such as network signal obstruction or temporary equipment malfunctions, the backend system may be unable to perform effective analysis due to missing trajectories, easily confusing normal signal interruptions with manual shutdowns. This could lead to missed detections of genuine violations or misjudgments of law-abiding vessels, resulting in insufficient supervision. Summary of the Invention
[0005] The main objective of this invention is to provide a big data-based method for identifying illegal declarations of vessels passing through locks. This method addresses the problems of reliance on manual reporting and verification for illegal supervision, significant deficiencies in basic trajectory preprocessing, inability to distinguish between objective signal interruptions and manual shutdown, difficulty in achieving real-time automated identification of illegal activities throughout the entire process from vessel declaration to lock waiting, low identification accuracy, and low verification efficiency.
[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for identifying illegal declarations of vessels passing through locks based on big data, comprising the following steps: Obtain vessel lock passage declaration information, vessel positioning data, and positioning terminal power on / off log data, and preprocess the data; The time window for trajectory analysis before declaration is determined based on the time of the vessel's declaration to pass through the lock. The first trajectory data of the vessel within the time window before declaration is extracted. The first trajectory data is subjected to multi-dimensional judgment of continuous points. If it is determined that there is a positioning interruption and / or speed abnormality, the vessel is marked as having violated the declaration rules by combining the power on / off log data of the positioning terminal. Based on the vessel's lock passage status, when the vessel is in the state of having declared but not yet anchored, the second trajectory data of the vessel from the time of declaration to the current time is extracted. The second trajectory data is then subjected to multi-dimensional judgment of continuous points. If it is determined that there is a positioning interruption and / or speed abnormality, the vessel is identified as having violated the declaration by combining the positioning terminal power-on / off log data and marked as having violated the declaration. When the ship's status changes to "anchored", the actual anchoring time is recorded, the last valid positioning point outside the anchor line and the first valid positioning point inside the anchor line are extracted, and the instantaneous speed between the two points is calculated. If the instantaneous speed exceeds the preset speed threshold, the ship is judged to have illegally anchored by combining the number of valid positioning points within the preset distance range outside the anchorage area, and the ship is marked as having illegally anchored. When a vessel is in the state of having reached anchor but not passing through the lock, the third trajectory data of the vessel from the time of reaching anchor to the current time is extracted. The continuous point time interval of the third trajectory data is judged. If it is determined that there is a positioning interruption, the positioning terminal power-on / off log data is combined to identify the vessel that has violated the rules after reaching anchor and mark the vessel that violated the rules after reaching anchor. Output all violation identification results.
[0007] In a preferred embodiment, the data preprocessing includes: Segmented temporal neighborhood weighted smoothing denoising is employed: a sliding window is constructed using 10 consecutive temporal positioning points, and temporal weights are assigned based on the distance of each point from the current time. The weighted average of latitude, longitude, and speed within the window is calculated. If the deviation between the original latitude, longitude, and speed of the current point and the weighted average exceeds a preset threshold, the window weighted average is used to correct the noisy points. Tiered interpolation for filling time-series gaps: Differentiated processing based on three levels according to the duration of missing location data: missing duration ≤ normal location upload interval Standard linear interpolation is used. For missing durations less than 30 minutes, piecewise linear interpolation with channel boundary constraints is used, and the interpolation coordinates are limited to the channel waterway vector range; for missing durations ≥ 30 minutes, interpolation completion is not performed. Speed-Geographic Linkage Two-Layer Anomaly Screening: The first layer utilizes the waterway GIS vector boundary to directly eliminate drift points falling into land and tidal flat areas; the second layer uses a combination of speed and geographic information systems to filter out remaining points within the waterway. The principle is to screen out points with abnormal speeds; only points that simultaneously meet the criteria of drifting outside the channel and speed deviation of 3 times the standard deviation are judged as invalid anomalies and removed, while points with excessive speed within the channel are retained and marked as suspected anomalies.
[0008] In the preferred scheme, the continuous points are judged from multiple dimensions, including: Time interval judgment: Obtain the time difference between any two consecutive positioning points in the trajectory. If the time difference exceeds the custom time interval threshold Z, it is determined that there is a positioning interruption abnormality. Speed determination: The instantaneous speed of the ship is calculated based on the distance and time difference between two consecutive positioning points. If the instantaneous speed exceeds the legal navigation speed range for this river section, it is determined that there is a speed anomaly. The time interval judgment result and the speed judgment result are jointly analyzed. If both the positioning interruption abnormality and the speed abnormality exist at the same time, it is marked as a highly suspected violation. Wherein, the time interval threshold The calculation formula is: Let N be the total number of vessels that declare passage through the lock within one year, and n be the total number of location data points uploaded by all N vessels within the time period X. For the i-th data point and the i-th Calculate the average μ of all valid time intervals between a given data point: ; Calculate the sample standard deviation σ: ; According to the 3σ principle, the theoretical maximum upload interval =μ+3σ.
[0009] In the preferred scheme, the actual geographical distance S between the two locations is calculated using the Haversine formula, which is: ; In the formula, R is the average radius of the Earth (a fixed constant is taken for inland river scenarios). , These are the latitudes of the first and second points, respectively. These are the longitudes of the first and second points, respectively.
[0010] In the preferred embodiment, the step of identifying unauthorized shutdowns by combining power-on / off log data from the positioning terminal specifically involves: Obtain the start time of the location interruption period and the time when the location terminal is turned off. If the shutdown time is earlier than or equal to the location interruption start time... If the location interruption time matches the shutdown time, it is determined to be a human-induced illegal shutdown and marked as a violation vessel; otherwise, it is determined to be a location interruption caused by objective factors and is not marked as a violation.
[0011] In the preferred embodiment, the determination of the violation at the anchor specifically includes: Obtain the last valid positioning point outside the anchor line. And to the first effective positioning point within the anchor line Record the corresponding time , Calculated using the Haversine formula and The actual geographical distance between Then instantaneous velocity ; like If the speed is greater than the preset speed threshold V, then the preset distance range from outside the anchorage to the anchor line is extracted. The number of effective positioning points for vessels within the territory; If the number of valid positioning points is less than a preset threshold If so, it is determined to be a vessel that has arrived at anchor ahead of schedule in violation of regulations.
[0012] In the preferred embodiment, the time window for trajectory analysis before declaration is 2 to 6 hours before the declaration gate passage time.
[0013] In the preferred scheme, the vessel's lock passage status is obtained cyclically at preset time intervals. If the vessel's status is still "declared but not yet anchored", the abnormal trajectory detection and illegal shutdown identification after declaration are repeated. If the status changes to "anchored", the actual anchoring time is recorded and real-time monitoring is terminated, and the process is transferred to the post-anchoring processing flow. If the status changes to "passed through the lock" or "declaration withdrawn", the identification is terminated and the record is archived.
[0014] In a preferred embodiment, the ship positioning data includes AIS positioning data and / or BeiDou positioning data.
[0015] Secondly, the present invention provides a system for identifying and recognizing illegal declarations of vessels passing through locks based on big data, characterized in that the method for implementing the aforementioned method for identifying and recognizing illegal declarations of vessels passing through locks based on big data includes: The data acquisition module is used to acquire ship lock passage declaration information, ship positioning data, and positioning terminal power on / off log data; A data preprocessing module is used to preprocess the data; The pre-declaration violation identification module is used to determine the pre-declaration trajectory analysis time window based on the vessel's declared lock passage time, extract the vessel's first trajectory data within the pre-declaration time window and perform multi-dimensional judgment of continuous points, and combine the positioning terminal's power on / off log data to identify pre-declaration violation shutdown and mark vessels that violated pre-declaration regulations. The module for identifying violations of not arriving at anchor after declaration is used to extract the second trajectory data of the vessel from the time of declaration to the current time and make multi-dimensional judgments on continuous points when the vessel is in the state of having declared but not arriving at anchor. It combines the power-on / off log data of the positioning terminal to identify violations of power-off after declaration and marks the vessel that violated the declaration. The illegal anchoring identification module is used to record the actual anchoring time when the ship's status changes to anchored, extract the last valid positioning point outside the anchor line and the first valid positioning point inside the anchor line to calculate the instantaneous speed, and combine the number of valid positioning points within the preset distance range outside the anchor area to determine illegal anchoring and mark the ship that illegally anchored. The post-anchor violation identification module is used to extract the third trajectory data of the ship from the time of anchoring to the current time when the ship is in the state of anchoring but not passing through the lock. It performs continuous point time interval judgment and combines the positioning terminal power-on and power-off log data to identify post-anchor violation shutdown and mark the ship that violated the rules after anchoring. The output module is used to output all violation identification results.
[0016] This invention provides a method for identifying illegal declarations of vessels passing through locks based on big data, aiming to solve the problems of existing supervision relying on manual reporting and verification, trajectory analysis being easily confused with objective signal interruption and human-controlled shutdown, low identification accuracy, and poor verification efficiency. This invention first performs layered preprocessing on the raw positioning data, and then conducts violation identification in four stages: First, based on the trajectory window 2-6 hours before the declaration, it uses an adaptive 3σ threshold to determine positioning interruption, and combines the Haversine formula to calculate the speed to complete multi-dimensional trajectory analysis, linking the power on / off logs to identify vessels that illegally shut down before the declaration; Second, it continuously monitors the status of vessels that have declared but not yet reached anchor, and continuously extracts the trajectory to identify human shutdown behavior after the declaration; Third, when a vessel arrives at the anchorage, it extracts the instantaneous speed of key points inside and outside the anchor line, and uses double verification with the density of positioning points in the pre-anchor section to determine vessels that have illegally arrived at anchor ahead of schedule; Fourth, it continuously monitors the trajectory interruption of vessels that have reached anchor and are waiting for the lock, verifies illegal shutdown behavior after reaching anchor, and finally outputs all violation results uniformly. This improves the automatic identification rate of violations, reduces the verification time for a single vessel, improves the accuracy of distinguishing between objective anomalies such as equipment failure and signal obstruction and subjective violations, significantly reduces the false judgment rate, improves the degree of automation, and enhances the fairness and efficiency of the overall navigation scheduling of the hub. Attached Figure Description
[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation
[0018] Example 1 like Figure 1-2 As shown, this embodiment uses the Three Gorges-Gezhouba Dam section as the application scenario. It focuses on four core types of violations: cutting in line before vessel declaration, falsifying flight paths after declaration, forging anchor arrival before anchoring, and shutting down equipment after anchoring to evade lock gate supervision. Relying on multi-source big data such as vessel AIS, BeiDou positioning data, lock passage declaration data, terminal power-on / off logs, and waterway geographic information, it introduces continuous point multi-dimensional judgment algorithms, trajectory anomaly fitting algorithms, and threshold adaptive optimization algorithms to achieve full-process violation identification before, after, during, and after vessel declaration. The specific steps are as follows: Step 1: Data Preprocessing and Vessel Screening The first step in obtaining the ship's access data source includes: (1) Business data: Online declaration form for vessel passage through locks, including vessel name, MMSI identification number, declaration time, declaration route, and estimated time of arrival at anchor; (2) Positioning data: Ships use Beidou positioning + AIS to automatically report their location. The raw data includes longitude, latitude, timestamp, and instantaneous speed. (3) Equipment log: Location terminal power-on and power-off records, accurate to the minute level power-on and power-off timestamps; (4) Geographic data: GIS vector waterway boundary of river section, anchorage area range, and precise coordinates to anchor line.
[0019] In this embodiment, the ship passage status is polled in batches every 5 minutes, and all data is preprocessed. The preprocessing process includes data denoising, missing value imputation, and outlier removal, specifically: 1) Segmented temporal neighborhood weighted smoothing denoising: A sliding window is constructed using 10 consecutive temporal positioning points. Temporal weights are assigned based on the distance of the points from the current time, and the weighted average of latitude, longitude, and speed within the window is calculated. If the deviation between the original latitude, longitude, and speed of the current point and the weighted average exceeds a preset threshold, the window weighted average is used to correct the noisy points. This improves the integrity of the ship's true navigation trajectory outline, reduces the interference of positioning interval judgment, and reduces the basic probability of abnormal misjudgment.
[0020] 2) Time-series gap hierarchical interpolation filling: Based on the duration of missing location data, three levels of differentiated processing are applied: missing duration ≤ normal location upload interval Standard linear interpolation is used. When the missing duration is less than 30 minutes, piecewise linear interpolation with channel boundary constraints is used, and the interpolation coordinates are limited to the channel waterway vector range; when the missing duration is ≥30 minutes, interpolation completion is not performed, only the lost connection interval is marked and the terminal power-on / off log is linked to verify the human shutdown behavior, which improves data integrity.
[0021] 3) Speed-Geographic Linkage Two-Layer Anomaly Screening: The first layer uses the waterway GIS vector boundary to directly remove drifting points falling on land and tidal flats; the second layer uses the 3σ principle to screen for speed anomalies among the remaining points in the waterway; only points that simultaneously meet the criteria of drifting outside the waterway and speed deviation of 3 times the standard deviation are judged as invalid anomalies and removed, while speeding points in the waterway are retained and marked as suspected anomalies; this achieves thorough removal of invalid and dirty data and retention of suspected illegal speeding data, reduces the amount of computation, and improves data cleaning efficiency and the completeness of violation identification.
[0022] Step 2: Determining the time window for trajectory analysis before application Obtain the time for ships to declare their passage through the lock. To comprehensively capture the vessel's travel trajectory before declaration and avoid overlooking any violations, the starting time for uploading the vessel's location before declaration is determined to be ( - ). The window is typically set 2-6 hours before the declaration. This window covers the entire process of vessels in the upstream and downstream waters of the Three Gorges Dam entering the sensitive waiting area from regular navigation. It is a typical time period for vessels to accelerate, detour, or shut down their engines in advance.
[0023] This embodiment covers the entire process of a vessel navigating from the main line into a sensitive waterway awaiting lock, improving the completeness and accuracy of capturing various concealed violations before declaration.
[0024] Step 3: Multi-dimensional assessment of continuous locations before application right( - )to The complete trajectory of the ship within a time period is extracted, and a multi-dimensional judgment algorithm for continuous points is used to analyze all continuous points in the trajectory. First, a custom time interval is defined. (Based on analysis of historical ship data, this represents normal circumstances.) Within the time range, the maximum interval for uploading ship positioning data is specified, and at the same time, any two consecutive points in the trajectory are acquired sequentially. , Record the time corresponding to the location. , And the latitude and longitude coordinates of the point ( ), ( ).
[0025] The calculation formula is derived as follows: First, define N as the total number of vessels that declare passage through the lock within a year, and n as the number of vessels that declare passage through the lock within a year. The total number of AIS and BeiDou data points uploaded by all N ships within the time period; For the i-th data point and the i-th The time interval between one data point is: ; Calculate the sample standard deviation σ: ; Calculate the sample standard deviation σ: According to the 3σ principle, the theoretical maximum upload interval ( ): =μ+3σ. The same algorithm is reused to generate [the data] after the declaration and during the post-anchoring stage. It adapts to the differences in reported characteristics at different stages of navigation.
[0026] Continuous location determination is divided into three dimensions: first, time interval determination, calculation... and The difference, if > This indicates an abnormal interruption in positioning data between the two points, prompting further judgment in step 4; the second step is distance judgment, calculated using the Haversine formula. , The actual geographical distance S between the two points: ; In the formula, R is the average radius of the Earth (a fixed constant is taken for inland river scenarios). , These are the latitudes of the first and second points, respectively. These are the longitudes of the first and second points, respectively.
[0027] Secondly, speed is determined by calculating the ship's speed based on the distance and time interval between two points. ,like If the speed exceeds the legal navigation speed range for this section of the river, it is judged as abnormal speeding and marked as a suspected violation, with a joint judgment made based on abnormal time intervals. The normal navigation speed for vessels in the Three Gorges-Gezhouba section is generally controlled at 6-12 knots (approximately 11-22 km / h). If... If the signal strength is >12 (or exceeds 20 knots) and accompanied by a prolonged period without a positioning signal, it is highly suspected that the vessel accelerated and ran aground during the signal interruption in an attempt to reach anchor ahead of schedule, which is a typical violation of the rule of cutting in line before declaration. By using multi-dimensional continuous point verification, we can distinguish between normal signal fluctuations and positioning interruptions caused by human violations, thus reducing the false judgment rate.
[0028] This embodiment reduces the misjudgment rate for different vessels and at different times, improves the adaptability to vessel flow in the river section, and significantly enhances the accuracy of the initial screening of highly suspected illegal vessels.
[0029] Step 4: Identification of Unauthorized Power Off Before Submission Retrieve the terminal power-on / off logs for the positioning interruption interval and extract the power-on / off time points of the ship positioning terminal. Based on the location interruption time period determined in step 3, the unauthorized shutdown behavior is accurately identified. This indicates that at the time of the positioning interruption, the ship's positioning terminal was in a powered-off state, and the shutdown time perfectly matched the positioning interruption time. Combined with the abnormal distance and speed of the continuous points in step 3, equipment malfunction was ruled out as the cause of the shutdown. The ship was marked as a vessel that illegally shut down before declaration, and the start time, end time, and coordinates of the interruption point of the illegal shutdown were recorded to form a violation record. If This indicates that the location interruption was caused by a network signal problem, not by a user turning off the device. No violation will be marked; only the location interruption anomaly will be recorded.
[0030] Step 5: Determine if the status has not reached the anchor after submission. Based on the application time obtained in step 2 The starting time for uploading the vessel's location information after declaration is determined to be... Get the current system time The navigation management system allows for real-time monitoring of vessel lock passage status, which includes four types: declared but not yet at anchor, at anchor, passed through the lock, and declared cancelled. Status = Declared but not yet at anchor: Entering the trajectory monitoring process after declaration; Status = Anchor Reached: Real-time trajectory collection stops, and the entire trajectory is aggregated for the determination of illegal anchor reaching. Status = Passed through gate / Cancelled declaration: The identification process is terminated, and all abnormal records are archived in the database.
[0031] Step 6: Multi-dimensional judgment of continuous points after application right to The ship's trajectory within the time period is extracted, using the same continuous point multi-dimensional judgment algorithm as in step 3, with a custom time interval. ( and The method of value selection is the same, taking the empirical time range. (Maximum interval time for uploading ship positioning data), sequentially obtain any two consecutive points in the trajectory. Record the time corresponding to the location. , and latitude and longitude coordinates ( ), ( ).
[0032] Similarly, the judgment is made from the two dimensions of time and speed: calculation and The difference, if - > This indicates that the location service was interrupted after the application was submitted; calculate the driving speed. = distance / ( - ),like If the speed exceeds the legal range, it is considered an overspeed anomaly. If an anomaly occurs in either of the two dimensions, proceed to step 7 for further determination of whether it was caused by an unauthorized shutdown.
[0033] Step 7: Identification of Unauthorized Shutdowns After Reporting Obtain the power on / off time of the ship positioning terminal Based on the location interruption and overspeed anomalies identified in step 6, illegal shutdown is identified. This indicates that the terminal was powered off during the positioning interruption period, and the power-off behavior occurred simultaneously with the abnormal trajectory and speeding. Excluding objective factors such as network signal and equipment failure, the vessel is marked as a vessel that illegally powered off after reporting, and the violation information is recorded; if This indicates that the location interruption was caused by objective factors. Only abnormal information is recorded, and no violation is marked. At the same time, the interrupted trajectory is completed by using a trajectory fitting algorithm to ensure the completeness of subsequent analysis.
[0034] In this embodiment, the above steps reduce the consumption of ineffective computing power, improve the integrity of the behavior of shutting down positioning and speeding up anchoring after obtaining the ship's declaration, improve the accuracy of identifying the behavior of voluntarily shutting down to evade supervision during the declaration stage, and form a complete closed loop for screening violations of shutdown before and after the declaration, covering all violation windows from the ship's declaration to its arrival at anchor.
[0035] Step 8: Cyclic monitoring of gate passage status and anchor handling A cyclic monitoring mechanism is adopted, and the vessel's lock passage status is retrieved again every 5 minutes. If the vessel's status is still "declared but not yet anchored," steps 5 to 7 are repeated to continuously monitor the vessel's trajectory until the vessel's status changes. If the vessel's status changes to "anchored," the actual anchoring time M is immediately recorded, and real-time trajectory monitoring is stopped. The complete trajectory of the vessel from declaration to anchoring is summarized and analyzed, including trajectory integrity verification, anomaly point statistics, and speed analysis, to provide data support for subsequent identification of illegal anchoring. If the vessel's status changes to "passed the lock" or "declaration withdrawn," the identification process is terminated, and the recorded abnormal information is archived to the system database.
[0036] This embodiment reduces the cost of manual intervention and is suitable for real-time monitoring of massive numbers of vessels passing through the Three Gorges Dam.
[0037] Step 9: Determine the distance to consecutive points near the anchor line Extract two consecutive key positioning points of the vessel, one inside and one outside the anchor line. , and the corresponding time points , ,in To reach the last valid positioning point outside the anchor line, These two points are the first valid positioning points within the anchorage line, and are the core basis for determining whether a vessel has arrived at anchor prematurely in violation of regulations. First, the precise latitude and longitude range of the anchorage line is determined using the waterway geographic information system for accurate positioning. , The points were calculated using the Haversine formula. , The actual geographical distance between two points .
[0038] Instantaneous velocity .
[0039] like If the number of knots exceeds 20, it indicates that the vessel's trajectory changed when crossing the anchor line, suggesting that it may have arrived at the anchor ahead of schedule by falsifying the trajectory or using other equipment. Proceed to step 10 for further verification and count the number of valid positioning points within a 2km range outside the anchorage. If the number of points is less than 15, both conditions are met, and the vessel is officially marked as "arriving at the anchor ahead of schedule in violation of regulations," and the trajectory and speed evidence chain is preserved.
[0040] If the speed is less than or equal to 20 knots, it indicates that the ship's trajectory across the anchor line is normal, and the process proceeds to step 13 for post-anchor monitoring.
[0041] Step 10: Determining the density of positioning points outside the anchorage area and identifying violations at the anchorage. Custom distance range from outside the anchor zone to the anchor line (Value range is 1-3 kilometers). This range represents the normal waiting area for ships before anchoring. Ships should maintain continuous positioning within this area to ensure visibility for monitoring. Statistical analysis is performed through trajectory extraction. The number of valid positioning points for ships within the range, and a custom positioning point threshold. (according to Interval length and location terminal upload frequency (values range from 10 to 20). If in... Within the specified range, the number of ship positioning points is less than the defined value. This indicates that the vessel has a large number of missing locations in the waiting area. Combined with the abnormal speed in step 9, the vessel is determined to be a vessel that has anchored illegally. The violation type is marked as premature illegal anchoring, and detailed information such as the violation trajectory and speeding range is recorded.
[0042] This embodiment reduces misjudgments caused by single positioning drift and improves the accuracy of identifying illegal anchoring.
[0043] Step 11: Determine the time window for anchor positioning and monitoring. Based on the anchoring time M obtained in step 8, determine the starting time point for positioning upload after the ship anchors as M; obtain the current system time. The navigation management system determines the lock passage status of the vessel in real time. If the vessel status is still "anchored" (not yet entered the lock or completed the lock passage), it means that the vessel is in a waiting state and its behavior after anchoring needs to be monitored, proceeding to step 14. If the vessel status changes to "passed the lock", the identification process is terminated and all violation records are summarized and archived.
[0044] Step 12: Determine the interval of continuous positioning points after anchoring To M The ship's trajectory is extracted within a time period, using a continuous point time interval judgment algorithm, with a custom time interval. ( and The method of value selection is the same, taking the empirical time range. (Maximum interval time for uploading ship positioning data), sequentially obtain any two consecutive points in the trajectory. Record the time corresponding to the location. , .calculate and The difference, if This indicates that the vessel experienced a positioning interruption after anchoring. Proceed to step 15 to determine if it was due to an unauthorized shutdown. If the difference is less than or equal to... This indicates that the positioning is normal; continue monitoring until the vessel's status changes. Simultaneously, verify the distance and speed at continuous points after anchoring. If abnormal trajectories or speeding are detected, mark them as suspected violations and combine this information with a comprehensive assessment after shutting down the system.
[0045] Step 13: Identification of unauthorized shutdown after reaching the anchor Obtain the power on / off time of the ship positioning terminal Based on the positioning interruption situation determined in step 14, the system proceeds to identify unauthorized shutdown after anchoring. If - A value >0 indicates that the vessel's terminal was switched off during the positioning interruption period after anchoring, and the vessel attempted to circumvent monitoring during the lock-keeping period by switching off the terminal. This vessel will be marked as having illegally switched off the terminal after anchoring, and the time of the illegal shutdown and the coordinates of the lock-keeping area will be recorded. - ≤0 indicates that the positioning interruption was caused by objective factors. No violation will be marked, only abnormal information will be recorded, and continuous monitoring will continue until the vessel completes the passage through the lock.
[0046] In this embodiment, four types of violation records are summarized: shutdown before declaration, shutdown after declaration, illegal anchoring, and shutdown after anchoring. Standardized reports are automatically generated, which include vessel information, violation time period, trajectory coordinates, power on / off logs, and speeding data. These reports can be directly synchronized to the navigation integrity points system to automatically deduct integrity points and lower the priority of subsequent lock passage.
[0047] In this embodiment, continuous monitoring is performed on vessels that have reached anchor but have not yet passed through the lock. The adaptive positioning interval threshold identifies positioning interruptions during the waiting period and verifies whether the machine has been manually shut down by combining the power on / off logs. The system accurately identifies violations such as vessels turning off positioning after reaching anchor, moving arbitrarily without supervision, and disrupting the waiting order. This forms a full lifecycle violation identification system that covers all violation scenarios from vessel declaration to completion of lock passage.
[0048] Through the above steps, combined with complex algorithms such as multi-dimensional judgment of continuous points, trajectory fitting, and threshold adaptive optimization, violation identification is achieved throughout the entire process of a vessel's pre-declaration, post-declaration, and post-anchoring phases. This solves the problems of traditional technologies, such as reliance on positioning stability, low identification accuracy, and high false positive rates. This embodiment improves the automatic detection rate of violating vessels, reduces the time for verifying violations on a single vessel, increases the degree of automation, and enhances the fairness and efficiency of overall navigation scheduling at the hub.
[0049] Example 2 To further illustrate with reference to Embodiment 1, a system for identifying and recognizing illegal declarations of vessels passing through locks based on big data is characterized in that the method for identifying and recognizing illegal declarations of vessels passing through locks based on big data, as described in Embodiment 1, includes: The data acquisition module is used to acquire ship lock passage declaration information, ship positioning data, and positioning terminal power on / off log data; A data preprocessing module is used to preprocess the data; The pre-declaration violation identification module is used to determine the pre-declaration trajectory analysis time window based on the vessel's declared lock passage time, extract the vessel's first trajectory data within the pre-declaration time window and perform multi-dimensional judgment of continuous points, and combine the positioning terminal's power on / off log data to identify pre-declaration violation shutdown and mark vessels that violated pre-declaration regulations. The module for identifying violations of not arriving at anchor after declaration is used to extract the second trajectory data of the vessel from the time of declaration to the current time and make multi-dimensional judgments on continuous points when the vessel is in the state of having declared but not arriving at anchor. It combines the power-on / off log data of the positioning terminal to identify violations of power-off after declaration and marks the vessel that violated the declaration. The illegal anchoring identification module is used to record the actual anchoring time when the ship's status changes to anchored, extract the last valid positioning point outside the anchor line and the first valid positioning point inside the anchor line to calculate the instantaneous speed, and combine the number of valid positioning points within the preset distance range outside the anchor area to determine illegal anchoring and mark the ship that illegally anchored. The post-anchor violation identification module is used to extract the third trajectory data of the ship from the time of anchoring to the current time when the ship is in the state of anchoring but not passing through the lock. It performs continuous point time interval judgment and combines the positioning terminal power-on and power-off log data to identify post-anchor violation shutdown and mark the ship that violated the rules after anchoring. The output module is used to output all violation identification results.
[0050] This embodiment provides a method for identifying illegal declarations of vessels passing through locks based on big data. The working process, working details and technical effects can be found in Embodiment 1, and will not be repeated here.
[0051] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.
Claims
1. A method for identifying illegal declarations of vessels passing through locks based on big data, characterized in that, Includes the following steps: Obtain vessel lock passage declaration information, vessel positioning data, and positioning terminal power on / off log data, and preprocess the data; The time window for trajectory analysis before declaration is determined based on the time of the vessel's declaration to pass through the lock. The first trajectory data of the vessel within the time window before declaration is extracted. The first trajectory data is subjected to multi-dimensional judgment of continuous points. If it is determined that there is a positioning interruption and / or speed abnormality, the vessel is marked as having violated the declaration rules by combining the power on / off log data of the positioning terminal. Based on the vessel's lock passage status, when the vessel is in the state of having declared but not yet anchored, the second trajectory data of the vessel from the time of declaration to the current time is extracted. The second trajectory data is then subjected to multi-dimensional judgment of continuous points. If it is determined that there is a positioning interruption and / or speed abnormality, the vessel is identified as having violated the declaration by combining the positioning terminal power-on / off log data and marked as having violated the declaration. When the ship's status changes to "anchored", the actual anchoring time is recorded, the last valid positioning point outside the anchor line and the first valid positioning point inside the anchor line are extracted, and the instantaneous speed between the two points is calculated. If the instantaneous speed exceeds the preset speed threshold, the ship is judged to have illegally anchored by combining the number of valid positioning points within the preset distance range outside the anchorage area, and the ship is marked as having illegally anchored. When a vessel is in the state of having reached anchor but not passing through the lock, the third trajectory data of the vessel from the time of reaching anchor to the current time is extracted. The continuous point time interval of the third trajectory data is judged. If it is determined that there is a positioning interruption, the positioning terminal power-on / off log data is combined to identify the vessel that has violated the rules after reaching anchor and mark the vessel that violated the rules after reaching anchor. Output all violation identification results.
2. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, The data preprocessing includes: Segmented temporal neighborhood weighted smoothing denoising is employed: a sliding window is constructed using 10 consecutive temporal positioning points, and temporal weights are assigned based on the distance of each point from the current time. The weighted average of latitude, longitude, and speed within the window is calculated. If the deviation between the original latitude, longitude, and speed of the current point and the weighted average exceeds a preset threshold, the window weighted average is used to correct the noisy points. Tiered interpolation for filling time-series gaps: Differentiated processing based on three levels according to the duration of missing location data: missing duration ≤ normal location upload interval Standard linear interpolation is used. For missing durations less than 30 minutes, piecewise linear interpolation with channel boundary constraints is used, and the interpolation coordinates are limited to the channel waterway vector range; for missing durations ≥ 30 minutes, interpolation completion is not performed. Speed-Geographic Linkage Two-Layer Anomaly Screening: The first layer utilizes the waterway GIS vector boundary to directly eliminate drift points falling into land and tidal flat areas; the second layer uses a combination of speed and geographic information systems to filter out remaining points within the waterway. The principle is to screen out points with abnormal speeds; only points that simultaneously meet the criteria of drifting outside the channel and speed deviation of 3 times the standard deviation are judged as invalid anomalies and removed, while points with excessive speed within the channel are retained and marked as suspected anomalies.
3. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, Multi-dimensional judgment of continuous points, including: Time interval judgment: Obtain the time difference between any two consecutive positioning points in the trajectory. If the time difference exceeds the custom time interval threshold Z, it is determined that there is a positioning interruption abnormality. Speed determination: The instantaneous speed of the ship is calculated based on the distance and time difference between two consecutive positioning points. If the instantaneous speed exceeds the legal navigation speed range for this river section, it is determined that there is a speed anomaly. The time interval judgment result and the speed judgment result are jointly analyzed. If both the positioning interruption abnormality and the speed abnormality exist at the same time, it is marked as a highly suspected violation. Wherein, the time interval threshold The calculation formula is: Let N be the total number of vessels that declare passage through the lock within one year, and n be the total number of location data points uploaded by all N vessels within the time period X. For the i-th data point and the i-th Calculate the average μ of all valid time intervals between a given data point: ; Calculate the sample standard deviation σ: ; According to the 3σ principle, the theoretical maximum upload interval =μ+3σ.
4. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 3, characterized in that, The actual geographic distance S between two locations is calculated using the Haversine formula, which is: ; In the formula, R is the average radius of the Earth (a fixed constant is taken for inland river scenarios). , These are the latitudes of the first and second points, respectively. These are the longitudes of the first and second points, respectively.
5. The method for identifying illegal declarations of vessels passing through locks based on big data according to claim 3, characterized in that, The method of identifying unauthorized shutdowns by combining power-on / off log data from the positioning terminal specifically involves: Obtain the start time of the location interruption period and the time when the location terminal is turned off. If the shutdown time is earlier than or equal to the location interruption start time... If the location interruption time matches the shutdown time, it is determined to be a human-induced illegal shutdown and marked as a violation vessel; otherwise, it is determined to be a location interruption caused by objective factors and is not marked as a violation.
6. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, The specific criteria for determining violations at anchor include: Obtain the last valid positioning point outside the anchor line. And to the first effective positioning point within the anchor line Record the corresponding time , Calculated using the Haversine formula and The actual geographical distance between Then instantaneous velocity ; like If the speed is greater than the preset speed threshold V, then the preset distance range from outside the anchorage to the anchor line is extracted. The number of effective positioning points for vessels within the territory; If the number of valid positioning points is less than a preset threshold If so, it is determined to be a vessel that has arrived at anchor ahead of schedule in violation of regulations.
7. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, The time window for trajectory analysis before declaration is 2 to 6 hours before the declared gate passage time.
8. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, The system continuously acquires the vessel's lock passage status at preset time intervals. If the vessel's status is still "declared but not yet anchored," the system repeats the detection of abnormal trajectory after declaration and identification of illegal shutdown. If the status changes to "anchored," the system records the actual anchoring time and terminates real-time monitoring, then proceeds to the post-anchoring processing flow. If the status changes to "passed the gate" or "declaration withdrawn", the identification process will be terminated and the record will be archived.
9. The method for identifying illegal declarations of vessels passing through locks based on big data as described in claim 1, characterized in that, The ship positioning data includes AIS positioning data and / or BeiDou positioning data.
10. A system for identifying and recognizing illegal declarations of vessels passing through locks based on big data, characterized in that, The method for identifying illegal declarations of vessels passing through locks based on big data, as described in any one of claims 1-9, includes: The data acquisition module is used to acquire ship lock passage declaration information, ship positioning data, and positioning terminal power on / off log data; A data preprocessing module is used to preprocess the data; The pre-declaration violation identification module is used to determine the pre-declaration trajectory analysis time window based on the vessel's declared lock passage time, extract the vessel's first trajectory data within the pre-declaration time window and perform multi-dimensional judgment of continuous points, and combine the positioning terminal's power on / off log data to identify pre-declaration violation shutdown and mark vessels that violated pre-declaration regulations. The module for identifying violations of not arriving at anchor after declaration is used to extract the second trajectory data of the vessel from the time of declaration to the current time and make multi-dimensional judgments on continuous points when the vessel is in the state of having declared but not arriving at anchor. It combines the power-on / off log data of the positioning terminal to identify violations of power-off after declaration and marks the vessel that violated the declaration. The illegal anchoring identification module is used to record the actual anchoring time when the ship's status changes to anchored, extract the last valid positioning point outside the anchor line and the first valid positioning point inside the anchor line to calculate the instantaneous speed, and combine the number of valid positioning points within the preset distance range outside the anchor area to determine illegal anchoring and mark the ship that illegally anchored. The post-anchor violation identification module is used to extract the third trajectory data of the ship from the time of anchoring to the current time when the ship is in the state of anchoring but not passing through the lock. It performs continuous point time interval judgment and combines the positioning terminal power-on and power-off log data to identify post-anchor violation shutdown and mark the ship that violated the rules after anchoring. The output module is used to output all violation identification results.
