A method for identifying fuel cell vehicle stops and origins

Abstract

This invention discloses a method for identifying the stop and origin / destination points of fuel cell vehicles, belonging to the field of information processing and intelligent transportation technology. The method includes the following steps: S1, acquiring the vehicle's upload time, location, gear position, cumulative mileage, and operating status to obtain a trajectory record; S2, obtaining a discrimination index based on the trajectory record; S3, generating candidate trips based on the discrimination index; S4, extracting the mileage change and average driving state of the candidate trips to obtain stop results and trip results; S5, performing segment correction on the stop results and trip results to obtain stop locations and categorized stop results; S6, outputting the vehicle stop point and the origin and destination of each trip based on historical stop points, vehicle category, refueling behavior, and stop locations. This invention, by jointly incorporating average driving state, mileage change, and vehicle category into stop determination, adapts to the differences in operating intensity, refueling frequency, and stop characteristics between fuel cell passenger vehicles and freight vehicles.

Description

Technical Field

[0001] This invention relates to the fields of information processing and intelligent transportation technology, and in particular to a method for identifying the stopping and origin / destination points of fuel cell vehicles. Background Technology

[0002] With the national advocacy for energy conservation and environmental protection, green transportation has become a key development goal for the transportation, automobile manufacturing, and other related industries. Promoting the commercial application of hydrogen fuel cell vehicles is of great strategic significance for achieving green development and optimizing the operation and management of urban passenger and freight transport. The development focus of hydrogen fuel cell commercial vehicles is on large and medium-sized passenger and freight vehicles, especially semi-trailers and semi-trailer tractors. These types of commercial vehicles can typically guarantee daily operating mileage and reduce costs through relatively fixed usage rates, which is generally conducive to the promotion of point-to-point transportation scenarios.

[0003] In recent years, guided by national and local industrial policies, the entire hydrogen energy industry chain, encompassing production, storage, transportation, refueling, and utilization, has been continuously improved, and the technological level of fuel cell products has been steadily enhanced. my country's fuel cell vehicles are rapidly moving from technological research and development to commercialization. However, relevant transportation information technology still has certain limitations. Currently, the usage status of hydrogen fuel cell vehicles is difficult to monitor effectively, and large-scale, full-sample, and high-precision data monitoring and processing still need improvement.

[0004] Traditional methods for processing travel data and acquiring origin-destination (OD) matrices typically involve large-scale vehicle travel sampling surveys. Current OD acquisition for gasoline-powered vehicles relies heavily on traffic survey equipment, primarily high-definition intelligent checkpoint systems. These survey devices are usually installed on major urban thoroughfares and, based on road vehicle data, obtain travel data for each traffic zone through methods such as license plate recognition. Key methods include conjugate Bayesian OD backpropagation, dynamic OD identification, OD identification based on the K-law shortest path algorithm, OD identification based on the Kalman filter model, and automatic vehicle identification. However, these methods are more suitable for gasoline-powered vehicles, whose travel is more arbitrary and less constrained by refueling stations and other supply facilities. For hydrogen fuel cell vehicles with stable refueling needs, effective OD and parking identification and monitoring are difficult to implement. Furthermore, the potential of hydrogen fuel cell vehicles as information platforms, capable of acquiring data from location devices and other onboard monitoring equipment, is not fully realized. The potential for high-precision, high-quality big data from hydrogen fuel cell vehicles remains untapped.

[0005] Therefore, there is an urgent need to provide a method for identifying the stopping and origin / destination points of fuel cell vehicles to solve the above problems. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art, which is more suitable for fuel vehicles with arbitrary travel and less constraint from refueling facilities such as gas stations; for hydrogen fuel cell vehicles with stable hydrogen refueling needs, it is difficult to effectively implement OD and parking identification and monitoring, and the advantages of hydrogen fuel cell vehicles in terms of information platform potential and the ability to obtain data from location devices and other on-board monitoring devices are not fully utilized. The present invention provides a method for identifying the parking and origin and destination points of fuel cell vehicles.

[0007] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: to provide a method for identifying the stopping and origin / destination points of a fuel cell vehicle, comprising the following steps: S1. Obtain the vehicle's upload time, location, gear, cumulative mileage, and operating status, and perform time alignment, deduplication, anomaly removal, and single-day merging on the upload time, location, gear, cumulative mileage, and operating status to obtain the trajectory record; S2. Identify the stopping records with the same location and cumulative mileage but different upload times based on the trajectory records, and calculate the time difference and position difference between adjacent trajectory records to obtain the discrimination index; S3. Mark the start point candidate, end point candidate and breakpoint candidate according to the discrimination index, the first preset time threshold and the second preset time threshold, and generate candidate routes according to the start point candidate, the end point candidate and the breakpoint candidate; S4. Extract the mileage change and average driving status of the candidate trip, and determine the stop based on the mileage change and average driving status according to the preset speed threshold and preset mileage threshold to obtain the stop result and trip result; S5. Perform segment correction on the dwell results and the trip results, and divide the corrected dwell results into short dwell and long dwell based on the third preset time threshold. Combine the refueling time characteristics of the vehicle to identify the refueling behavior, and cluster the dwell locations to obtain dwell locations and classified dwell results. S6. Based on historical stop points, vehicle type, refueling behavior, and stop locations, update the first preset time threshold, the second preset time threshold, the speed threshold, the mileage threshold, and the third preset time threshold, and correct the classification stop results according to the updated thresholds, outputting vehicle stop points and the start and end points of each trip.

[0008] The present invention is further configured such that the operating state includes two or more of the following: vehicle state, recharge state, operating mode, vehicle speed, charge state, accelerator pedal state, and brake pedal state.

[0009] The present invention is further configured such that: the process of time alignment, deduplication, anomaly removal, and daily merging of upload time, location, gear position, cumulative mileage, and operating status includes: filtering original records according to the target date; deleting duplicate records based on the collection time; establishing a correspondence between different data tables based on a unified timestamp; removing location anomalies, speed anomalies, and time out-of-order points; and merging the cleaned records daily to form the trajectory record.

[0010] The present invention is further configured such that the marking of start point candidates, end point candidates, and breakpoint candidates includes: assigning a first judgment condition when the time difference between a record and the previous record is greater than a first preset time threshold; assigning a second judgment condition when the time difference between a record and the next record is greater than a second preset time threshold; directly assigning the first judgment condition when the first record lacks a previous record, and directly assigning the second judgment condition when the last record lacks a next record; and identifying start point candidates, end point candidates, and breakpoint candidates based on the combination result of the two judgment conditions.

[0011] The present invention is further configured such that: the mileage change of the candidate trip is obtained by the cumulative mileage difference corresponding to the start and end positions of the candidate trip; the average driving state is obtained by the average speed within the time period corresponding to the candidate trip; and the position difference is calculated based on the position coordinates of adjacent trajectory records.

[0012] The present invention is further configured such that: the stop determination includes: when the average driving state is greater than the speed threshold and the mileage change is less than the mileage threshold, it is determined to be a stop; when the average driving state is greater than the speed threshold and the mileage change is greater than or equal to the mileage threshold, it is determined to be a trip; when the average driving state is less than or equal to the speed threshold, it is determined to be a stop; the speed threshold and the mileage threshold are adaptively adjusted according to the vehicle category.

[0013] The present invention is further configured such that the segment correction includes: merging adjacent stop results into continuous stop segments; and incorporating breakpoint candidates formed by data loss into adjacent stop segments or adjacent travel segments that are continuous with its time and have smaller mileage changes, so as to reduce the impact of scattered points on the identification results.

[0014] The present invention is further configured such that: the division of short stays and long stays includes: taking stays with a duration less than the third time threshold as short stay candidates, and verifying them in conjunction with the refueling time characteristics to identify the refueling behavior; and clustering the stay locations to obtain the stay locations.

[0015] The present invention is further configured such that: the stopping location is obtained by merging stopping locations that are close in location and recur; when a starting point candidate or an ending point candidate falls within the neighborhood of the stopping location, the starting point candidate or the ending point candidate is corrected to the center position of the corresponding stopping location.

[0016] The present invention is further configured such that updating the first time threshold, the second time threshold, the speed threshold, the mileage threshold, and the third time threshold includes updating the first time threshold, the second time threshold, the speed threshold, the mileage threshold, and the third time threshold based on the historical stop points, the vehicle category, the refueling behavior, and the stop location.

[0017] The beneficial effects of this invention are as follows: 1. This invention marks point data as candidate start point, candidate end point, and candidate breakpoint by combining the results of dual time thresholds, and then performs a secondary judgment on the candidate trip by combining mileage changes and average driving status, thereby improving the logical integrity and recognition accuracy of start and end point identification. 2. This invention incorporates average driving status, mileage change and vehicle category into the dwell determination, so that dwell identification is no longer limited to static threshold judgment, and can better adapt to the differences in operating intensity, refueling frequency and dwell characteristics of fuel cell passenger vehicles and freight vehicles; 3. By incorporating candidate breakpoints caused by data loss into adjacent stop segments or adjacent travel segments that are continuous in time and have smaller mileage changes, this invention can correct the phenomenon of misclassification of trajectory scattered points and breakpoints, and improve the stability of trajectory segmentation results. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is an example diagram of repeated data entry according to the present invention; Figure 3 This is an example diagram of point data recognition according to the present invention; Figure 4 This is an example diagram showing the freight vehicle travel trajectory recognition results of the present invention; Figure 5 This is an example diagram of the freight vehicle parking recognition result of the present invention; Figure 6 This is an example diagram showing the passenger vehicle travel trajectory recognition results of the present invention; Figure 7 This is an example diagram of the passenger vehicle parking recognition result of the present invention. Detailed Implementation

[0019] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0020] Please see Figures 1-7 A method for identifying the stopping and origin / destination points of a fuel cell vehicle includes the following steps: S1. Obtain the vehicle's upload time, location, gear, cumulative mileage, and operating status, and perform time alignment, deduplication, anomaly removal, and single-day merging on the upload time, location, gear, cumulative mileage, and operating status to obtain the trajectory record; S2. Identify stopping records with the same location and accumulated mileage but different upload times based on the trajectory records, and calculate the time difference and position difference between adjacent trajectory records to obtain the discrimination index; S3. Mark the starting point candidate, ending point candidate, and breakpoint candidate according to the discrimination index, the first preset time threshold, and the second preset time threshold, and generate candidate routes based on the starting point candidate, ending point candidate, and breakpoint candidate. S4. Extract the mileage change and average driving status of the candidate trip, and determine the stop based on the mileage change and average driving status according to the preset speed threshold and preset mileage threshold to obtain the stop result and trip result; S5. Perform segment correction on the dwell results and trip results, and divide the corrected dwell results into short dwell and long dwell based on the third preset time threshold. Combine the refueling time characteristics of the vehicle to identify refueling behavior, and cluster the dwell locations to obtain dwell locations and classified dwell results. S6. Based on historical stop points, vehicle type, refueling behavior, and stop locations, update the first preset time threshold, second preset time threshold, speed threshold, mileage threshold, and third preset time threshold, and correct the classification stop results according to the updated thresholds, outputting the vehicle stop points and the start and end points of each trip.

[0021] In this embodiment, the vehicle is a fuel cell vehicle equipped with onboard monitoring and driving parameter recording equipment. The monitoring equipment uploads operational data every ten seconds. The raw data includes at least basic fields such as data upload time t, latitude and longitude location 1, vehicle gear, and cumulative mileage s, and may also include operational fields such as vehicle status, refueling status, operating mode, vehicle speed, state of charge, accelerator pedal travel value, and brake pedal status.

[0022] For step S1, given that the original uploaded data may contain issues such as abnormal maximum / minimum values, time errors, spatial errors, duplicate data entry, and inconsistencies between different database tables, quality control is first performed on the data. Specifically, this includes: filtering original records by target date and removing data from abnormal dates; deleting duplicate records using the collection time as the primary key, retaining only unique records; establishing correspondences between multiple source tables (location data, fuel cell data, vehicle data, energy storage device data, and extreme value data) using vehicle collection time as the indicator; removing location anomalies, speed anomalies, and out-of-order time points; and merging the cleaned multi-source records by day to obtain the trajectory record. An example of duplicate data entry is shown below. Figure 2 As shown.

[0023] For step S2, after completing data cleaning and trajectory record construction, the point data in the trajectory records are further processed. If the location, cumulative mileage, and upload time are all the same, they are considered redundant records and deleted; if the location and cumulative mileage are the same but the upload time is different, adjacent records are identified as consecutive stopping records at the same point, and the stopping records are iteratively updated.

[0024] In a preferred embodiment, the following relationship can be used when iteratively updating the continuously paused records:

[0025]

[0026] in, This indicates the updated mileage value of the previous record. This indicates the mileage value of the next record; This indicates the updated time amount. and These represent the time values ​​corresponding to two adjacent records.

[0027] Further, the time difference T and position difference L between adjacent records are calculated. For the time difference T, the time data in hour-minute-second format is converted to seconds, and the time difference between adjacent records is used as the time difference T. For the position difference L, it is calculated based on the position coordinates of adjacent records, preferably using the spherical distance formula:

[0028] Where L is the distance between the two points, in km; The radius of the Earth; , The longitudes of the two points; , The latitude of the two points is 2.

[0029] For step S3, set the first time threshold t. a Second time threshold tb This is used to identify the continuity of point data. If the time difference between a point and the previous point exceeds t... a Then the first judgment condition is assigned; if the time difference between a certain point and the next point exceeds t... b If the first record is true, then the second judgment condition is assigned; the first record is directly assigned the first judgment condition, and the last record is directly assigned the second judgment condition; based on the combination of the two judgment conditions, candidate start points, candidate end points, and candidate breakpoints are identified. An example of point data identification is shown below. Figure 3 .

[0030] In a preferred embodiment, point data type 1 represents a starting point candidate, point data type 2 represents an ending point candidate, and point data type 3 represents a breakpoint candidate.

[0031] For step S4, candidate trips are generated based on candidate start points, candidate end points, and candidate breakpoints. The mileage change S of the candidate trip is obtained through the cumulative mileage difference between the start and end positions of the candidate trip, and the average driving state V of the candidate trip is obtained through the average speed within the corresponding time period of the candidate trip. This is based on a speed threshold V. i and mileage threshold S j The determination of when to stop is based on changes in mileage and average driving conditions.

[0032] For step S5, the stop results and trip results are segmented and corrected. If two adjacent segments are both stops, they are merged into a continuous stop segment; if a breakpoint candidate is formed due to data loss, it is merged into an adjacent stop segment or adjacent trip segment that is continuous in time and has smaller mileage changes. Then, based on the third time threshold t... k The corrected dwell times are divided into short dwell times and long dwell times, and the refueling behavior is identified by combining the refueling time characteristics. At the same time, the dwell locations are clustered to obtain the dwell locations and the classified dwell times.

[0033] For step S6, based on historical stop points, vehicle type, refueling behavior, and stop location, the first time threshold t is... a Second time threshold t b Speed ​​threshold V i Mileage threshold S j and the third time threshold t k The system is updated, and the classification and dwell results are corrected based on the updated thresholds. The system outputs the vehicle dwell points and the start and end points of each trip.

[0034] Figure 4 , Figure 5 , Figure 6 and Figure 7 Examples of freight vehicle travel trajectory recognition results, freight vehicle parking recognition results, passenger vehicle travel trajectory recognition results, and passenger vehicle parking recognition results are provided.

[0035] The operating status includes two or more of the following: vehicle status, recharge status, operating mode, vehicle speed, charge status, accelerator pedal status, and brake pedal status.

[0036] In this embodiment, the operating status field can be selected from two or more of the following: vehicle status, recharging status, operating mode, vehicle speed, state of charge, accelerator pedal status, and brake pedal status. Using multiple status fields in conjunction improves the reliability of dwell time identification, recharging behavior identification, and origin / end point correction.

[0037] The process involves time alignment, deduplication, anomaly removal, and daily merging of uploaded time, location, gear position, cumulative mileage, and operating status. This includes: filtering original records according to the target date; deleting duplicate records based on the collection time; establishing correspondence between different data tables based on a unified timestamp; removing location anomalies, speed anomalies, and time out-of-order points; and merging the cleaned records daily to form trajectory records.

[0038] In this embodiment, time alignment, deduplication, anomaly removal, and single-day merging are performed in the following order: First, the original records are filtered based on the target date to remove records spanning multiple days, misaligned days, and abnormal dates; second, duplicate records are identified and deleted using the collection time as the primary key; third, the correspondence between location data, fuel cell data, vehicle data, energy storage device data, and extreme value data is established based on a unified timestamp; then, anomaly removal is performed on location data, vehicle speed data, and out-of-order time data that clearly exceed the normal range; finally, the cleaned records are merged on a single-day basis to obtain the trajectory record.

[0039] The process of marking start point candidates, end point candidates, and breakpoint candidates includes: assigning a first judgment condition when the time difference between a record and the previous record is greater than a first preset time threshold; assigning a second judgment condition when the time difference between a record and the next record is greater than a second preset time threshold; assigning the first judgment condition directly when the first record lacks a previous record, and assigning the second judgment condition directly when the last record lacks a next record; and identifying start point candidates, end point candidates, and breakpoint candidates based on the combination of the two judgment conditions.

[0040] In this embodiment, the first determination condition and the second determination condition respectively characterize the temporal continuity between the current record and the previous and next records. When a record only satisfies the first determination condition, it can be identified as a starting point candidate; when a record only satisfies the second determination condition, it can be identified as an ending point candidate; when a record satisfies both the first and second determination conditions, it can be identified as a breakpoint candidate.

[0041] Among them, the mileage change of the candidate trip is obtained by the cumulative mileage difference corresponding to the start and end positions of the candidate trip, the average driving status is obtained by the average speed within the corresponding time period of the candidate trip, and the position difference is calculated based on the position coordinates of adjacent trajectory records.

[0042] In this embodiment, the position difference is calculated based on the position coordinates of adjacent trajectory records; the mileage change of the candidate trip is obtained through the cumulative mileage difference corresponding to the start and end positions of the candidate trip; and the average driving status is obtained through the average speed within the corresponding time period of the candidate trip. The aforementioned position difference, mileage change, and average driving status are used as point-level and segment-level discrimination indicators, respectively, for subsequent stop and start / end point identification.

[0043] The determination of a stop includes: when the average driving state is greater than the speed threshold and the mileage change is less than the mileage threshold, it is determined to be a stop; when the average driving state is greater than the speed threshold and the mileage change is greater than or equal to the mileage threshold, it is determined to be a trip; when the average driving state is less than or equal to the speed threshold, it is determined to be a stop; the speed threshold and mileage threshold are adaptively adjusted according to the vehicle category.

[0044] In this embodiment, the determination of a stop includes: when the average driving state is greater than the speed threshold and the mileage change is less than the mileage threshold, it is determined as a stop; when the average driving state is greater than the speed threshold and the mileage change is greater than or equal to the mileage threshold, it is determined as a trip; when the average driving state is less than or equal to the speed threshold, it is determined as a stop. The speed threshold and mileage threshold can be adaptively adjusted according to the vehicle category to adapt to the differences in operating intensity, stop patterns, and refueling frequency of different vehicles.

[0045] The segment merging correction includes: merging adjacent stop results into continuous stop segments; and merging candidate breakpoints caused by data loss into adjacent stop segments or adjacent travel segments that are continuous in time and have smaller mileage changes, so as to reduce the impact of scattered points on the identification results.

[0046] In this embodiment, segment merging correction is used to eliminate the impact of scattered data and local data loss on the identification results. For adjacent stop results, they are merged into continuous stop segments; for breakpoint candidates formed by data loss, they are merged into adjacent stop segments or adjacent travel segments that are continuous with their time and have smaller mileage changes, so as to improve the stability of trajectory segmentation.

[0047] The process of distinguishing between short and long stays includes: considering stays with a duration less than a third time threshold as short stay candidates and verifying them in conjunction with refueling time characteristics to identify refueling behavior; and clustering stay locations to obtain stay locations and reduce location errors.

[0048] In this embodiment, stays with a duration less than a third time threshold are considered as short stay candidates and are verified in conjunction with refueling time characteristics to identify refueling behavior; at the same time, the stay locations are clustered to obtain the stay locations, thereby reducing errors caused by location drift.

[0049] The stopping point is obtained by merging the stopping points that are close in location and appear repeatedly; when the starting point candidate or the ending point candidate falls within the neighborhood of the stopping point, the starting point candidate or the ending point candidate is corrected to the center position of the corresponding stopping point.

[0050] In this embodiment, the stopping point is obtained by merging the stopping points that are close in location and recur. When the starting point candidate or the ending point candidate falls within the neighborhood of the stopping point, the starting point candidate or the ending point candidate can be corrected to the center position of the corresponding stopping point to improve the spatial stability of the output result.

[0051] The updating of the first time threshold, second time threshold, speed threshold, mileage threshold, and third time threshold includes updating the first time threshold, second time threshold, speed threshold, mileage threshold, and third time threshold based on historical stop points, vehicle type, refueling behavior, and stop location, in order to improve the accuracy of stop and origin-end point identification in subsequent trajectory records.

[0052] The specific steps for updating the first preset time threshold, the second preset time threshold, the speed threshold, the mileage threshold, and the third time threshold include: a. Based on historical stop points, vehicle type, refueling behavior, and stop locations, extract the categorized stop results corresponding to the target vehicle type from a preset statistical period. Then, organize the categorized stop results according to the recurrence frequency of stop locations, the frequency of refueling behavior, and the correlation between the start and end points of each trip to form a threshold update sample set. The threshold update sample set includes time difference samples, mileage change samples, average driving status samples, and stop duration samples. The time difference samples correspond to the recording interval before and after the historical stop points, the mileage change samples correspond to the cumulative mileage difference between the start and end points of each trip, the average driving status samples correspond to the average speed between the start and end points of each trip, and the stop duration samples correspond to the duration of each stop within the stop location.

[0053] b. Based on the time difference samples, mileage change samples, average driving status samples, and dwell time samples in the threshold update sample set, and combined with the recurrence frequency of dwell locations and the occurrence frequency of refueling behavior, different samples are merged and sorted to obtain candidate intervals for time difference, mileage change, average driving status, and dwell time. Among them, samples corresponding to dwell locations with a high recurrence frequency are given a larger update weight, and dwell time samples corresponding to refueling behavior are given a larger dwell classification weight, so that the candidate intervals for time difference, mileage change, average driving status, and dwell time can better reflect the actual operating rules of the target vehicle category.

[0054] c. Based on the candidate time difference interval, the stop location, and the refueling behavior, update the first preset time threshold and the second preset time threshold respectively. Specifically, the time difference samples in the candidate time difference interval that are before the stop location are taken as forward time difference samples, and the time difference samples in the candidate time difference interval that are after the stop location are taken as backward time difference samples. The forward time difference samples and backward time difference samples are arranged in ascending order of time length. The boundary time that minimizes the confusion between adjacent stop results and adjacent trip results is taken as the updated first preset time threshold and the updated second preset time threshold. When the number of times the stop location corresponding to the refueling behavior occurs is higher than the preset number, the value range of the updated first preset time threshold and the updated second preset time threshold is appropriately increased.

[0055] d. Based on the candidate intervals for mileage change, the candidate intervals for average driving status, the updated first preset time threshold, and the updated second preset time threshold, update the speed threshold and the mileage threshold respectively. Among them, the candidate trips that meet the continuous conditions corresponding to the updated first preset time threshold and the updated second preset time threshold are taken as valid candidate trips. In the valid candidate trips, the average driving status samples and mileage change samples corresponding to the stop results are taken as stop samples, and the average driving status samples and mileage change samples corresponding to the trip results are taken as trip samples. The speed threshold that minimizes the overlap between the stop samples and the trip samples is taken as the updated speed threshold, and the mileage threshold that minimizes the overlap between the stop samples and the trip samples is taken as the updated mileage threshold.

[0056] e. Based on the candidate interval of dwell time, the updated speed threshold, the updated mileage threshold, and the refueling behavior, update the third time threshold; wherein, the dwell time samples in the candidate interval of dwell time are arranged in ascending order of dwell time, and the effective dwell samples that meet the dwell determination conditions are selected by combining the updated speed threshold and the updated mileage threshold. Then, ordinary dwell samples and refueling dwell samples are distinguished from the effective dwell samples. The dwell time boundary value that makes the distinction between ordinary dwell samples and refueling dwell samples the greatest is taken as the updated third time threshold, so that the division results of short dwell and long dwell are consistent with the refueling behavior.

[0057] f. Based on the updated first preset time threshold, the updated second preset time threshold, the updated speed threshold, the updated mileage threshold, and the updated third time threshold, a target threshold group is formed. The updated first preset time threshold, the updated second preset time threshold, the updated speed threshold, the updated mileage threshold, and the updated third time threshold are compared with the thresholds corresponding to the previous statistical period. When the change in any updated threshold relative to the threshold corresponding to the previous statistical period exceeds a preset allowable range, the change in the updated threshold is reduced according to a preset correction ratio before being included in the target threshold group, thereby improving the stability of the target threshold group between consecutive statistical periods.

[0058] g. Correct the classification stop results based on the target threshold group, and output the vehicle stop point and the start and end points of each trip; wherein, the target threshold group is used to re-determine the start point candidate, end point candidate and breakpoint candidate, re-identify the stop results and trip results of the candidate trips, re-divide short stop and long stop, and correct the start point candidate or end point candidate falling within the neighborhood of the stop point to the center position of the corresponding stop point, thereby obtaining the corrected vehicle stop point, the corrected start point of each trip and the corrected end point of each trip.

[0059] In this embodiment, the threshold update is based on historical stop points, vehicle type, refueling behavior, and stop location. By updating the first time threshold, second time threshold, speed threshold, mileage threshold, and third time threshold, the accuracy of stop and origin / end point identification in subsequent trajectory recordings can be improved.

[0060] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for identifying the stopping and origin / destination points of a fuel cell vehicle, characterized in that: Includes the following steps: S1. Obtain the vehicle's upload time, location, gear, cumulative mileage, and operating status, and perform time alignment, deduplication, anomaly removal, and single-day merging on the upload time, location, gear, cumulative mileage, and operating status to obtain the trajectory record; S2. Identify the stopping records with the same location and cumulative mileage but different upload times based on the trajectory records, and calculate the time difference and position difference between adjacent trajectory records to obtain the discrimination index; S3. Mark the start point candidate, end point candidate and breakpoint candidate according to the discrimination index, the first preset time threshold and the second preset time threshold, and generate candidate routes according to the start point candidate, the end point candidate and the breakpoint candidate; S4. Extract the mileage change and average driving status of the candidate trip, and determine the stop based on the mileage change and average driving status according to the preset speed threshold and preset mileage threshold to obtain the stop result and trip result; S5. Perform segment correction on the dwell results and the trip results, and divide the corrected dwell results into short dwell and long dwell based on the third preset time threshold. Combine the refueling time characteristics of the vehicle to identify the refueling behavior, and cluster the dwell locations to obtain dwell locations and classified dwell results. S6. Based on historical stop points, vehicle type, refueling behavior, and stop locations, update the first preset time threshold, the second preset time threshold, the speed threshold, the mileage threshold, and the third preset time threshold, and correct the classification stop results according to the updated thresholds, outputting vehicle stop points and the start and end points of each trip.

2. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 1, characterized in that: The operating status includes two or more of the following: vehicle status, recharge status, operating mode, vehicle speed, charge status, accelerator pedal status, and brake pedal status.

3. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 2, characterized in that: The process of aligning, deduplicating, removing anomalies, and merging records by day for uploaded time, location, gear position, cumulative mileage, and operating status includes: filtering original records according to the target date; deleting duplicate records based on the collection time; establishing a correspondence between different data tables based on a unified timestamp; removing location anomalies, speed anomalies, and time out-of-order points; and merging the cleaned records by day to form the trajectory record.

4. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 3, characterized in that: The marking of start point candidates, end point candidates, and breakpoint candidates includes: assigning a first judgment condition when the time difference between a record and the previous record is greater than a first preset time threshold; assigning a second judgment condition when the time difference between a record and the next record is greater than a second preset time threshold; directly assigning the first judgment condition when the first record lacks a previous record, and directly assigning the second judgment condition when the last record lacks a next record; and identifying start point candidates, end point candidates, and breakpoint candidates based on the combination result of the two judgment conditions.

5. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 4, characterized in that: The mileage change of the candidate trip is obtained by the cumulative mileage difference corresponding to the start and end positions of the candidate trip, the average driving status is obtained by the average speed within the time period corresponding to the candidate trip, and the position difference is calculated based on the position coordinates of adjacent trajectory records.

6. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 5, characterized in that: The stop determination includes: when the average driving state is greater than the speed threshold and the mileage change is less than the mileage threshold, it is determined as a stop; when the average driving state is greater than the speed threshold and the mileage change is greater than or equal to the mileage threshold, it is determined as a trip; when the average driving state is less than or equal to the speed threshold, it is determined as a stop; the speed threshold and the mileage threshold are adaptively adjusted according to the vehicle category.

7. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 6, characterized in that: The segment merging correction includes: merging adjacent stop results into continuous stop segments; and merging candidate breakpoints caused by data loss into adjacent stop segments or adjacent travel segments that are continuous in time and have smaller mileage changes.

8. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 7, characterized in that: The process of classifying short and long stays includes: identifying stays with a duration less than the third time threshold as short stay candidates, and verifying them in conjunction with refueling time characteristics to identify the refueling behavior; and clustering the stay locations to obtain the stay locations.

9. The method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 8, characterized in that: The stopping point is obtained by merging stopping points that are close in location and recur; when a starting point candidate or an ending point candidate falls within the neighborhood of the stopping point, the starting point candidate or the ending point candidate is corrected to the center position of the corresponding stopping point.

10. A method for identifying the stopping and origin / destination points of a fuel cell vehicle according to claim 9, characterized in that: The step of updating the first time threshold, the second time threshold, the speed threshold, the mileage threshold, and the third time threshold includes updating the first time threshold, the second time threshold, the speed threshold, the mileage threshold, and the third time threshold based on the historical stop points, the vehicle category, the refueling behavior, and the stop location.

Images