A ship lock passing behavior rule analysis method based on trajectory data driving

Abstract

The application discloses a ship lock behavior rule analysis method based on trajectory data driving and relates to the technical field of ship lock behavior analysis, which comprises the following steps: ship lock link analysis is carried out by using ship trajectory and ship stay area to obtain a ship lock link time consumption analysis result; according to ship trajectory and channel center line fusion multi-parameter constraint and grid division, feature analysis is carried out to obtain a ship stay feature analysis result; according to ship trajectory data and ship lock record, multi-parameter coupled ship following behavior analysis is carried out to obtain a ship following behavior analysis result; and the ship lock link time consumption analysis result, the ship stay feature analysis result and the ship following behavior analysis result are taken as the ship lock behavior rule analysis result. The application can quickly identify the area with abnormal motion characteristics in the ship lock process, and the ship following feature parameter time sequence analysis is helpful to efficiently and intuitively analyze abnormal conditions affecting the ship lock passing efficiency.

Description

Technical Field

[0001] This invention relates to the field of ship lock passage behavior analysis technology, and in particular to a method for analyzing ship lock passage behavior patterns based on trajectory data. Background Technology

[0002] With the widespread use of AIS (Automatic Identification System) and BeiDou Navigation Satellite System, an increasing number of ship trajectories are being recorded. Deep analysis and semantic mining of massive trajectory data can provide a better understanding of ship motion characteristics and reveal behavioral patterns, which is of significant research value for ship navigation safety management. Ship locks, as box-shaped navigation structures used to overcome concentrated water level differences and ensure smooth ship passage, not only change the way and speed of ships pass through waterways but also create a high density of ships upstream and downstream, increasing the complexity of ship navigation and safety management. Lock throughput capacity is a key indicator for measuring lock performance, and it is affected by various factors such as track layout, dimensional design, draft, and loading status. Mining the implicit ship motion patterns from trajectory data and identifying lock-passing behavior characteristics can provide a solid theoretical basis and scientific decision support for optimizing ship scheduling, improving lock throughput capacity, and ensuring safe ship passage through locks.

[0003] Currently, extensive research has been conducted on ship motion pattern analysis and ship behavior feature identification using methods such as data analysis, pattern recognition, and artificial intelligence. Many studies have achieved the extraction of key trajectory features and the effective detection of abnormal trajectories, contributing to a better understanding of ship motion patterns and improving ship behavior awareness. However, current research largely focuses on ships in offshore, near-shore, port, and ferry crossing waters, with relatively little analysis of ship trajectories within locks and their upstream and downstream areas. Existing methods are difficult to directly apply to the analysis of ship lock passage behavior patterns. This is specifically manifested in the following aspects:

[0004] (1) Ships generally need to go through multiple segmented scheduling when passing through locks. Due to the influence of the lock scheduling mechanism, ships will stop in areas such as anchorage, waiting area, berthing pier and lock chamber. The movement trajectory of ships is significantly different from the stopping behavior of ships when sailing normally in the channel and berthing in the port.

[0005] (2) The special nature of the lock structure changes the way ships navigate, making ships only able to navigate in narrow waters with limited width. At the same time, due to the constraints of specific navigation rules and human intervention, the ship's route, speed, course and berthing position will change significantly. These changes are significantly different from when the ship is sailing in a wide sea area or a specific channel.

[0006] (3) When navigating in narrow and restricted waterways, ships need to simultaneously consider the requirements of safe navigation and efficient lock passage. Therefore, ships will follow closely behind the ship in front of them while ensuring a safe distance, exhibiting typical following behavior characteristics.

[0007] In summary, ship lock passage behavior exhibits unique patterns. Analyzing and mining the motion trajectory data of passing ships to reveal these patterns remains a key unresolved issue. Therefore, a trajectory data-driven method for analyzing ship lock passage behavior patterns is urgently needed to address the shortcomings of existing technologies. Summary of the Invention

[0008] The purpose of this invention is to propose a method for analyzing the behavior of ships passing through locks based on trajectory data. Starting from two dimensions, namely microscopic ship motion and macroscopic trajectory aggregation, it reveals the inherent laws of ship passing through locks, analyzes the links that restrict the efficiency of passing through locks and abnormal ship behaviors in the operation of large locks, and provides theoretical support and practical guidance for optimizing lock scheduling strategies and improving lock throughput capacity.

[0009] To achieve the above objectives, this invention provides a method for analyzing the patterns of ship lock passage behavior based on trajectory data, comprising the following steps:

[0010] S1. Analyze the time consumption of ship lock passage by using ship trajectory and ship stopping area;

[0011] S2. Based on the ship trajectory and the channel centerline, multi-parameter constraints and grid division are fused to perform feature analysis and obtain the ship's dwelling feature analysis results;

[0012] S3. Based on the ship trajectory data and the ship lock passage records, perform multi-parameter coupled ship following behavior analysis to obtain the ship following behavior analysis results;

[0013] S4. Obtain the analysis results of the time consumption of the ship passing through the lock, the analysis results of the ship's dwelling characteristics, and the analysis results of the ship's following behavior as the analysis results of the ship's lock passage behavior pattern.

[0014] Optionally, the analysis of ship lock passage time can be obtained by analyzing ship trajectories and ship dwell areas, including:

[0015] S1-1. Obtain a set of ship trajectory points using the ship trajectory and the ship's stopping area;

[0016] S1-2. Obtain the set of ship crossing points based on the set of ship trajectory points;

[0017] S1-3. Obtain a set of candidate critical points based on the set of ship crossing points;

[0018] S1-4. Filter and select the candidate critical point set sequentially using the ship dwelling information to obtain the ship lock passage process node set.

[0019] S1-5. Based on the set of nodes in the ship lock passage process, perform ship lock passage analysis to obtain the time consumption pattern of the ship in the stopping area and the time consumption pattern of the ship lock passage process as the time consumption analysis result of the ship lock passage process.

[0020] The vessel dwelling information includes the vessel dwelling duration and the average speed of the vessel during the dwelling period.

[0021] Optionally, obtaining a set of ship trajectory points using the ship trajectory and the ship's stopping area includes:

[0022] S1-1-1. Perform buffer calculations based on the ship's dwelling area to obtain the ship's extended dwelling area;

[0023] S1-1-2. Perform spatial intersection calculation on the ship trajectory and the ship extended stopping area to obtain the set of trajectory points to be analyzed in the ship stopping area;

[0024] S1-1-3. Obtain the set of trajectory points to be analyzed in the ship's dwelling area as the ship's trajectory point set;

[0025] The formula for calculating the extended stay area of ​​the ship is as follows:

[0026]

[0027] The formula for calculating the set of ship trajectory points is as follows:

[0028]

[0029] In the formula, D' represents the extended dwelling area of ​​the ship. ST_Buffer ( ▪ ) represents buffered operations. D This is a ship anchorage area. dis Let P' be the set of trajectory points to be analyzed in the ship's dwell area, representing the buffer distance. ST_Intersects ( ▪ () represents the spatial intersection operation. P For ship trajectory.

[0030] Optionally, obtaining the set of ship crossing points based on the set of ship trajectory points includes:

[0031] S1-2-1. Obtain initial ship trajectory points using the set of ship trajectory points;

[0032] S1-2-2. Obtain the corresponding forward and backward centroid points based on the initial ship trajectory points as the centroid points of the ship trajectory.

[0033] S1-2-3. Obtain the set of ship crossing points based on the center of gravity of the ship's trajectory and the ship's stopping area;

[0034] The formula for calculating the forward centroid is as follows:

[0035]

[0036] The formula for calculating the backward centroid point is as follows:

[0037]

[0038] In the formula, D' represents the extended dwelling area of ​​the ship. ST_Buffer ( ▪ ) represents buffered operations. D This is a ship anchorage area. dis Let P' be the set of trajectory points to be analyzed in the ship's dwell area, representing the buffer distance. ST_Intersects ( ▪ () represents the spatial intersection operation. P For ship trajectory.

[0039] Optionally, obtaining the set of ship crossing points based on the centroid of the ship's trajectory and the area where the ship stops includes:

[0040] S1-2-3-1. Obtain the spatial relationship between the forward center of gravity and the ship's stationary area, and the spatial relationship between the rearward center of gravity and the ship's stationary area, based on the center of gravity of the ship's trajectory and the ship's stationary area.

[0041] S1-2-3-2: Determine whether the spatial relationship between the forward center of gravity and the ship's dwelling area changes from not including to including based on the first judgment condition. If yes, obtain the forward center of gravity as the ship's entry point and execute S1-2-3-3. Otherwise, return to S1-2-1.

[0042] S1-2-3-3: Determine whether the spatial relationship between the rearward center of gravity and the ship's dwelling area changes from inclusion to exclusion based on the second judgment condition. If yes, obtain the rearward center of gravity as the ship's departure point and execute S1-2-3-4; otherwise, return to S1-2-1.

[0043] S1-2-3-4. Use the ship's entry point and the ship's departure point as the set of ship crossing points;

[0044] The calculation formula for the first judgment condition is as follows:

[0045] The formula for calculating the second judgment condition is as follows:

[0046] In the formula, ST_Within ( ▪ ) represents the space containing the operation function. O ( a,k ) is the first a A forward center of gravity, O ( a-1,k ) is the first a-1 A forward center of gravity, d j For the first j The nth ship anchorage area, O'(b,k) is the nth ship anchorage area. b The backward centroid, O'(b,k+1) is the th backward centroid. b+1 A rearward center of gravity, True To be correct, False This is incorrect. a , b , j All are constants.

[0047] Optionally, feature analysis results of ship dwelling characteristics can be obtained by fusing multi-parameter constraints and mesh generation based on the ship trajectory and the channel centerline, including:

[0048] S2-1. Obtain the ship channel by segmenting the channel centerline;

[0049] S2-2. Obtain the ship trajectory to be processed by mapping the trajectory points to the channel and the ship trajectory.

[0050] S2-3. Assign attribute values ​​to the ship trajectory points of the ship trajectory to obtain the ship trajectory point attributes;

[0051] S2-4. Based on the ship trajectory to be processed and the attributes of the ship trajectory points, perform data cleaning and trajectory smoothing processing in sequence to obtain the target ship trajectory;

[0052] S2-5. Determine whether the target vessel trajectory is a single vessel trajectory. If yes, proceed to S2-6; otherwise, proceed directly to S2-7.

[0053] S2-6. Obtain ship dwelling characteristics by fusing multi-parameter constraints based on the target ship trajectory;

[0054] S2-7. Obtain the vessel's stopping characteristics by using network segmentation based on the target vessel's trajectory and the channel centerline;

[0055] S2-8. Based on the ship dwelling characteristics, perform feature analysis to obtain ship dwelling point distribution information and ship top current index information as the results of the ship dwelling characteristic analysis;

[0056] The ship trajectory includes several ship trajectory points.

[0057] Optionally, obtaining ship dwelling characteristics based on the target ship trajectory through multi-parameter constraint fusion includes:

[0058] S2-6-1. Set the ship running distance threshold, ship running time threshold, and ship running speed threshold as multi-parameter constraints.

[0059] S2-6-2. Based on the Stop / Move model, the multi-parameter constraints are used to detect the trajectory of the target ship and obtain a set of candidate stopping points;

[0060] S2-6-3. Perform spatial intersection operation between the candidate stopping point set and the ship stopping area to obtain the ship's normal stopping points and abnormal stopping points as the ship stopping features.

[0061] Optionally, obtaining the vessel's dwelling characteristics based on the target vessel's trajectory and the channel centerline using network segmentation includes:

[0062] S2-7-1. Obtain the basic unit for density calculation by dividing the grid according to the centerline of the waterway;

[0063] S2-7-2. Obtain the set of ship trajectory points contained in the grid cell based on the ship trajectory mapping points of the target ship trajectory and the grid cell of the density calculation;

[0064] S2-7-3. Obtain the density weight parameter of the grid cell based on the set of ship trajectory points contained in the grid cell;

[0065] S2-7-4. Calculate the dwell index of the grid cell using the density weight parameter of the grid cell;

[0066] S2-7-5. Obtain the density threshold based on the dwell index of the grid cell;

[0067] S2-7-6. Determine whether the dwell index of the grid cell is lower than the density threshold. If yes, then the grid cell is identified as a smooth flow point for ships; otherwise, the grid cell is identified as a congested point for ships.

[0068] S2-7-7. Based on the dwell index of the grid cells, obtain the dwell index thematic map and the traffic flow index thematic map using the hierarchical color scheme.

[0069] S2-7-8. The vessel dwelling characteristics are based on the vessel congestion points, the vessel bottleneck points, the dwelling index thematic map, and the smooth flow index thematic map.

[0070] The basic unit for density calculation is a number of equally spaced grid cells.

[0071] Optionally, the dwell index of the grid cell is calculated as follows:

[0072] In the formula, σ ( m ) represents the first m The dwell index of each grid cell d The length of the grid cell. w i For the first i The density weight parameter of each grid cell. k It is a constant.

[0073] Optionally, the ship following behavior analysis results can be obtained by performing multi-parameter coupling between ship trajectory data and ship lock passage records, including:

[0074] S3-1. Obtain the lock passage record for each lock session using the aforementioned lock passage record;

[0075] S3-2. Obtain the corresponding set of vessel trajectory lines as the vessel trajectory for each lock based on the lock passage records and the vessel trajectory data.

[0076] S3-3. Preprocess the vessel trajectory according to the lock sequence to obtain the corresponding vessel trajectory mapping point;

[0077] S3-4. Perform linear interpolation and mean resampling on the trajectory attributes of the ship trajectory mapping points to obtain the resampled time series and ship trajectories in the same lock;

[0078] S3-5. Based on the ship trajectory mapping points, perform time series analysis of following characteristic parameters to obtain the time series analysis results of following characteristic parameters of ships in the same lock.

[0079] S3-6. Based on the resampled time series and the trajectory of the vessel in the same lock, perform a time series analysis of the vessel following distance to obtain the time series analysis results of the vessel following distance;

[0080] S3-7. Based on the time series analysis results of the following characteristic parameters and the time series analysis results of the ship following distance, perform a coupling relationship analysis of the following characteristic parameters to obtain the following distance-velocity coupling relationship and the following distance-acceleration coupling relationship as the results of the following characteristic parameter coupling relationship analysis.

[0081] S3-8. Obtain the time series analysis results of the following characteristic parameters and the coupling relationship analysis results of the following characteristic parameters as the analysis results of the ship following behavior.

[0082] Compared with the closest existing technology, the present invention has the following advantages:

[0083] This invention employs a multi-parameter constrained trajectory dwell point extraction method, which can quickly identify areas of abnormal motion characteristics during ship passage through locks, providing a more targeted basis for lock scheduling and management. Through dwell index analysis and congestion index analysis, this invention can analyze the spatiotemporal aggregation of ship trajectories, intuitively displaying the operational status of the lock and surrounding waterways. The time-series analysis of ship following characteristic parameters used in this invention, by analyzing the changing trends of characteristic parameters such as position, speed, acceleration, and following distance of ships in the same lock cycle, helps to efficiently and intuitively analyze abnormal situations affecting lock throughput efficiency, such as delayed ship starts, slow sailing speeds, and disordered and loose entry and exit from the lock. Combining coupling relationship curves and data density distribution, this invention can intuitively display the distribution of abnormal data in ship following behavior during lock passage, providing important data support for improving lock throughput capacity. Starting from both microscopic ship motion and macroscopic trajectory aggregation dimensions, this invention reveals the inherent laws of ship passage behavior, analyzes the links that restrict passage efficiency and abnormal ship behaviors in the operation of large locks, and provides theoretical support and practical guidance for optimizing lock scheduling strategies and improving lock throughput capacity. Attached Figure Description

[0084] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0085] Figure 1 This is a flowchart illustrating a method for analyzing ship lock passage behavior patterns based on trajectory data, according to an embodiment of the present invention.

[0086] Figure 2 The flowchart of the stopping area crossing recognition algorithm proposed in the embodiments of the present invention is as follows: (a) is a flowchart for selecting a set of ship trajectory points based on the ship stopping area; (b) is a flowchart for recognizing a set of crossing points based on the center point; and (c) is a flowchart for merging candidate crossing points based on multi-condition judgment.

[0087] Figure 3 This is a schematic diagram of the ship dwelling area time consumption analysis results proposed in an embodiment of the present invention;

[0088] Figure 4 This is a schematic diagram showing the time consumption analysis results of each stage of the gate passage proposed in the embodiments of the present invention;

[0089] Figure 5 This is a flowchart of the method for extracting the dwelling features of ships passing through locks proposed in an embodiment of the present invention;

[0090] Figure 6 This is a schematic diagram of the distribution of ship stopping points proposed in an embodiment of the present invention, wherein (a) shows the spatiotemporal distribution of stopping points, (b) shows the berthing situation at the upstream anchorage, (c) shows the berthing situation at the downstream waiting area, and (d) shows the berthing situation at the downstream anchorage.

[0091] Figure 7 The results of extracting trajectory dwell features using the LINE-STING algorithm proposed in this embodiment of the invention are as follows: (a) is a dwell index thematic map with the number of trajectory points as the weight, (b) is a congestion index thematic map with the average speed (including stationary points) as the weight, and (c) is a congestion index thematic map with the average speed (excluding stationary points) as the weight.

[0092] Figure 8 This is a flowchart of the car-following behavior analysis algorithm proposed in an embodiment of the present invention;

[0093] Figure 9 This is a time-series analysis diagram of the following characteristic parameters of ships in the same lock according to an embodiment of the present invention, wherein (a) is the change of ship position, (b) is the change of ship speed, (c) is the change of ship acceleration, and (d) is the change of ship following distance.

[0094] Figure 10 The figure shows the analysis results of the coupling relationship of the following characteristic parameters proposed in the embodiment of the present invention, where (a) is the coupling relationship of following distance and velocity, and (b) is the coupling relationship of following distance and acceleration. Detailed Implementation

[0095] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0096] The terminology used in the embodiments section of this invention is for the purpose of explaining specific embodiments of the invention only, and is not intended to limit the invention.

[0097] Currently, a large amount of research has been conducted on ship motion law analysis and ship behavior characteristic identification using methods such as data analysis, pattern recognition, and artificial intelligence.

[0098] Stop points, which divide continuous trajectories into segments with similar shapes, can reflect the purpose and intent of ship activities and have become a hot topic in the study of ship motion patterns. Currently, stop point analysis algorithms can be broadly categorized into three types: motion feature-based algorithms, geometric feature-based methods, and density clustering-based methods. The Stop / Move model is a classic method for extracting stop points based on trajectory motion features. It abstracts the trajectory into a sequence of Stop / Move objects and analyzes trajectory stop points based on motion states, but it neglects the influence of velocity parameters on trajectory stop points. Numerous studies have extended the Stop / Move model by fusing motion parameters such as distance, time, speed, and topological relationships, proposing stop point algorithms adaptable to different scenarios. Furthermore, geometric feature-based methods are widely used in scenarios where motion trajectories have obvious spatial distribution patterns. These methods are often combined with motion feature-based algorithms. For example, analyzing the trajectory characteristics of individual and group refueling behaviors from the perspectives of trajectory motion features and geometric patterns can establish a stop behavior detection model applicable to crowdsourced vehicle trajectories. Based on large-scale ship trajectory data, the alpha-shape algorithm was used to extract channel boundaries. Density-based clustering methods, on the other hand, start from the spatiotemporal distribution characteristics of trajectories. They discover spatiotemporally concentrated dwelling areas by clustering dense areas of trajectory points. Common clustering methods include the DBSCAN algorithm, minimum covering circle, dwelling index analysis, and kernel density estimation. These algorithms have achieved significant results in identifying trajectory dwelling points, providing a data foundation for subsequent ship behavior feature identification. However, given the diversity of practical application scenarios and the complexity of trajectory data characteristics, especially the uniqueness of ship lock passage trajectories, the universality of existing algorithms is limited to some extent, making it difficult to fully adapt to the complex and ever-changing needs of trajectory analysis.

[0099] Currently, methods for detecting anomalies in ship trajectories include those based on statistical analysis, predictive models, and machine learning. For example, kernel density estimation is used to statistically analyze and mine the behavioral patterns of inland river ferries, leading to the establishment of an algorithm for detecting ferry anomalies based on position and speed anomalies. A semantic cognitive computing model for ship behavior is used to model and represent the behavior of ships with different motion characteristics and spatial topological features in port waters by fusing spatiotemporal trajectory data and navigation environment information. Based on the extraction of trajectory semantic information, a ship activity knowledge graph is constructed based on the SAM ship activity ontology model, providing a relatively complete representation of ship activities at sea, from behavior to navigation process. In summary, ship trajectory anomaly detection exhibits a diversified technological development trend and is gradually moving towards intelligentization.

[0100] The aforementioned research can extract key trajectory features and effectively detect abnormal trajectories, contributing to a better understanding of ship motion patterns and improving ship behavior cognition. However, current research largely focuses on ships in waterways such as the sea, nearshore, ports, and ferry crossings, with relatively little analysis of ship trajectories within locks and their upstream and downstream specific waterways. Existing methods are difficult to directly apply to the analysis of ship lock passage behavior patterns. Furthermore, compared to ships in sea, nearshore, inland waterway, and port environments, ships passing through locks exhibit unique behavioral patterns. Although numerous studies have focused on ship behavior analysis in various waterways, no directly relevant research has yet been found to analyze ship lock passage behavior patterns from massive trajectory data.

[0101] Therefore, this invention focuses on exploring the patterns of ship lock passage behavior. Based on an in-depth analysis of the motion characteristics and navigation patterns of ships passing through locks, it proposes a trajectory data-driven method for analyzing ship lock passage behavior patterns. First, a method for identifying stopping areas that takes into account center of gravity drift is proposed, enabling accurate identification and time-consuming analysis of each stage of ship lock passage. Second, a stopping feature extraction method integrating multi-feature constraints and grid partitioning is proposed, analyzing the trajectory stopping and spatiotemporal distribution characteristics of ships passing through locks from two dimensions: individual stopping behavior and group aggregation phenomena. Then, a multi-parameter coupled method for analyzing ship following behavior is proposed, analyzing the temporal variation patterns and coupling relationships between various following feature parameters.

[0102] Changzhou Ship Lock is located on the Xunjiang section of the Xijiang River, a major tributary of the Xijiang River. It is the last cascade ship lock on the Xijiang shipping route and is known as the "throat" of the Xijiang waterway. Changzhou Ship Lock has four bidirectional single-stage ship locks. There are two types of anchorages, one for dry season and one for flood season, both upstream and downstream. To facilitate scheduling, waiting areas are set up near the upstream and downstream gate areas, and berthing piers are set up on both sides of the approach channel.

[0103] This embodiment analyzes ship passage records and ship movement trajectories from the Changzhou Ship Lock in June 2021, as shown in Table 1. The ship movement trajectories were collected by a shipborne BeiDou terminal at a frequency of 30 seconds. Compared to AIS data, the data provided by the BeiDou terminal has a higher collection frequency and higher positional accuracy.

[0104] Table 1 Main Data Content

[0105] Data preprocessing steps: First, the ship trajectory point data is converted into trajectory line data. Then, shorter trajectories are discarded based on the expected minimum duration of the trajectory. Simultaneously, the trajectory is segmented into multiple independent trajectories by utilizing the time gaps between consecutive trajectory points. Trajectory motion parameters include velocity, direction, and turning angle, among which velocity, direction, and their changes play a crucial role in dwell extraction and semantic recognition. To facilitate trajectory analysis, auxiliary information such as velocity, acceleration, direction, direction difference, time increment, and distance increment are added to each trajectory based on the numerical changes of consecutive trajectory points.

[0106] like Figure 1 As shown, this embodiment of the invention provides a method for analyzing the behavior patterns of ships passing through locks based on trajectory data, including the following steps:

[0107] S1. Analyze the ship's lock passage process using the ship's trajectory and the ship's stopping area to obtain the time consumption analysis results of the ship's lock passage process. That is, use the stopping area crossing identification method that takes into account the center of gravity drift to analyze the ship's lock passage process and obtain the time consumption pattern of the ship in each stopping area and the time consumption pattern of each stage of the ship's lock passage.

[0108] Ship passage through locks typically involves five main stages: from anchorage to the waiting area, from the waiting area to the berthing pier, from the berthing pier to the lock chamber, the lock operation phase, and the departure phase from the lock chamber. Accurately identifying these passage stages is a prerequisite for discovering the time patterns of each stage from massive amounts of ship trajectory data.

[0109] The process of identifying ship passage through locks essentially involves identifying the ship's crossing behavior at key lock nodes, dividing the ship's trajectory into multiple independent segments using the crossing points as boundaries. Key lock nodes refer to specific stopping areas such as waiting areas, berthing piers, and lock chambers; these are relatively fixed spatial ranges where ships are required to follow certain navigation rules. To effectively identify these lock passage points, geofencing technology can be used to determine the ship's entry or exit point and crossing time. Ray methods, convex hull polygons, and triangulation methods are commonly used geofencing crossing identification algorithms. However, these algorithms have limitations when applied to identifying ship lock nodes. First, ships often exhibit complex berthing and unberthing behaviors at lock nodes, causing trajectory data to drift, especially at the boundaries of berthing areas, where trajectory points may jump repeatedly, affecting algorithm accuracy. Second, the berthing areas chosen by ships are not fixed; for example, berthing positions may differ between upstream and downstream journeys, and ships may use different waiting areas during dry and flood seasons. Furthermore, the direct connection between berthing areas and waterways increases the risk of misjudgment. Therefore, determining geofence crossing is quite complex and cannot be based solely on spatial relationships to identify crossing behavior.

[0110] To address the aforementioned issues, a method for identifying transit points in a stationary area that takes into account center of gravity drift is proposed. This method comprises three steps: selecting a set of ship trajectory points based on the ship's stationary area, identifying a set of transit points based on the center of gravity, and merging candidate transit points based on multi-condition judgment. Figure 2 As shown, specifically:

[0111] S1-1. Obtain a set of ship trajectory points using the ship's trajectory and the ship's stopping area, such as... Figure 2 (a) illustrates the steps for selecting a set of ship trajectory points based on the ship's dwelling area, where the ship trajectory is represented as... n A set of continuous trajectory points ordered by time P= { p 1 , p 2 ,⋯, p n The area where ships stay is represented as follows: m A set of polygons D= { d 1 , d 2 ,⋯, d m}

[0112] S1-1-1. Perform buffer calculations based on the ship's dwelling area to obtain the ship's extended dwelling area;

[0113] To reduce the impact of trajectory point drift at the boundary of the ship's dwelling area, while ensuring the integrity of the trajectory at the boundary as much as possible, a buffer calculation is performed on the ship's dwelling area to obtain the ship's extended dwelling area. The calculation method is as follows:

[0114] in, ST_Buffer ( ▪ ) represents buffered operations. dis This is for buffer distance.

[0115] S1-1-2. Perform spatial intersection calculation on the ship trajectory and the ship extended stopping area to obtain the set of trajectory points to be analyzed in the ship stopping area;

[0116] ship trajectory P Spatial intersection operations are performed with the ship's extended dwelling area D' to obtain the set of trajectory points to be analyzed for entering the ship's dwelling area. The calculation method is as follows:

[0117] in, ST_Intersects ( ▪ ) represents the spatial intersection operation.

[0118] S1-1-3. Obtain the set of trajectory points to be analyzed in the area where the ship stays as the ship trajectory point set.

[0119] S1-2. Obtain the centroid of the ship's trajectory based on the set of ship trajectory points. Figure 2 (b) demonstrates the steps for identifying the set of crossing points based on the center of gravity, and the change in the position of the center of the circle intuitively shows the process of judging the ship's entry point and departure point. The boundary points are judged by the method of convergence of the center of gravity of the trajectory, which effectively eliminates the trajectory drift error caused by the change in speed.

[0120] S1-2-1. Obtain initial ship trajectory points using the aforementioned set of ship trajectory points. , The first of the ship dwelling areas i One trajectory point to be analyzed;

[0121] S1-2-2. Obtain the corresponding forward and backward centroid points based on the initial ship trajectory points as the centroid points of the ship trajectory.

[0122] definition Before and after the point k The centroids of the trajectory points are respectively the forward centroids. O ( i,k and backward center of gravity Iterate through the set of trajectory points P' to be analyzed and calculate sequentially. Corresponding forward centroid O ( i,k and backward center of gravity The calculation method is as follows:

[0123]

[0124]

[0125] In the formula, ST_Centroid ( ▪ () is the geometric centroid calculation function, used to calculate the geometric centroid of the set of points on the input trajectory. k It is a constant;

[0126] Iterate through the set of trajectory points P' to be analyzed and calculate them sequentially. Corresponding forward centroid O ( i,k and backward center of gravity .

[0127] S1-2-3. Obtain the set of ship crossing points based on the center of gravity of the ship's trajectory and the ship's stopping area;

[0128] S1-2-3-1. Obtain the spatial relationship between the forward center of gravity and the ship's stationary area, and the spatial relationship between the rearward center of gravity and the ship's stationary area, based on the center of gravity of the ship's trajectory and the ship's stationary area.

[0129] S1-2-3-2: Based on the first judgment condition, determine whether the spatial relationship between the forward center of gravity and the ship's stopping area changes from non-inclusion to inclusion. If yes, then obtain the forward center of gravity as the ship's entry point and execute S1-2-3-3; otherwise, return to S1-2-1. Specifically:

[0130] The spatial relationship between the current centroid and the dwelling area is not included ( Without ) becomes containing ( Within When ), the trajectory points Define the ship's entry point as the forward center point; otherwise, return to S1-2-1 to traverse the set of trajectory points P' to be analyzed and calculate sequentially. Corresponding forward centroid O ( i,k ), and determine the spatial relationship between the forward center of gravity and the stopping area, where, a It is a constant;

[0131] S1-2-3-3: Based on the second judgment condition, determine whether the spatial relationship between the rearward center of gravity and the ship's stopping area changes from inclusion to exclusion. If yes, then obtain the rearward center of gravity as the ship's departure point and execute S1-2-3-4; otherwise, return to S1-2-1. Specifically:

[0132] When the spatial relationship between the backward center of gravity and the dwelling area is comprised of ( Within ) becomes not containing ( Without When ), the trajectory points Define the ship's departure point as the rearward center of gravity; otherwise, return to S1-2-1 to traverse the set of trajectory points P' to be analyzed and calculate sequentially. Corresponding backward center of gravity And determine the spatial relationship between the backward center of gravity and the dwelling area, where, b It is a constant;

[0133] The formula for calculating the first judgment condition is as follows:

[0134] The formula for calculating the second judgment condition is as follows:

[0135] In the formula, ST_Within ( ▪ () is a spatial containment function used to determine whether the center point is completely located within the dwelling area. O ( a,k ) is the first a A forward center of gravity, O ( a-1,k ) is the first a- One forward center of gravity, d j For the first j A ship mooring area For the first b A rearward center of gravity, For the first b+ One rearward center of gravity, True To be correct, False This is incorrect. a , b , j All are constants.

[0136] S1-2-3-4. Use the ship's entry point and the ship's departure point as the set of ship crossing points;

[0137] S1-3. Obtain a set of candidate critical points based on the set of ship crossing points. Figure 2 (c) demonstrates the steps for merging candidate crossing points based on multi-condition judgment, obtaining a set of candidate critical points by calculating the ship's entry point and departure point. ;

[0138] S1-4. Filter and select the candidate critical point set sequentially using ship dwelling information to obtain the ship lock passage process node set. The ship dwelling information includes the ship dwelling duration and the average speed during the dwelling segment. Specifically:

[0139] Because ships need to repeatedly adjust their attitude to accurately dock at the boundary of the stopping area, the trajectory points may jump back and forth at the boundary, resulting in redundant crossing points. To solve this problem, the set of candidate critical points is filtered and selected sequentially based on the ship's stopping time and the average speed of the stopping section, ultimately obtaining the accurate set of nodes for the lock passage process. .

[0140] S1-5. Based on the set of ship lock passage process nodes, perform ship lock passage link analysis to obtain the time consumption pattern of the ship in the stopping area and the time consumption pattern of the ship lock passage link as the time consumption analysis result of the ship lock passage link. Specifically:

[0141] Based on the set of nodes in the ship lock passage process, the analysis of the ship lock passage process reveals the time consumption patterns of ships in each stopping area and the time consumption patterns of each stage of the ship lock passage process, such as... Figure 3-4 As shown, vessels with higher gray values ​​are those going upstream, while those with lower gray values ​​are those going downstream.

[0142] Depend on Figure 3 It can be seen that the vessel paused in the designated areas before passing through the lock, and then entered free navigation after passing through the lock. The time spent in the pause areas, from longest to shortest, is as follows: berthing piers (A1, B1), lock chamber (C), waiting area (A2, B2), and anchorage (C1, C2), with the majority of the pause time concentrated at the berthing piers and lock chambers. The vessel spent the longest time at the berthing piers, with the largest variation in data, and the time spent upstream was significantly longer than downstream. In contrast, the overall time spent in the lock chambers did not vary much, with the downstream time slightly longer than the upstream time.

[0143] Depend on Figure 4 It can be seen that in the time distribution of upstream vessels, S5, S6, and S7 account for the majority, and the average time of these three stages is similar. Although the average time is similar, it is worth noting that there are significant differences in the channel mileage of these three stages. The channel mileage of S7 is much longer than that of S5 and S6. Therefore, S7 cannot be simply regarded as the main stage affecting the time of vessel passage through the lock. For downstream vessels, stages S2 and S3 account for the majority, and are generally shorter than those for upstream vessels. In addition, the time fluctuation of upstream vessels in stage S7 is significantly higher than that of downstream vessels in stage S1.

[0144] Based on the above results, it can be concluded that the period from a vessel entering the waiting area to passing through the lock is the main stage in terms of lock passage time, especially the stage from the berthing pier to the lock chamber, which is short in distance but long in time, making it the most time-consuming stage in lock passage. Considering the influence of channel mileage, the order of the impact of each stage on lock passage efficiency is as follows: from berthing pier to lock chamber, from waiting area to berthing pier, lock chamber operation, from downstream anchorage to waiting area, and from upstream anchorage to waiting area.

[0145] S2. Based on the ship trajectory and the channel centerline, multi-parameter constraints and grid division are fused to perform feature analysis to obtain the ship dwelling feature analysis results. That is, based on the ship trajectory and the ship channel, the dwelling feature extraction method of multi-parameter constraints and grid division is used to analyze the ship dwelling features and obtain the ship dwelling point distribution and ship dwelling index thematic map.

[0146] Ships passing through locks exhibit stable speed changes and relatively fixed trajectories. Therefore, methods for analyzing dwelling characteristics based on motion and geometric features are insufficient to comprehensively and accurately summarize the dwelling behavior of ships during lock passage. While density clustering methods perform well in extracting dwelling features from vehicles or ships moving freely in open environments, their advantages are not significant in extracting features clustered along their motion trajectories. Therefore, considering the data characteristics of ship lock passage trajectories, this embodiment proposes a dwelling feature extraction method that integrates multi-parameter constraints and grid partitioning. This method analyzes individual dwelling behavior and group clustering phenomena of passing ships by integrating multiple parameter constraints based on the ship's motion parameters and positional distribution, thereby extracting the ship dwelling features. The specific process is as follows: Figure 5 As shown, specifically:

[0147] S2-1. Obtain the vessel channel by dividing the channel centerline into segments, that is, dividing the channel centerline into five parts according to the lock passage process: upstream channel, upstream approach channel, lock chamber, downstream approach channel and downstream channel.

[0148] S2-2. Obtain the ship trajectory to be processed by mapping the trajectory points to the channel using the ship channel and the ship trajectory. That is, establish a matching relationship between the ship trajectory points and the ship channel based on the ship channel and the ship trajectory points, and map the ship trajectory points to the nearest channel segment. The mapping point is the vertical projection point of the ship trajectory point to the center line of the channel, thereby obtaining the ship trajectory to be processed. The ship trajectory consists of several ship trajectory points.

[0149] S2-3. Assign attributes to the ship trajectory points of the ship trajectory to obtain ship trajectory point attributes, that is, assign new attributes to the ship trajectory points: ① Segment number, that is, the segment identifier corresponding to the trajectory point; ② Mileage value, that is, the actual distance from the mapping point to the starting point of the segment; ③ Offset, that is, the straight-line distance between the trajectory point and its mapping point.

[0150] S2-4. Based on the ship trajectory to be processed and the attributes of the ship trajectory points, perform data cleaning and trajectory smoothing processing sequentially to obtain the target ship trajectory. Specifically:

[0151] Data cleaning: ① When the speed exceeds a preset speed threshold, a data cleaning method based on speed anomalies is used to eliminate spikes in the trajectory; ② When the offset exceeds a preset distance threshold, a data cleaning method based on distance anomalies is used to eliminate drift points in the trajectory.

[0152] Trajectory smoothing: A Kalman filter with a constant velocity model is used to filter and smooth the trajectory, and the missing positions in the trajectory are fitted.

[0153] S2-5. Determine whether the target vessel trajectory is a single vessel trajectory. If yes, proceed to S2-6; otherwise, proceed directly to S2-7.

[0154] S2-6. Based on the target ship trajectory, multi-parameter constraints are fused to obtain ship dwelling characteristics, i.e., individual dwelling feature extraction. For a single ship trajectory, based on the Stop / Move model, a multi-parameter constraint-based trajectory dwelling point extraction method is introduced. By setting ship travel distance, duration, and speed thresholds, stationary or low-speed trajectory points within a specified area are detected, resulting in a candidate dwelling point set. Then, the spatial intersection relationship between the dwelling points and the dwelling area is determined to identify abnormal and normal ship dwelling points. Specifically:

[0155] S2-6-1. Set the ship running distance threshold, ship running time threshold, and ship running speed threshold as multi-parameter constraints, namely distance constraint, time constraint, and speed constraint;

[0156] S2-6-2. Based on the Stop / Move model, the target ship trajectory is detected using the multi-parameter constraints to obtain the stopping center point, stopping duration and stopping point set, and then to obtain the candidate stopping point set.

[0157] S2-6-3. Perform spatial intersection operation between the candidate stopping point set and the ship stopping area to obtain the ship's normal stopping points and abnormal stopping points as the ship stopping features.

[0158] S2-7. Based on the target vessel trajectory and the channel centerline, the vessel's dwelling characteristics are obtained through network partitioning, i.e., group clustering feature extraction. For the trajectory point set, drawing on the STING (Statistical Information Grid) algorithm, a LINE-STING algorithm is proposed for trajectory dwelling feature extraction. This algorithm calculates statistical information on each linear grid cell and performs dwelling feature analysis based on this information. Specifically:

[0159] S2-7-1. Based on the channel centerline, the basic unit for density calculation is obtained by dividing the channel centerline into n equally spaced grid units as the basic unit for density calculation.

[0160] S2-7-2. Obtain the set of ship trajectory points contained in the grid cell based on the ship trajectory mapping points of the target ship trajectory and the grid cell of the density calculation. That is, determine the set of ship trajectory points contained in the grid cell based on the spatial relationship between the ship trajectory mapping points of the target ship trajectory and the grid cell.

[0161] S2-7-3. Obtain the density weight parameter of the grid cell based on the set of ship trajectory points contained in the grid cell, that is, select the number of trajectory points or the average speed as the density weight parameter of the grid cell based on the set of ship trajectory points contained in the grid cell.

[0162] S2-7-4. Calculate the dwell index of the grid cell using the density weight parameter of the grid cell. The calculation method is as follows:

[0163] In the formula, σ(m) Represented as the first m The dwell index of each grid cell d The length of the grid cell. w i For the first i The density weight parameter of each grid cell. k It is a constant, starting from the first m Take consecutive values ​​forward or backward from each grid cell. k Each grid cell participates in the stay index calculation, increasing the... k Values ​​can reduce data errors, but they may also reduce data accuracy.

[0164] S2-7-5. Obtain the density threshold based on the dwell index of the grid cell, that is, set the density threshold based on the dwell index of the grid cell. Different weight parameters correspond to different density thresholds.

[0165] S2-7-6. Determine whether the dwell index of the grid cell is lower than the density threshold. If yes, then the grid cell is identified as a smooth flow point for ships; otherwise, the grid cell is identified as a congested point for ships.

[0166] S2-7-7. Based on the dwell index of the grid unit, the dwell index thematic map and the smooth flow index thematic map are obtained using the hierarchical color method. That is, based on the dwell index of the grid unit, the dwell index thematic map and the smooth flow index thematic map are drawn using the hierarchical color method, which makes it easier to intuitively display the characteristics of ship dwelling.

[0167] S2-7-8. The vessel dwelling characteristics are based on the vessel smooth flow points, the vessel congestion points, the dwelling index thematic map, and the smooth flow index thematic map.

[0168] S2-8. Based on the aforementioned ship dwelling characteristics, perform feature analysis to obtain ship dwelling point distribution information and ship top current index information as the results of the ship dwelling characteristic analysis. That is, analyze the propagation dwelling characteristics to obtain the ship dwelling point distribution and ship dwelling index thematic maps, such as... Figure 6-7 As shown.

[0169] Figure 6 It visually illustrates the distribution of stopping points along the waterway and the duration of each stopping point. Figure 6 In (a)-(d), the triangle marks represent abnormal stopping points, and the circle marks represent normal stopping points. It can be seen that abnormal stopping points are mainly distributed in the surrounding areas of the upstream and downstream anchorages (① and ③) and the adjacent area of ​​the downstream waiting area (②). Most of these stopping points are short-term stopping points, with a stopping time of about half an hour. Stopping points with a stopping time of more than 1 hour are mainly concentrated near the downstream waiting area (②).

[0170] Figure 7 (a) is a thematic map of the dwell index calculated and drawn with the number of trajectory points as the weight parameter. The results show that trajectory points exhibit obvious clustering in dwell areas such as lock chambers, berthing piers, waiting areas, and anchorage areas. This result is highly consistent with the analysis results of the dwell point distribution. Figure 6 This verifies the effectiveness of the method. Specifically, the dwell index of ships at the downstream berthing pier is slightly higher than that at the lock chamber and upstream berthing pier, and the dwell index of the downstream waiting area is slightly higher than that of the upstream waiting area, which is consistent with the results of the dwell area time consumption analysis. Figure 4 The upstream travel time is generally longer than the downstream travel time. The high vessel dwell time and long dwell time at the downstream dock significantly affect the efficiency of vessel passage through the lock.

[0171] Figure 7 (b) and Figure 7 (c) is a thematic map of the congestion index calculated and plotted using average speed as the weighting parameter, where, Figure 7 (b) is the result calculated based on all trajectory points. Figure 7 (c) is a thematic map of the congestion index calculated and drawn based on trajectory points after filtering out stationary and low-speed points. The results show that the congestion index is highest at and around the lock, followed by the upstream anchorage. In the channel without designated berthing areas, there are multiple congestion phenomena of varying degrees between the downstream pilot channel and the downstream anchorage. In contrast, the ship speed changes in the upstream pilot channel are relatively stable. This indicates that the downstream channel is relatively long, the ship speed changes unevenly, and there are abnormal stopping behaviors. Strengthening the management of downstream ship navigation is beneficial to improving the efficiency of ship passage through the lock. Compared with the thematic map of the congestion index including stationary points ( Figure 7 (b) Congestion index thematic map excluding stationary points ( Figure 7(c) This more accurately reflects the movement patterns of ships when they are not waiting for dispatch. For ships in motion, the upstream and downstream channels are generally relatively smooth, with an average speed of over 8 km / h, but the congestion index upstream is higher than downstream. The congestion level of locks and pilot channels is relatively high, with an average speed below 5 km / h, and the average speed near the lock chamber is as low as about 3 km / h. According to the relevant provisions in the "General Design Specification for Locks" (JTJ 305-2001), when the lock is operating in both directions, the reference average speed for a single motorized ship entering the lock is 3.6 m / s, and the reference average speed for exiting the lock is 5.04 m / s. Therefore, in the process of ships entering and exiting the lock, under the premise of ensuring safety, increasing the average speed of ships can effectively shorten the time for ships to pass through the lock and further improve the ship throughput capacity.

[0172] S3. Based on the ship trajectory data and the ship lock passage records, perform multi-parameter coupled ship following behavior analysis to obtain the ship following behavior analysis results. That is, based on the ship trajectory data and the ship lock passage records, use the multi-parameter coupled ship following behavior analysis method to analyze the ship lock passage following behavior and obtain the time series analysis results of following characteristic parameters and the coupling relationship analysis results of following characteristic parameters.

[0173] Car-following models study the dynamic processes of changes in motion states between a leading vessel and its followers. Ship car-following models, in essence, study the changes in distance, speed, and acceleration between leading and following vessels under certain constraints, as well as their interaction mechanisms. It is worth noting that ship car-following models differ significantly from vehicle car-following models. Vehicle car-following models typically assume vehicles travel on one-way roads with overtaking restrictions. However, ships passing through locks require multiple scheduling adjustments. Although ship travel in the channel is also one-way with overtaking restrictions, the car-following order of ships may change at lock-passing nodes such as waiting areas, berthing piers, and lock chambers. For example... Figure 8 As shown, the scheduling and gear shifting alters the berthing and departure order of ships, making their following behavior more complex. Furthermore, compared to vehicles, ships have longer starting and braking distances, and typically maintain low speeds and relatively small accelerations during lock passage, resulting in a relatively constant following distance. This contrasts sharply with the characteristics of vehicles, which have shorter braking distances, more frequent acceleration fluctuations, and larger speed variations.

[0174] Therefore, this embodiment takes vessels passing through locks as the research object and proposes a multi-parameter coupled method for analyzing vessel following behavior. This method reveals the characteristics of vessel following behavior when passing through locks by studying the variation patterns and coupling relationships of vessel following characteristic parameters such as following distance, speed, and acceleration. The specific process is as follows: Figure 8 As shown, specifically:

[0175] S3-1. Obtain the lock passage records for each lock session using the aforementioned lock passage records. The lock passage time in the lock passage records is based on the gate closing time. The gate closing time is the same for ships in the same lock session. Based on this, the lock passage records for each lock session can be extracted.

[0176] S3-2. Obtain the corresponding set of ship trajectory lines as the ship trajectory for the lock according to the lock passage record and the ship trajectory data. That is, according to the list of ships in the same lock, retrieve the set of ship movement trajectory points from the ship trajectory data and generate the corresponding set of ship trajectory lines to obtain the ship trajectory for the lock.

[0177] S3-3. Preprocess the vessel trajectory according to the lock to obtain the corresponding vessel trajectory mapping point. The preprocessing includes steps such as trajectory point channel mapping, data cleaning and trajectory smoothing.

[0178] S3-4. Perform linear interpolation and mean resampling on the trajectory attributes of the ship trajectory mapping points to obtain a resampled time series and the ship trajectory of the same lock, i.e., a unified time series. Perform linear interpolation and mean resampling on the trajectory attributes (such as point velocity and acceleration) of the ship trajectory mapping points to ensure the consistency of data collection frequency.

[0179] S3-5. Based on the ship trajectory mapping points, perform time-series analysis of the following characteristic parameters to obtain the time-series analysis results of the following characteristic parameters for ships in the same lock session. This includes: ① Based on the spatial position of the trajectory mapping points, compare and analyze the position distribution and changing trends of ships in the same lock session. ② Combine the mapped point velocity and acceleration to compare and analyze the changing trends of the velocity and acceleration of ships in the same lock session.

[0180] S3-6. Based on the resampled time series and the trajectory of the vessels in the same lock, perform a time series analysis of the vessel following distance to obtain the vessel following distance time series analysis results, i.e., the vessel following distance time series analysis, including: for cases where the following order of vessels changes after passing through the lock node, retrieve the vessel position distribution information within the sampling interval based on the resampled time series; then, reorder the vessels according to their chronological order along the direction of navigation to correct the vessel following order; finally, calculate the straight-line distance between the preceding and following vessels, and compare and analyze the trend of the changing vessel following distance in the same lock.

[0181] S3-7. Based on the time series analysis results of the following characteristic parameters and the time series analysis results of the following distance, perform a coupling relationship analysis of the following characteristic parameters to obtain the following distance-speed coupling relationship and the following distance-acceleration coupling relationship as the results of the following characteristic parameter coupling relationship analysis, that is, the following characteristic parameter coupling relationship analysis includes: based on the results of the time series analysis of the following characteristic parameters and the time series analysis of the following distance of ships in the same lock, analyze the coupling relationship between the following characteristic parameters and the following distance of ships under a unified time series, and fit the coupling relationship function of following distance-speed under different headings and different routes;

[0182] S3-8. Obtain the time-series analysis results of the following characteristic parameters and the coupling relationship analysis results of the following characteristic parameters as the analysis results of the ship following behavior, such as... Figure 9-10 As shown.

[0183] Figure 9 (a)-(d) show the comparative analysis results of different following characteristic parameters under the same time series. The ship entry sequence is ①②③④⑤. After passing through the waiting area scheduling, berthing pier scheduling, and ship exit, the following sequence changes, and the exit sequence becomes ②③⑤①④. This result illustrates that the ship gearing affects the driving sequence. The waiting time at the berthing pier is the longest, greatly exceeding the normal waiting time. When entering the lock, the ship travels at a low speed, but the speed and acceleration fluctuate greatly. The channel distance from the berthing pier to the lock chamber is short, which means that the following ship has only just started moving when the preceding ship has already entered the lock chamber. Due to the untimely start of the ship in the waiting area, the entry process is not orderly and compact, resulting in a short berthing pier scheduling stage with a long time consumption, which reduces the efficiency of ship passage through the lock.

[0184] Compared to the waiting area scheduling, the speed and acceleration fluctuations during the exit scheduling phase are more stable. Especially when vessels enter the berthing pier from the waiting area, their speed drops rapidly, and through continuous adjustments to speed and position, they eventually stop smoothly at the designated berthing position. This process is time-consuming, and although the following distance of vessels generally remains within the normal range, some vessels' following distances significantly exceed the normal range, which greatly affects the efficiency of vessel passage through the lock. Therefore, strengthening the organization and scheduling of vessels in the waiting area and berthing pier to ensure more orderly and compact vessel passage through the lock is crucial for further maximizing the lock's throughput capacity.

[0185] Figure 10 This diagram illustrates the coupling relationship between ship following characteristics parameters and following distances under different headings and routes. The horizontal axis represents following distance, and the vertical axis represents speed or acceleration. In the diagram, smaller gray values ​​and more concentrated data distribution indicate stronger coupling between variables. The fitted coupling function shows that the relationship between following distance and speed follows a power function distribution with an exponent between 0 and 1. f ( x)= x a (0< a <1), the power exponent value varies depending on the direction of travel and the lock line. With the same line, the power exponent value for upstream vessels is smaller than that for downstream vessels, meaning that upstream vessels have a longer following distance or slower speed. With the same direction of travel, the power exponent values ​​for the first and second lines are greater than those for the third and fourth lines, meaning that vessels in the first and second lines have a shorter following distance or faster speed.

[0186] Figure 10 (a) shows that when the speed exceeds 3 m / s, the ship exhibits significant car-following characteristics. As the car-following distance increases, the speed also increases accordingly, but the rate of increase in speed gradually decreases and tends towards the normal operating speed. Based on the coupling relationship function and data density distribution, the reference speed of the ship can be estimated, providing a reference for tapping the potential of the lock's throughput capacity. Figure 10 (a) The situation of large following distance is common, especially those data with large following distance and low speed, which have a greater impact on the efficiency of ship passing through locks.

[0187] Figure 10 (b) illustrates the coupling relationship between following distance and acceleration, exhibiting a conical divergence distribution centered on the origin. With the same course, the upward conical shape is sharper than the downward conical shape, indicating more stable acceleration for upward-bound vessels. With the same direction of travel, the conical shape of the first and second lines is sharper than that of the third and fourth lines, indicating more stable acceleration for vessels in the first and second lines.

[0188] By analyzing the coupling relationship between speed, acceleration, and following distance, the distribution of abnormal data in the following behavior of ships passing through locks can be displayed intuitively. By fitting a power function, the coupling relationship between ship characteristic parameters can be analyzed, and the degree of coupling can be quantitatively evaluated by combining the data density distribution.

[0189] S4. Obtain the analysis results of the time consumption of the ship passing through the lock, the analysis results of the ship's dwelling characteristics, and the analysis results of the ship's following behavior as the analysis results of the ship's lock passage behavior pattern.

[0190] This embodiment analyzes vessel passage records and movement trajectories from the Changzhou Ship Lock in June 2021, with data sourced from the Guangxi Xijiang Ship Lock Joint Dispatch Center. Experiments and analysis at the Changzhou Ship Lock verified the effectiveness of the proposed method, and the following conclusions were drawn based on the experimental results:

[0191] (1) The time spent by ships passing through the lock consists mainly of the time spent from the waiting area to the exit of the lock chamber. Among these, the longest time is spent by ships staying at the berthing pier. Whether in the lock passage or the stopping area, the time spent going upstream is generally longer than that going downstream. According to the degree of impact on the efficiency of ship passage through the lock, the links of the Changzhou Lock are as follows: from the berthing pier to the lock chamber, from the waiting area to the berthing pier, the operation of the lock chamber, from the downstream anchorage to the waiting area, and from the upstream anchorage to the waiting area.

[0192] (2) The trajectory dwell point extraction method with multi-parameter constraints can quickly identify areas with abnormal motion characteristics during the passage of ships through the lock, providing a more targeted basis for lock scheduling and management. The LINE-STING trajectory dwell feature extraction method can analyze the spatiotemporal aggregation of ship trajectories through dwell index analysis and congestion index analysis, and intuitively show the operating status of the lock and surrounding waterways. Using the method proposed in this embodiment, it was found that: the original design capacity of the waiting anchorage of the Changzhou Lock is seriously insufficient, and it is recommended to strengthen the reconstruction and expansion of the anchorage area and long-term planning; there are abnormal dwelling behaviors in the downstream waterway, and navigation management needs to be further strengthened; although the relative congestion of the lock and the pilot channel is relatively high, the average speed of ships in the entry and exit of the lock is still lower than the design reference speed, which indicates that the throughput capacity of the Changzhou Lock still has some room for improvement.

[0193] (3) Time-series analysis of vessel following characteristics: By analyzing the changing trends of characteristic parameters such as position, speed, acceleration, and following distance of vessels in the same lock cycle, it is helpful to efficiently and intuitively analyze abnormal situations affecting the lock's throughput efficiency, such as untimely vessel start-up, slow sailing speed, and disorderly and loose entry and exit from the lock. The coupling relationship between vessel following distance and speed exhibits a power function distribution characteristic with a power exponent between 0 and 1, and the power exponent fluctuates depending on the sailing direction and lock route. Combining the coupling relationship curve and data density distribution, the abnormal data distribution in vessel following behavior can be intuitively displayed, which can provide important data support for improving the lock's throughput capacity.

[0194] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0195] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0196] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0197] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0198] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for analyzing the behavior patterns of ships passing through locks based on trajectory data, characterized in that, Specifically, the following steps are included: S1. Analyze the time consumption of ship lock passage by using ship trajectory and ship stopping area; S2. Based on the fusion of multi-parameter constraints and mesh generation between the ship trajectory and the channel centerline, feature analysis is performed to obtain the ship's dwelling feature analysis results, including: S2-1. Obtain the ship channel by segmenting the channel centerline; S2-2. Obtain the ship trajectory to be processed by mapping the trajectory points to the channel and the ship trajectory. S2-3. Assign attribute values ​​to the ship trajectory points of the ship trajectory to obtain the ship trajectory point attributes; S2-4. Based on the ship trajectory to be processed and the attributes of the ship trajectory points, perform data cleaning and trajectory smoothing processing in sequence to obtain the target ship trajectory; S2-5. Determine whether the target vessel trajectory is a single vessel trajectory. If yes, proceed to S2-6; otherwise, proceed directly to S2-7. S2-6. Obtain ship dwelling characteristics by fusing multi-parameter constraints based on the target ship trajectory; S2-7. Obtain the vessel's stopping characteristics by using network segmentation based on the target vessel's trajectory and the channel centerline; S2-8. Based on the ship dwelling characteristics, perform feature analysis to obtain ship dwelling point distribution information and ship top current index information as the results of the ship dwelling characteristic analysis; The ship trajectory includes several ship trajectory points; S3. Obtain the ship following behavior analysis results by performing multi-parameter coupling analysis of ship trajectory data and ship lock passage records, including: S3-1. Obtain the lock passage record for each lock session using the aforementioned lock passage record; S3-2. Obtain the corresponding set of vessel trajectory lines as the vessel trajectory for each lock based on the lock passage records and the vessel trajectory data. S3-3. Preprocess the vessel trajectory according to the lock sequence to obtain the corresponding vessel trajectory mapping point; S3-4. Perform linear interpolation and mean resampling on the trajectory attributes of the ship trajectory mapping points to obtain the resampled time series and ship trajectories in the same lock; S3-5. Based on the ship trajectory mapping points, perform time series analysis of following characteristic parameters to obtain the time series analysis results of following characteristic parameters of ships in the same lock. S3-6. Based on the resampled time series and the trajectory of the vessel in the same lock, perform a time series analysis of the vessel following distance to obtain the time series analysis results of the vessel following distance; S3-7. Based on the time series analysis results of the following characteristic parameters and the time series analysis results of the ship following distance, perform a coupling relationship analysis of the following characteristic parameters to obtain the following distance-velocity coupling relationship and the following distance-acceleration coupling relationship as the results of the following characteristic parameter coupling relationship analysis. S3-8. Obtain the time series analysis results of the following characteristic parameters and the coupling relationship analysis results of the following characteristic parameters as the analysis results of the ship following behavior; S4. Obtain the analysis results of the time consumption of the ship passing through the lock, the analysis results of the ship's dwelling characteristics, and the analysis results of the ship's following behavior as the analysis results of the ship's lock passage behavior pattern.

2. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 1, characterized in that, The analysis of ship lock passage time using ship trajectory and ship dwell area includes: S1-1. Obtain a set of ship trajectory points using the ship trajectory and the ship's stopping area; S1-2. Obtain the set of ship crossing points based on the set of ship trajectory points; S1-3. Obtain a set of candidate critical points based on the set of ship crossing points; S1-4. Filter and select the candidate critical point set sequentially using the ship dwelling information to obtain the ship lock passage process node set. S1-5. Based on the set of nodes in the ship lock passage process, perform ship lock passage analysis to obtain the time consumption pattern of the ship in the stopping area and the time consumption pattern of the ship lock passage process as the time consumption analysis result of the ship lock passage process. The vessel dwelling information includes the vessel dwelling duration and the average speed of the vessel during the dwelling period.

3. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 2, characterized in that, Obtaining a set of ship trajectory points using the ship's trajectory and the ship's stopping area includes: S1-1-1. Perform buffer calculations based on the ship's dwelling area to obtain the ship's extended dwelling area; S1-1-2. Perform spatial intersection calculation on the ship trajectory and the ship extended stopping area to obtain the set of trajectory points to be analyzed in the ship stopping area; S1-1-3. Obtain the set of trajectory points to be analyzed in the ship's dwelling area as the ship's trajectory point set; The formula for calculating the extended stay area of ​​the ship is as follows: The formula for calculating the set of ship trajectory points is as follows: In the formula, D' represents the extended dwelling area of ​​the ship. ST_Buffer ( ▪ ) represents buffered operations. D This is a ship anchorage area. dis Let P' be the set of trajectory points to be analyzed in the ship's dwell area, representing the buffer distance. ST_Intersects ( ▪ () represents the spatial intersection operation. P For ship trajectory.

4. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 2, characterized in that, The set of ship crossing points is obtained from the set of ship trajectory points, including: S1-2-1. Obtain initial ship trajectory points using the set of ship trajectory points; S1-2-2. Obtain the corresponding forward and backward centroid points based on the initial ship trajectory points as the centroid points of the ship trajectory. S1-2-3. Obtain the set of ship crossing points based on the center of gravity of the ship's trajectory and the ship's stopping area; The formula for calculating the forward centroid is as follows: The formula for calculating the backward centroid point is as follows: In the formula, O ( i,k () is the forward center of gravity. ST_Centroid ( ▪ ) is the geometric centroid operation function. The first of the ship dwelling areas i One trajectory point to be analyzed. k It is a constant. It is the rearward center of gravity.

5. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 4, characterized in that, The set of ship crossing points obtained based on the centroid of the ship's trajectory and the area where the ship stopped includes: S1-2-3-1. Obtain the spatial relationship between the forward center of gravity and the ship's stationary area, and the spatial relationship between the rearward center of gravity and the ship's stationary area, based on the center of gravity of the ship's trajectory and the ship's stationary area. S1-2-3-2: Determine whether the spatial relationship between the forward center of gravity and the ship's dwelling area changes from not including to including based on the first judgment condition. If yes, obtain the forward center of gravity as the ship's entry point and execute S1-2-3-3. Otherwise, return to S1-2-1. S1-2-3-3: Determine whether the spatial relationship between the rearward center of gravity and the ship's dwelling area changes from inclusion to exclusion based on the second judgment condition. If yes, obtain the rearward center of gravity as the ship's departure point and execute S1-2-3-4; otherwise, return to S1-2-1. S1-2-3-4. Use the ship's entry point and the ship's departure point as the set of ship crossing points; The calculation formula for the first judgment condition is as follows: The formula for calculating the second judgment condition is as follows: In the formula, ST_Within ( ▪ ) represents the space containing the operation function. O ( a,k ) is the first a A forward center of gravity, O ( a-1,k ) is the first a-1 A forward center of gravity, d j For the first j A ship mooring area For the first b A rearward center of gravity, For the first b+1 A rearward center of gravity, True To be correct, False This is incorrect. a , b , j All are constants.

6. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 1, characterized in that, The ship's dwelling characteristics are obtained by fusing multi-parameter constraints based on the target ship's trajectory, including: S2-6-1. Set the ship running distance threshold, ship running time threshold, and ship running speed threshold as multi-parameter constraints. S2-6-2. Based on the Stop / Move model, the multi-parameter constraints are used to detect the trajectory of the target ship and obtain a set of candidate stopping points; S2-6-3. Perform spatial intersection operation between the candidate stopping point set and the ship stopping area to obtain the ship's normal stopping points and abnormal stopping points as the ship stopping features.

7. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 1, characterized in that, The vessel's stopping characteristics are obtained by network segmentation based on the target vessel's trajectory and the channel centerline, including: S2-7-1. Obtain the basic unit for density calculation by dividing the grid according to the centerline of the waterway; S2-7-2. Obtain the set of ship trajectory points contained in the grid cell based on the ship trajectory mapping points of the target ship trajectory and the grid cell of the density calculation; S2-7-3. Obtain the density weight parameter of the grid cell based on the set of ship trajectory points contained in the grid cell; S2-7-4. Calculate the dwell index of the grid cell using the density weight parameter of the grid cell; S2-7-5. Obtain the density threshold based on the dwell index of the grid cell; S2-7-6. Determine whether the dwell index of the grid cell is lower than the density threshold. If yes, then the grid cell is identified as a smooth flow point for ships; otherwise, the grid cell is identified as a congested point for ships. S2-7-7. Based on the dwell index of the grid cells, obtain the dwell index thematic map and the traffic flow index thematic map using the hierarchical color scheme. S2-7-8. The vessel dwelling characteristics are based on the vessel congestion points, the vessel bottleneck points, the dwelling index thematic map, and the smooth flow index thematic map. The basic unit for density calculation is a number of equally spaced grid cells.

8. The method for analyzing ship lock passage behavior patterns based on trajectory data as described in claim 7, characterized in that, The formula for calculating the dwell index of the grid cell is as follows: In the formula, σ(m) Represented as the first m The dwell index of each grid cell d The length of the grid cell. w i For the first i The density weight parameter of each grid cell. k It is a constant.

Images