A method for analyzing GPS trajectory of suspicious refined oil transport vehicles

Abstract

The present invention discloses a method for analyzing the GPS trajectory of a suspicious vehicle transporting refined oil. The method for analyzing the GPS trajectory of a suspicious vehicle transporting refined oil includes the following steps: obtaining GPS trajectory data of the transport vehicles involved in the illegal transportation of refined oil and the GPS trajectory data of all tank trucks; cleaning the GPS trajectory data; extracting features from the cleaned GPS trajectory data; analyzing the potential patterns and laws of illegal transportation of refined oil to obtain corresponding pattern analysis results; and mining potential illegal vehicles transporting refined oil and suspicious places of stay of illegal vehicles transporting refined oil based on the pattern analysis results. The present invention provides a method for analyzing the GPS trajectory of a suspicious vehicle transporting refined oil, which can effectively guide business departments to further refine and carry out GPS trajectory analysis and judgment work of illegal vehicles transporting refined oil in a targeted manner, thereby improving the effectiveness of GPS trajectory analysis and judgment work of illegal vehicles transporting refined oil.

Description

Technical Field

[0001] The present invention relates to the technical field of refined oil transportation, and in particular to a method for analyzing the GPS trajectory of a suspicious refined oil transportation vehicle. Background Technology

[0002] Refined oil plays a vital role in modern society. However, the illegal transportation and storage of refined oil is rampant, which has brought great harm to social security and environmental protection. Traditional monitoring methods require manual monitoring of a large number of tank trucks, which has low monitoring efficiency, high monitoring costs, and is prone to causing personal injury. How to quickly, accurately and cost-effectively monitor the illegal transportation and storage of refined oil has become an urgent problem to be solved.

[0003] Illegal transport vehicles of refined oil have always been a transportation medium that is difficult to focus on in the field of illegal refined oil operations. Illegal transport vehicles of refined oil have certain characteristics and rules in terms of driving trajectory and driving time, but it is difficult to find clues of illegal transportation by relying on manual trajectory analysis, and it is difficult to effectively explore and curb illegal transportation of refined oil. SUMMARY OF THE INVENTION

[0004] In order to solve the technical problem that it is difficult to find the slightest trace of illegal transportation of refined oil vehicles by relying on manual trajectory analysis, and it is difficult to effectively explore and curb the illegal transportation of refined oil, the present invention provides a GPS trajectory analysis method for suspicious refined oil transportation vehicles. Through multiple mode analysis methods, it can effectively guide business departments to further refine and carry out GPS trajectory analysis and judgment work of illegal refined oil transportation vehicles in a targeted manner, thereby improving the effectiveness of GPS trajectory analysis and judgment work of illegal refined oil transportation vehicles.

[0005] The present invention provides a method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle, comprising the following steps:

[0006] Step 1: Obtain GPS track data of vehicles involved in illegal transportation of refined oil and GPS track data of all tank trucks from the data lake;

[0007] Step 2: Clean the GPS track data of the transport vehicle involved in the case and the GPS track data of all tanker trucks to generate new GPS track data of finished oil transport vehicles;

[0008] Step 3: Extract features from the GPS trajectory data of the refined oil transportation vehicle through pattern analysis, and construct a pattern analysis model;

[0009] Step 4: Analyze the potential patterns and rules of illegal transportation and storage of refined oil products, the hidden patterns of illegal transportation and storage of refined oil products, as well as the rules and trends of illegal transportation and storage of refined oil products in different time periods according to the pattern analysis model, and obtain the corresponding pattern analysis results;

[0010] Step 5: According to the pattern analysis results, dig out potential illegal refined oil transportation vehicles and suspicious staying locations of illegal transportation vehicles, form a list of suspicious transportation vehicles and a list of suspicious locations that need to be key verified, and repeat the digging process.

[0011] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, in the second step, cleaning the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks includes the following steps:

[0012] In the actual monitoring process, perform missing value processing on the missing values existing in the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks;

[0013] In the data collection process, identify and remove or replace the outliers existing in the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks by using box plots and Z-score methods;

[0014] Check and delete duplicate records existing in the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks;

[0015] Unify and convert the formats of the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks.

[0016] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, in the third step, the pattern analysis methods include cluster analysis, association analysis, and time series analysis.

[0017] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, the cluster analysis includes the following steps:

[0018] Extract the features of the GPS trajectory data of the refined oil transportation vehicles to obtain longitude and latitude data;

[0019] Define the distance between different GPS trajectories of refined oil transportation vehicles, calculate the distance matrix, each trajectory data contains longitude and latitude, speed, and direction, calculate the distances respectively, and then add them according to a certain ratio to obtain the final distance, and obtain the distance matrix between different GPS trajectories of refined oil transportation vehicles;

[0020] Perform K-means clustering on the distance matrix and obtain the final clustering analysis model using the optimal number of clusters.

[0021] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, the association analysis includes the following steps:

[0022] Preprocess and segment the GPS trajectory data of the refined oil transportation vehicles, and extract relevant features from the trajectory data as the input for association rule analysis.

[0023] Use the Apriori association rule mining algorithm to perform association rule analysis on the relevant features of the GPS trajectory data of the refined oil transportation vehicles, find the item sets and frequent item sets in the GPS trajectory data of the refined oil transportation vehicles, calculate the association degree and support degree between the item sets, and then screen out the rules with a certain degree of relevance according to the two association rule evaluation indicators of confidence and lift set in advance.

[0024] Interpret the mined association rules, understand the behaviors and patterns in the trajectory data, and compare and verify them with the real driving behaviors of the transportation vehicles.

[0025] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, the relevant features include trajectory length, speed, and direction.

[0026] In a preferred embodiment of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention, the time series analysis includes the following steps:

[0027] Clean and preprocess the time series trajectory data to form a standard data set containing timestamp and spatial point coordinate information. At the same time, perform corresponding operations on all the GPS trajectory data of the oil tank trucks to ensure the atomicity and timeliness of all the GPS trajectory data of the oil tank trucks.

[0028] Perform sequence pattern law analysis on the constructed trajectory time series data, and summarize and extract the common time series laws of the vehicles in different regions and at different times.

[0029] Establish a classification prediction model based on the regional multi-time series feature set, and at the same time use sample data to evaluate the fitting degree and prediction ability of the model. Further train and iterate the classification prediction performance of the model by enhancing historical positive and negative samples, and predict future trajectory data based on the historical time series trajectory data of suspicious transportation vehicles.

[0030] In a preferred embodiment of the method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle provided by the present invention, the corresponding operations on the GPS trajectory data of all tanker trucks to ensure the atomicity and time sequence of the GPS trajectory data of all tanker trucks include: time series alignment, outlier processing, trajectory compression, time series segmentation and trajectory segmentation of the GPS trajectory data of all tanker trucks.

[0031] In a preferred embodiment of the method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle provided by the present invention, the sequence pattern analysis of the constructed trajectory time series data includes: using a variety of time series analysis techniques to mine and analyze the time series regularity patterns, and summarizing and extracting the common time series regularities of vehicles at different times and in different regions.

[0032] In a preferred embodiment of the method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle provided by the present invention, the time series analysis technology includes moving average, exponential smoothing, seasonal decomposition, autoregressive model and exponential smoothing.

[0033] Compared with the prior art, the GPS trajectory analysis method for suspicious refined oil transportation vehicles provided by the present invention has the following beneficial effects:

[0034] 1. Through comprehensive and three-dimensional analysis of the GPS trajectory data of illegal refined oil transport vehicles, using multiple modes of analysis to analyze the GPS trajectory characteristics and transportation modes of illegal refined oil transport vehicles, outlining the driving behavior characteristics of illegal refined oil transport vehicles from multiple angles, and constructing an overall illegal refined oil transport vehicle trajectory risk portrait, it can effectively guide business departments to further refine and carry out the analysis and judgment of the GPS trajectory of illegal refined oil transport vehicles in a targeted manner, thereby improving the effectiveness of the analysis and judgment of the GPS trajectory of illegal refined oil transport vehicles.

[0035] Second, through cluster analysis, summarize and discover the potential patterns and rules of illegal transportation of refined oil, determine the trajectory of transportation vehicles, identify the rationality of vehicle transportation behavior, and promptly discover the inconsistencies in the transportation trajectory of transportation vehicles; through association analysis, explore the association relationship in the trajectory data of illegal transportation of refined oil vehicles, and discover the hidden patterns of illegal transportation of refined oil; through time series analysis, discover the rules and trends of illegal transportation of refined oil in different time periods, and identify the abnormal situation after the transportation behavior is associated with time. Brief Description of the Figures

[0036] ​To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings, where:

[0037] Figure 1 is the flowchart of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention;

[0038] Figure 2 is Figure 1 the logic diagram of the method for judging the GPS trajectory of the suspicious refined oil transportation vehicle shown; Detailed implementation manners

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0040] Please refer to Figure 1 and Figure 2 together, where Figure 1 is the flowchart of the method for judging the GPS trajectory of suspicious refined oil transportation vehicles provided by the present invention; Figure 2 is Figure 1 the logic diagram of the method for judging the GPS trajectory of the suspicious refined oil transportation vehicle shown;

[0041] The method for judging the GPS trajectory of the suspicious refined oil transportation vehicle includes the following steps:

[0042] Step 1: Obtain the GPS trajectory data of the involved transportation vehicles in the illegal refined oil transportation cases and the GPS trajectory data of all oil tank trucks from the data lake respectively;

[0043] Step 2: Clean the GPS trajectory data of the involved transportation vehicles and the GPS trajectory data of all oil tank trucks to generate new GPS trajectory data of refined oil transportation vehicles;

[0044] Further, cleaning the GPS trajectory data of the involved transportation vehicles and the GPS data of all oil tank trucks includes the following steps:

[0045] During the actual monitoring process, missing value processing is performed on the missing values in the GPS trajectory data of the involved transport vehicles and the GPS data of all oil tank trucks; the missing value processing specifically includes interpolation method, mean filling, and median filling; by processing the missing values, the integrity of the data is improved to ensure the accuracy of subsequent analysis;

[0046] During the data collection process, outliers in the GPS trajectory data of the involved transport vehicles and the GPS data of all oil tank trucks are identified using box plots and the Z-score method and either removed or replaced to further improve the reliability and accuracy of the data;

[0047] Check for and delete duplicate records in the GPS trajectory data of the involved transport vehicles and the GPS data of all oil tank trucks to ensure the independence and effectiveness of the data set;

[0048] Unify and convert the formats of the GPS trajectory data of the involved transport vehicles and the GPS data of all oil tank trucks for subsequent processing and analysis.

[0049] Step 3: Extract features from the GPS trajectory data of the refined oil transport vehicles through pattern analysis and construct a pattern analysis model;

[0050] Furthermore, the pattern analysis method includes cluster analysis, association analysis, and time series analysis;

[0051] Furthermore, the cluster analysis includes the following steps:

[0052] Extract features from the GPS trajectory data of the refined oil transport vehicles to obtain longitude and latitude data;

[0053] Define the distance between different GPS trajectories of refined oil transport vehicles, calculate the distance matrix. Each trajectory data includes longitude, latitude, speed, and direction, and calculate the distances respectively, and then add them according to a certain ratio to obtain the final distance, and get the distance matrix between different GPS trajectories of refined oil transport vehicles;

[0054] Perform K-means clustering on the distance matrix and use the optimal number of clusters to obtain the final cluster analysis model.

[0055] Specifically, the K-means clustering specifically includes: first, initialize the cluster centers using kmeans++, randomly select a trajectory as the first center, and then select the next center according to the following steps:

[0056] (1) For each data point, calculate its shortest distance to all the centers that have been selected

[0057] (2) Select a new center point according to the weights of these distances, as far as possible from the center points that have already been selected.

[0058] (3) Repeat steps (1) and (2) until k center points are selected.

[0059] (4) Subsequently, perform clustering according to the K-means++ algorithm, assign each data point to the cluster of the center point closest to it, then determine the optimal number of clusters based on the sum of squared errors and the within-cluster variance, and use the optimal number of clusters to obtain the final clustering model.

[0060] Further, the association analysis includes the following steps:

[0061] Preprocess and segment the GPS trajectory data of the refined oil transportation vehicles, and extract relevant features from the trajectory data as the input for association rule analysis.

[0062] Use the Apriori association rule mining algorithm to perform association rule analysis on the relevant features of the GPS trajectory data of the refined oil transportation vehicles, find the item sets and frequent item sets in the GPS trajectory data of the refined oil transportation vehicles, calculate the association degree and support degree between item sets, and then screen out the rules with a certain degree of association according to two association rule evaluation indicators of confidence and lift set in advance.

[0063] Interpret the mined association rules, understand the behaviors and patterns in the trajectory data and compare them with the real driving behaviors of the transportation vehicles to provide rule support for the mining of hidden patterns such as the transportation time rules and GPS online / offline rules of illegal refined oil transportation vehicles.

[0064] Specifically, the relevant features include trajectory length, speed, and direction.

[0065] Further, the time series analysis includes the following steps:

[0066] Clean and preprocess the time series trajectory data to form a standard data set containing timestamp and spatial point coordinate information, and perform corresponding operations on all the GPS trajectory data of the oil tankers to ensure the atomicity and timeliness of all the GPS trajectory data of the oil tankers.

[0067] Perform sequence pattern law analysis on the constructed trajectory time series data, and summarize and extract the common time series laws of vehicles in different regions at different times.

[0068] A classification prediction model is established based on the regional multivariate time series feature set, and sample data is used to evaluate the model's fit and prediction ability. The classification prediction performance of the iterative model is further trained by enhancing historical positive and negative samples. Future trajectory data is predicted based on the historical time series trajectory data of suspicious transport vehicles; dynamic warnings are given to vehicles with such transport characteristics using rolling predictions and dynamic model updates.

[0069] Specifically, the corresponding operations on all the GPS track data of the tanker trucks to ensure the atomicity and temporality of all the GPS track data of the tanker trucks include: time series alignment, outlier processing, track compression, time series segmentation and track segmentation of all the GPS track data of the tanker trucks.

[0070] Specifically, the sequence pattern analysis of the constructed trajectory time series data includes: using a variety of time series analysis techniques to mine and analyze the time series regularity patterns, and summarizing and extracting the common time series regularities of vehicles at different times and in different regions.

[0071] Specifically, the time series analysis techniques include moving average, exponential smoothing, seasonal decomposition, autoregressive model and exponential smoothing.

[0072] Step 4: Analyze the potential patterns and rules of illegal transportation and storage of refined oil products, the hidden patterns of illegal transportation and storage of refined oil products, and the rules and trends of illegal transportation and storage of refined oil products in different time periods according to the pattern analysis model, and obtain corresponding pattern analysis results;

[0073] Specifically, by aggregating information from the GPS trajectory data of the refined oil transport vehicle, the oil storage or oil loading and unloading base location data of the suspected illegal refined oil transport terminal is obtained; the association rules existing in the GPS trajectory data of the refined oil transport vehicle are analyzed according to the constructed association analysis model, and the relationship between different variables is analyzed. Through this method, the hidden pattern of illegal transportation and storage of refined oil is revealed; the time series analysis model is used to analyze the GPS trajectory data time of the refined oil transport vehicle, and the rules and trends of illegal transportation and storage of refined oil in different time periods are discovered. Based on the identified time series pattern of illegal refined oil transportation, an important basis is provided for subsequent illegal refined oil transportation warning and decision-making.

[0074] Step 5: Based on the analysis results of the model, potential illegal transport vehicles of refined oil and suspicious places where illegal transport vehicles stay are discovered, a list of suspicious transport vehicles and suspicious places that need to be checked is formed, and the discovery process is repeated.

[0075] Specifically, the analysis of illegal transportation tracks of refined oil products includes:

[0076] 1. Discover the potential patterns and rules of illegal transportation and storage of refined oil through cluster analysis method:

[0077] Cluster the driving routes of vehicles with the distance between vehicle trajectories as the dimension: With the distance between vehicle trajectories as the dimension, define the distance between different trajectories according to the longitude and latitude data of the transport vehicle trajectories, calculate the distance matrix. Each trajectory data includes longitude, latitude, speed, and direction. Calculate the distances respectively and add them according to a certain ratio to obtain the final distance. After obtaining the distance matrix between different trajectories, perform K-means clustering to form the identification of multiple central points with the distance weight as the main judgment basis, and output the location information of the central points where the vehicle trajectories gather.

[0078] 2. Discover the hidden patterns of illegal transportation and storage of refined oil through association analysis method:

[0079] For the transportation trajectory information of oil tanker vehicles, use the association analysis method to mine the association rules existing in the data and analyze the relationships between different variables. Discover the transportation time rules and GPS offline / online rules of illegal refined oil transportation vehicles, and reveal the hidden patterns of illegal transportation and storage of refined oil.

[0080] 3. Discover the rules and trends of illegal transportation and storage of refined oil in different time periods through time series analysis method:

[0081] Conduct a time-dimensional sequence analysis on the trajectories of illegal refined oil transportation vehicles, make pre-judgments on different vehicle trajectories and the same vehicle trajectory within the same time interval. For the trajectory data that meets the specific time interval, identify and determine it as the transportation behavior of illegal refined oil transportation vehicles, give early warnings to the vehicles with such transportation characteristics, and form an important basis for subsequent decision-making.

[0082] The method for judging the GPS trajectories of suspicious refined oil transportation vehicles provided by the present invention has the following beneficial effects:

[0083] First, through a comprehensive and three-dimensional analysis of the GPS trajectory data of illegal refined oil transportation vehicles, use various mode analysis methods to analyze the GPS trajectory characteristics and transportation modes of illegal refined oil transportation vehicles, outline the driving behavior characteristics of illegal refined oil transportation vehicles from multiple angles, construct an overall risk portrait of the trajectories of illegal refined oil transportation vehicles, and can effectively guide the business department to further carry out the analysis and judgment work of the GPS trajectories of illegal refined oil transportation vehicles in a more refined and targeted manner, improving the effectiveness of the analysis and judgment work of the GPS trajectories of illegal refined oil transportation vehicles.

[0084] Second, through cluster analysis, summarize and discover potential patterns and rules of illegal refined oil transportation, determine the trajectories of transportation vehicles, identify the rationality of vehicle transportation behaviors, and timely detect inconsistent phenomena in the transportation trajectories of transportation vehicles; through association analysis, mine the association relationships existing in the trajectory data of illegal refined oil transportation vehicles and discover hidden patterns of illegal refined oil transportation; through time series analysis, discover the rules and trends of illegal refined oil transportation and storage in different time periods, and identify abnormal situations after the transportation behaviors are associated with time.

[0085] The above are only embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A method for analyzing GPS tracks of suspicious refined oil transport vehicles, characterized in that: The following steps are involved: Step 1: Obtain the GPS track data of the transport vehicles involved in the illegal transportation of refined oil and the GPS track data of all tank trucks from the data lake; Step 2: Clean the GPS track data of the transport vehicle involved in the case and the GPS track data of all the tanker trucks to generate new GPS track data of the finished oil transport vehicle; Among them, cleaning the GPS track data of the transport vehicle involved in the case and the GPS data of all tanker trucks includes the following steps: In the actual monitoring process, the missing values ​​in the GPS trajectory data of the transport vehicles involved in the case and the GPS data of all tank trucks are processed; During the data collection process, the outliers in the GPS trajectory data of the transport vehicles involved in the case and the GPS data of all tanker trucks were identified using box plots and Z-score methods and then removed or replaced; Check and delete the duplicate records in the GPS track data of the transport vehicle involved in the case and the GPS data of all tank trucks; Unify and convert the format of the GPS track data of the transport vehicles involved in the case and the GPS data of all tanker trucks; Step 3: extracting features from the GPS trajectory data of the refined oil transport vehicle through pattern analysis, and constructing a pattern analysis model; Step 4: Analyze the potential patterns and rules of illegal transportation and storage of refined oil products, the hidden patterns of illegal transportation and storage of refined oil products, and the rules and trends of illegal transportation and storage of refined oil products in different time periods according to the pattern analysis model, and obtain corresponding pattern analysis results; Step 5: Based on the results of the pattern analysis, potential illegal refined oil transport vehicles and suspicious stopover locations of illegal transport vehicles are mined to form a list of suspicious transport vehicles and suspicious locations that need to be checked, and repeat the mining process.

2. The method for analyzing GPS tracks of suspicious refined oil transport vehicles according to claim 1 is characterized in that: In the step three, the pattern analysis methods include cluster analysis, association analysis and time series analysis.

3. The method for analyzing GPS tracks of suspicious refined oil transport vehicles according to claim 2 is characterized in that: The cluster analysis includes the following steps: Extract features from the GPS trajectory data of the refined oil transportation vehicle to obtain longitude and latitude data; Define the distance between different GPS tracks of refined oil transport vehicles and calculate the distance matrix. Each track data contains longitude and latitude, speed, and direction. The distance is calculated separately and then added according to a certain ratio to obtain the final distance. The distance matrix between the GPS tracks of different refined oil transport vehicles is obtained. K-means clustering is performed on the distance matrix, and the optimal number of clusters is used to obtain the final cluster analysis model.

4. The method for analyzing GPS tracks of suspicious refined oil transport vehicles according to claim 2 is characterized in that: The association analysis comprises the following steps: Preprocessing and segmenting the GPS trajectory data of the refined oil transportation vehicle, and extracting relevant features from the trajectory data as input for association rule analysis; By using the Apriori association rule mining algorithm, the relevant features of the GPS trajectory data of the refined oil transportation vehicle are analyzed for association rules, the item sets and frequent item sets in the GPS trajectory data of the refined oil transportation vehicle are found, and the association and support between the item sets are calculated. Then, the rules with certain association are screened out according to the two association rule evaluation indicators of confidence and lift set in advance; Interpret the mined association rules to understand behaviors and patterns in trajectory data and compare and verify with the actual driving behavior of transport vehicles.

5. The method for analyzing GPS trajectory of suspicious refined oil transport vehicles according to claim 4 is characterized in that: The relevant features include trajectory length, speed and direction.

6. The method for analyzing GPS tracks of suspicious refined oil transport vehicles according to claim 2 is characterized in that: The time series analysis includes the following steps: Clean and preprocess the time series trajectory data to form a standard data set containing timestamp and spatial point coordinate information, and perform corresponding operations on all the GPS trajectory data of the tanker trucks to ensure the atomicity and temporal sequence of all the GPS trajectory data of the tanker trucks; Analyze the sequence pattern of the constructed trajectory time series data, and summarize and extract the common time series rules of vehicles at different times and in different regions; A classification prediction model is established based on the regional multivariate time series feature set, and sample data is used to evaluate the model's fit and prediction ability. The classification prediction performance of the iterative model is further trained by enhancing historical positive and negative samples, and future trajectory data is predicted based on the historical time series trajectory data of suspicious transport vehicles.

7. The method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle according to claim 6 is characterized in that: The corresponding operations on all the GPS trajectory data of the tanker trucks to ensure the atomicity and temporality of all the GPS trajectory data of the tanker trucks include: time series alignment, outlier processing, trajectory compression, time series segmentation and trajectory segmentation of all the GPS trajectory data of the tanker trucks.

8. The method for analyzing GPS tracks of suspicious refined oil transport vehicles according to claim 6 is characterized in that: The sequence pattern analysis of the constructed trajectory time series data includes: using a variety of time series analysis techniques to mine and analyze the time series regularity patterns, and summarizing and extracting the common time series regularities of vehicles at different times and in different regions.

9. The method for analyzing the GPS trajectory of a suspicious refined oil transport vehicle according to claim 8, characterized in that: The time series analysis techniques include moving average, exponential smoothing, seasonal decomposition, autoregressive models, and exponential smoothing.

Images