LNG tanker and waybill identification method and system based on traffic big data

Abstract

The present invention discloses a method and system for identifying LNG tank trucks and waybills based on traffic big data, and the method includes: S1, constructing a known tank truck vehicle traffic data model, the known tank truck vehicle traffic data model includes known tank truck historical traffic big data and fusion annotation of liquid contact points and user terminals containing longitude and latitude information; S2, constructing a tank truck identification algorithm model, collecting unknown vehicle trajectory data, and using the tank truck identification algorithm model to identify the tank truck: S3, constructing a tank truck waybill identification algorithm model, and using the tank truck waybill identification algorithm model to generate waybills. The present invention can identify LNG tank trucks from the vehicle monitoring data of dangerous goods transport vehicles, and then extract, optimize and key mark the stop point data of the tank truck, screen and obtain a stop point set, and effectively monitor the stop of the LNG tank truck; it can monitor the liquid connection and unloading, determine and count waybills, and accurately identify the logical waybills of the tank truck, which is convenient for the effective safety management of the LNG tank truck.

Description

Technical Field

[0001] The present invention relates to the field of traffic big data, and in particular to a method and system for identifying LNG tank trucks and waybills based on traffic big data. Background Art

[0002] As an important energy material in my country, liquefied natural gas (LNG) has been widely used in various fields such as industrial production, transportation, and people's livelihood security. With the development of navigation and positioning technologies such as Beidou and GPS positioning, a large amount of data support has been provided for the location positioning and operation trajectory of various vehicles. LNG (liquefied natural gas) tank trucks are important vehicles for transporting liquefied natural gas, involving the connection, transportation, and unloading of tank trucks. LNG tank trucks are a type of dangerous goods transport vehicle (dangerous goods transport vehicles also include transporting dangerous chemicals, fireworks, civilian explosives and other highly dangerous items, among which dangerous chemicals include natural gas). In order to strengthen the supervision of these vehicles, transportation companies need to install satellite positioning devices that meet the "Technical Requirements for On-board Terminals of Satellite Positioning Systems for Road Transport Vehicles" as required. Therefore, how to identify LNG tank trucks from the vehicle monitoring data of dangerous goods transport vehicles based on the transportation behavior and characteristics of LNG tank trucks, and then conduct LNG tank truck parking monitoring, liquid connection and unloading monitoring, waybill determination, statistics, and safety management and operation decision-making for LNG tank trucks have practical significance. Summary of the invention

[0003] The purpose of the present invention is to solve the technical problems pointed out by the background technology, and to provide a LNG tank truck and waybill identification method and system based on traffic big data, which can perform liquid receiving and unloading monitoring and waybill judgment and statistics, and can accurately identify the logical waybill of the tank truck, so as to facilitate the effective safety management of the LNG tank truck.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A method for identifying LNG tankers and waybills based on traffic big data, the method comprising:

[0006] S1. Construct a known tank truck vehicle traffic data model, which includes known tank truck historical traffic big data with positioning longitude and latitude information and fused annotations of contact points and user terminals containing longitude and latitude information;

[0007] S2. Construct a tank truck identification algorithm model and collect unknown vehicle trajectory data. The tank truck identification algorithm model performs tank truck identification according to the following method:

[0008] A1. Based on the unknown vehicle trajectory data, a stop point set of the unknown vehicle is screened out, and a liquid contact point determination box is drawn according to the latitude and longitude information of the liquid contact point. If a stop point in the stop point set of the unknown vehicle falls within the liquid contact point determination box and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the unknown vehicle trajectory data screened out and determined to be a tank truck is recorded as the determined tank truck trajectory data;

[0009] S3. Construct a tank truck waybill recognition algorithm model. The tank truck waybill recognition algorithm model generates waybill according to the following method:

[0010] B1. Identify two adjacent liquid receiving points from the set of stop points in the tank truck trajectory data and group them into a waybill data group; determine whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, filter out the stop points suspected of being user terminals, and then filter out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading;

[0011] B2. The first liquid receiving point in the waybill data group and each user terminal for unloading liquid are counted in sequence as a logical waybill.

[0012] In order to better implement the present invention, in step A1, based on the built-in positioning module of the unknown vehicle, the unknown vehicle trajectory data containing the longitude and latitude information reported by the vehicle is obtained, and the time interval between adjacent longitude and latitude information reported is T1. The time interval before and after T1 is the adjacent point position in the unknown vehicle trajectory data; the method for obtaining the stop point set of the unknown vehicle is as follows:

[0013] The points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data are within the threshold Q range are merged into suspected stay points A, and they are merged in rounds in sequence. The threshold Q range of each round of merging is increased by ΔQ, and finally a set of suspected stay points is obtained; the user terminal judgment box is drawn with the latitude and longitude information of the user terminal, and the suspected stay points A with a duration greater than T4 and falling into the user terminal judgment box are marked, and the suspected stay points A with a duration greater than T2 and less than T3 in the suspected stay point set are screened and retained, and the marked suspected stay points A are retained, and the remaining suspected stay points A are retained as stop points in the stop point set.

[0014] Preferably, in combination with the traffic data collected corresponding to the unknown vehicle trajectory data, before being determined as a stop point, the remaining suspected stop points A near traffic lights and traffic jams are deleted; at the same time, the longitude and latitude data of the service area are collected, and the service areas are marked for the remaining suspected stop points A that fall within the longitude and latitude data of the service area.

[0015] Preferably, in step A1, the liquid receiving point includes a receiving station and a liquid plant, and a receiving station determination box is defined by the longitude and latitude information of the receiving station, the receiving station determination box includes the longitude and latitude data of the receiving station and the longitude and latitude data of the safety enclosure of the receiving station, and a liquid plant determination box is defined by the longitude and latitude information of the liquid plant, the liquid plant determination box includes the longitude and latitude data of the liquid plant and the longitude and latitude data of the safety enclosure of the liquid plant;

[0016] A11. If a stop point in the stop point set of the unknown vehicle falls within the receiving station determination box and the stop time is more than 40 minutes and less than 300 minutes, the unknown vehicle is determined to be a tank truck;

[0017] A12. If a stop point in the stop point set of the unknown vehicle falls within the liquid plant determination box and the stop time is more than 40 minutes and less than 200 minutes, the unknown vehicle is determined to be a tank truck.

[0018] Preferably, in step S1, the stop points of the known tank truck historical traffic big data are screened to obtain a stop point set of known vehicles, the stop point frequency of each stop point in the stop point set of the known vehicles is counted, and the stop point frequency greater than P2 is screened as the stop point historical data set, and the stop point historical data set is retrieved for reference when the stop point set of the unknown vehicle is obtained.

[0019] Preferably, the waybill data group in step B2 is stored in a known tank truck vehicle traffic data model as historical data of the known tank truck. The known tank truck vehicle traffic data model has a data input module, and the data input module is used for data input of liquid contact points and user terminals.

[0020] Preferably, in step B2, the last user terminal for unloading liquid in the waybill data group is recorded as the final unloading point. If the waybill data group only includes the first liquid receiving point and the final unloading point, it is defined as a direct waybill. If the waybill data group also includes other user terminals for unloading liquid, it is defined as a split waybill, and the other user terminals for unloading liquid in the waybill data group are used as split task points.

[0021] Preferably, in step S3, if the user terminal for unloading liquid is not found in the waybill data group, a judgment is made as to whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, and the stop point of the suspected user terminal is screened out as the replacement unloading point and marked as the user terminal for unloading liquid, and the predicted logical waybill is statistically analyzed.

[0022] An LNG tanker and waybill identification system based on traffic big data includes a known tanker vehicle traffic data model, a tanker identification algorithm model and a tanker waybill identification algorithm model. The known tanker vehicle traffic data model includes known tanker historical traffic big data with positioning longitude and latitude information and fused labeled contact points and user terminals containing longitude and latitude information; the tanker identification algorithm model collects unknown vehicle trajectory data and performs tanker identification according to the following method:

[0023] A1. Based on the unknown vehicle trajectory data, a stop point set of the unknown vehicle is screened out, and a liquid contact point determination box is drawn according to the latitude and longitude information of the liquid contact point. If a stop point in the stop point set of the unknown vehicle falls within the liquid contact point determination box and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the unknown vehicle trajectory data screened out and determined to be a tank truck is recorded as the determined tank truck trajectory data;

[0024] The tank truck waybill recognition algorithm model generates waybill according to the following method:

[0025] B1. Identify two adjacent liquid receiving points from the set of stop points in the tank truck trajectory data and group them into a waybill data group; determine whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, screen out the stop points suspected of being user terminals, and then screen out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading;

[0026] B2. The first liquid receiving point in the waybill data group and each user terminal for unloading liquid are counted in sequence as a logical waybill, and then output.

[0027] Preferably, the known tank truck vehicle traffic data model includes a stop point history data set and a data input module, and the stop point history data set is obtained by: performing stop point screening on the known tank truck historical traffic big data to obtain a stop point set of known vehicles, counting the stop point frequencies of each stop point set of the known vehicles and screening the stop point frequencies greater than P2 as the stop point history data set, and referencing the stop point history data set when the stop point set of unknown vehicles is obtained; the data input module is used for data input of the liquid contact point and the user terminal; the tank truck identification algorithm model also includes a stop point set acquisition module, and the stop point set acquisition module is used for the following method to obtain the stop point set:

[0028] The points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data or the known tank truck historical traffic big data are within the threshold Q range are merged into suspected stay points A, and they are merged in rounds in sequence. The threshold Q range of each round of merging is increased by ΔQ, and finally a set of suspected stay points is obtained; the user terminal judgment box is drawn with the latitude and longitude information of the user terminal, and the suspected stay points A with a duration greater than T4 and falling into the user terminal judgment box are marked, and the suspected stay points A with a duration greater than T2 and less than T3 in the suspected stay point set are screened and retained, and the marked suspected stay points A are retained, and the remaining suspected stay points A are retained as stop points in the stop point set.

[0029] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0030] (1) The present invention can identify LNG tank trucks from the vehicle monitoring data of dangerous goods transport vehicles, and then extract, optimize and mark the tank truck's stop point data, screen out a stop point set, and effectively monitor the stop of LNG tank trucks; it can monitor liquid receiving and unloading and determine and count waybills, and can accurately identify the tank truck's logical waybills, which is convenient for the effective safety management of LNG tank trucks, and has practical significance for the operation decision-making of LNG tank trucks, thereby providing data support for subsequent LNG resource market analysis, supply and demand analysis, etc.

[0031] (2) The present invention can be applied to the identification and analysis of historical data of LNG tank trucks, and can also be applied to the identification and analysis of real-time data transmitted by LNG tank trucks. It is of great significance to the post-supervision and real-time safety management of LNG tank trucks, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A method flow chart of the LNG tanker and waybill identification method of the present invention;

[0033] Figure 2 The figure is a principle structure diagram of the LNG tanker and waybill identification system of the present invention. DETAILED DESCRIPTION

[0034] The present invention is further described in detail below in conjunction with embodiments:

[0035] Example

[0036] like Figure 1 As shown, a method for identifying LNG tank trucks and waybills based on traffic big data includes:

[0037] S1. Construct a known tank truck vehicle traffic data model (i.e., it has been determined that the dangerous goods transport vehicle is an LNG tank truck and its vehicle traffic data has been collected). The known tank truck vehicle traffic data model includes the known tank truck historical traffic big data with positioning longitude and latitude information and the fused annotations of the liquid contact points and user terminals containing longitude and latitude information (the liquid contact points and user terminals can be statistically re-input based on the study area, and for newly added liquid contact points or user terminals, a data input module can be set for data input).

[0038] In step S1, preferably, the known tank truck historical traffic big data is screened for stop points to obtain a stop point set of known vehicles (obtained according to the stop point set acquisition method of unknown vehicles in the present invention), the stop point frequencies of each stop point in the stop point set of known vehicles are counted and the stop point frequencies greater than P2 are screened as the stop point historical data set, and the stop point historical data set is referenced and retrieved when the stop point set of unknown vehicles is obtained (the present invention constructs a tank truck stop point acquisition model, and the tank truck stop point acquisition model is trained and extracted according to the tank truck stop point features and the stop point acquisition method, and the tank truck stop point features can be extracted based on the stop point historical data set, so that the stop point set acquisition method of unknown vehicles is also synchronously extracted and obtained through the tank truck stop point acquisition model).

[0039] S2. Construct a tank truck identification algorithm model and collect unknown vehicle trajectory data. The tank truck identification algorithm model performs tank truck identification according to the following method:

[0040] A1. Based on the unknown vehicle trajectory data (LNG tank trucks are identified from the vehicle monitoring data of dangerous goods transport vehicles), a set of stop points for unknown vehicles is screened out, and a liquid contact point determination box is drawn using the latitude and longitude information of the liquid contact point. If a stop point in the stop point set of the unknown vehicle falls within the liquid contact point determination box and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the trajectory data of the unknown vehicle screened out and determined to be a tank truck is recorded as the determined tank truck trajectory data.

[0041] In some preferred embodiments, in step A1, the unknown vehicle trajectory data containing the longitude and latitude information reported by the vehicle is obtained based on the built-in positioning module of the unknown vehicle, and the time interval between adjacent reported longitude and latitude information is T1 (generally the interval time for the satellite positioning device to report positioning data, such as 3 seconds or 5 seconds, or 10 seconds or 20 seconds or 30 seconds), and the time interval before and after T1 is the adjacent point in the unknown vehicle trajectory data. The method for obtaining the stop point set of the unknown vehicle is as follows (the method for obtaining the stop point set of the unknown vehicle is also applicable to the acquisition of the stop points of the known tank truck historical traffic big data):

[0042] Merge the points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data are within the threshold Q range (for example, set to 80 meters) into suspected stop points A, and merge them in rounds in turn. The threshold Q range of each round of merging increases by ΔQ (the ΔQ increase in each round can be set to be different, for example, the ΔQ increased in the second round is 40 meters, the ΔQ increased in the second round is 20 meters, ...; of course, the ΔQ superimposed in each round can be the same, for example, it is set to 10 meters, that is, if the threshold Q range in the first round is 80 meters, the second round is 90 meters, the third round is 100 meters, ...; Generally speaking, when merging suspected stop points A in rounds, it will end in a few rounds, that is, there is no new stop point in the last round. However, there is also a situation where there is a traffic jam for several kilometers for a long time, and the moving threshold Q plus the superposition of the number of response rounds ΔQ within a certain period of time, such as 80 meters or 100 meters or 120 meters, will produce more suspected stop points A within a certain distance range, which cannot be merged according to the rules. The present invention provides a more preferred technical solution in the following technical description, that is, it involves traffic data based on traffic jam points. The traffic data of traffic jam points can be derived from satellite positioning devices or from traffic road network information. Further reduction of suspected stop points A can effectively reduce the number of invalid suspected stop points A and improve processing efficiency), and finally obtain a suspected stop point set. The user terminal judgment frame is drawn with the latitude and longitude information of the user terminal, and the suspected stop points A with a duration greater than T4 and falling into the user terminal judgment frame are highlighted. The suspected stop points A with a duration greater than T2 and less than T3 in the suspected stop point set are screened and retained, and the highlighted suspected stop points A are retained, and the remaining suspected stop points A are retained as stop points in the stop point set. In some embodiments, based on the time interval T1 and longitude and latitude reported by the unknown vehicle trajectory data, the distance (based on the previous and subsequent longitude and latitude), duration (based on the number of times including the time interval T1) and speed of adjacent points are calculated, and the distance is set to <Y meters (for example, 100 meters) and duration > T2 (for example, 30 minutes), and it is selected as the suspected stop point A or stop point.

[0043] In some embodiments, in step A1, the liquid contact point includes a receiving station and a liquid plant. The longitude and latitude information of the receiving station is used to draw a receiving station determination box, and the receiving station determination box includes the longitude and latitude data of the receiving station and the longitude and latitude data of the safety boundary of the receiving station. The longitude and latitude information of the liquid plant is used to draw a liquid plant determination box, and the liquid plant determination box includes the longitude and latitude data of the liquid plant and the longitude and latitude data of the safety boundary of the liquid plant.

[0044] A11. If a stop point in the stop point set of the unknown vehicle falls within the receiving station determination box and the stop time is more than 40 minutes and less than 300 minutes, the unknown vehicle is determined to be a tank truck.

[0045] A12. If a stop point in the stop point set of the unknown vehicle falls within the liquid plant determination box and the stop time is more than 40 minutes and less than 200 minutes, the unknown vehicle is determined to be a tank truck.

[0046] S3. Construct a tank truck waybill recognition algorithm model. The tank truck waybill recognition algorithm model generates waybill according to the following method:

[0047] B1. Identify two adjacent liquid receiving points from the set of stop points in the tanker track data and group them into a waybill data group. Determine whether the stop point is a user terminal based on the tanker track data in the waybill data group, filter out the stop points suspected of being user terminals, and then filter out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading.

[0048] B2. The first liquid connection point and each liquid unloading user terminal in the waybill data group are counted in sequence as a logical waybill. Preferably, the last liquid unloading user terminal in the waybill data group is recorded as the final liquid unloading point. If the waybill data group only includes the first liquid connection point and the final liquid unloading point, it is defined as a direct waybill. If the waybill data group also includes other liquid unloading user terminals, it is defined as a split unloading waybill, and the other liquid unloading user terminals in the waybill data group are used as split unloading task points. In some embodiments, if the liquid unloading user terminal is not found in the waybill data group, a judgment is made as to whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, and the stop point of the suspected user terminal is screened out as the supplementary unloading point and marked as the supplementary unloading user terminal, and the predicted logical waybill is counted.

[0049] In some embodiments, in combination with the traffic data collected corresponding to the unknown vehicle trajectory data, before determining as a stop point, the remaining suspected stop points A near the traffic light points and the traffic jam points are deleted (the number of suspected stop points A can be further effectively reduced, facilitating subsequent faster processing); at the same time, the longitude and latitude data of the service area are collected, and the suspected stop points A among the remaining suspected stop points A that fall into the longitude and latitude data of the service area are marked as service areas. These service areas are marked and retained, so as to facilitate knowing which service areas the tank truck has stopped at (of course, some service areas may also be user terminal points of the tank truck. The present invention also marks the user terminals as key points, so such service areas are double-marked).

[0050] In some embodiments, the waybill data set in step B2 is stored in the known tank truck vehicle traffic data model as the historical data of the known tank truck, which is determined to be a tank truck by the present invention and the waybill data set of the tank truck is obtained, and then stored to facilitate data supplement (in the case of less data or earlier layout of the known tank truck vehicle traffic data model). The known tank truck vehicle traffic data model has a data input module, which is used for data input of the liquid contact point and the user terminal.

[0051] like Figure 2 As shown, a LNG tanker and waybill identification system based on traffic big data includes a known tanker vehicle traffic data model, a tanker identification algorithm model and a tanker waybill identification algorithm model. The known tanker vehicle traffic data model includes known tanker historical traffic big data with positioning longitude and latitude information and fused annotations of contact points and user terminals containing longitude and latitude information. The tanker identification algorithm model collects unknown vehicle trajectory data and performs tanker identification according to the following method:

[0052] A1. A stop point set of an unknown vehicle is screened out based on the unknown vehicle trajectory data, and a liquid contact point determination box is drawn according to the latitude and longitude information of the liquid contact point. If a stop point in the stop point set of the unknown vehicle falls within the liquid contact point determination box and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the unknown vehicle trajectory data screened and determined to be a tank truck is recorded as the determined tank truck trajectory data.

[0053] The tank truck waybill recognition algorithm model generates waybill according to the following method:

[0054] B1. Identify two adjacent liquid receiving points from the set of stop points in the tanker track data and group them into a waybill data group. Determine whether the stop point is a user terminal based on the tanker track data in the waybill data group, filter out the stop points suspected of being user terminals, and then filter out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading.

[0055] B2. The first liquid receiving point in the waybill data group and each user terminal for unloading liquid are counted in sequence as a logical waybill, and then output.

[0056] The known tank truck vehicle traffic data model of the present invention includes a stop point history data set and a data input module. The method for obtaining the stop point history data set is as follows: the known tank truck historical traffic big data is subjected to stop point screening to obtain a stop point set of known vehicles, the stop point frequencies of each stop point set of known vehicles are counted and the stop point frequencies greater than P2 are screened as the stop point history data set, and the stop point history data set is referenced and retrieved when the stop point set of unknown vehicles is obtained. The data input module is used for data input of the liquid contact point and the user terminal. The tank truck identification algorithm model also includes a stop point set acquisition module, and the stop point set acquisition module is used for the following method for obtaining the stop point set:

[0057] Merge the points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data or the known tank truck historical traffic big data within the threshold Q range into suspected stop points A, and merge them in rounds in turn. The threshold Q range of each round of merging is increased by ΔQ, and finally a set of suspected stop points is obtained. The user terminal judgment box is drawn with the latitude and longitude information of the user terminal, and the suspected stop points A with a duration greater than T4 and falling into the user terminal judgment box are highlighted. The suspected stop points A with a duration greater than T2 and less than T3 in the suspected stop point set are screened and retained. At the same time, the highlighted suspected stop points A are retained, and the remaining suspected stop points A are retained as stop points in the stop point set.

[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying LNG tankers and waybills based on traffic big data, characterized in that: The methods include: S1. Construct a known tank truck vehicle traffic data model, which includes known tank truck historical traffic big data with positioning longitude and latitude information and fused annotations of contact points and user terminals containing longitude and latitude information; S2. Construct a tank truck identification algorithm model and collect unknown vehicle trajectory data. The tank truck identification algorithm model performs tank truck identification according to the following method: A1. Based on the unknown vehicle trajectory data, a stop point set of the unknown vehicle is screened out, and a liquid connection point determination frame is drawn out with the latitude and longitude information of the liquid connection point. The liquid connection point includes a receiving station and a liquid plant. A receiving station determination frame is drawn out with the latitude and longitude information of the receiving station. The receiving station determination frame includes the latitude and longitude data of the receiving station and the latitude and longitude data of the receiving station safety enclosure. A liquid plant determination frame is drawn out with the latitude and longitude information of the liquid plant. The liquid plant determination frame includes the latitude and longitude data of the liquid plant and the latitude and longitude data of the liquid plant safety enclosure. If a stop point in the stop point set of the unknown vehicle falls into the liquid connection point determination frame and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the unknown vehicle trajectory data screened out and determined to be a tank truck is recorded as the determined tank truck trajectory data; S3. Construct a tank truck waybill recognition algorithm model. The tank truck waybill recognition algorithm model generates waybill according to the following method: B1. Identify two adjacent liquid receiving points from the set of stop points in the tank truck trajectory data and group them into a waybill data group; determine whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, screen out the stop points suspected of being user terminals, and then screen out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading; B2. The first liquid receiving point and each user terminal for unloading liquid in the waybill data group are counted in sequence as a logical waybill; the last user terminal for unloading liquid in the waybill data group is recorded as the final unloading point. If the waybill data group only includes the first liquid receiving point and the final unloading point, it is defined as a direct waybill; if the waybill data group also includes other user terminals for unloading liquid, it is defined as a split waybill, and the other user terminals for unloading liquid in the waybill data group are used as split task points.

2. The LNG tanker and waybill identification method based on traffic big data according to claim 1 is characterized in that: In step A1, based on the built-in positioning module of the unknown vehicle, the unknown vehicle trajectory data containing the longitude and latitude information reported by the vehicle is obtained. The time interval between adjacent longitude and latitude information reported is T1. The points before and after the time interval T1 are adjacent points in the unknown vehicle trajectory data. The method for obtaining the stop point set of the unknown vehicle is as follows: The points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data are within the threshold Q range are merged into suspected stay points A, and they are merged in rounds in sequence. The threshold Q range of each round of merging is increased by ΔQ, and finally a set of suspected stay points is obtained; the user terminal judgment box is drawn with the latitude and longitude information of the user terminal, and the suspected stay points A with a duration greater than T4 and falling into the user terminal judgment box are marked, and the suspected stay points A with a duration greater than T2 and less than T3 in the suspected stay point set are screened and retained, and the marked suspected stay points A are retained, and the remaining suspected stay points A are retained as stop points in the stop point set.

3. The LNG tanker and waybill identification method based on traffic big data according to claim 2 is characterized in that: Combined with the traffic data collected corresponding to the unknown vehicle trajectory data, before being determined as a stop point, the remaining suspected stop points A near traffic lights and traffic jams are deleted; at the same time, the longitude and latitude data of the service area are collected, and the remaining suspected stop points A that fall into the longitude and latitude data of the service area are marked as service areas.

4. The LNG tanker and waybill identification method based on traffic big data according to claim 1 is characterized in that: In step A1, the following method is included: A11. If a stop point in the stop point set of the unknown vehicle falls within the receiving station determination box and the stop time is more than 40 minutes and less than 300 minutes, the unknown vehicle is determined to be a tank truck; A12. If a stop point in the stop point set of the unknown vehicle falls within the liquid plant determination box and the stop time is more than 40 minutes and less than 200 minutes, the unknown vehicle is determined to be a tank truck.

5. The LNG tanker and waybill identification method based on traffic big data according to claim 1 is characterized in that: In step S1, the stop points of known tank truck historical traffic big data are screened to obtain a stop point set of known vehicles, the stop point frequencies of the known vehicle stop point set are counted, and the stop point frequencies greater than P2 are screened as the stop point historical data set. When the stop point set of unknown vehicles is obtained, the stop point historical data set is retrieved for reference.

6. The LNG tanker and waybill identification method based on traffic big data according to claim 1 is characterized in that: The waybill data group in step B2 is stored in the known tank truck vehicle traffic data model as the historical data of the known tank truck. The known tank truck vehicle traffic data model has a data input module, and the data input module is used for data input of the liquid contact point and the user terminal.

7. The LNG tanker and waybill identification method based on traffic big data according to claim 1 is characterized in that: In step S3, if the user terminal for unloading liquid is not found in the waybill data group, the stop point is judged to be a user terminal based on the tank truck trajectory data in the waybill data group, and the stop point suspected to be the user terminal is selected as the replacement unloading point and marked as the user terminal for replacement unloading, and the predicted logical waybill is counted.

8. An LNG tanker and waybill identification system based on traffic big data that implements the LNG tanker and waybill identification method of claim 1, characterized in that: It includes a known tank truck vehicle traffic data model, a tank truck identification algorithm model and a tank truck waybill identification algorithm model. The known tank truck vehicle traffic data model includes known tank truck historical traffic big data with positioning longitude and latitude information and fused annotations of contact points and user terminals containing longitude and latitude information; the tank truck identification algorithm model collects unknown vehicle trajectory data and performs tank truck identification according to the following method: A1. Based on the unknown vehicle trajectory data, a stop point set of the unknown vehicle is screened out, and a liquid contact point determination box is drawn according to the latitude and longitude information of the liquid contact point. If a stop point in the stop point set of the unknown vehicle falls within the liquid contact point determination box and the stop time exceeds M1 and does not exceed N1, the unknown vehicle is determined to be a tank truck, and the unknown vehicle trajectory data screened out and determined to be a tank truck is recorded as the determined tank truck trajectory data; The tank truck waybill recognition algorithm model generates waybill according to the following method: B1. Identify two adjacent liquid receiving points from the set of stop points in the tank truck trajectory data and group them into a waybill data group; determine whether the stop point is a user terminal based on the tank truck trajectory data in the waybill data group, screen out the stop points suspected of being user terminals, and then screen out the stop points suspected of being user terminals whose stop time is longer than P1 and determine them as user terminals for unloading; B2. The first liquid receiving point in the waybill data group and each user terminal for unloading liquid are counted in sequence as a logical waybill, and then output.

9. The LNG tanker and waybill identification system based on traffic big data according to claim 8 is characterized in that: The known tank truck vehicle traffic data model includes a stop point history data set and a data input module. The stop point history data set is obtained by filtering the known tank truck historical traffic big data to obtain a stop point set of known vehicles, counting the stop point frequencies of each stop point set of the known vehicles and filtering the stop point frequencies greater than P2 as the stop point history data set, and referencing the stop point history data set when the stop point set of unknown vehicles is obtained; the data input module is used for data input of the liquid contact point and the user terminal; the tank truck identification algorithm model also includes a stop point set acquisition module, and the stop point set acquisition module is used for obtaining the stop point set as follows: The points whose latitude and longitude information of adjacent points in the unknown vehicle trajectory data or the known tank truck historical traffic big data are within the threshold Q range are merged into suspected stay points A, and they are merged in rounds in sequence. The threshold Q range of each round of merging is increased by ΔQ, and finally a set of suspected stay points is obtained; the user terminal judgment box is drawn with the latitude and longitude information of the user terminal, and the suspected stay points A with a duration greater than T4 and falling into the user terminal judgment box are marked, and the suspected stay points A with a duration greater than T2 and less than T3 in the suspected stay point set are screened and retained, and the marked suspected stay points A are retained, and the remaining suspected stay points A are retained as stop points in the stop point set.

Images