A method and system for predicting urban traffic demand

Abstract

The present invention discloses a method and system for predicting traffic demand in a metropolitan area, which belongs to the technical field of traffic data processing. The method is based on the four-stage theory to efficiently obtain the travel demand OD matrix of the central area granularity of the entire metropolitan area in the future years, so as to grasp the overall traffic demand pattern of the metropolitan area from a macro perspective; based on the travel chain theory, the method is based on accurately obtaining the travel demand OD matrix of the urban area traffic district granularity with higher temporal and spatial accuracy in the future years, which can adapt to more application scenarios and reflect traffic forecasts and evaluation analyses of traffic demand details; through the traffic allocation algorithm of motor vehicles and public transportation, the traffic flow index carried by the traffic supply facilities of the metropolitan area is obtained, so that the current and future traffic supply facilities, travel demand, road flow, public transportation passenger volume and other indicators are maintained in continuity and consistency in temporal and spatial distribution, thereby ensuring that the traffic demand forecast and evaluation index results of the metropolitan area in the future years can scientifically meet the needs of users.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic data processing, and in particular to a method and system for predicting traffic demand in a metropolitan area. Background Art

[0002] Within the current landscape of urban development, the integrated development of metropolitan areas has become a key path for megacities to achieve transformational development. The growing demand for intercity and long-distance travel has placed higher demands on the planning and construction of intercity transportation facilities. To meet these new development requirements, metropolitan area transportation infrastructure is continuously improving, with intercity railway and expressway networks continuously increasing in size. The cross-city integration of urban roads and urban rail transit is accelerating, significantly boosting the development of integrated transportation within metropolitan areas and injecting new momentum into the integrated and high-quality development of cities within them.

[0003] To promote the development of metropolitan area transportation infrastructure towards a rational scale, scientific layout, and efficient and orderly operation, a key step in achieving this goal is to construct a multimodal transportation demand forecasting model based on modeling travel demand for different purposes and multiple modes within the metropolitan area. By comprehensively utilizing multiple data sources, such as mobile phone signaling data, resident travel surveys, and departmental statistics, we can accurately grasp the current travel characteristics of residents. Furthermore, we fully consider the impact of various factors on the evolution of transportation demand, such as economic development, population growth, urban planning, and the development of green transportation. Regional transportation demand forecasts are analyzed over different timeframes. This provides scientific advice for identifying key development corridors, formulating plans for the planning and construction of major road and rail transportation infrastructure, and assisting in the research of regional transportation development strategies. However, existing technologies for transportation data planning struggle to comprehensively consider current data and accurately grasp future development data, resulting in metropolitan area transportation demand forecasts that fail to meet practical needs.

[0004] Therefore, in the process of traffic data planning in the existing technology, there is a problem that the results of urban circle traffic demand forecasting are difficult to meet actual needs. Summary of the Invention

[0005] In view of this, it is necessary to provide a method and system for predicting urban circle traffic demand to solve the problem that the results of urban circle traffic demand prediction are difficult to meet actual needs in the process of traffic data planning in the existing technology.

[0006] In order to solve the above problems, the present invention provides a method for predicting traffic demand in a metropolitan area, comprising:

[0007] Constructing a supply facility model of the metropolitan area based on the basic data of the metropolitan area;

[0008] Predict the global travel demand OD matrix of the metropolitan area based on the four-stage theory;

[0009] Predict the OD matrix of urban travel demand in the core cities of the metropolitan area based on the travel chain theory;

[0010] The global travel demand OD matrix and the city travel demand OD matrix are integrated to obtain a data integration result, and the road traffic facility traffic flow index and public transportation passenger flow index of the metropolitan area are determined based on the supply facility model and the data integration result.

[0011] In one possible implementation, the basic data includes physical transportation facility data, traffic management and traffic operation data, current travel characteristic data, current road flow data, current public transportation passenger flow data, physical roads, virtual roads, public transportation networks, and public transportation operation data; the supply facility model includes a motor vehicle supply facility model and a public transportation supply facility model; and constructing the supply facility model of the metropolitan area based on the basic data of the metropolitan area includes:

[0012] constructing a road network structure according to the physical roads and the virtual roads;

[0013] Establishing the motor vehicle supply facility model of the metropolitan area based on the road network structure, the physical traffic facility data, the traffic management and traffic operation data, the current travel characteristic data, and the current road traffic data;

[0014] A public transportation connection network is constructed based on the public transportation network, and the public transportation supply facility model of the metropolitan area is established based on the public transportation connection network and the current public transportation passenger flow data.

[0015] In a possible implementation, the prediction of the global travel demand OD matrix of the metropolitan area based on the four-stage theory includes:

[0016] Dividing the metropolitan area into regions to obtain multiple traffic centers;

[0017] Based on the data of population, job position and trip purpose, a trip generation and attraction model is established through the cross-classification generation rate method to obtain the total number of trips and the total number of trips attracted by different groups and different trip purposes in different traffic centers.

[0018] Establishing a travel distribution model of the total amount of trips generated and the total amount of trips attracted by using a traffic impedance function, and calculating the full-day travel demand OD matrix of different traffic centers based on the travel distribution model;

[0019] Based on the travel utility parameters and travel mode division model, the travel mode sharing ratio among different traffic centers based on different traffic demand choices is determined;

[0020] Determine the travel time probability ratio among different traffic centers based on the full-day travel demand according to the travel time division model;

[0021] According to the full-day travel demand OD matrix, the travel mode share rate and the travel probability ratio of the time period, the full-area travel demand OD matrix of the time period and travel mode between the multiple traffic centers in the metropolitan area is determined.

[0022] In one possible implementation, the trip generation and attraction model includes a trip generation model and a trip attraction model. The trip generation and attraction model is established by using a cross-classification generation rate method to obtain the total trip generation and total trip attraction for different groups of people and different travel purposes in different traffic centers, including:

[0023] Determining the total amount of trips in the traffic center area based on the trip generation model according to the population classification data and the trip purpose classification data;

[0024] According to the job type data and the travel purpose classification data, the total amount of travel attraction in the traffic center is determined based on the travel attraction model.

[0025] In a possible implementation, the calculation formula of the trip generation model is:

[0026] ;

[0027] Where, is the amount of traffic generated in the middle zone i and the travel purpose P; is the cross-classification generation rate of zone i, group c, and travel purpose p; is the number of people in category c in zone i; m is the total number of categories of people in zone i;

[0028] The calculation formula of the travel attraction model is:

[0029] ;

[0030] Where: is the attraction of the traffic center area i and the travel purpose P; The regression coefficient related to travel purpose P and job type w; is the number of positions of position type w in zone i corresponding to travel purpose P.

[0031] In a possible implementation, establishing a travel distribution model of the total amount of trips generated and the total amount of trips attracted by using a traffic impedance function, and calculating the full-day travel demand OD matrix of different traffic centers based on the travel distribution model includes:

[0032] Calculating the comprehensive traffic impedance of the traffic center area based on the distance data, time data and cost data of the metropolitan area;

[0033] Determining a traffic impedance function of the traffic center area based on the comprehensive traffic impedance and a gamma function;

[0034] According to the traffic impedance function, the travel distribution model is established based on the double-constraint gravity model, and the full-day travel demand OD matrix of different traffic centers is calculated according to the travel distribution model.

[0035] In one possible implementation, the OD matrix of urban travel demand of the core cities in the metropolitan area is predicted based on the travel chain theory, including:

[0036] Dividing the urban area of ​​the core city of the metropolitan area into regions to obtain a plurality of traffic zones;

[0037] Integrate the population classification, job type, income level and motor vehicle data of the core cities in the metropolitan area to obtain the cross-classification results of the urban population;

[0038] According to the cross-classification results of the urban population, the total number of trip chains of different trip chains with the traffic zone as the starting point of a day's trip is calculated by the generation rate method;

[0039] Determining the destination selection probability and mode selection probability of the trip chain according to the trip chain parameters and the discrete choice model;

[0040] Calculate the time-divided travel probability of the travel chain by probability method;

[0041] The urban travel demand OD matrix of the core city of the metropolitan area is determined according to the total amount of the travel chain, the destination selection probability, the mode selection probability and the travel probability in different time periods.

[0042] In one possible implementation, the discrete choice model includes a single-constraint destination discrete choice model and a double-constraint destination discrete choice model; the trip chain parameters include at least the distance data, the time data, and the fare data; and determining the destination selection probability of the trip chain based on the trip chain parameters and the discrete choice model includes:

[0043] Determining a single trip utility function between the traffic zones according to the distance data, the time data, and the cost data, and updating a travel chain utility function of a destination selection model according to the single trip utility function;

[0044] determining the destination selection probabilities of the different travel chains respectively according to the destination selection model, the single-constraint destination discrete selection model, and the double-constraint destination discrete selection model;

[0045] When the destination of the travel chain includes multiple stopover points, the destination selection probabilities corresponding to the multiple stopover points are determined respectively according to the single-constraint destination discrete choice models corresponding to the stopover points.

[0046] In one possible implementation, integrating the global travel demand OD matrix and the city travel demand OD matrix to obtain a data integration result, and determining the road traffic facility traffic flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result, includes:

[0047] Based on the preset data integration rules, the global travel demand OD matrix and the city travel demand OD matrix are integrated to obtain a high-precision travel demand OD matrix for the metropolitan area at different time periods and different transportation modes;

[0048] Calculating the traffic flow index of the road traffic facilities in the metropolitan area based on the motor vehicle travel demand OD matrix, the road delay function, the motor vehicle comprehensive cost function and the motor vehicle traffic allocation algorithm;

[0049] The public transportation passenger flow index of the metropolitan area is calculated according to the public transportation travel demand OD matrix, the public transportation in-vehicle time function, the public transportation generalized cost function and the public transportation passenger flow allocation algorithm.

[0050] In order to solve the above problems, the present invention further provides a metropolitan area traffic demand forecasting system, comprising:

[0051] A supply facility model building module, configured to build a supply facility model of the metropolitan area based on the basic data of the metropolitan area;

[0052] A global travel demand OD matrix prediction module, used to predict the global travel demand OD matrix of the metropolitan area based on the four-stage theory;

[0053] The urban travel demand OD matrix prediction module is used to predict the urban travel demand OD matrix of the core cities in the metropolitan area based on the travel chain theory;

[0054] The vehicle flow and passenger flow allocation module is used to integrate the data of the global travel demand OD matrix and the urban travel demand OD matrix to obtain the data integration result, and determine the road traffic facility vehicle flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result.

[0055] The beneficial effects of adopting the above embodiment are as follows: the present invention provides a method for predicting traffic demand in a metropolitan area, which constructs a supply facility model of a metropolitan area and performs digital modeling on traffic facilities and operation management information; in the link of traffic demand prediction, it gives full play to the advantages of the two traffic modeling theories of the four-stage theory and the travel chain theory; based on the four-stage theory, it efficiently obtains the OD matrix of the travel demand of the central area granularity of the whole traffic of the metropolitan area in the future, and grasps the overall traffic demand pattern of the metropolitan area from a macro perspective; based on the travel chain theory, it accurately obtains the OD matrix of the travel demand of the urban traffic district granularity with higher spatiotemporal accuracy in the future, which can adapt to more applications. Traffic forecast and evaluation analysis based on scenarios and reflecting the details of traffic demand; integrating the OD matrix of travel demand in the central area of ​​the metropolitan area and the OD matrix of travel demand in the urban area of ​​the core city; on the basis of the metropolitan area supply facility model, through the traffic allocation algorithm of motor vehicles and public transportation, obtaining the traffic flow index carried by the metropolitan area traffic supply facilities, achieving the continuity and consistency of the temporal and spatial distribution of the current and future traffic supply facilities, travel demand, road flow, bus passenger volume and other indicators, thereby ensuring that the traffic demand forecast and evaluation index results of the metropolitan area in the future can scientifically meet the needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flow chart of an embodiment of a method for predicting urban area traffic demand provided by the present invention;

[0057] Figure 2 A schematic flow chart of an embodiment of a supply facility model for a metropolitan area provided by the present invention;

[0058] Figure 3 A schematic diagram of a process for predicting the global travel demand OD matrix of a metropolitan area according to an embodiment of the present invention;

[0059] Figure 4 A schematic diagram of the results of an embodiment of a transportation mode selection framework provided by the present invention;

[0060] Figure 5 A flow chart of an embodiment of an OD matrix for predicting urban travel demand in a core city of a metropolitan area provided by the present invention;

[0061] Figure 6 A schematic diagram of the framework of an embodiment of the urban travel chain model provided by the present invention;

[0062] Figure 7 A schematic diagram of a flow chart of an embodiment of determining a road traffic facility traffic flow index and a public transportation passenger flow index in a metropolitan area provided by the present invention;

[0063] Figure 8This is a structural block diagram of an embodiment of the metropolitan area traffic demand forecasting system provided by the present invention. DETAILED DESCRIPTION

[0064] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0065] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in the present invention illustrate operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flowcharts can be implemented out of sequence, and steps that have no logical contextual relationship can be reversed in order or implemented simultaneously. In addition, those skilled in the art, guided by the content of the present invention, can add one or more other operations to the flowcharts or remove one or more operations from the flowcharts. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0066] The terms "first" and "second" in the embodiments of the present invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature designated as "first" or "second" may explicitly or implicitly include at least one such feature.

[0067] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0068] In order to solve the problem in the prior art that the results of metropolitan area traffic demand prediction cannot meet actual needs during traffic data planning, the present invention provides a metropolitan area traffic demand prediction method and system, which are described in detail below.

[0069] like Figure 1 As shown, Figure 1 A flow chart of an embodiment of a method for predicting urban area traffic demand provided by the present invention includes:

[0070] S101: Construct a metropolitan area supply facility model based on the basic data of the metropolitan area;

[0071] S102: Predict the Origin-Destination matrix (OD matrix) of the metropolitan area's overall travel demand based on the four-stage theory;

[0072] S103: Predicting the OD matrix of urban travel demand in core cities of metropolitan areas based on travel chain theory;

[0073] S104: Integrate the data of the global travel demand OD matrix and the city travel demand OD matrix to obtain the data integration results, and determine the road traffic facility traffic flow indicators and public transportation passenger flow indicators of the metropolitan area based on the supply facility model and the data integration results.

[0074] In some embodiments of the present invention, the four-stage theory is an important theoretical framework that divides a day's transportation travel activities into several independent travel itineraries with clear starting and ending points, and divides the determination process of each itinerary into four consecutive decision-making stages. Each of the four stages has specific goals and tasks, which respectively determine the purpose of each trip, the starting and ending points, the choice of travel mode, and the choice of travel route. This division method enables people to understand and analyze various complex processes more deeply. It provides a systematic thinking method to help people gradually advance and fully grasp each link of the entire process. The four-stage theory has the advantages of relatively simple theory, relatively low modeling data requirements, and efficient modeling, and has a certain universality for general application scenarios.

[0075] In some embodiments of the present invention, a trip chain refers to a series of travel behaviors that begins at a starting point, passes through several stopover points, and ultimately returns to the starting point. The trip chain model breaks down a resident's daily transportation activities into several trip chains and models and analyzes these trip chains. The trip chain model meets the requirements of mode-time-space continuity and is more scientific and accurate than the four-stage model. It also better considers the impact of urban spatial patterns, economic development, population policies, and traffic management policy optimization on travel behavior. Trip chain theory has broader applications in fields such as transportation planning and urban traffic management. For example, trip chain-based travel theory can explain peak periods and uncertainty in traffic systems, providing a scientific basis for transportation planning and urban traffic management. Furthermore, trip chain theory helps understand travelers' travel behaviors and needs, providing a reference for the formulation and implementation of urban transportation policies. The architectural design of this embodiment implements a two-way interactive feedback mechanism between transportation demand and transportation infrastructure supply, meeting requirements such as architectural flexibility, supply-demand balance, and operational stability.

[0076] In this embodiment, by constructing a supply facility model of the metropolitan area, the transportation facilities and operation management information are digitally modeled; in the traffic demand forecasting link, the advantages of the two traffic modeling theories, the four-stage theory and the travel chain theory, are fully utilized; by taking the four-stage theory as the basis, the global travel demand OD matrix of the metropolitan area in the future years is efficiently obtained, and the overall traffic demand pattern of the metropolitan area is grasped from a macro perspective; by taking the travel chain theory as the basis, the urban travel demand OD matrix of the core cities of the metropolitan area in the future years is accurately obtained, which can adapt to more application scenarios and reflect the traffic forecast and evaluation analysis of traffic demand details with higher temporal and spatial accuracy; the global travel demand OD matrix and the urban travel demand OD matrix are integrated, and on the basis of the metropolitan area supply facility model, the traffic flow index carried by the metropolitan area transportation supply facilities is obtained through the traffic allocation algorithm of motor vehicles and public transportation, so as to achieve the continuity and consistency of the temporal and spatial distribution of the current and future traffic supply facilities, travel demand, road flow, bus passenger volume and other indicators, thereby ensuring that the traffic demand forecast and evaluation index results of the metropolitan area in the future years can scientifically meet the needs of users.

[0077] In a specific embodiment, the travel demand OD matrix first divides the city into several traffic centers, and then uses a two-dimensional table to describe in detail the travel demand between each starting traffic center and the end traffic center. The travel demand is presented as quantitative indicators of the number of trips and vehicles, reflecting the scale of personnel or vehicle flow between different zones within a specific time period (such as the morning rush hour on weekdays), which is an important basis for conducting demand analysis for transportation facility construction; by integrating data from the metropolitan area's entire travel demand OD matrix and the city's travel demand OD matrix, a high-precision demand OD matrix for different time periods and different transportation modes in the metropolitan area in the future years is obtained. Through the traffic allocation algorithm of motor vehicle traffic and public transportation passenger flow, the traffic flow index carried by the transportation supply facilities is obtained, which realizes the continuity and consistency of the temporal and spatial distribution of indicators such as transportation supply facilities, travel demand, road flow, and public transportation passenger volume in the current and future years, thereby ensuring that the traffic demand forecast and evaluation index results of the metropolitan area in the future years can scientifically meet the needs of users.

[0078] In some embodiments of the present invention, in S101, the basic data includes physical traffic facility data, traffic management and traffic operation data, current travel characteristic data, current road flow data, current public transportation passenger flow data, physical roads, virtual roads, and public transportation networks; Figure 2 As shown, Figure 2 This is a flow chart of an embodiment of the supply facility model of a metropolitan area provided by the present invention. The supply facility model of the metropolitan area includes a motor vehicle supply facility model and a public transportation supply facility model.

[0079] First, in order to construct a motor vehicle supply facility model for a metropolitan area, a road network structure consisting of physical roads and virtual roads was constructed, and a road network database system was built to establish a road supply facility model for a metropolitan area. The specific steps include:

[0080] (1) Build a basic road network, including physical roads that carry different modes of transportation, such as cars, public transportation, and slow traffic (electric vehicles, bicycles, and walking).

[0081] (2) Generate virtual road connections. This includes traffic zone links, rail transit entry and exit links, and transfer links between rail stations. Traffic zone links are the connecting lines between traffic zones and road network nodes.

[0082] (3) Build a road network database system. Implement hierarchical data management based on road sections and nodes. Road network section data should include road grade, number of motor vehicle lanes, free flow speed, toll collection, road delay function, and available car / bus / slow-moving traffic modes. The node database should include data such as node type, intersection control mode, and turn management.

[0083] Then, the public transportation network of the metropolitan area is obtained, a public transportation connection network is constructed, and a public transportation database is built to obtain a public transportation supply facility model of the metropolitan area. The specific steps of constructing the public transportation supply facility model of the metropolitan area include:

[0084] (1) Build a public transportation network. The public transportation network consists of public transportation lines built on physical roads and rail lanes, including rail transit, conventional buses, trams, intercity railways, national railways and other public transportation modes.

[0085] (2) Generate a bus connection network. Use various virtual roads to simulate the end-to-end connection and bus transfer connection of public transportation. The end-to-end connection of public transportation is achieved through the centroid link of the traffic area, the rail station link, and the slow road.

[0086] (3) Public transportation database. This includes route data and node data. The bus route database includes route type, route name, departure information, fare, and other data. The bus node database mainly includes station type and stop information data.

[0087] In this embodiment, by constructing a supply facility model for the metropolitan area, unified data planning and processing for the metropolitan area is achieved, so as to facilitate subsequent unified data analysis and adjustment and avoid data disorder problems.

[0088] In some embodiments of the present invention, in S102, the OD matrix of the global travel demand of the metropolitan area is predicted according to the four-stage theory, such as Figure 3As shown, Figure 3 The flowchart of an embodiment of the OD matrix for predicting global travel demand in a metropolitan area provided by the present invention includes:

[0089] S301: Divide the metropolitan area into regions to obtain multiple traffic centers;

[0090] S302: Based on the demographic data, job data, and trip purpose classification data, a trip generation and attraction model is established using a cross-classification generation rate method to obtain the total number of trips generated and the total number of trips attracted for different demographic groups and different trip purposes in different traffic centers.

[0091] S303: Establish a travel distribution model of the total amount of trips generated and the total amount of trips attracted by the traffic impedance function, and calculate the full-day travel demand OD matrix of different traffic centers based on the travel distribution model;

[0092] S304: Determine travel mode sharing rates among different traffic centers based on different traffic demand choices based on travel utility parameters and travel mode classification models;

[0093] S305: Determine the travel probability ratio of different traffic centers based on the full-day travel demand according to the travel time division model;

[0094] S306: Determine the global travel demand OD matrix of different time periods and travel modes among multiple traffic centers in the metropolitan area based on the full-day travel demand OD matrix, the travel mode share rate, and the travel probability ratio of different time periods.

[0095] Furthermore, in S302, the travel generation and attraction model includes a travel generation model and a travel attraction model. In order to obtain the total amount of travel generation and the total amount of travel attraction, the total amount of travel generation in the traffic center is first determined based on the travel generation model according to the population classification data and the travel purpose classification data; then, the total amount of travel attraction in the traffic center is determined based on the travel attraction model according to the job type data and the travel purpose classification data.

[0096] Specifically, based on the population classification data and job classification data, the generation rate method of cross-classification is used to obtain the travel generation and travel attraction of different groups of people c and different travel purposes p in different traffic centers.

[0097] In some embodiments of the present invention, the calculation formula of the trip generation model is:

[0098]

[0099] Where: Central Transportation District i , travel purpose P The amount of production; For zone i, c Group of people, travel purpose p The cross-classification yield rate; for i District c the number of people in the category; m for i The total number of classifications of population types in the district.

[0100] In some embodiments of the present invention, the calculation formula of the travel attraction model is:

[0101]

[0102] Where: for i District, travel purpose P The amount of attraction; For travel purposes P and job type w The relevant regression coefficients; for i The number of positions of position type w corresponding to the travel purpose P in the area.

[0103] In some embodiments of the present invention, in S303, in order to establish a travel distribution model of the total amount of trips generated and the total amount of trips attracted through the traffic impedance function, and calculate the full-day travel demand OD matrix of different traffic centers based on the travel distribution model, first, the comprehensive traffic impedance of the traffic center is calculated based on the distance data, time data and cost data of the metropolitan area; then, based on the comprehensive traffic impedance, the traffic impedance function of the traffic center is determined based on the gamma function; finally, based on the traffic impedance function, a travel distribution model is established based on the double-constraint gravity model, and the full-day travel demand OD matrix of different traffic centers is calculated based on the travel distribution model.

[0104] Specifically, the travel distribution model is used to establish the spatial relationship between the origin and destination points of each traffic center. A dual-constraint gravity model is used to consider regional socioeconomic growth factors, travel distance, and time impedance factors. The travel distribution model is constructed to calculate the travel demand between traffic centers using the following calculation expression:

[0105]

[0106] Where: for i District to j Travel volume in the district; for i The occurrence volume of the area; for j The attraction of the area; The inter-county connection factor in the metropolitan area represents the factors that cannot be explained by travel impedance in terms of the connection strength between different districts and counties, such as the degree of socioeconomic dependence. for i District to j The comprehensive traffic impedance of the area; is the traffic impedance function, and the function value is related to the community i and community j is inversely proportional to the traffic impedance.

[0107] The traffic impedance function adopts the gamma function form, and the function form is:

[0108] ;

[0109] Where: a 、 b 、 c For function parameters.

[0110] The comprehensive traffic impedance of the traffic center is obtained by weightedly averaging the traffic impedance of different modes according to the proportion of the mode structure. The traffic impedance of different modes is generally the total cost composed of travel distance, time, and cost. The calculation method of comprehensive traffic impedance is as follows:

[0111] ;

[0112] Where: for i District to j District, using travel mode k travel time; For transportation k travel expenses; VOT is the average time value of the travel group (yuan / minute); for i District to j The mode of transportation used in the district k proportion.

[0113] In some embodiments of the present invention, in S304, traffic mode classification is to predict the probability of using a certain traffic mode between traffic zones. Figure 4 As shown, Figure 4 This is a schematic diagram of the results of an embodiment of the transportation mode selection framework provided by the present invention. It adopts a layered architecture. The first layer is the primary mode layer. Walking is first separated according to a certain probability distribution based on the distance between transportation zones. Then, the LOGIT model is used to obtain the share of different primary transportation modes. The second layer is the public transportation sub-mode selection. The public transportation sub-mode is competitively selected during the public transportation passenger flow allocation stage.

[0114] The LOGIT model is a probabilistic competition model based on the utility values ​​of different transportation modes. The functional form of the model is expressed as:

[0115]

[0116] Where: for i district, j Between districts m The travel mode share; It is the utility function of different transportation modes. The utility value is basically similar to the traffic impedance algorithm. The difference is that in addition to considering travel time and travel cost parameters, the utility function can also add mode selection preference parameters for different populations and travel purposes.

[0117] In some embodiments of the present invention, in S305, the distribution probability of the travel volume of commuting and non-commuting purposes at different time periods in a day is identified based on big data, and the time-divided travel probability is obtained. .

[0118] In some embodiments of the present invention, in S306, the travel demands of different travel purposes are aggregated to obtain the travel demands of three different modes of transportation, namely private cars, shared cars (including taxis), and public transportation, at different time periods. .

[0119] The traffic demand model calculation for different periods such as morning peak and evening peak is as follows:

[0120] ;

[0121] Where: for t Time period (morning peak, evening peak, off-peak), i District and j Between districts, travel mode m traffic demand value.

[0122] In this embodiment, by predicting the OD matrix of the global travel demand of the metropolitan area based on the four-stage theory, it is possible to integrate various data related to residents' travel: regional connection data, traffic impedance, travel attraction, population type, job type, etc., which greatly improves the reliability of the OD matrix of the global travel demand.

[0123] In some embodiments of the present invention, in S103, in order to predict the urban travel demand OD matrix of the core city of the metropolitan area according to the travel chain theory, such as Figure 5 As shown, Figure 5 The flowchart of an embodiment of the OD matrix for predicting urban travel demand in a core city of a metropolitan area provided by the present invention includes:

[0124] S501: Divide the urban area of ​​the core city of the metropolitan area into regions to obtain multiple traffic zones;

[0125] S502: Integrate the population classification, job type, income level, and motor vehicle data of the core cities in the metropolitan area to obtain the cross-classification results of the city population;

[0126] S503: Based on the cross-classification results of the urban population, the total number of trip chains of different trip chains with the traffic zone as the starting point of the daily trip is calculated by the generation rate method;

[0127] S504: Determine the destination selection probability and mode selection probability of the trip chain according to the trip chain parameters and the discrete choice model;

[0128] S505: Calculate the time-divided travel probability of the travel chain by probability method;

[0129] S506: Determine the urban travel demand OD matrix of the core cities in the metropolitan area based on the total number of travel chains, destination selection probability, mode selection probability and travel probability in different time periods.

[0130] In a specific embodiment, based on travel survey data, an analysis of the main travel chain types in the city is obtained, including:

[0131] Based on population classification, job type, income level, and motor vehicle data, a cross-classification of the urban population is obtained. Specifically, based on mobile phone big data and resident travel surveys, the main stopover points and destinations of residents are identified, and typical travel chains representing more than 95% of trips are extracted as modeling objects, such as "home-work-home" (HWH), "home-work-dining-home" (HWEH), etc. The original starting point of the travel chain is generally home or work. It should be noted that the total number of people in various population categories is first obtained based on population age, household income, and vehicle ownership data, as shown in the following table. The following table shows the results of an embodiment of the urban travel chain model population classification method provided by the present invention:

[0132]

[0133] The total amount of different travel chains with the traffic zone as the starting point for a day's trip is obtained through the generation rate method, including: the frequency of different groups c using different travel chains p based on the travel survey , based on the cross-classification method, the total amount of all trip starting points, traffic zones, different groups of people, and different trip chains is calculated:

[0134] ;

[0135] Where: i is the starting traffic area of ​​the trip chain; p is the type of trip chain; c is the crowd category.

[0136] In some embodiments of the present invention, S505-S506 need to consider the factors that affect the traveler's destination selection and travel mode selection, and use the travel time, cost and distance of different modes as input to establish a travel chain utility function, such as Figure 6 As shown, Figure 6 This is a schematic diagram of the framework of an embodiment of the urban travel chain model provided by the present invention.

[0137] The single trip utility function considering different factors is defined as follows:

[0138] ;

[0139] in: The utility value of a single trip for transportation mode m between transportation zones i and j with travel purpose p and group type c; 、 、 、 、 The utility function parameters to be calibrated for different transportation modes; 、 、 These are the parameters to be calibrated for public transportation modes.

[0140] Based on the comprehensive utility function, the travel chain utility function of the travel chain mode choice model is defined to calculate the mode choice probability, which is defined as follows:

[0141] ;

[0142] Where: The utility of a closed travel chain with a travel purpose of p, a crowd type of c, a starting point of the travel chain in traffic area i, a stopover in traffic area j, and a travel mode selected; , represent the travel utility of the outbound and return trips respectively.

[0143] By performing logarithmic summation on the travel chain utilities of the mode choices, the travel chain utility function of the destination choice model is defined as follows:

[0144] ;

[0145] Where: It represents the utility of a closed trip chain with the trip purpose being p, the group type being c, the trip starting point being the transportation zone i, and the trip stopping point being the transportation zone j; are model parameters.

[0146] Based on the trip chain parameters and discrete choice model, the probability of trip chain destination selection is determined, including: in general, the trip starting point (usually home or work), a series of intermediate destinations, and then returning to the trip starting point (home / work) constitute a complete trip chain; in some embodiments of the present invention, trip chain destination selection includes the selection of the main destination and the destination of other stopover points; the destination selection adopts the disaggregated discrete choice (LOGIT) algorithm framework, and the destination selection is further divided into a single-constraint destination selection probability function and a double-constraint destination selection probability function according to the trip chain type.

[0147] Flexible travel (stable destination) adopts a single-constraint destination selection model, and the single-constraint destination selection probability function is as follows:

[0148] ;

[0149] in: From the original cell of the chain i Departure, choice j probability of being a destination; i The transportation zone (TAZ) where the origin zone (home / workplace) of the trip chain is located; j The TAZ of the travel destination; p For travel purposes; c to categorize the population; Select a scaling factor for your destination; for TAZ(i) arrive TAZ(j) The utility function value between .

[0150] Commuting uses a dual-constraint destination selection model that requires satisfying both the home and workplace / school totals. In addition to satisfying the origin total constraint, the destination total constraint must also be satisfied. The formula for calculating the dual-constraint destination selection probability function is as follows:

[0151] ;

[0152] in: For destination j and travel purpose is p (work / school) balance factor.

[0153] The model for the probability of destination selection for other stopovers assumes that each traveler aims to minimize total travel costs when determining stopovers, using a "rubber band" algorithm. Taking the "HWEH" travel chain model of employees eating / shopping on their way home from get off work as an example, we first define the spatial offset between the return stopover and the original return route as the utility function for the probability function of selecting the other stopover points q on the return trip. The function for other stopovers is defined as follows:

[0154] ;

[0155] Where: The utility from the destination to the return stopover point; is the impedance from the stopover point to the departure point on the return journey; is the impedance from the origin to the destination excluding stopovers.

[0156] After the utility calculation of the stopover point q is completed, the single-constraint destination selection probability function as described above is used to determine the traffic zone corresponding to the stopover point q.

[0157] Based on the trip chain parameters and discrete choice model, the probability of mode selection of the trip chain is determined, including: trip mode selection of the trip chain, including the mode selection of the outbound trip and the mode selection of the return trip, and allowing the return trip to choose a different mode of transportation from the outbound trip according to different transition probabilities. The mode transition probability matrix is ​​obtained through traffic surveys. The probability of mode selection of the outbound trip is defined as follows:

[0158] ;

[0159] In the above formula: The probability of choosing transportation mode m for the outbound trip; m For the outbound mode of transportation; k For optional transportation; is the scaling factor corresponding to different modes of transportation; For the travel chain utility.

[0160] Through travel survey data, we can get the travel purpose as p , travel time is t The proportion of the trip chain of the destination to the total number of trip chains generated throughout the day, let The total amount of travel chains, destination selection probability, mode selection probability, and travel probability by time period are comprehensively considered to predict the OD matrix of urban travel demand in the core cities of the metropolitan area. m , Transportation area at the starting point of travel i , the travel destination is a transportation zone j The total amount of trip chains between is calculated as follows:

[0161] ;

[0162] In this embodiment, the OD matrix of urban travel demand of the core cities in the metropolitan area is predicted based on the travel chain theory, which is more in line with the actual travel behavior decision-making model. At the same time, it can reflect the impact of factors such as urban spatial structure, economic development level, traffic charges, and changes in travel habits on urban travel demand, and can better meet the needs of high-precision quantitative analysis in different usage scenarios.

[0163] In some embodiments of the present invention, in S104, in order to integrate the global travel demand OD matrix and the city travel demand OD matrix to obtain the data integration result, and determine the road traffic facility traffic flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result, such as Figure 7 As shown, Figure 7 A flow chart of an embodiment of determining a road traffic facility traffic flow index and a public transportation passenger flow index in a metropolitan area provided by the present invention includes:

[0164] S701: Based on the preset data integration rules, the global travel demand OD matrix and the city travel demand OD matrix are integrated to obtain a high-precision travel demand OD matrix for the metropolitan area at different time periods and different transportation modes;

[0165] S702: Calculate the traffic flow index of the road traffic facilities in the metropolitan area based on the motor vehicle travel demand OD matrix, the road delay function, the motor vehicle comprehensive cost function, and the motor vehicle traffic assignment algorithm;

[0166] S703: Calculate the public transportation passenger flow index of the metropolitan area based on the public transportation travel demand OD matrix, the public transportation in-vehicle time function, the public transportation generalized cost function and the public transportation passenger flow allocation algorithm.

[0167] Specifically, based on the preset data integration rules, the travel demand OD matrix at the granularity of the metropolitan area's overall traffic center and the travel demand OD matrix at the granularity of the metropolitan area's core city traffic district are integrated to obtain a high-precision travel demand OD matrix for the metropolitan area in different time periods (peak and off-peak) and different modes of transportation (cars, shared cars including taxis, and public transportation).

[0168] In one specific embodiment, the basic principles for integrating traffic demand OD data at the metropolitan area traffic center granularity and the urban area traffic district granularity are as follows: For trips within the urban area, the urban area traffic district travel demand is used; for trips with both origin and destination outside the urban area, the metropolitan area traffic demand is used; and for cross-city trips with one end in the urban area and the other end outside the urban area, the OD of cross-city traffic demand is spatially segmented at the trip end within the urban area (the trip end within the urban area is segmented using the traffic district, and the trip end outside the urban area is segmented using the traffic center district). Taking the spatial segmentation of traffic demand for trips with origin within the urban area and destination outside the urban area as an example, the method of locating the trip origin center area to the urban area traffic district according to a certain ratio is explained:

[0169] ;

[0170] Where: IJ is the traffic center area number of the starting and ending points of the trip within the metropolitan area; is the demand between the urban traffic center area I and the center area J; i is the traffic area included in the urban traffic center area I; is the demand breakdown result between the urban traffic zone i and the metropolitan traffic center zone J; m is the traffic mode; t is the traffic time division; 、 Model the population and jobs within a metropolitan area that are accessible by car or public transportation within 45 minutes for transportation districts.

[0171] Based on the OD matrix of automobile travel demand, road delay function, comprehensive cost function of motor vehicles, and motor vehicle traffic distribution algorithm, the flow index of urban circle road traffic facilities is obtained; based on the OD matrix of public transportation travel demand, public transportation in-vehicle time function, generalized cost function of public transportation, and public transportation passenger flow distribution algorithm, the passenger flow index of the urban circle public transportation system is obtained.

[0172] Using a traffic distribution model, the high-precision travel demand OD matrix of different time periods and modes within the metropolitan area is distributed to the traffic network according to the path search algorithm. It is divided into motor vehicle traffic distribution and public transportation passenger traffic distribution according to the traffic mode. It is necessary to first define the comprehensive cost function of motor vehicle traffic in S702 and the generalized cost function of public transportation passenger flow in S703; then use commercial software to implement the traffic distribution algorithm. The commercial platforms that can be used include EMME, TRANCAD, etc.

[0173] Using the motor vehicle traffic allocation algorithm, the OD matrix of car demand and the OD matrix of shared car (including taxi) demand are allocated to the road network facilities to obtain the motor vehicle flow index; the motor vehicle flow allocation algorithm can adopt a variety of allocation algorithms such as the multi-path probability allocation method and the equilibrium allocation method. The motor vehicle flow allocation model uses the comprehensive cost function of motor vehicle traffic as the impedance parameter of the road section to participate in the allocation. The comprehensive cost takes into account travel time, vehicle operating costs and road tolls, where travel time is defined by the road delay function; different road delay functions are used for highway and urban road sections. The highway delay function adopts the commonly used BPR model, which is suitable for roads with continuous flow; the delay function of urban roads is the Akcelik model, which is mainly suitable for general urban main and secondary roads and branches with intermittent flow affected by intersections and entrances and exits; the road delay function of the Akcelik model is as follows:

[0174] ;

[0175] Where: T is the travel time of the road section, is the free flow of traffic on the road, unit (min); is the one-way traffic volume and capacity of the road, unit (pcu / h); voc is the saturation (min); a, b, c, DelayZero are the parameters to be determined in the model.

[0176] In terms of the OD allocation order of motor vehicle demand, vehicles running on fixed routes and departure frequencies, such as regular buses and passenger bus lines, are first converted into standard vehicles and pre-loaded on the network. Then, the demand for cars is allocated first. After the delay function is updated based on road traffic, the traffic demand for shared cars is allocated. Finally, the motor vehicle flow on the road section is obtained in total.

[0177] The public transportation passenger flow allocation model uses an allocation model based on departure frequency and congestion penalty to obtain indicators such as passenger flow for different bus routes, bus sections, and bus stops. S703 expresses the generalized cost function of public transportation passenger flow as follows:

[0178] ;

[0179] Where: GT is the generalized time for public transportation; TTF is the time cost in the bus; WALK is the walking connection and transfer time; WAIT is the waiting time; BP is the waiting penalty; x, y, z are the weights used in the model calibration process. The form of the time cost function in public transportation is as follows:

[0180] ;

[0181] Where: TTF is the time function in the vehicle (people's perceived time); us1 is the congestion penalty coefficient calculated based on the passenger flow of the bus line section; DF is the damping factor; us3 is the time perceived by people in a non-congested state, which is equivalent to the actual travel time of the vehicle.

[0182] It is a demand-facility supply cycle iteration and convergence algorithm for traffic demand forecasting and traffic distribution. Traffic distribution adopts a multiple-iteration method. The flow results obtained by the traffic distribution model are the parameters for calculating the travel chain utility function in the demand forecasting process. According to the output indicators of the traffic distribution model, the travel utility and travel delay functions are updated, and the results will be fed back to the next round of metropolitan area demand forecasting, city demand forecasting, and traffic distribution calculation process. The convergence condition of the distribution algorithm is that the difference in road network travel time between two consecutive iterations is less than 0.5% or the set maximum number of iterations is reached (the default is 10 times). After the model has converged through multiple "demand calculation-traffic distribution" cycle iterations, the final motor vehicle and bus passenger flow distribution results are output.

[0183] In summary, this application constructs a supply facility model of the metropolitan area to conduct digital modeling of transportation facilities and operation management information; in the traffic demand forecasting link, it gives full play to the advantages of the two traffic modeling theories, the four-stage theory and the travel chain theory; based on the four-stage theory, it efficiently obtains the travel demand OD matrix of the central area granularity of the entire metropolitan area in the future years, and grasps the overall traffic demand pattern of the metropolitan area from a macro perspective; based on the travel chain theory, it accurately obtains the travel demand OD matrix of the urban traffic district granularity with higher temporal and spatial accuracy in the future years, which can realize traffic forecasting and evaluation analysis that adapts to more application scenarios and reflects the details of traffic demand; the travel demand OD matrix of the central area granularity of the entire metropolitan area and the travel demand OD matrix of the urban traffic district granularity of the core city are integrated, and on the basis of the metropolitan area supply facility model, the traffic flow index carried by the metropolitan area traffic supply facilities is obtained through the traffic allocation algorithm of motor vehicles and public transportation, so as to achieve the continuity and consistency of the temporal and spatial distribution of the current and future traffic supply facilities, travel demand, road flow, bus passenger volume and other indicators, thereby ensuring that the traffic demand forecast and evaluation index results of the metropolitan area in the future years can scientifically meet the needs of users. By integrating and modeling data on multiple factors related to traffic elements, the reliability of traffic demand forecast results is improved, which can effectively guide traffic management departments to make adaptive adjustments to traffic facilities.

[0184] In order to better implement the metropolitan area traffic demand prediction method in the embodiment of the present invention, the embodiment of the present invention also provides a metropolitan area traffic demand prediction system, such as Figure 8 As shown, Figure 8 This is a structural block diagram of an embodiment of a metropolitan area traffic demand forecasting system provided by the present invention. The metropolitan area traffic demand forecasting system 800 includes:

[0185] A supply facility model building module 801 is used to build a supply facility model of a metropolitan area based on basic data of the metropolitan area;

[0186] The global travel demand OD matrix prediction module 802 is used to predict the global travel demand OD matrix of the metropolitan area according to the four-stage theory;

[0187] The urban travel demand OD matrix prediction module 803 is used to predict the urban travel demand OD matrix of the core cities in the metropolitan area based on the travel chain theory;

[0188] The vehicle flow and passenger flow allocation module 804 is used to integrate the data of the global travel demand OD matrix and the urban travel demand OD matrix to obtain a data integration result, and determine the road traffic facility vehicle flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result.

[0189] Those skilled in the art will appreciate that all or part of the process flow of the above-described method embodiment can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0190] The above is a detailed introduction to the urban circle traffic demand prediction method, device, electronic device and storage medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for predicting traffic demand in a metropolitan area, characterized in that: include: Constructing a supply facility model of the metropolitan area based on the basic data of the metropolitan area; Predict the global travel demand OD matrix of the metropolitan area based on the four-stage theory; Predict the OD matrix of urban travel demand in the core cities of the metropolitan area based on the travel chain theory; Integrate the global travel demand OD matrix and the city travel demand OD matrix to obtain a data integration result, and determine the road traffic facility traffic flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result; The OD matrix for predicting urban travel demand in core cities of a metropolitan area based on the travel chain theory includes: Dividing the urban area of ​​the core city of the metropolitan area into regions to obtain a plurality of traffic zones; Integrate the population classification, job type, income level and motor vehicle data of the core cities in the metropolitan area to obtain the cross-classification results of the urban population; According to the cross-classification results of the urban population, the total number of trip chains of different trip chains with the traffic zone as the starting point of a day's trip is calculated by the generation rate method; Determining the destination selection probability and mode selection probability of the trip chain according to the trip chain parameters and the discrete choice model; Calculate the time-divided travel probability of the travel chain by probability method; The urban travel demand OD matrix of the core city of the metropolitan area is determined according to the total amount of the travel chain, the destination selection probability, the mode selection probability and the travel probability in different time periods.

2. The method for predicting urban area traffic demand according to claim 1, characterized in that: The basic data includes physical transportation facility data, traffic management and traffic operation data, current travel characteristics data, current road traffic data, current public transportation passenger flow data, physical roads, virtual roads, public transportation networks, and public transportation operation data; the supply facility model includes a motor vehicle supply facility model and a public transportation supply facility model; the supply facility model of the metropolitan area constructed based on the basic data of the metropolitan area includes: constructing a road network structure according to the physical roads and the virtual roads; Establishing the motor vehicle supply facility model of the metropolitan area based on the road network structure, the physical traffic facility data, the traffic management and traffic operation data, the current travel characteristic data, and the current road traffic data; A public transportation connection network is constructed based on the public transportation network, and the public transportation supply facility model of the metropolitan area is established based on the public transportation connection network and the public transportation operation data.

3. The method for predicting urban area traffic demand according to claim 1, characterized in that: The OD matrix of the global travel demand of the metropolitan area predicted according to the four-stage theory includes: Dividing the metropolitan area into regions to obtain multiple traffic centers; Based on the data of population, job position and trip purpose, a trip generation and attraction model is established through the cross-classification generation rate method to obtain the total number of trips and the total number of trips attracted by different groups and different trip purposes in different traffic centers. Establishing a travel distribution model of the total amount of trips generated and the total amount of trips attracted by using a traffic impedance function, and calculating the full-day travel demand OD matrix of different traffic centers based on the travel distribution model; Based on the travel utility parameters and travel mode division model, the travel mode sharing ratio among different traffic centers based on different traffic demand choices is determined; Determine the travel time probability ratio among different traffic centers based on the full-day travel demand according to the travel time division model; According to the full-day travel demand OD matrix, the travel mode share rate and the travel probability ratio of the time period, the full-area travel demand OD matrix of the time period and travel mode between the multiple traffic centers in the metropolitan area is determined.

4. The method for predicting urban area traffic demand according to claim 3, characterized in that: The trip generation and attraction model includes a trip generation model and a trip attraction model. The trip generation and attraction model is established by the cross-classification generation rate method to obtain the total trip generation and total trip attraction for different groups of people and different travel purposes in different traffic centers, including: Determining the total amount of trips in the traffic center area based on the trip generation model according to the population classification data and the trip purpose classification data; According to the job type data and the travel purpose classification data, the total amount of travel attraction in the traffic center is determined based on the travel attraction model.

5. The method for predicting urban area traffic demand according to claim 4, characterized in that: The calculation formula of the trip generation model is: ； Where, is the amount of traffic generated in the middle zone i and the travel purpose P; is the cross-classification generation rate of zone i, group c, and travel purpose p; is the number of people in category c in zone i; m is the total number of categories of people in zone i; The calculation formula of the travel attraction model is: ； Where: is the attraction of the traffic center area i and the travel purpose P; The regression coefficient related to travel purpose P and job type w; is the number of positions of position type w in zone i corresponding to travel purpose P.

6. The method for predicting urban area traffic demand according to claim 3, characterized in that: The travel distribution model of the total amount of trips generated and the total amount of trips attracted is established by using the traffic impedance function, and the full-day travel demand OD matrix of different traffic centers is calculated according to the travel distribution model, including: Calculating the comprehensive traffic impedance of the traffic center area based on the distance data, time data and cost data of the metropolitan area; Determining a traffic impedance function of the traffic center area based on the comprehensive traffic impedance and a gamma function; According to the traffic impedance function, the travel distribution model is established based on the double-constraint gravity model, and the full-day travel demand OD matrix of different traffic centers is calculated according to the travel distribution model.

7. The method for predicting urban area traffic demand according to claim 6, characterized in that: The discrete choice model includes a single-constraint destination discrete choice model and a double-constraint destination discrete choice model; the travel chain parameters include at least the distance data, the time data and the cost data; Determining the destination selection probability of the trip chain according to the trip chain parameters and the discrete choice model includes: Determining a single trip utility function between the traffic zones according to the distance data, the time data, and the cost data, and updating a travel chain utility function of a destination selection model according to the single trip utility function; determining the destination selection probabilities of the different travel chains respectively according to the destination selection model, the single-constraint destination discrete selection model, and the double-constraint destination discrete selection model; When the destination of the travel chain includes multiple stopover points, the destination selection probabilities corresponding to the multiple stopover points are determined respectively according to the single-constraint destination discrete choice models corresponding to the stopover points.

8. The method for predicting urban area traffic demand according to claim 2, characterized in that: The data integration of the global travel demand OD matrix and the city travel demand OD matrix to obtain a data integration result, and determining the road traffic facility traffic flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result, includes: Based on the preset data integration rules, the global travel demand OD matrix and the city travel demand OD matrix are integrated to obtain a high-precision travel demand OD matrix for the metropolitan area at different time periods and different transportation modes; Calculating the traffic flow index of the road traffic facilities in the metropolitan area based on the motor vehicle travel demand OD matrix, the road delay function, the motor vehicle comprehensive cost function and the motor vehicle traffic allocation algorithm; The public transportation passenger flow index of the metropolitan area is calculated according to the public transportation travel demand OD matrix, the public transportation in-vehicle time function, the public transportation generalized cost function and the public transportation passenger flow allocation algorithm.

9. A metropolitan area traffic demand forecasting system, characterized in that: include: A supply facility model building module, configured to build a supply facility model of the metropolitan area based on the basic data of the metropolitan area; A global travel demand OD matrix prediction module, used to predict the global travel demand OD matrix of the metropolitan area based on the four-stage theory; The urban travel demand OD matrix prediction module is used to predict the urban travel demand OD matrix of the core cities in the metropolitan area based on the travel chain theory; a vehicle flow and passenger flow allocation module, configured to integrate the global travel demand OD matrix and the city travel demand OD matrix to obtain a data integration result, and determine the road traffic facility vehicle flow index and public transportation passenger flow index of the metropolitan area based on the supply facility model and the data integration result; The specific implementation of the urban travel demand OD matrix prediction module for predicting the urban travel demand OD matrix of the core cities of the metropolitan area based on the travel chain theory is as follows: Dividing the urban area of ​​the core city of the metropolitan area into regions to obtain a plurality of traffic zones; Integrate the population classification, job type, income level and motor vehicle data of the core cities in the metropolitan area to obtain the cross-classification results of the urban population; According to the cross-classification results of the urban population, the total number of trip chains of different trip chains with the traffic zone as the starting point of a day's trip is calculated by the generation rate method; Determining the destination selection probability and mode selection probability of the trip chain according to the trip chain parameters and the discrete choice model; Calculate the time-divided travel probability of the travel chain by probability method; The urban travel demand OD matrix of the core city of the metropolitan area is determined according to the total amount of the travel chain, the destination selection probability, the mode selection probability and the travel probability in different time periods.

Images