A travel characteristics analysis method based on provincial multi-source spatial data

Abstract

The present invention provides a travel characteristics analysis method based on provincial multi-source spatial data, including: collecting basic data; for mobile phone data with high user proportion and high triggering, using big data mining to generate user travel station data that meets the characteristics of time-space travel activities; conducting travel characteristics analysis of different transportation modes according to different data characteristics; using multi-dimensional space differentiated hierarchical analysis and verification methods to analyze the travel characteristics of data sources; completing the multi-level travel OD data verification from macro to micro provincial-bay area-city; verifying the data of trunk traffic corridors; generating the final provincial multi-dimensional space travel characteristics data analysis results. The present invention combines the advantages of big data and new technologies, utilizes the coupling relationship of provincial hierarchical and differentiated characteristic data, realizes the mutual verification between provincial space big data and traditional survey data, further improves the accuracy of regional multi-dimensional space data analysis, and provides data support for regional multi-dimensional space planning.

Description

Technical Field

[0001] The present invention relates to the technical field of provincial multi-dimensional spatial traffic planning and traffic demand management, and specifically to a travel characteristic analysis method based on provincial multi-source spatial data, which can provide big data support for provincial multi-dimensional spatial traffic planning and traffic demand management. Background Art

[0002] With the application of big data and GIS spatial analysis technology, great progress has been made in data mining and spatial structure analysis, but it still faces the challenge of extracting effective information from massive dynamic and redundant big data, and adapting large sample data to small sample survey data. At the same time, in the context of globalization, urban agglomerations have become the core units participating in global competition, and the formation and development of urban agglomerations also play an important role in enhancing the competitive position of the global network. Therefore, with the research on regional integration, urbanization and bay area coordinated development strategies, urban transportation planning decisions have broken through administrative boundaries, and big data analysis also needs to adapt to regional integration analysis, realize deep data mining of macro-micro nested analysis, and complete the analysis of large-scale complex spatiotemporal problems between transportation and spatial structure from a multi-dimensional perspective of time and space, promote regional transportation integration development, and improve transportation operation efficiency.

[0003] Traditional provincial transportation spatial travel analysis mainly relies on traffic surveys, highway toll data and hub operation data. The data obtained is of coarse granularity, small sample size, and lacks effective data verification. With the support of big data, data acquisition channels are more diverse, and individual travel data can be obtained, with richer sample size and finer granularity. Through the integration of traditional survey data and big data, each data can be mutually verified and calibrated, and data analysis is more comprehensive and reliable. Therefore, it is necessary to combine the data analysis capabilities of big data for the overall macro situation and large samples with the fine small sample data analysis mechanism to achieve the technical integration of macro situation and fine mechanism.

[0004] Since regional spatial big data covers all-round and different granularity data sources such as mobile phone signaling, Internet location, road traffic information, traditional survey data, etc., it has problems such as inconsistent data sources, data structures, data sampling rates, data spatial scopes, and data granularity. Therefore, it cannot be obtained by expanding the sample of a single data. Instead, it is necessary to combine the characteristics of multi-source data and the regional spatial scope, and perform spatiotemporal multi-source data feature processing in layers to support regional multi-dimensional spatial travel feature analysis and verification. Summary of the invention

[0005] Taking into account that big data with wide coverage and high triggering can make up for the shortcomings of traditional survey data, the present invention proposes a travel characteristic analysis method based on provincial multi-source spatial data, and utilizes the coupling relationship of provincial hierarchical and differentiated characteristic data to realize mutual verification between provincial spatial big data and traditional survey data, further improve the accuracy of regional multi-dimensional spatial data analysis, and provide data support for regional multi-dimensional spatial planning.

[0006] The present invention provides a travel characteristics analysis method based on provincial multi-source spatial data, and the specific steps are as follows:

[0007] Step 1: Collect basic information: Collect multi-source data of different spatial dimensions from the national, provincial to municipal levels;

[0008] Step 2: For mobile phone data with high user ratio and high triggering, big data mining and processing are used, combined with the displacement state judgment and cluster analysis of mobile phone user travel time series, to judge the travel residence state of the displacement point and generate user travel residence point data that meets the characteristics of time-space travel activities;

[0009] Step 3: For Internet and traditional survey data, conduct provincial multi-dimensional space analysis of different transportation modes and travel characteristics of different transportation modes according to different data characteristics;

[0010] Step 4: For the provincial spatial analysis and verification of multi-source data, the hierarchical analysis and verification method of multi-dimensional space differentiation is used to first analyze the travel characteristics of different data sources in multi-dimensional space;

[0011] Step 5: Conduct multidimensional data mining and travel characteristics analysis on mobile phone data in terms of spatial scope and user type, expand the sample according to the age and gender structure of mobile phone users and user proportion, and combine the multidimensional spatial data travel OD data of different transportation modes of Internet and survey data to complete the multi-level travel OD data verification from macro to micro level of province-bay area-city area;

[0012] The spatial scope includes provinces, bay areas, metropolitan areas, city areas, districts, streets and towns, and transportation communities; the user types include permanent residents, non-permanent residents, short-term floating population, hub arrivals and departures, and transit population;

[0013] In terms of space, the multi-level travel OD data verification model from macro to micro level, from province to bay area to city, is expressed as follows:

[0014]

[0015] In the provincial multidimensional spatial data analysis and verification, α refers to the traffic flow of the road sections that constitute the traffic corridor. In the spatial dimension, the path flow data of the provincial external travel corridor within a certain time range must be realized. Path flow data of provincial inter-city travel corridors and route flow data of travel corridors within provincial cities Hierarchical and differentiated data fusion and verification;

[0016] Based on the provincial multidimensional spatial travel data, combined with the provincial travel characteristic data results α of different transportation modes generated by the Internet and traditional survey data, the process of minimizing the difference of multi-source data in the multidimensional space is obtained within a certain time range, that is, min α F α (α,x(α)), that is, to realize the spatial travel data verification at different levels and generate provincial multi-dimensional spatial travel characteristic data;

[0017] Step 6: Based on the verification results of the provincial multidimensional space OD travel characteristic data, the provincial traffic model is used to perform road network traffic allocation for different traffic modes in the province. Then, the data analysis results of different provincial traffic modes based on the Internet and survey data are combined to verify the data of arterial traffic corridors. The model expression of the arterial traffic corridor data verification in road network traffic allocation is as follows:

[0018]

[0019] In step 6, the verification of the traffic data of the main corridor based on the provincial multidimensional spatial data traffic distribution requires the realization of the traffic data x of the impedance weight a of different provincial traffic modes within a certain time range in the spatial dimension. a (α) hierarchical and differentiated data fusion and verification; among them, C a is the impedance of different transportation modes, w is bounded by Different traffic data;

[0020] Based on the road traffic network, the traffic flow data of all roads included in the traffic corridor a of different transportation modes in the province within a certain time range is combined Route selection information of all roads included in the transportation corridor a with different transportation modes in the province within a certain time range The process of minimizing the flow difference of multi-source data in the road network allocation corridor within a certain time range, namely minZ L (α), that is, to realize the verification of road flow data at different levels and generate provincial multi-dimensional spatial trunk road traffic flow data;

[0021] Step 7: Generate the final provincial multi-dimensional space travel characteristic data analysis results through a top-down hierarchical data fusion and feature feedback system from macro to micro provincial-bay area-city multi-dimensional space.

[0022] Preferably, the multi-source data is obtained from:

[0023] In terms of spatial scope, data of different spatial dimensions are collected from the national, provincial to municipal levels;

[0024] At the national level, it includes Tencent migration data at the national level, provincial to national railway arrival and departure data, and airport passenger flow distribution data;

[0025] At the provincial level, it includes provincial highway network toll data, Tencent travel data, mobile phone signaling data, Amap vehicle speed and GPS data;

[0026] At the city level, it includes population portrait data, resident travel characteristics data, checkpoint vehicle identification data, bus IC card and subway gate data, road traffic operation survey data, and population census data;

[0027] In terms of data types, it includes big data and traditional survey data, among which big data accounts for more than 70%.

[0028] Preferably, in step 2, the stationary point identification model expression based on time-space displacement and cluster analysis is as follows:

[0029]

[0030] For the time series data of the same user number, the K-means time-space clustering method with the minimum mean square error of the objective function in the continuous time range is used. j That is, it is the center of the jth feature time series vector cluster, and p is the index that determines the weight effect. C j arrive The distance It is the cluster center distance weight to perform spatiotemporal clustering analysis of user trajectory points; combined with the time, distance and speed judgment of each cluster point cluster, an effective user travel residence point set is formed to generate an individual user travel OD table based on mobile phone data analysis.

[0031] Preferably, step 3 is specifically as follows: based on highway toll data and hub ticketing data, OD analysis of trunk traffic corridors and hub travel OD analysis of different spatial dimensions are performed; based on Internet data, flow, speed, and saturation analysis of trunk traffic corridors of different spatial dimensions are performed; based on traditional survey data, road flow, spatiotemporal distribution, and vehicle travel OD analysis, as well as spatiotemporal distribution and travel OD analysis of subways and buses are performed; based on resident travel surveys, population censuses, and socioeconomic data, population work-residence characteristics analysis of different spatial dimensions, travel characteristics analysis of different age structures, and travel proportion analysis of different modes of transportation are performed.

[0032] Preferably, in step 7, after generating the final provincial multi-dimensional space travel characteristic data analysis results, the provincial multi-dimensional space dimensions and user characteristics can be combined to complete the provincial multi-dimensional travel characteristic analysis of different spatial dimensions and different users. The provincial multi-dimensional travel characteristic analysis includes provincial travel characteristic analysis, provincial commuting characteristic analysis, metropolitan area analysis or same-city analysis.

[0033] The present invention combines the characteristics of provincial spatial multi-source big data and the regional multi-dimensional spatial scope, and proposes a travel characteristics analysis method based on provincial multi-source spatial data to realize the processing and verification of provincial spatial multi-dimensional spatiotemporal data, which greatly makes up for the extensiveness of traditional survey data, further broadens the spatial boundary of data analysis, improves the accuracy of data analysis, and provides data support for provincial multi-dimensional spatial planning and decision-making; the travel characteristics analysis method based on provincial multi-source spatial data of the present invention has the following beneficial effects:

[0034] 1) The present invention is based on provincial spatial big data analysis. Compared with traditional traffic surveys, it can greatly make up for the defects of small samples and long-period static data in traditional surveys, and realize real-time dynamic update of data in a short period of time;

[0035] 2) Based on the hierarchical processing and verification technology of provincial spatial big data and from the perspective of regional coordinated development, the present invention conducts characteristic analysis and verification of provincial multi-dimensional spatiotemporal and multi-source data, which is helpful for the study of transportation integration under the background of regional coordination and provides a data basis for regional systematic research and data analysis.

[0036] 3) The present invention combines provincial multi-dimensional spatial data analysis methods, uses big data and new technologies to break through administrative boundaries, and supports regional spatial strategic planning, which is conducive to improving the scientific nature of provincial multi-dimensional spatial planning decisions, promoting regional transportation connections, and improving the utilization rate and traffic efficiency of transportation facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The present invention discloses a flow chart of a travel characteristics analysis method based on provincial multi-source spatial data. DETAILED DESCRIPTION

[0038] The specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Figure 1 The present invention provides a travel characteristics analysis method based on provincial multi-source spatial data, and the specific implementation steps are as follows:

[0039] Step 1: In order to obtain comprehensive and wide-coverage basic data resources, in terms of basic data collection, in terms of spatial scope, the collected information resources include Tencent migration data at the national level, provincial to national railway arrival and departure data, airport passenger flow distribution data, provincial highway network toll data, Tencent travel data, mobile phone signaling data, AutoNavi vehicle speed and GPS data, city-level population portrait data, resident travel characteristics data, checkpoint vehicle identification data, bus IC card and subway gate data, road traffic operation survey data, population census data, etc.; in terms of type, it covers traditional survey data, intelligent collection data and new big data, of which big data accounts for more than 70%, as follows:

[0040] Mobile phone data: 29 billion pieces of mobile phone data per day, and 2.1 billion pieces of mobile phone data per day for China Unicom;

[0041] Internet data: 1.35 billion pieces of Tencent positioning data and Amap navigation data per day;

[0042] Road traffic information data: 816 expressway toll stations, about 200 million expressway toll flow data, more than 7,900 high-definition road checkpoints, and an average of 25 million checkpoint flow data per day;

[0043] IC data: 150 million card swipe data and 140 million station announcement data were obtained for regular buses, and 110 million subway card swipe data were obtained for subways;

[0044] Taxi data: 9 billion pieces of GPS data, 22 million pieces of taximeter data.

[0045] Traditional survey data: resident travel survey data, the seventh national census data, socioeconomic data, railway hub and airport ticketing data, etc.

[0046] Step 2: For mobile phone data with high user ratio and high triggering, we mainly rely on big data mining and processing technology, combined with the displacement state judgment and cluster analysis of mobile phone user travel time series, to realize the judgment of the travel residence state of the displacement point, so as to generate user travel residence point data that meets the characteristics of time-space travel activities. The expression of the residence point identification model based on time-space displacement and cluster analysis is as follows:

[0047]

[0048] For the time series data of the same user number, the K-means time-space clustering method with the minimum sum of the squared errors (SSE) of the objective function within the continuous time range is used. jThat is, it is the center of the jth feature time series vector cluster, and p is the index that determines the weight effect, so as to perform time-space clustering analysis of user trajectory points. Then, the time, distance and speed judgment of each cluster point cluster are combined to form an effective user travel residence point set, and generate an individual user travel OD table based on mobile phone data analysis.

[0049] Step 3: For Internet and survey data, considering the differences in analysis accuracy and analysis dimensions, conduct provincial multi-dimensional space analysis of different transportation modes and travel characteristics analysis of different transportation modes respectively.

[0050] For highway toll data and hub ticketing data with higher accuracy, the main tasks are OD analysis of trunk traffic corridors and hub travel OD analysis in different spatial dimensions.

[0051] For Internet data with relatively small sampling numbers and relatively lower accuracy, such as vehicle GPS data and Amap vehicle speed data, the main work is to analyze the flow, vehicle speed, and saturation of trunk traffic corridors in different spatial dimensions.

[0052] For traditional survey data with higher data accuracy, such as checkpoint data, subway and bus data, the main tasks are road flow, spatiotemporal distribution and vehicle travel OD analysis, as well as spatiotemporal distribution and travel OD analysis of subways and buses.

[0053] For resident travel surveys, population censuses and socioeconomic data with more refined data, the main activities include analysis of population work-residence characteristics in different spatial dimensions, analysis of travel characteristics of different age structures, and analysis of travel proportions by different modes of transportation.

[0054] Step 4: Considering the differences in data analysis content and accuracy in the multi-level spatial dimensions of province, region and city, a single-level data analysis model cannot meet the data analysis needs. Therefore, for multi-source provincial spatial data, multi-dimensional spatial differentiated hierarchical analysis and verification technology is adopted to realize travel characteristics analysis of different data sources in multi-dimensional space.

[0055] Step 5: The hierarchical analysis and verification technology of multi-dimensional space differentiation mainly utilizes the coupling relationship of provincial hierarchical and differentiated characteristic data to realize the analysis and verification of different data sources in multi-dimensional space.

[0056] Considering the characteristics of mobile phone data covering many users and a wide sampling range, firstly, multi-dimensional data mining and travel characteristics analysis are carried out on mobile phone data in terms of spatial scope (province, bay area, metropolitan area, city area, district, street town and traffic area) and user type (permanent population, non-resident population, short-term floating population, hub arrival and transit population), and the sample is expanded according to the age and gender structure of mobile phone users and the proportion of users;

[0057] Combined with the multi-dimensional spatial data of travel OD data of different modes of transportation from the Internet and survey data, the multi-level travel OD data verification from macro to micro level of province-bay area-city is completed. In terms of space, the multi-level travel OD data verification model from macro to micro level of province-bay area-city is expressed as follows:

[0058]

[0059] Provincial multi-dimensional spatial data analysis and verification. In terms of spatial dimension, provincial outbound travel data within a certain time range needs to be realized. Provincial inter-city travel data and provincial city travel data Hierarchical and differentiated data fusion and verification.

[0060] Based on the provincial multidimensional spatial travel data, combined with the provincial travel characteristic data results α of different transportation modes generated by the Internet and traditional survey data, the process of minimizing the difference of multi-source data in the multidimensional space is obtained within a certain time range, that is, min α F α (α,x(α)), then spatial travel data verification at different levels can be realized and provincial multi-dimensional spatial travel characteristic data can be generated.

[0061] Step 6: Based on the verification results of multi-dimensional space OD travel characteristic data, it is mainly to use big data and new technologies to realize data verification of trunk traffic corridors.

[0062] Based on the verification results of multi-dimensional space OD travel characteristic data, provincial traffic model technology is used to carry out road network traffic allocation for different transportation modes in the province;

[0063] Combined with the provincial data analysis results of different traffic modes from the Internet and survey data, the data of trunk traffic corridors are verified. The model expression of trunk traffic corridor data verification in road network traffic distribution is as follows:

[0064]

[0065] The verification of the flow data of the main corridor based on the provincial multi-dimensional spatial data traffic distribution requires the realization of the flow data x of the impedance weight A of different provincial transportation modes within a certain time range in the spatial dimension. a (α) Hierarchical and differentiated data fusion and verification.

[0066] Based on the road traffic network, the traffic flow data of all roads included in the traffic corridor a of different transportation modes in the province within a certain time range is combined Route selection information of all roads included in the transportation corridor a with different transportation modes in the province within a certain time range The process of minimizing the flow difference of multi-source data in the road network allocation corridor within a certain time range, namely minZ L (α), the road flow data at different levels can be verified and the provincial multi-dimensional spatial trunk road traffic flow data can be generated.

[0067] Step 7: Finally, a hierarchical data fusion and feature feedback system is used from top to bottom in the province-bay area-city multi-dimensional space from macro to micro to generate the final provincial multi-dimensional space travel feature data analysis results. Combined with the provincial multi-dimensional space dimensions and user characteristics, provincial multi-dimensional travel feature analysis of different spatial dimensions and different users can be performed (such as provincial travel feature analysis, provincial commuting feature analysis, metropolitan area analysis or same-city analysis, etc.).

[0068] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0069] First, in terms of basic data collection, in terms of spatial scope, the collected information resources include Tencent migration data at the national level, provincial to national railway arrival and departure data, and airport passenger flow distribution data; at the provincial level, highway network toll data, Tencent travel data, mobile phone signaling data, Amap vehicle speed and GPS data; at the city level, population portrait data, resident travel characteristics data, checkpoint vehicle identification data, bus IC card and subway gate data, road traffic operation survey data, population census data, etc.; in terms of type, it covers traditional survey data, intelligent collection data and new big data, such as the acquired data resources shown in Table 1.

[0070]

[0071] Table 1 Schematic diagram of travel characteristics analysis based on different dimensions of multi-source data

[0072] In order to facilitate the study of the spatial distribution characteristics of travel between traffic areas, mobile phone data must correspond to traffic zones, and then use traffic model technology to analyze and verify the data. At the same time, in order to analyze and verify the traffic travel characteristics between regions as much as possible, traffic zones should try to take into account both breadth and depth. Therefore, this zoning system intends to adopt a hierarchical zoning system, that is, divided into traffic districts, traffic middle districts and traffic districts, to adapt to traffic analysis work at different levels. At the traffic district level, it is based on roads, river boundaries, administrative districts, street town boundaries and traffic research boundaries; at the traffic middle district level, it is based on street towns, administrative districts and research scope; at the traffic district level, it is based on street towns and administrative considerations.

[0073] Based on the provincial multi-dimensional spatial travel data, combined with the provincial travel characteristic data results of different transportation modes generated by the Internet and traditional survey data, the process of minimizing the differences of multi-source data in the multi-dimensional space is obtained within a certain time range, and spatial travel data verification at different levels is realized to generate provincial multi-dimensional spatial travel characteristic data.

[0074] Based on the verification results of multi-dimensional space OD travel characteristic data, the provincial traffic model technology is used to perform road network traffic distribution for different modes of transportation in the province, and then the data of trunk traffic corridors are verified by combining the provincial data analysis results of different modes of transportation from the Internet and survey data. In the spatial dimension, it is necessary to realize the hierarchical and differentiated data fusion and verification of the flow data of the impedance weights of different modes of transportation in the province within a certain time range, and based on the road traffic network, combined with the flow data and path selection information of all roads contained in the traffic corridors of different modes of transportation in the province within a certain time range, the process of minimizing the flow difference of multi-source data in the road network distribution corridor within a certain time range can be obtained, so that the road flow data at different levels can be verified and the provincial multi-dimensional space trunk traffic flow data can be generated.

[0075] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A travel characteristics analysis method based on provincial multi-source spatial data, characterized in that: The specific steps are as follows: Step 1: Collect basic information: Collect multi-source data of different spatial dimensions from the national, provincial to municipal levels; Step 2: For mobile phone data with high user ratio and high triggering, big data mining and processing are used, combined with the displacement state judgment and cluster analysis of mobile phone user travel time series, to judge the travel residence state of the displacement point and generate user travel residence point data that meets the characteristics of time-space travel activities; Step 3: For Internet and traditional survey data, conduct provincial multi-dimensional space analysis of different transportation modes and travel characteristics of different transportation modes according to different data characteristics; Step 4: For the provincial spatial analysis and verification of multi-source data, the hierarchical analysis and verification method of multi-dimensional space differentiation is used to first analyze the travel characteristics of different data sources in multi-dimensional space; Step 5: Conduct multidimensional data mining and travel characteristics analysis on mobile phone data in terms of spatial scope and user type, expand the sample according to the age and gender structure of mobile phone users and user proportion, and combine the multidimensional spatial data travel OD data of different transportation modes of Internet and survey data to complete the multi-level travel OD data verification from macro to micro level of province-bay area-city area; The spatial scope includes provinces, bay areas, metropolitan areas, city areas, districts, streets and towns, and transportation communities; the user types include permanent residents, non-permanent residents, short-term floating population, hub arrivals and departures, and transit population; In terms of space, the multi-level travel OD data verification model from macro to micro level, from province to bay area to city, is expressed as follows: In the provincial multidimensional spatial data analysis and verification, in the spatial dimension, it is necessary to realize the path flow data of the provincial external travel corridor within a certain time range. Path flow data of provincial inter-city travel corridors and route flow data of travel corridors within provincial cities Hierarchical and differentiated data fusion and verification; Based on the provincial multidimensional space travel data, combined with the provincial travel characteristic data results α of different transportation modes generated by the Internet and traditional survey data, the path flow x(α) is constructed, and the process of minimizing the difference of multi-source data in the multidimensional space is obtained within a certain time range, that is, min α F α (α,x(α)), that is, to realize the spatial travel data verification at different levels and generate provincial multi-dimensional spatial travel characteristic data; Step 6: Based on the verification results of the provincial multidimensional space OD travel characteristic data, the provincial traffic model is used to perform road network traffic allocation for different traffic modes in the province. Then, the data analysis results of different provincial traffic modes based on the Internet and survey data are combined to verify the data of arterial traffic corridors. The model expression of the arterial traffic corridor data verification in road network traffic allocation is as follows: The verification of the flow data of the main corridor based on the provincial multi-dimensional spatial data traffic distribution requires the realization of the flow data x of the impedance weight a of different transportation modes in the province within a certain time range in the spatial dimension. a (α) hierarchical and differentiated data fusion and verification; among them, C a is the impedance of different transportation modes, w is bounded by Different traffic data, α g is the adjustment factor for the flow data in the arterial corridor; Based on the road traffic network, the traffic flow data of all roads included in the traffic corridor a of different transportation modes in the province within a certain time range is combined Route selection information of all roads included in the transportation corridor a with different transportation modes in the province within a certain time range The process of minimizing the flow difference of multi-source data in the road network allocation corridor within a certain time range, namely minZ L (α), that is, to realize the verification of road flow data at different levels and generate provincial multi-dimensional spatial trunk road traffic flow data; Step 7: Generate the final provincial multi-dimensional space travel characteristic data analysis results through a top-down hierarchical data fusion and feature feedback system from macro to micro provincial-bay area-city multi-dimensional space.

2. A travel characteristics analysis method based on provincial multi-source spatial data according to claim 1, characterized in that: The acquisition sources of the multi-source data are: In terms of spatial scope, data of different spatial dimensions are collected from the national, provincial to municipal levels; At the national level, it includes Tencent migration data at the national level, provincial to national railway arrival and departure data, and airport passenger flow distribution data; At the provincial level, it includes provincial highway network toll data, Tencent travel data, mobile phone signaling data, Amap vehicle speed and GPS data; At the city level, it includes population portrait data, resident travel characteristics data, checkpoint vehicle identification data, bus IC card and subway gate data, road traffic operation survey data, and population census data; In terms of data types, it includes big data and traditional survey data, among which big data accounts for more than 70%.

3. A travel characteristics analysis method based on provincial multi-source spatial data according to any one of claims 1-2, characterized in that: In step 2, the stationary point identification model based on time-space displacement and cluster analysis is expressed as follows: For the time series data of the same user number, the K-means time-space clustering method with the minimum mean square error of the objective function in the continuous time range is used. j That is, it is the center of the jth feature time series vector cluster, and p is the index that determines the weight effect. is the center of the vector cluster c j To time series The distance It is the cluster center distance weight to perform spatiotemporal clustering analysis of user trajectory points; combined with the time, distance and speed judgment of each cluster point cluster, an effective user travel residence point set is formed to generate an individual user travel OD table based on mobile phone data analysis.

4. The travel characteristics analysis method based on provincial multi-source spatial data according to claim 3 is characterized in that: Step 3 is as follows: Based on highway toll data and hub ticketing data, conduct OD analysis of trunk traffic corridors and hub travel OD analysis in different spatial dimensions; based on Internet data, conduct flow, speed, and saturation analysis of trunk traffic corridors in different spatial dimensions; based on traditional survey data, conduct road flow, spatiotemporal distribution, and vehicle travel OD analysis, as well as spatiotemporal distribution and travel OD analysis of subways and buses; based on resident travel surveys, population censuses, and socioeconomic data, conduct population work-residence characteristics analysis in different spatial dimensions, travel characteristics analysis of different age structures, and travel proportion analysis of different modes of transportation.

5. The travel characteristics analysis method based on provincial multi-source spatial data according to claim 4 is characterized in that: In step 7, after generating the final provincial multi-dimensional spatial travel characteristic data analysis results, the provincial multi-dimensional spatial dimensions and user characteristics are combined to complete the provincial multi-dimensional travel characteristic analysis of different spatial dimensions and different users. The provincial multi-dimensional travel characteristic analysis includes provincial travel characteristic analysis, provincial commuting characteristic analysis, metropolitan area analysis or same-city analysis.

Images