CN115995149A - Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system - Google Patents
Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system Download PDFInfo
- Publication number
- CN115995149A CN115995149A CN202211579767.5A CN202211579767A CN115995149A CN 115995149 A CN115995149 A CN 115995149A CN 202211579767 A CN202211579767 A CN 202211579767A CN 115995149 A CN115995149 A CN 115995149A
- Authority
- CN
- China
- Prior art keywords
- parking
- data
- time
- vehicle
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for dynamically evaluating parking supply and demand characteristics based on multi-source data, wherein the method comprises the following steps: dividing parking communities and determining the total supply of parking resources of each parking community; cleaning the acquired vehicle access data of the sample parking lot, and acquiring parking space occupancy indexes of the parking lot; based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property; carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district; drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period. The method dynamically identifies the time-varying characteristics of the supply and demand of the object area, evaluates and analyzes the supply and demand characteristics of the area, and provides a reference for optimizing the regional parking strategy.
Description
Technical Field
The invention relates to the field of traffic management, in particular to a method and a system for dynamically evaluating parking supply and demand characteristics based on multi-source data.
Background
With the continuous and rapid development of economy and the continuous rising of the number of motor vehicles, the contradiction between urban parking supply and demand is prominent year by year, and the problem of difficult parking is an urgent civil problem to be solved in each large city. The urban parking difficulty is relieved, and contradictory characteristics of urban current parking supply and demand, including characteristics of parking scale, space-time distribution and the like, must be accurately identified and analyzed. In reality, it is difficult to analyze the parking supply and demand characteristics by gathering all the parking lot data of different types in the area, and in practice, the traditional method relies on manual investigation to collect or access part of intelligent parking system data to analyze the parking supply and demand characteristics, but the method has the defects of lower precision, unsustainable property and the like. Therefore, based on a multisource space-time data fusion and analysis technology, by utilizing general survey data of parking facilities, high-definition road bayonet data and different types of parking lot distribution characteristic data of different areas, and combining part of available sample parking lot access data, the time-varying supply and demand scale and space-time distribution characteristics of an object area are dynamically identified, and the parking supply and demand characteristics of an analysis area are evaluated, so that references are provided for optimizing parking strategies of the area, and the method is the key point and the purpose of the research of the invention.
At present, the limitation of the current parking demand data source is limited, and the research on the parking supply and demand characteristics is focused on two aspects: on the other hand, although the parking supply and demand scale and distribution characteristics of a target area at a certain time point are estimated by the current resident travel survey or other parking characteristic sampling survey, there are drawbacks in terms of persistence and comprehensiveness due to restrictions of the survey time point and the survey sample size. On the other hand, with the establishment of an intelligent parking data acquisition system, part of researches can be performed on parking lot dynamic data which is accessed through a platform, and fine and continuous parking time-varying feature analysis can be performed, but the method is limited by the sample size of the parking lot data, and has defects in the aspect of the comprehensiveness of regional assessment. In recent years, students try to analyze regional parking requirements by using license plate recognition data, so that the problems caused by insufficient sample size and space-time precision of traditional parking data sources are solved to a great extent, but the research is limited to time-varying parking requirements, the research on time-varying supply and demand matching conditions is insufficient, and the overall consideration of recognition and evaluation of regional parking supply and demand time-varying characteristics is lacking.
Chen Jun et al summarize the parking demand prediction model and propose a model of static traffic occurrence rate and a solution algorithm; nie Tingting and the like are combined with the current situation data of the parking facility investigation, and a travel attraction model is used for carrying out demand prediction and carrying out parking supply and demand gap analysis; guan Hongzhi, etc., performs individual-based parking analysis using a Logit model, and performs price sensitivity analysis using SP survey data; cui Honghai and the like, to perform parking demand prediction and supply and demand analysis based on traffic survey data. The above identification and evaluation of the parking supply and demand characteristics are mainly based on the prediction and evaluation of the parking demand of the traditional investigation, and are static identification and analysis of the parking supply and demand of a target area at a certain time point, and the evaluation is continuously and comprehensively insufficient.
Xu Benying et al propose a method for solving the problem of supply and demand of parking facilities by using traffic cells and plots; tang Keshuang and the like perform time-varying feature analysis on the occupancy rate of parking places by using parking place detection data based on BP neural network and other methods; li Linbo and the like, based on historical data of a parking lot of a certain object, a night parking demand prediction model is established by adopting a survival analysis method, and features such as vehicle parking time period, duration and the like are estimated from a microscopic level. The continuous parking time-varying feature analysis is carried out by using the platform-interposed parking lot dynamic data, but the analysis and the evaluation of the parking lot supply and demand features are limited to a single parking lot or a part of parking lots by the limitation of the parking lot data sample size, and the whole area evaluation has defects.
Yu Jie et al. To calculate the parking requirement based on a license plate track recognition method of the road gate monitoring video; Sha Zhiren proposes a parking demand analysis method based on road junction data; ma Xiao and the like propose a set of target area parking demand estimation and a method for identifying a time-space characteristic mode by utilizing the multi-element data such as traffic bayonets and the like; Zan Yuyao et al. A parking behavior recognition method based on GBDT algorithm is proposed by combining license plate recognition data from road checkpoints, parking lot entry and exit vehicle data, and taxi satellite positioning data. The method for estimating the parking demand by using the license plate recognition data of the road gate solves the problems that the traditional parking data source is insufficient in sample size and time-varying characteristics cannot be obtained, but related researches focus on recognition and evaluation of the time-varying characteristics of the parking demand, and the analysis of the time-varying characteristics of the parking space supply is lacking, so that the contradictory characteristics of the regional supply and demand cannot be comprehensively reflected.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention firstly provides a parking supply and demand characteristic dynamic evaluation method based on multi-source data.
The invention further provides a parking supply and demand characteristic dynamic evaluation system based on the multi-source data.
A parking supply and demand characteristic dynamic evaluation method based on multi-source data comprises the following steps:
STEP1: dividing parking communities and determining the total supply of parking resources of each parking community;
STEP2: cleaning the acquired vehicle access data of the sample parking lot, and acquiring the parking space occupancy rate index of the sample parking lot;
STEP3: based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property;
STEP4: carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district;
STEP5: drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period.
Preferably, the STEP1, dividing parking cells, and determining the total supply of parking resources in each parking cell comprises the following specific implementation processes:
dividing a target city area into a plurality of parking cellsBased on general survey data of all-city parking resources, analyzing and acquiring space distribution data of parking spaces of various types of parking lots in each parking district; marking the information sequence of the j parking lot of the i parking district as P ij ，P ij ＝{Location ij ,Type ij ,Land ij ,Num ij Location in which ij Is the position of the parking lot; type (Type) ij Type of parking lot ij ＝[1,2，3]1 represents a building matched parking lot, 2 represents a social public parking lot, and 3 represents an intra-road parking lot; land ij Land property of parking lot ij ＝[1,2,3,4,5]1 represents residence, 2 represents office, 3 represents business, 4 represents public service, 5 represents recreational entertainment; num (Num) ij Is the total number of berths in the parking lot.
Preferably, the STEP2 cleans the acquired vehicle access data of the sample parking lot, and the specific implementation process is as follows:
the sample parking data comprise social parking data and in-road parking data, wherein the social parking data mainly comprise charging data of a parking lot and a public parking lot matched with a building; the road parking data comprises high-order video charging data and POS machine charging data;
social parking data cleaning is to clean a vehicle V in a sample parking lot P ij The multiple access information of the vehicle V is subjected to data splicing, cleaning, judging and matching to obtain the vehicle V in the sample parking lot P ij The steps of the access field information of (a) are as follows:
1) Extracting vehicles V in sample parking lot P ij The license plate number is taken as the unique ID to splice a plurality of pieces of parking data to form a vehicle entrance-exit information sequence
Respectively representing the nth approach time and the nth departure time of the vehicle V;
2) Performing duplicate removal and null value removal processing on an incoming time sequence and an outgoing time sequence in the incoming and outgoing time sequence respectively;
3) Sequencing the incoming time sequence and the outgoing time sequence according to time, and decomposing the sequence into a plurality of incoming and outgoing parking data after matching
The cleaning of the parking data in the road comprises cleaning of high-order video charging data and cleaning of POS machine charging data;
wherein the high-order video charging data cleaning: the high-order video charging data is to record the time of entering and exiting each berth vehicle through the high-order video to obtain timing charging data; the data cleaning method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data are ordered according to the time of entering and exiting the parking place by taking the license plate number as a unique ID;
2) Removing abnormal data caused by batch approach and departure by using a maximum error interval of high-order video shooting; setting T g Maximum error interval for high-order video shooting if
Judging that the data is abnormal, the nth parking data and the (n+1) th parking data entering and exiting the parking space are recorded in the same parking space, and combining the nth parking data and the (n+1) th parking data entering and exiting the parking space into +.>
Cleaning charging data of a POS machine: the POS machine charging data is timing charging data obtained by recording the parking time of the vehicle in a charging period through a POS machine charging system; the data cleaning method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data are ordered according to the time of entering and exiting the parking place by taking the license plate number as a unique ID;
2) Judging and complementing parking data of vehicles entering and exiting the parking lot according to the charging time of the parking lot and the maximum error interval of a charging system; setting the charging starting time of a parking lot as h s The charging stop time is h e The maximum error interval of the POS charging system is T s According to the time logic, the following three judging and complementing conditions exist:
①
and->
Judging that the vehicle is overnight in the parking lot, and splicing the n-th parking data and the n+1-th parking data of the parking lot into the parking record of the same time and the parking data of the vehicle entering and exiting the parking lot into the parking record of +.>
②
Or if no next parking data exists, judging that the vehicle is out of the road in a non-charging period, and marking the actual out time of the vehicle as +.>
The parking data of the vehicle entering and exiting the parking place is +.>
③
Or no last parking data, < >>
Then it is determined that the vehicle is approaching in the non-charging period, the actual approach time of the vehicle is marked +.>
The parking data of the vehicle entering and exiting the parking place is +.>
Preferably, the calculating process of the berth occupancy rate of the sample parking lot in STEP2 in each period is as follows:
counting the berth occupation amount M of a sample parking lot in a period h by using the cleaned parking data ij，h And occupancy rate R ij，h An index; the steps of the berth occupancy rate calculation are as follows:
1) If it is
Then it is determined that the vehicle V is in the sample parking lot P in the period h ij Parking;
2) Statistical sample parking lot P ij The total number of vehicles parked in the period h is the occupied amount M of the berth ij,h ；
3) Calculating the occupancy rate R of berths ij,h ，R ij,h ＝M ij,h /Num ij ；
Because the parking data in the road has the condition that the time of entering and exiting the road in the non-charging period can not be accurately inferred, the inference result of the berth occupancy index in the non-charging period is smaller.
Preferably, the STEP3 calculates the parking cell parking occupancy scale in the initial period by:
selecting an initial period, and calculating the parking cell parking space occupation amount by using the sample parking space occupation rate of the initial period h=0; the method comprises the following specific steps:
1) Selecting an initial period of time
Drawing a time-varying distribution map of the occupancy of the entrance parking space and the exit parking space of a sample parking lot, selecting a period with good overall trend, less travel vehicles and stable parking behavior as an initial period, and simultaneously avoiding selecting a non-charging period for parking in a road as the initial period; the information sequence of the sample parking lot is denoted as P ij ＝{Location ij ,Type ij ,Land ij ,Num ij ,M ij,h ,R ij,h }；
2) Inferring non-sample parking lot initial period berth occupancy
Setting the equal-type and same-use parking lot berth utilization balance in a parking district, wherein the average value of the equal-type and same-use parking lots in the range of the peripheral distance l of the non-sample parking lot is as follows:
wherein R is ij，0 For an initial period berth utilization of a sample parking lot,
the average value of the berth utilization rate of the same type and the same-use property parking lot in the range of the peripheral distance l of the sample parking lot;
3) Calculating the occupied amount of berths in the initial period of each parking lot
Parking space occupation M of ith parking district and jth parking lot initial period ij,0 ＝R ij,0 ×Num ij ；
4) Calculating the occupied amount of berths in the initial period of a parking district
(1) Total supply of berths in the ith parking area:
(2) parking space occupation amount of the initial period of the ith parking district:
preferably, the STEP4 judges the parking requirement of the driving vehicle in each period of the parking district, and the specific process is as follows:
fusing the road high-definition bayonet and the parking operation data, judging and acquiring the traveling OD of the parking district in each period of the area, and studying and judging the parking requirements of the vehicle in each period of the area; the method comprises the following specific steps:
1) Acquiring a bayonet sensing sequence; preprocessing the bayonet data, and removing records with completely consistent information such as license plates, bayonet points and passing time; removing records of the same vehicle with adjacent record time less than 60 seconds in the same gateThe method comprises the steps of carrying out a first treatment on the surface of the Removing abnormal license plates and record of identification results such as no license plates; removing abnormal data that the space moving speed between adjacent bayonets is greater than the maximum speed threshold of the road network; extracting a bayonet-aware sequence of a vehicle V
Wherein K is v,q For the detected vehicle V q-th time, T v,q For the time of the vehicle V q-th time detected by the bayonet, m k Is the total number of times detected by the bayonet;
2) The parking data is converted into a parking sensing sequence; parking data after cleaning sample parking lot
Conversion to a perception sequence (K, T) v ＝{(P ij ,P ij ),(T en ,T ex )}；
3) The bayonet sensing sequence is fused with parking sensing data; taking a license plate number as a unique ID, inserting the parking sensing sequence into the bayonet sensing sequence according to sensing time, and obtaining the bayonet sensing sequence with parking lot access information;
4) Calculating a bayonet perception time information constraint threshold; selecting a bayonet pair with the identification number of average daily times larger than a constraint threshold as a bayonet pair, and calculating a bayonet time constraint threshold by utilizing a statistical box graph abnormal value discrimination principle:
wherein:
V max -maximum running speed of the road network, and taking a value according to actual running conditions of the road network;
V min -minimum running speed of the road network, and taking a value according to actual running conditions of the road network;
5) Judging parking behavior according to the gate time constraint threshold and the parking sensing sequence:
(1) judging and identifying the parking behavior of the vehicle according to the parking sensing sequence; if the q-th record and the q+1-th record in the sensing sequence of the vehicle V are parking records, judging that the vehicle has parking behaviors by taking the q-th record as the last travel end point and the q+1-th record as the next travel start point;
(2) judging and identifying the traveling and parking behaviors of the vehicle according to the space-time information constraint; if the interval time of adjacent bayonets in the sensing sequence of the vehicle V exceeds a threshold value, taking the last record as an end point, taking the next record as a starting point of the next trip, and judging that the vehicle stops;
(3) removing the passing travel demands according to the boundary streets; screening main external roads in the urban area, drawing corresponding external check lines, and considering the travel start point or the travel end point as the transit travel requirement if the travel start point or the travel end point is out of the check lines;
6) Extracting vehicle departure and arrival area information according to time intervals, wherein the vehicle departure quantity of the ith parking cell in the h time interval, namely the driving-off demand and the newly-increased berth supply quantity O i,h The vehicle arrival amount of the ith parking cell in the ith period is the parking requirement D of the running vehicle i,h 。
Preferably, the STEP5, the comprehensive evaluation process of parking supply and demand in each period of the parking district is as follows:
calculating the supply quantity and demand of berths in each period of a parking districtThe quantity is comprehensively evaluated to comprehensively evaluate the parking supply and demand conditions of each period of the parking district; berth demand of parking cell i and period h
Total berth supply Num i Occupancy rate R of berth i,h ＝Q i,h /Num i The method comprises the steps of carrying out a first treatment on the surface of the And drawing a time-varying chart of the parking supply and demand ratios of each parking district, evaluating the parking supply quantity and the demand quantity of each parking district in each period, and providing reference for optimizing the parking strategy.
A multi-source data based parking supply and demand feature dynamic assessment system, the system comprising:
the parking community parking resource total supply obtaining module: dividing parking communities and determining the total supply of parking resources of each parking community;
and a data cleaning module: cleaning the acquired vehicle access data of the sample parking lot, and acquiring the parking space occupancy rate index of the sample parking lot;
the large-scale acquisition module of the occupied parking berth of the parking community: based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property;
fusion analysis module: carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district;
and an evaluation module: drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the method for dynamically evaluating parking supply and demand characteristics based on multi-source data.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the method for dynamic assessment of parking supply and demand characteristics based on multi-source data.
Compared with the prior art, the invention has the beneficial effects that: the method is based on a multisource space-time data fusion and analysis technology, utilizes general survey data of parking facilities, high-definition road bayonet data and different types of parking lot distribution characteristic data of different areas, combines part of available sample parking lot access data, dynamically identifies time-varying supply and demand scale and space-time distribution characteristics of an object area, evaluates and analyzes parking supply and demand characteristics of the area, and provides reference for optimizing regional parking strategies. The invention has the following characteristics:
(1) The data is detailed and easy to obtain, the coverage area of the sample is wide, and time-varying parking supply and demand characteristic analysis in different areas can be realized.
(2) The method has the advantages of data driving performance, simple and mature algorithm and key technology and high processing efficiency.
(3) Because of the limitation of the distribution density of the bayonets and the available sample parking data, the analysis precision is limited, but the fault tolerance is higher.
Drawings
FIG. 1 is a general framework diagram of a method for dynamically evaluating parking supply and demand characteristics based on multi-source data.
Fig. 2 is a flow chart of parking cell travel OD calculation.
FIG. 3 is a diagram showing examples of the social parking lot parking data before and after cleaning in the embodiment.
Fig. 4 is a diagram showing an example of a vehicle entry and exit time distribution in a parking lot before and after cleaning according to an embodiment.
Fig. 5 is a diagram illustrating an example of parking cell berth occupancy distribution in an embodiment.
Detailed Description
The invention is further described below with reference to the drawings and detailed description.
The invention provides a multi-source data-based parking supply and demand characteristic dynamic evaluation method, which dynamically identifies the supply and demand time-varying characteristics of an object region, evaluates and analyzes the supply and demand characteristics of the region, provides reference for regional parking strategy optimization, and has an overall research framework shown in figure 1. STEP1: dividing parking communities and determining the total supply of parking resources of each parking community based on parking census data; STEP2: cleaning the acquired vehicle access data of the sample parking lot, and acquiring the parking space occupancy rate index of the sample parking lot; STEP3: based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property; STEP4: carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district; STEP5: drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period.
The specific implementation process of each step is as follows:
STEP1 divides parking cells and determines the parking resource supply amount of each parking cell
Dividing the urban area into a plurality of parking areas, and analyzing and acquiring berth space distribution data of various parking lots in each parking area based on general survey data of all-city parking resources. Marking the information sequence of the j parking lot of the i parking district as P ij ，P ij ＝{Location ij ,Type ij ,Land ij ,Num ij Location in which ij Is the position of the parking lot; type (Type) ij Type of parking lot ij ＝[1,2，3]1 represents a building matched parking lot, 2 represents a social public parking lot, and 3 represents an intra-road parking lot; land ij Land property of parking lot ij ＝[1,2,3,4,5]1 represents residence, 2 represents office, 3 represents business, 4 represents public service, 5 represents recreational entertainment; num (Num) ij Is the total number of berths in the parking lot.
In this embodiment, the classification of the type of the parking lot and the nature of the land mainly refers to the related specifications, and the classification of the parking lot based on the general survey data of the parking lot and the classification of the actual use condition of the parking lot can be combined in practical application.
STEP2 obtains occupancy index of parking lot berth
And acquiring vehicle access data of the sample parking lots, cleaning and analyzing the data, and calculating the berth occupancy indexes of each sample parking lot in each period. The sample parking data comprises social parking data and intra-road parking data, wherein the social parking data mainly comprises charging data of a parking lot and a public parking lot matched with a building, and the data is directly accessed into a traffic department intelligent management platform by a parking lot enterprise; the in-road parking data includes high-order video charging data and POS charging data, and the data is directly regulated by traffic departments.
(1) Social parking data cleaning
Description of data: the vehicle enters and exits the building to build a parking lot and a public parking lot, the vehicle entering time and the vehicle exiting time are recorded, and the vehicle V is marked in a sample parking lot P ij The nth parking data of (a) is
At present, as the charging running water of the parking lot consists of inlet running water and outlet running water, the condition of non-uniformity of the inlet running water and the outlet running water possibly exists, and the collected vehicle parking data has the following three display conditions:
3)
The departure time is not uploaded to be a null value, and the arrival time is erroneously uploaded.
Data cleaning: since there are situations of no upload and erroneous upload, it is necessary to place the vehicle V in the sample parking lot P ij Multiple entry and exit fields of (a)The information is subjected to data splicing, cleaning and judging matching to obtain a vehicle V in a sample parking lot P ij Is provided with access information. The method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The license plate number is taken as the unique ID to splice a plurality of pieces of parking data to form a vehicle entrance-exit information sequence
2) And performing duplicate removal and null value removal processing on the incoming time sequence and the outgoing time sequence in the incoming and outgoing time sequence respectively.
3) Sequencing the incoming time sequence and the outgoing time sequence according to time, and decomposing the sequence into a plurality of incoming and outgoing parking data after matching
(2) Cleaning of parking data in road
The in-road parking data comprises high-order video charging data and POS charging data.
High-order video charging data cleaning: the time charging data are recorded through high-order video to record the entrance and exit time of each berth vehicle. Due to the characteristic of the sectional charging mode of the parking in the road, the data of the data often has abnormal data phenomenon caused by batch approach and departure, and the information of the vehicles on the multiple approach and departure in the parking lot needs to be spliced according to time to acquire corrected vehicle approach and departure parking data. The method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data is ordered according to the time of entering and exiting the parking place by taking the license plate number as the unique ID.
2) And eliminating abnormal data caused by batch approach and departure by using a maximum error interval of high-order video shooting. Setting T g Maximum error interval for high-order video shooting if
Then determine that the data is abnormal, nth andthe n+1th parking data in and out are the same parking record, and the n and n+1th parking data in and out are combined into +.>
Cleaning charging data of a POS machine: the POS machine charging data is timing charging data realized by recording the parking time of the vehicle in a charging period through a POS machine charging system. Because the POS machine charging system only records the use condition of the berths in the charging period and lacks the information of entering and exiting the parking place in the non-charging period, the condition of entering and exiting the parking place in batches exists in the charging starting period and the charging ending period, and the information of entering and exiting the parking place for multiple times of vehicles in the parking lot is required to be judged, so that the completed and pushed vehicle entering and exiting parking data are obtained. The method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data is ordered according to the time of entering and exiting the parking place by taking the license plate number as the unique ID.
2) And judging and complementing the parking data of the vehicles entering and exiting the parking lot according to the charging time of the parking lot and the maximum error interval of the charging system. Setting the charging starting time of a parking lot as h s The charging stop time is h e The maximum error interval of the POS charging system is T s According to the time logic, the following three judging and complementing conditions exist:
①
and->
Judging that the vehicle is overnight in the parking lot, and splicing the n-th parking data and the n+1-th parking data of the parking lot into the parking record of the same time and the parking data of the vehicle entering and exiting the parking lot into the parking record of +.>
②
(or no next parking data), determining that the vehicle is coming out in the non-charging period, and marking the actual coming-out time of the vehicle as +.>
The parking data of the vehicle entering and exiting the parking place is +.>
③
(or no last parking data),>
then it is determined that the vehicle is approaching in the non-charging period, the actual approach time of the vehicle is marked +.>
The parking data of the vehicle entering and exiting the parking place is +.>
(3) Calculation of berth occupancy rate of sample parking lot at each time interval
Counting the berth occupation amount M of a sample parking lot in a period h by using the cleaned parking data ij，h And occupancy rate R ij，h And (5) an index. The steps of the berth occupancy rate calculation are as follows:
1) If it is
Then it is determined that the vehicle V is in the sample parking lot P in the period h ij And (5) parking.
2) Statistical sample parking lot P ij The total number of vehicles parked in the period h is the occupied amount M of the berth ij,h 。
3) Calculating the occupancy rate R of berths ij,h ，R ij,h ＝M ij,h /Num ij 。
Because the parking data in the road has the condition that the time of entering and exiting the road in the non-charging period can not be accurately inferred, the inference result of the berth occupancy index in the non-charging period is smaller.
In this embodiment, the social parking data cleaning is based on the acquired parking data of the social parking lot, and in practical application, cleaning is performed by combining the acquired data format of the sample parking lot and the charging feature setting parameters, for example, fig. 3 before and after cleaning the parking data of the social parking lot;
secondly, in this embodiment, the parking data cleaning of the parking lot in the road is based on the acquired high-level video and charging data of the POS machine, the charging start time of the parking lot is 7:30, the charging stop time is 21:30, the maximum error interval is ±30min, the actual application should be cleaned in combination with the acquired data format and charging feature setting parameters of the sample parking lot, and the vehicle access time distribution of a certain parking lot before and after cleaning is shown in fig. 4;
STEP3 calculates the occupied scale of parking cell berth in the initial period
And selecting an initial period, and calculating the parking cell parking space occupation amount by using the sample parking space occupation rate of the initial period h=0. The method comprises the following specific steps:
1) Selecting an initial period of time
And drawing a time-varying distribution diagram of the occupancy of the entrance parking space and the exit parking space of the sample parking lot, selecting a time period with good overall trend, less travel vehicles and stable parking behavior as an initial time period, and simultaneously avoiding selecting a non-charging time period for parking in a road as the initial time period. The information sequence of the sample parking lot can be represented as P ij ＝{Location ij ,Type ij ,Land ij ,Num ij ,M ij,h ,R ij,h }。
2) Inferring non-sample parking lot initial period berth occupancy
Setting the utilization balance of the parking spaces with the same type and the same type in the parking area, wherein the utilization ratio of the parking spaces with the same type and the same type in the non-sample parking area in the initial period is the average value of the parking spaces with the same type and the same type in the peripheral distance l
Wherein R is ij，0 For the initial period berth utilization of the sample parking lot, < >>
The average value of the berth utilization rate of the same type and the same-use property parking lot in the range of the peripheral distance l of the sample parking lot.
3) Calculating the occupied amount of berths in the initial period of each parking lot
Parking space occupation M of ith parking district and jth parking lot initial period ij,0 ＝R ij,0 ×Num ij 。
4) Calculating the occupied amount of berths in the initial period of a parking district
(1) Total supply of berths in the ith parking area:
(2) parking space occupation amount of the initial period of the ith parking district:
in this embodiment, a method for setting the parking lot berth utilization balance of the same type and the same type in the parking area and calculating the parking berth utilization rate of an unknown parking lot, wherein the peripheral distance l is the radiation range of the parking area, and other applicable methods can be selected during practical application, and the parking area berth occupancy distribution is shown in fig. 5;
STEP4 study and judge parking requirements of vehicles running in each period of parking district
And merging the road high-definition bayonets and the parking operation data, judging and acquiring the traveling OD of the parking district in each period of the area, and studying and judging the parking requirements of the vehicle in each period of the area. The method comprises the following specific steps:
1) And acquiring a bayonet sensing sequence. Preprocessing the bayonet data, and removing records with completely consistent information such as license plates, bayonet points, vehicle passing time and the like; removing records of the same vehicle with adjacent record time less than 60 seconds in the same gate; removing abnormal license plates, license-free identification result records and the like; removing between adjacent bayonetAbnormal data of which the space moving speed is greater than the maximum speed threshold of the road network; finally extracting bayonet perception sequence of vehicle V
Wherein K is v,q For the detected vehicle V q-th time, T v,q For the time of the vehicle V q-th time detected by the bayonet, m k Is the total number of times that is detected by the bayonet.
2) The parking data is converted into a parking awareness sequence. Parking data after cleaning sample parking lot
Conversion to a perception sequence (K, T) v ＝{(P ij ,P ij ),(T en ,T ex )}。
3) The bayonet perception sequence is fused with the parking perception data. And taking the license plate number as a unique ID, inserting the parking sensing sequence into the bayonet sensing sequence according to the sensing time, and obtaining the bayonet sensing sequence with the parking lot access information.
4) And calculating a constraint threshold of the bayonet perception on the time information. Selecting a bayonet pair with the identification number of average daily times larger than a constraint threshold as a bayonet pair, and calculating a bayonet time constraint threshold by utilizing a statistical box graph abnormal value discrimination principle:
wherein:
V max -maximum running speed of the road network, and taking a value according to actual running conditions of the road network;
V min -minimum running speed of the road network, and taking value according to actual running condition of the road network.
5) Judging parking behavior according to the gate time constraint threshold and the parking sensing sequence:
(1) and judging and identifying the parking behavior of the vehicle according to the parking sensing sequence. If the q-th record and the q+1-th record in the sensing sequence of the vehicle V are parking records, the q-th record is taken as the last trip end point, and the q+1-th record is taken as the next trip start point, so that the parking behavior of the vehicle is judged.
(2) And judging and identifying the traveling and parking behaviors of the vehicle according to the space-time information constraint. If the interval time of adjacent bayonets in the sensing sequence of the vehicle V exceeds the threshold value, the last record is taken as an end point, the next record is taken as the starting point of the next trip, and the parking behavior of the vehicle is judged.
(3) And eliminating the passing travel requirement according to the boundary street. Screening the urban area mainly comprises the steps of drawing corresponding external check lines on external roads, and considering the external check lines as the transit travel requirement if the travel starting point or the travel terminal point is out of the check lines.
6) Extracting vehicle departure and arrival area information according to time intervals, wherein the vehicle departure quantity of the ith parking cell in the h time interval, namely the driving-off demand and the newly-increased berth supply quantity O i,h The vehicle arrival amount of the ith parking cell in the ith period is the parking requirement D of the running vehicle i,h 。
In this embodiment, the method of calculating the parking requirement of the driving vehicle by calculating the frequent traffic pattern sensing time-space information constraint threshold value selects the traffic pattern pair with the average number of times of day greater than 1000 times per day to be combined into the traffic pattern pair, the lowest running speed in the road network is 5km/h, the highest running speed is 120km/h, and other applicable methods can be selected in practical application.
STEP5 parking district parking supply and demand comprehensive evaluation at each time period
Calculating the berth supply and demand of each period of a parking district, and comprehensively evaluating the parking supply and demand conditions of each period of the parking district; berth demand of parking cell i and period h
Total berth supply Num i Occupancy rate R of berth i,h ＝Q i,h /Num i The method comprises the steps of carrying out a first treatment on the surface of the And drawing a time-varying chart of the parking supply and demand ratios of each parking district, evaluating the parking supply quantity and the demand quantity of each parking district in each period, and providing reference for optimizing the parking strategy.
In this embodiment, the supply and demand conditions of the parking cells are evaluated by drawing a time-varying map of the parking supply and demand ratio of each parking cell, and other applicable indexes may be selected in practical application.
The embodiments of the present invention described above do not limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit principles of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A method for dynamically evaluating parking supply and demand characteristics based on multi-source data, comprising the steps of:
STEP1: dividing parking communities and determining the total supply of parking resources of each parking community;
STEP2: cleaning the acquired vehicle access data of the sample parking lot, and acquiring the parking space occupancy rate index of the sample parking lot;
STEP3: based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property;
STEP4: carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district;
STEP5: drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period.
2. The method of claim 1, wherein the STEP1, dividing the parking cells, and determining the total supply of parking resources of each parking cell is implemented by:
dividing a target city area into a plurality of parking cells, and analyzing and acquiring berth space distribution data of various parking lots in each parking cell based on general survey data of all-city parking resources; marking the information sequence of the j parking lot of the i parking district as P ij ，P ij ＝{Location ij ,Type ij ,Land ij ,Num ij Location in which ij Is the position of the parking lot; type (Type) ij Type of parking lot ij ＝[1,2，3]1 represents a building matched parking lot, 2 represents a social public parking lot, and 3 represents an intra-road parking lot; land ij Land property of parking lot ij ＝[1,2,3,4,5]1 represents residence, 2 represents office, 3 represents business, 4 represents public service, 5 represents recreational entertainment; num (Num) ij Is the total number of berths in the parking lot.
3. The method according to claim 1 or 2, wherein the STEP2 cleans the acquired vehicle entry and exit data of the sample parking lot, and the specific implementation process is as follows:
the sample parking data comprise social parking data and in-road parking data, wherein the social parking data mainly comprise charging data of a parking lot and a public parking lot matched with a building; the road parking data comprises high-order video charging data and POS machine charging data;
social parking data cleaning is to clean a vehicle V in a sample parking lot P ij The multiple access information of the vehicle V is subjected to data splicing, cleaning, judging and matching to obtain the vehicle V in the sample parking lot P ij The steps of the access field information of (a) are as follows:
1) Extracting vehicles V in sample parking lot P ij The license plate number is taken as the unique ID to splice a plurality of pieces of parking data to form a vehicle entrance-exit information sequence
Respectively representing the nth approach time and the nth departure time of the vehicle V;
2) Performing duplicate removal and null value removal processing on an incoming time sequence and an outgoing time sequence in the incoming and outgoing time sequence respectively;
3) Sequencing the incoming time sequence and the outgoing time sequence according to time, and decomposing the sequence into a plurality of incoming and outgoing parking data after matching
The cleaning of the parking data in the road comprises cleaning of high-order video charging data and cleaning of POS machine charging data;
wherein the high-order video charging data cleaning: the high-order video charging data is to record the time of entering and exiting each berth vehicle through the high-order video to obtain timing charging data; the data cleaning method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data are ordered according to the time of entering and exiting the parking place by taking the license plate number as a unique ID;
2) Removing abnormal data caused by batch approach and departure by using a maximum error interval of high-order video shooting; setting T g Maximum error interval for high-order video shooting if
Judging that the data is abnormal, the nth and the (n+1) th parking data entering and exiting the parking space are recorded in the same parking space, and combining the nth and the (n+1) th parking data entering and exiting the parking spaceIs->
Cleaning charging data of a POS machine: the POS machine charging data is timing charging data obtained by recording the parking time of the vehicle in a charging period through a POS machine charging system; the data cleaning method comprises the following specific steps:
1) Extracting vehicles V in sample parking lot P ij The parking data are ordered according to the time of entering and exiting the parking place by taking the license plate number as a unique ID;
2) Judging and complementing parking data of vehicles entering and exiting the parking lot according to the charging time of the parking lot and the maximum error interval of a charging system; setting the charging starting time of a parking lot as h s The charging stop time is h e The maximum error interval of the POS charging system is T s According to the time logic, the following three judging and complementing conditions exist:
①
and->
Judging that the vehicle is overnight in the parking lot, and splicing the n-th parking data and the n+1-th parking data of the parking lot into the parking record of the same time and the parking data of the vehicle entering and exiting the parking lot into the parking record of +.>
②
Or if no next parking data exists, judging that the vehicle is out of the road in a non-charging period, and marking the actual out time of the vehicle as +.>
The parking data of the vehicle entering and exiting the parking place is +.>
③
Or no last parking data, < >>
Then it is determined that the vehicle is approaching in the non-charging period, the actual approach time of the vehicle is marked +.>
The parking data of the vehicle entering and exiting the parking place is +.>
4. The method of claim 3, wherein the calculating process of the occupancy of the berths of the sample parking lot in each period in STEP2 is:
counting the occupied quantity M and the occupied rate R of the berths in the period h of the sample parking lot by using the cleaned parking data
ij, h ij, h index; the steps of the berth occupancy rate calculation are as follows:
1) If it is
Then it is determined that the vehicle V is in the sample parking lot P in the period h ij Parking;
2) Statistical sample parking lot P ij The total number of vehicles parked in the period h is the occupied amount M of the berth ij,h ；
3) Calculating the occupancy rate R of berths ij,h ，R ij,h ＝M ij,h /Num ij ；
Because the parking data in the road has the condition that the time of entering and exiting the road in the non-charging period can not be accurately inferred, the inference result of the berth occupancy index in the non-charging period is smaller.
5. The method of claim 4, wherein the STEP3 calculates the parking cell parking occupancy scale for the initial period by:
selecting an initial period, and calculating the parking cell parking space occupation amount by using the sample parking space occupation rate of the initial period h=0; the method comprises the following specific steps:
1) Selecting an initial period of time
Drawing a time-varying distribution map of the occupancy of the entrance parking space and the exit parking space of a sample parking lot, selecting a period with good overall trend, less travel vehicles and stable parking behavior as an initial period, and simultaneously avoiding selecting a non-charging period for parking in a road as the initial period; the information sequence of the sample parking lot is denoted as P ij ＝{Location ij ,Type ij ,Land ij ,Num ij ,M ij,h ,R ij,h }；
2) Inferring non-sample parking lot initial period berth occupancy
Setting the equal-type and same-use parking lot berth utilization balance in a parking district, wherein the average value of the equal-type and same-use parking lots in the range of the peripheral distance l of the non-sample parking lot is as follows:
wherein R is ij，0 For an initial period berth utilization of a sample parking lot,
the average value of the berth utilization rate of the same type and the same-use property parking lot in the range of the peripheral distance l of the sample parking lot;
3) Calculating the occupied amount of berths in the initial period of each parking lot
Ith parking cell, jth parking cellBerth occupation amount M of initial period of parking lot ij,0 ＝R ij,0 ×Num ij ；
4) Calculating the occupied amount of berths in the initial period of a parking district
(1) Total supply of berths in the ith parking area:
(2) parking space occupation amount of the initial period of the ith parking district:
6. the method according to claim 5, wherein said STEP4 is used for studying and judging the parking requirements of the vehicles in each period of the parking area, and comprises the following specific STEPs:
fusing the road high-definition bayonet and the parking operation data, judging and acquiring the traveling OD of the parking district in each period of the area, and studying and judging the parking requirements of the vehicle in each period of the area; the method comprises the following specific steps:
1) Acquiring a bayonet sensing sequence; preprocessing the bayonet data, and removing records with completely consistent information such as license plates, bayonet points and passing time; removing records of the same vehicle with adjacent record time less than 60 seconds in the same gate; removing abnormal license plates and record of identification results such as no license plates; removing abnormal data that the space moving speed between adjacent bayonets is greater than the maximum speed threshold of the road network; extracting a bayonet-aware sequence of a vehicle V
Wherein K is v,q For the detected vehicle V q-th time, T v,q For the time of the vehicle V q-th time detected by the bayonet, m k Is the total number of times detected by the bayonet;
2) The parking data is converted into a parking sensing sequence; parking data after cleaning sample parking lot
Conversion to a perception sequence (K, T) v ＝{(P ij ,P ij ),(T en ,T ex )}；
3) The bayonet sensing sequence is fused with parking sensing data; taking a license plate number as a unique ID, inserting the parking sensing sequence into the bayonet sensing sequence according to sensing time, and obtaining the bayonet sensing sequence with parking lot access information;
4) Calculating a bayonet perception time information constraint threshold; selecting a bayonet pair with the identification number of average daily times larger than a constraint threshold as a bayonet pair, and calculating a bayonet time constraint threshold by utilizing a statistical box graph abnormal value discrimination principle:
wherein:
V max -maximum running speed of the road network, and taking a value according to actual running conditions of the road network;
V min -minimum running speed of the road network, and taking a value according to actual running conditions of the road network;
5) Judging parking behavior according to the gate time constraint threshold and the parking sensing sequence:
(1) judging and identifying the parking behavior of the vehicle according to the parking sensing sequence; if the q-th record and the q+1-th record in the sensing sequence of the vehicle V are parking records, judging that the vehicle has parking behaviors by taking the q-th record as the last travel end point and the q+1-th record as the next travel start point;
(2) judging and identifying the traveling and parking behaviors of the vehicle according to the space-time information constraint; if the interval time of adjacent bayonets in the sensing sequence of the vehicle V exceeds a threshold value, taking the last record as an end point, taking the next record as a starting point of the next trip, and judging that the vehicle stops;
(3) removing the passing travel demands according to the boundary streets; screening main external roads in the urban area, drawing corresponding external check lines, and considering the travel start point or the travel end point as the transit travel requirement if the travel start point or the travel end point is out of the check lines;
6) Extracting vehicle departure and arrival area information according to time intervals, wherein the vehicle departure quantity of the ith parking cell in the h time interval, namely the driving-off demand and the newly-increased berth supply quantity O i,h The vehicle arrival amount of the ith parking cell in the ith period is the parking requirement D of the running vehicle i,h 。
7. The method of claim 6, wherein the STEP5, the parking supply and demand comprehensive evaluation process for each period of the parking cell is as follows:
calculating the berth supply and demand of each period of a parking district, and comprehensively evaluating the parking supply and demand conditions of each period of the parking district; berth demand of parking cell i and period h
Total berth supply Num i Occupancy rate R of berth i,h ＝Q i,h /Num i The method comprises the steps of carrying out a first treatment on the surface of the Drawing a time-varying chart of the parking supply and demand ratios of each parking district, and evaluating each parking districtAnd the parking supply and demand of each period of the zone provide references for optimizing the parking strategy.
8. A multi-source data based parking supply and demand feature dynamic assessment system, the system comprising:
the parking community parking resource total supply obtaining module: dividing parking communities and determining the total supply of parking resources of each parking community;
and a data cleaning module: cleaning the acquired vehicle access data of the sample parking lot, and acquiring the parking space occupancy rate index of the sample parking lot;
the large-scale acquisition module of the occupied parking berth of the parking community: based on the type distribution and land feature data of each parking lot in the parking census data, calculating the occupied parking scale of each parking district by using the parking space occupancy rate indexes of the sample parking lots with the same type and land property;
fusion analysis module: carrying out fusion analysis on the road high-definition bayonet data and the parking operation data, judging and obtaining the traveling OD of each time period of the parking district, and judging the vehicle parking requirement of each time period h in each parking district;
and an evaluation module: drawing a time-varying graph of the supply and demand of the parking, measuring and calculating supply and demand evaluation indexes, and comprehensively evaluating the supply and demand conditions of the parking in each region in each period.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the multi-source data based parking supply and demand feature dynamic assessment method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements a method for dynamic evaluation of parking supply and demand characteristics based on multi-source data as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211579767.5A CN115995149A (en) | 2022-12-06 | 2022-12-06 | Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211579767.5A CN115995149A (en) | 2022-12-06 | 2022-12-06 | Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115995149A true CN115995149A (en) | 2023-04-21 |
Family
ID=85991471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211579767.5A Pending CN115995149A (en) | 2022-12-06 | 2022-12-06 | Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115995149A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117173898A (en) * | 2023-11-03 | 2023-12-05 | 深圳市城市交通规划设计研究中心股份有限公司 | Method for extracting travel OD (optical density) based on parking lot flow data in time-division and destination-division manner |
| CN117274886A (en) * | 2023-08-04 | 2023-12-22 | 中铁城市规划设计研究院有限公司 | Data analysis method for current parking supply and demand resource relationship |
-
2022
- 2022-12-06 CN CN202211579767.5A patent/CN115995149A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117274886A (en) * | 2023-08-04 | 2023-12-22 | 中铁城市规划设计研究院有限公司 | Data analysis method for current parking supply and demand resource relationship |
| CN117274886B (en) * | 2023-08-04 | 2024-03-19 | 中铁城市规划设计研究院有限公司 | Data analysis method for current parking supply and demand resource relationship |
| CN117173898A (en) * | 2023-11-03 | 2023-12-05 | 深圳市城市交通规划设计研究中心股份有限公司 | Method for extracting travel OD (optical density) based on parking lot flow data in time-division and destination-division manner |
| CN117173898B (en) * | 2023-11-03 | 2024-02-27 | 深圳市城市交通规划设计研究中心股份有限公司 | Method for extracting travel OD (optical density) based on parking lot flow data in time-division and destination-division manner |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111243277B (en) | Commuting vehicle space-time trajectory reconstruction method and system based on license plate recognition data | |
| Sun et al. | Analyzing spatiotemporal traffic line source emissions based on massive didi online car-hailing service data | |
| CN108346292B (en) | Urban expressway real-time traffic index calculation method based on checkpoint data | |
| CN115995149A (en) | Multi-source data-based parking supply and demand characteristic dynamic evaluation method and system | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| CN111008505B (en) | Urban ramp driving condition construction method and application | |
| CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
| CN107729938B (en) | Rail station classification method based on bus connection radiation zone characteristics | |
| CN101710448A (en) | Road traffic state detecting device based on omnibearing computer vision | |
| Tiedemann et al. | Concept of a data thread based parking space occupancy prediction in a berlin pilot region | |
| CN116631186B (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
| Lin et al. | Path travel time estimation using attribute-related hybrid trajectories network | |
| CN114091581A (en) | Vehicle operation behavior type identification method based on sparse track | |
| CN112767686B (en) | Road network automobile emission estimation method based on multi-source data fusion | |
| Zhao et al. | A machine learning based approach for predicting usage efficiency of shared e-scooters using vehicle availability data | |
| CN116415756B (en) | Urban virtual scene experience management system based on VR technology | |
| CN116863744A (en) | Intelligent berth scheduling management method and system | |
| CN110659774B (en) | Big data method driven parking demand prediction method | |
| Tian et al. | Spatial analysis of commuting carbon emissions in main urban area of Beijing: A GPS trajectory-based approach | |
| Zhao et al. | Understanding the spatio-temporally heterogeneous effects of built environment on urban travel emissions | |
| Naji et al. | Towards understanding the impact of human mobility patterns on taxi drivers' income based on GPS data: A case study in Wuhan—China | |
| CN114407661A (en) | Data-driven electric vehicle energy consumption prediction method, system, device and medium | |
| CN117473398B (en) | Urban dust pollution source classification method based on slag transport vehicle activity | |
| Zhao et al. | A Parking Utilization Prediction Approach Based on NARNN Model | |
| Riveros et al. | Mobility Studies in Villavicencio: Situation and Prospects Towards a Smart Mobility. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |