Microscopic three-dimensional simulation method and system for macro-micro integration of composite traffic network

Abstract

The invention discloses a macro-micro integrated microscopic three-dimensional simulation method and system for a composite traffic network, and relates to the technical field of traffic simulation. The method comprises the following steps: constructing a real-time linkage global dynamic traffic scene; carrying out multi-source traffic agent parameter fusion calibration; a front-end and back-end bidirectional integrated communication rendering module; and development of a composite traffic network scene reproduction analysis response dynamic visualization demonstration interaction and evaluation system and full-view simulation video stream archiving return visit analysis. On the basis of macro-micro integration of a composite traffic network, a micro three-dimensional simulation system is constructed through an improved centralized learning-distributed execution depth multi-source agent algorithm, micro simulation model parameter multi-dimensional calibration, a micro traffic flow individual interaction model and a micro traffic evaluation index calculation system; and a reliable and practical means is provided for three-dimensional simulation of urban traffic from a microscopic angle.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic simulation, and in particular to a microscopic three-dimensional simulation method and system for macro-micro integration of a composite traffic network. Background Art

[0002] With the continuous improvement of social economy, the improvement of residents' living conditions and the vigorous development of the new energy vehicle industry, the number of motor vehicles and travel demand have increased significantly, and most areas in my country are more or less facing urban traffic congestion problems. Urban traffic conditions are complex, with many emergencies, and various traffic conditions often occur, such as bad weather scenarios, road construction scenarios, isolated ramp bottleneck sections, active intervention traffic accident risk scenarios, bottleneck section traffic flow fluctuation scenarios, high truck ratio traffic scenarios, occasional continuous ramp multiple bottleneck sections scenarios, frequent continuous ramp multiple bottleneck sections scenarios, large-scale emergency incidents and accidents, and emergency lane management scenarios. The above traffic scenarios will cause serious and large-scale congestion, greatly affecting urban traffic functions and residents' daily life and travel.

[0003] Traffic simulation technology can simulate the traffic congestion conditions that often occur in cities, restore traffic conditions in a low-cost, low-manpower and high-efficiency manner, and provide technical support for cities to specify traffic management and control strategies. It is an indispensable part of urban traffic congestion management. However, the current microscopic three-dimensional traffic simulation technology relies on a single special scenario to calibrate parameters of a specific single data type, focusing on the behavior of individual vehicles, but failing to fully consider the interaction between vehicles. It also lacks a systematic microscopic traffic evaluation index calculation system and cannot effectively adapt to the needs of different traffic scenarios, resulting in reduced reliability and accuracy of simulation results, and a lack of scientific basis for the formulation of traffic management and control strategies.

[0004] To this end, we propose a microscopic three-dimensional simulation method and system for macro-micro integration of composite transportation networks. Summary of the invention

[0005] The purpose of the present invention is to provide a microscopic three-dimensional simulation method and system for macro-micro integration of a composite transportation network to solve the problems raised in the above-mentioned background technology.

[0006] According to the first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a microscopic three-dimensional simulation method for macro-micro integration of a composite transportation network, specifically comprising the following steps:

[0007] Build a real-time linkage global dynamic traffic scene: Establish a real-time linkage global dynamic traffic scene interface, develop a high-precision editable three-dimensional traffic scene twin module based on the high-precision GIS map information platform API, and build a full-traffic four-dimensional scene model platform by online editing of weather and road property models and adding the time change dimension on the basis of three-dimensional space;

[0008] Multi-source traffic agent parameter fusion calibration: construct control unit agent, traveler agent, road sensor agent and communication unit agent, and through the collaborative information collection and transmission of multiple agents, process the actually collected multi-source micro travel data, perform gradient update, and realize multi-dimensional calibration of micro simulation model parameters through the multi-dimensional fusion algorithm matrix of traffic scene parameters.

[0009] Front-end and back-end two-way integrated communication rendering module: Based on the mature API of the micro-simulation platform, a front-end and back-end two-way integrated data communication module is developed to achieve macro-simulation evolution while simultaneously rendering the vehicle trajectory and micro-simulation results in real time in the three-dimensional scene;

[0010] Reproduction, analysis and response of complex traffic network scenarios: For large-scale complex traffic network scenarios, a real-time model of typical traffic scenarios in urban road networks and a rapid analysis and calculation model are constructed to achieve timely response of control strategies in the micro dimension;

[0011] Development of dynamic visualization demonstration interaction and evaluation system: Based on the front-end and back-end two-way integrated communication rendering module, a microscopic traffic evaluation index calculation system is constructed based on the principal component fuzzy fusion algorithm to realize the visualization of simulation analysis results on the screen;

[0012] Full-view simulation video stream archiving and revisit analysis: While the simulation is in progress, key data is stored in the background for revisit analysis.

[0013] Furthermore, the global dynamic traffic scene interface is linked in real time, and the time change dimension is added on the basis of three-dimensional space to build a four-dimensional traffic scene model platform, as follows:

[0014] (11) Linking with the GIS map platform API, high-precision real-time conversion and positioning of spatial geographic location information of road property models around road scenes, where the road property models include roadside greening facilities, buildings, signs and markings, and roadside equipment;

[0015] (12) By specifying the model size, border, color and multi-dimensional information through parameters, the full traffic scene model itself is digitally and dynamically defined;

[0016] (13) The interface supports seamless integration of real-time data streams and dynamic updates of the entire scene model module. Examples of dynamic scene updates include real-time changes to road alignment, real-time changes to road signs, and real-time addition or deletion of static obstacles.

[0017] Furthermore, the multi-source traffic agent parameter fusion calibration specifically includes the following:

[0018] (21) Constructing a signal control unit agent, a traveler agent, a road sensor agent, and a communication unit agent; the signal control unit agent is used to control traffic signals, the traveler agent is used to make intelligent predictions of traffic conditions based on historical traffic data, the road sensor agent is used to make intelligent and accurate collaborative perceptions of road traffic volume, and the communication unit agent is used to perform preliminary processing and transmission of traffic flow data;

[0019] (22) Through the collaborative information collection and transmission of multiple intelligent agents, the collected information is subjected to the improved centralized learning-distributed execution deep multi-source intelligent agent algorithm:

[0020] θ=(θ1,…,θ i ) represents the parameters of the signal control unit agent, the traveler agent, the road sensor agent, and the communication unit agent. The gradient is updated through the strategy optimization function and gradient function of each traffic agent. The optimization function and its gradient function are:

[0021]

[0022] Where: i Represents the discount factor of each agent, which represents the degree of influence of each agent on the traffic conditions, usually between 0 and 1; r i,t Represents the reward function of each agent for improving traffic conditions at time t; β i The correction bias parameter representing the impact of each agent on the traffic conditions; π i Represents the strategy of each agent; t i represents the state value currently collected by each traffic agent; α=(t1,…,t i ) represents the state value vector collected by each traffic agent through centralized learning; a i,n Represents the nth control strategy action executed by each traffic agent; Represents the traffic impact-control strategy action function under the improved centralized learning-distributed execution of each traffic agent;

[0023] (23) Multi-dimensional calibration refers to the three dimensions of scene, vehicle type, and traffic flow parameters. The underlying layer selects various traffic conditions in complex urban traffic as scenes. The above traffic scenes realize the multi-dimensional calibration of micro-simulation model parameters through the multi-dimensional fusion algorithm matrix of traffic scene parameters;

[0024] There are various types of vehicles corresponding to each scenario. In complex urban traffic, there are many types of vehicles, including family cars, taxis, buses, trucks, trailers and non-motor vehicles. At the same time, different types of vehicles have different traffic parameters. Following, lane changing and speed are selected as the matrix parameters of the last layer. The three traffic scene parameters are multi-dimensionally integrated to form a set of traffic scene parameter multi-dimensional fusion algorithm matrix:

[0025]

[0026]

[0027] micropra ij =[CF LC CS]

[0028] Among them, Microsim p is a multidimensional microscopic parameter matrix, scene n is the dimensional parameter matrix of a typical traffic scene, micropra ij is the vehicle model dimension parameter matrix under the typical traffic scenario dimension, CF is the vehicle following parameter model library, LC is the vehicle lane changing parameter model library, and CS is the vehicle speed model library.

[0029] Furthermore, various traffic conditions in complex urban traffic are selected as scenarios, which are specifically divided into severe weather scenarios, road construction scenarios, isolated ramp bottleneck section scenarios, active intervention traffic accident risk scenarios, bottleneck section traffic flow fluctuation scenarios, high truck ratio traffic scenarios, occasional continuous ramp multiple bottleneck section scenarios, frequent continuous ramp multiple bottleneck section scenarios, large-scale sudden emergency events and accidents, and emergency lane management scenarios.

[0030] Furthermore, the front-end and back-end bidirectional integrated communication rendering module uses JSON format data, which specifically includes the following:

[0031] Long polling technology used to simulate real-time communication: the client sends a request to the server, and the server keeps the connection open until there is data to send to the client or it times out, achieving two-way communication;

[0032] Front-end implementation: Use JavaScript to send long polling requests to the server and process the data received from the server;

[0033] Backend implementation: Write code in Python to process requests from the frontend and send data to the frontend as needed. At the same time, set a timeout on the server and send a response to the client when there is new data.

[0034] Message format and protocol: When transmitting data between the front-end and back-end, define a unified message format and communication protocol to ensure that the front-end and back-end can correctly parse and process the data.

[0035] Furthermore, a real-time model and a rapid analysis and calculation model for typical traffic scenarios in urban road networks are constructed, including the following:

[0036] (41) Call the digital definition interface of the full traffic scene model, link the GIS map platform to locate the scene construction content, and start updating the road three-dimensional modeling in real time;

[0037] (42) Use information collection equipment to monitor road conditions in real time, receive multi-source micro-travel data through the front-end and back-end two-way integrated communication rendering module, including road event type, event occurrence time, event generation space coordinates, and traffic flow scene information around the event occurrence point, and perform real-time rendering simulation of the traffic flow module;

[0038] (43) Form a fast computational analysis model for micro-simulation to predict simulation results for management and control decisions.

[0039] Furthermore, the fast computational analysis model for microscopic simulation mainly includes traffic flow simulation at the microscopic level. In order to reflect the diversity of interactions between vehicles and driving behaviors, a microscopic traffic flow individual interaction model is constructed. Each vehicle is regarded as an independent intelligent agent and simulated according to its individual characteristics and behavior rules. The details are as follows:

[0040] (431) Individual state definition: Each vehicle is defined as an individual, and its state includes information such as position, speed, acceleration, lane position, etc. The state formula is as follows:

[0041] s i (t) = (x i (t),v i (t),a i (t),l i (t))

[0042] Where: x i (t) represents the position of vehicle i at time t, v i (t) represents speed, a i (t) represents acceleration, l i (t) represents the lane position;

[0043] (432) Individual behavior rules: define the situations in which the vehicle follows the speed of the vehicle in front, maintains a safe following distance, and limits the maximum acceleration. In addition, lane position information is introduced into the behavior rule function and used to adjust the vehicle's acceleration. The formula of the individual behavior rule is as follows:

[0044] a i (t) = min(a max ,max(0,f desired (v i (t), Δx i (t))+f safe (Δv i (t),Δx i (t))+f lane (l i (t),l desired (i))))

[0045] In the formula, a max is the maximum acceleration limit of the vehicle; Δv i (t) is the speed difference between vehicle i and the preceding vehicle at time t; Δx i (t) is the distance difference between vehicle i and the preceding vehicle at time t; f desired is a function of the vehicle's drag speed, which depends on the relative speed of the vehicle and the preceding vehicle, f safe It is a function of the safe following distance maintained by the vehicle, which depends on the distance and speed difference between the vehicle and the vehicle ahead; l i (t) is the lane position of vehicle i at time t; l desired (i) is the desired lane position of vehicle i; f lane is the lane position adjustment function that adjusts the vehicle acceleration, depending on the difference between the current lane and the desired lane position, using cubic spline interpolation.

[0046] (433) Define individual interaction rules. The interaction rules defined here adjust the speed according to the vehicle spacing, speed difference, acceleration difference and prediction of other vehicle behaviors to avoid collision or maintain an appropriate spacing. The formula of the individual interaction rule function is as follows:

[0047]

[0048] In the formula, v desired is the desired speed function of the vehicle, which is adjusted according to the distance to the vehicle ahead and the acceleration of the current vehicle; v safe is the speed function when maintaining a safe distance, which is adjusted according to the distance to the vehicle in front; T is a time constant used to control the safe distance between vehicles; T pis a prediction time, used to predict the speed of other vehicles; α, β, γ, δ, η, θ are adjustment parameters with a value range of [0, 1], used to control the impact of different factors on vehicle speed;

[0049] (434) Define the overall system behavior function of the traffic flow, which is specifically expressed as:

[0050]

[0051] Where: ρ(t) is the density of traffic flow at time t, which can be calculated based on the number of vehicles and road length; represents the acceleration change rate of vehicle i, represents the rate of change of traffic flow density; ω1, ω2, ω3, ω4, ω5 are the weights of various factors, which are used to adjust the impact of each factor on the overall behavior; other factors include weather factors and road control.

[0052] Furthermore, a microscopic traffic evaluation index calculation system is constructed based on the principal component fuzzy fusion algorithm, which specifically includes the following:

[0053] (51) Specify the traffic evaluation object system: For complex urban traffic scenarios, select vehicle delay time, number of stops, travel speed, travel time and fuel consumption indicators to construct an evaluation index set, P = {u1,u2,…,u i},u i = {u i1 ,u i2 ,…,u ij}(i=1,2,…n)(j=1,2,…m) is a single set of various traffic evaluation indicators;

[0054] (52) The traffic index standardization processing method is expressed as:

[0055]

[0056] Where: is the average of the set of i traffic evaluation indicators, s i is the standard deviation of the set of i-item traffic evaluation indicators; α ij is u ij Indicator relative to the overall u i The scaling index of the indicator set is between 0 and 1; β i It is the indicator factor of each traffic indicator, representing the degree of indication of each traffic indicator on the traffic conditions; ε i is the dimensional conversion parameter of each traffic index, representing the sensitivity of the dimensional indication of traffic conditions; the correction bias parameter δ i Correction dimension conversion parameters;

[0057] (53) Principal component analysis of traffic evaluation index data: Analyze the index set, extract the traffic evaluation index component load matrix, score matrix and weight coefficient, and obtain the traffic evaluation index weight vector C = {c1, c2, …, c i}, representing vehicle delay time Vt, number of stops Vs, travel speed Vv, travel time VT and fuel consumption Ve;

[0058] (54) Determine the fuzzy evaluation level of traffic indicators and establish the fuzzy evaluation matrix R of traffic indicators: Determine the fuzzy evaluation level of traffic indicators as needed into i layers: V = {v1, v2, …, v i}, from high to low;

[0059] According to the fuzzy evaluation level of traffic indicators, the evaluation indicator set P is fuzzy judged, and the traffic indicator fuzzy evaluation matrix R is established according to the frequency to complete the fuzzy evaluation of indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption:

[0060]

[0061] Where: γ i,i is the discount factor of the i-th traffic index to the i-th fuzzy evaluation level, usually between 0 and 1; θ i,i is the fuzzy weight of the i-th level fuzzy evaluation level for the i-th traffic index pair;

[0062] (55) Synthesize the evaluation result of the principal component fuzzy fusion algorithm: The traffic evaluation index weight vector C and the traffic index fuzzy evaluation matrix R are synthesized to obtain the evaluation result S of the principal component fuzzy fusion algorithm:

[0063]

[0064] The evaluation result S is evaluated to complete the evaluation of traffic indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption.

[0065] According to a second aspect of the present invention, the present invention provides a microscopic three-dimensional simulation system for macro-micro integration of a composite transportation network, which is used to implement the above-mentioned microscopic three-dimensional method for macro-micro integration of a composite transportation network, comprising:

[0066] Build a module to establish a real-time linkage global dynamic traffic scene interface. Based on the high-precision GIS map information platform API, develop a high-precision editable three-dimensional traffic scene twin module. Through real-time online editing of weather and road property models, add the time change dimension on the basis of three-dimensional space to build a full traffic four-dimensional scene model platform;

[0067] The multi-dimensional calibration module is used to construct control unit agents, traveler agents, road sensor agents and communication unit agents. Through the collaborative information collection and transmission of multiple agents, the multi-source micro travel data actually collected is processed and gradient updated. The multi-dimensional fusion algorithm matrix of traffic scene parameters is used to realize the multi-dimensional calibration of micro simulation model parameters.

[0068] The communication rendering module is used to develop a two-way integrated data communication module for the front and back ends based on the mature API of the micro-simulation platform, so as to achieve real-time rendering of the vehicle trajectory and micro-simulation results in the three-dimensional scene while realizing the evolution of the macro-simulation;

[0069] The analysis and response module is used to build a real-time model and a rapid analysis and calculation model for typical traffic scenarios in urban road networks for large-scale complex traffic network scenarios, so as to achieve timely response of management and control strategies in the micro dimension;

[0070] Visualization module, which is used to build a microscopic traffic evaluation index calculation system based on the main component fuzzy fusion algorithm based on the front-end and back-end two-way integrated communication rendering module, and realize the visualization of simulation analysis results on the screen;

[0071] The archiving module is used to store key data in the background while the simulation is in progress, for follow-up analysis.

[0072] According to the third aspect of the present invention, the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned macro-micro integrated microscopic three-dimensional simulation method of a composite transportation network is adopted.

[0073] The present invention has at least the following beneficial effects:

[0074] The present invention uses a macro-micro integrated model to more comprehensively consider various traffic scenarios and factors, improve the accuracy of traffic flow simulation, and reflect the actual traffic conditions. At the same time, through multi-dimensional data calibration and micro traffic flow individual interaction models, it can effectively respond to a variety of complex traffic scenarios, such as severe weather, emergencies, etc., provide more flexible and real-time simulation support, help traffic management departments formulate emergency response measures, and provide a scientific basis for traffic management.

[0075] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a schematic diagram of the process of the method of the present invention;

[0077] Figure 2 It is a schematic diagram of the three-dimensional traffic scene twin module in the present invention;

[0078] Figure 3 This is a schematic diagram of the screen end of the high-precision model platform for all traffic scenarios in the present invention;

[0079] Figure 4 It is a schematic diagram of a road construction scene in the present invention;

[0080] Figure 5 A schematic diagram of an isolated ramp bottleneck section scenario in the present invention;

[0081] Figure 6 A schematic diagram of a three-dimensional simulation demonstration of a road construction scene in the present invention;

[0082] Figure 7 A schematic diagram of a three-dimensional simulation demonstration of an isolated ramp bottleneck section scenario in the present invention;

[0083] Figure 8 It is a schematic diagram of the simulation analysis results in the present invention. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0085] Embodiment 1:

[0086] See also Figure 1-Figure 8 The present invention provides a technical solution: a microscopic three-dimensional simulation method for a composite transportation network with macro-micro integration, comprising the following steps:

[0087] S10. Real-time linkage global dynamic traffic scene construction: A real-time linkage global dynamic traffic scene interface is proposed. Based on the high-precision GIS map information platform API, a high-precision editable three-dimensional traffic scene twin module is developed. Through real-time online editing of weather and road property models, the time change dimension is added on the basis of three-dimensional space to complete the four-dimensional traffic scene construction function. Based on the precise road geometry line information, a two-dimensional underlying road network editing module is developed to build a multi-dimensional simulation scene integrated restoration function to improve the dynamic and global nature of the simulation environment. The details are as follows:

[0088] (S11) Linking with the GIS map platform API to perform high-precision real-time conversion and positioning of spatial geographic location information of road property models around the road scene, wherein the road property model includes roadside greening facilities, buildings, signs and markings, and roadside equipment;

[0089] (S12) defining the full traffic scene model digitally and dynamically by specifying model size, border, color and multi-dimensional information through parameters;

[0090] (S13) The interface supports seamless integration of real-time data streams to dynamically update the overall module of the scene model, where examples of dynamic scene updates include real-time changes in road alignment, real-time changes in road signs, and real-time addition or deletion of static obstacles;

[0091] S20. Multi-source traffic agent parameter fusion calibration: Construct control unit agent, traveler agent, road sensor agent and communication unit agent, and through the coordinated information collection and transmission of multiple agents, process the multi-source micro travel data actually collected, perform gradient update, and realize the multi-dimensional calibration of micro simulation model parameters through the multi-dimensional fusion algorithm matrix of traffic scene parameters, as follows:

[0092] (S21) constructing a signal control unit agent, a traveler agent, a road sensor agent, and a communication unit agent; the signal control unit agent is used to control traffic signals, the traveler agent is used to make intelligent predictions of traffic conditions based on historical traffic data, the road sensor agent is used to make intelligent and accurate collaborative perceptions of road traffic volume, and the communication unit agent is used to perform preliminary processing and transmission of traffic flow data;

[0093] (S22) Through the collaborative information collection and transmission of multiple agents, the collected information is subjected to the improved centralized learning-distributed execution deep multi-source agent algorithm:

[0094] θ=(θ1,…,θ i ) represents the parameters of the signal control unit agent, the traveler agent, the road sensor agent, and the communication unit agent. The gradient is updated through the strategy optimization function and gradient function of each traffic agent. The optimization function and its gradient function are:

[0095]

[0096] Where: i Represents the discount factor of each agent, which represents the degree of influence of each agent on the traffic conditions, usually between 0 and 1; r i,t Represents the reward function of each agent for improving traffic conditions at time t; β i The correction bias parameter representing the impact of each agent on the traffic conditions; π i Represents the strategy of each agent; t i represents the state value currently collected by each traffic agent; α=(t1,…,t i) represents the state value vector collected by each traffic agent through centralized learning; a i,n Represents the nth control strategy action executed by each traffic agent; Represents the traffic impact-control strategy action function under the improved centralized learning-distributed execution of each traffic agent;

[0097] (S23) Multi-dimensional calibration refers to three dimensions: scenario, vehicle type, and traffic flow parameters. The bottom layer selects various traffic conditions in complex urban traffic as scenarios, which are specifically divided into severe weather scenarios, road construction scenarios, isolated ramp bottleneck sections scenarios, active intervention traffic accident risk scenarios, bottleneck section traffic flow fluctuation scenarios, high truck ratio traffic scenarios, occasional continuous ramp multi-bottleneck sections scenarios, frequent continuous ramp multi-bottleneck sections scenarios, large-scale emergency incidents and accidents, and emergency lane management scenarios. The above traffic scenarios realize multi-dimensional calibration of micro-simulation model parameters through a multi-dimensional fusion algorithm matrix of traffic scenario parameters;

[0098] There are various types of vehicles corresponding to each scenario. In complex urban traffic, there are many types of vehicles, including family cars, taxis, buses, trucks, trailers and non-motor vehicles. At the same time, different types of vehicles have different traffic parameters. Following, lane changing and speed are selected as the matrix parameters of the last layer. The three traffic scene parameters are multi-dimensionally integrated to form a set of traffic scene parameter multi-dimensional fusion algorithm matrix:

[0099]

[0100] micropra ij =[CF LC CS]

[0101] Among them, Microsim p is a multidimensional microscopic parameter matrix, scene n is the dimensional parameter matrix of a typical traffic scene, micropra ij is the vehicle model dimension parameter matrix under the typical traffic scenario dimension, CF is the vehicle following parameter model library, LC is the vehicle lane changing parameter model library, and CS is the vehicle speed model library;

[0102] S30. Front-end and back-end two-way integrated communication rendering module: Based on the mature API of the micro-simulation platform, a front-end and back-end two-way integrated data communication module is developed to integrate the reception of macro-simulation data results and the sending of micro-simulation results. While realizing the evolution of macro-simulation, the micro-simulation results such as vehicle trajectories and specific control information are synchronously rendered in real time in the three-dimensional scene. Specifically, it includes the following:

[0103] Long Polling Technology: Long polling technology is a method of simulating real-time communication. The client sends a request to the server, and the server keeps the connection open until there is data to send to the client or it times out, thus achieving two-way communication.

[0104] Front-end implementation: Use JavaScript to send long polling requests to the server and process the data received from the server;

[0105] Backend implementation: Write code in Python to process requests from the frontend and send data to the frontend as needed. At the same time, set an appropriate timeout on the server and send a response to the client when there is new data.

[0106] Message format and protocol: When transmitting data between the front-end and back-end, a unified message format and communication protocol need to be defined to ensure that both parties can correctly parse and process the data;

[0107] This module uses JSON format data;

[0108] S40. Reproduction analysis and response of complex traffic network scenarios: For large-scale complex traffic network scenarios, a real-time model of typical traffic scenarios in urban road networks is constructed: road construction scenarios, isolated ramp bottleneck road scenarios, such as Figure 4 , Figure 5 As shown in the figure, as well as the rapid analysis and calculation model, timely response of the control strategy in the micro dimension is achieved, as follows:

[0109] (S41) calling the networking digital definition interface of the full traffic scene model, linking the GIS map platform to locate the scene construction content, and starting to update the road three-dimensional modeling in real time;

[0110] (S42) using information collection equipment to monitor road conditions in real time, receiving multi-source micro-travel data through a front-end and back-end two-way integrated communication rendering module, including road event type, event occurrence time, event generation space coordinates, and traffic flow scene information around the event occurrence point, and performing real-time rendering simulation of the traffic flow module;

[0111] (S43) Forming a fast calculation and analysis model for microscopic simulation to predict the simulation results of control decisions, mainly including microscopic traffic flow simulation. In order to reflect the interaction between vehicles and the diversity of driving behaviors, a microscopic traffic flow individual interaction model is constructed. Each vehicle is regarded as an independent intelligent agent and simulated according to its individual characteristics and behavior rules. The details are as follows:

[0112] (431) Individual state definition: Each vehicle is defined as an individual, and its state includes information such as position, speed, acceleration, lane position, etc. The state formula is as follows:

[0113] s i (t) = (x i (t),v i (t),a i (t),l i (t))

[0114] Where: x i (t) represents the position of vehicle i at time t, v i (t) represents speed, a i (t) represents acceleration, l i (t) represents the lane position;

[0115] (432) Individual behavior rules: define the situations in which the vehicle follows the speed of the vehicle in front, maintains a safe following distance, and limits the maximum acceleration. In addition, under normal circumstances, the vehicle will try to stay in its lane, but may also consider changing lanes. In the behavior rule function, lane position information is introduced and used to adjust the vehicle's acceleration. The formula for the individual behavior rule is as follows:

[0116] a i (t) = min(a max ,max(0,f desired (v i (t), Δx i (t))+f safe (Δv i (t),Δx i (t))+f lane (l i (t),l desired (i))))

[0117] In the formula, a max is the maximum acceleration limit of the vehicle; Δv i (t) is the speed difference between vehicle i and the preceding vehicle at time t; Δx i (t) is the distance difference between vehicle i and the preceding vehicle at time t; f desired is a function of the vehicle's drag speed, which depends on the relative speed of the vehicle and the preceding vehicle, f safe It is a function of the safe following distance maintained by the vehicle, which depends on the distance and speed difference between the vehicle and the vehicle in front. These two functions are usually fitted based on experience or actual data.

[0118] Among them, l i (t) is the lane position of vehicle i at time t; l desired (i) is the desired lane position of vehicle i; f lane is the lane position adjustment function that adjusts the vehicle acceleration, depending on the difference between the current lane and the desired lane position, using cubic spline interpolation.

[0119] (433) Define individual interaction rules. The interaction rules defined here adjust the speed according to the vehicle spacing, speed difference, acceleration difference and prediction of other vehicle behaviors to avoid collision or maintain an appropriate spacing. The formula of the individual interaction rule function is as follows:

[0120]

[0121] In the formula, v desired is the desired speed function of the vehicle, which is adjusted according to the distance to the vehicle ahead and the acceleration of the current vehicle; v safe is the speed function when maintaining a safe distance, which is adjusted according to the distance to the vehicle in front; T is a time constant used to control the safe distance between vehicles; T p is a prediction time, used to predict the speed of other vehicles; α, β, γ, δ, η, θ are adjustment parameters with a value range of [0, 1], used to control the impact of different factors on vehicle speed;

[0122] (434) Define the overall system behavior function of the traffic flow, which is specifically expressed as:

[0123]

[0124] Where: ρ(t) is the density of traffic flow at time t, which can be calculated based on the number of vehicles and road length; represents the acceleration change rate of vehicle i, represents the rate of change of traffic flow density; ω1, ω2, ω3, ω4, ω5 are the weights of various factors, which are used to adjust the impact of each factor on the overall behavior; other factors include weather factors and road control;

[0125] S50. Development of dynamic visualization demonstration interaction and evaluation system: Based on the front-end and back-end two-way integrated communication rendering module, a microscopic traffic evaluation index calculation system is constructed based on the principal component fuzzy fusion algorithm to realize the visualization of simulation analysis results on the screen, as follows:

[0126] (51) Specify the traffic evaluation object system: For complex urban traffic scenarios, select vehicle delay time, number of stops, travel speed, travel time and fuel consumption indicators to construct an evaluation index set, P = {u1,u2,…,u i},u i = {u i1 ,u i2 ,…,u ij}(i=1,2,…n)(j=1,2,…m) is a single set of various traffic evaluation indicators;

[0127] (52) The traffic index standardization method is expressed as:

[0128]

[0129] Where: is the average of the set of i traffic evaluation indicators, s i is the standard deviation of the set of i-item traffic evaluation indicators; α ij is u ij Indicator relative to the overall u i The scaling index of the indicator set is between 0 and 1; β i It is the indicator factor of each traffic indicator, representing the degree of indication of each traffic indicator on the traffic conditions; ε i is the dimensional conversion parameter of each traffic index, representing the sensitivity of the dimensional indication of traffic conditions; the correction bias parameter δ i Correction dimension conversion parameters;

[0130] (53) Principal component analysis of traffic evaluation index data: Analyze the index set, extract the traffic evaluation index component load matrix, score matrix and weight coefficient, and obtain the traffic evaluation index weight vector C = {c1, c2, …, c i}, representing vehicle delay time Vt, number of stops Vs, travel speed Vv, travel time VT and fuel consumption Ve;

[0131] (54) Determine the fuzzy evaluation level of traffic indicators and establish the fuzzy evaluation matrix R of traffic indicators: Determine the fuzzy evaluation level of traffic indicators as needed into i layers: V = {v1, v2, …, v i}, from high to low;

[0132] According to the fuzzy evaluation level of traffic indicators, the evaluation indicator set P is fuzzy judged, and the traffic indicator fuzzy evaluation matrix R is established according to the frequency to complete the fuzzy evaluation of indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption:

[0133]

[0134] Where: γ i,i is the discount factor of the i-th traffic index to the i-th fuzzy evaluation level, usually between 0 and 1; θ i,i is the fuzzy weight of the i-th level fuzzy evaluation level for the i-th traffic index pair;

[0135] (55) Synthesize the evaluation result of the principal component fuzzy fusion algorithm: The traffic evaluation index weight vector C and the traffic index fuzzy evaluation matrix R are synthesized to obtain the evaluation result S of the principal component fuzzy fusion algorithm:

[0136]

[0137] Evaluate the evaluation result S and complete the evaluation of traffic indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption;

[0138] S60. Archiving and revisiting analysis of full-view simulation video streams: While the simulation is in progress, key data is stored in the background for revisiting and analysis.

[0139] Next, the present invention is further described in conjunction with specific embodiments:

[0140] like Figure 1 The present embodiment describes a microscopic three-dimensional simulation method based on macro-micro integration of a composite transportation network, the flow chart of which is as follows: Figure 1 As shown, the steps are as follows:

[0141] Step 1: Based on the real-time linkage global dynamic traffic scene interface and the high-precision GIS map information platform API, develop a high-precision editable three-dimensional traffic scene twin module, through real-time online editing of weather and road property models, such as Figure 2 As shown;

[0142] On the basis of three-dimensional space, the time change dimension is added to complete the four-dimensional scene construction function of the whole traffic. Based on the precise road geometry information, a two-dimensional underlying road network editing module is developed to build a high-precision model platform for the whole traffic scene, such as Figure 3 As shown;

[0143] Step 2: Based on the multi-source micro-travel data actually collected, the multi-agent collaborative sensing interface of the improved centralized learning-distributed execution deep multi-source agent algorithm is used, including the signal control unit agent, the traveler agent, the road sensor agent, and the communication unit agent, to perform gradient updates on the optimization function and gradient function of each traffic agent strategy. The optimization function and its gradient function are:

[0144]

[0145] Various traffic conditions that often occur in complex urban traffic are selected as scenarios, such as road construction scenarios and isolated ramp bottleneck sections. The micro-simulation model parameter scenario, vehicle type, and traffic flow parameter calibration are realized through the multi-dimensional fusion algorithm matrix of traffic scenario parameters.

[0146] According to the various types of vehicles corresponding to the scene, as well as the different traffic parameters of following, lane changing and speed of each type of vehicle, a set of traffic scene parameter multi-dimensional fusion algorithm matrix is ​​formed:

[0147]

[0148] micropra ij =[CF LC CS]

[0149] Step 3: Based on the mature API of the micro-simulation platform, a front-end and back-end two-way data communication module is developed to integrate the reception of macro-simulation data results and the transmission of micro-simulation results. While realizing the evolution of macro-simulation, the micro-simulation results such as vehicle trajectories and specific control information are synchronously rendered in real time in the three-dimensional scene;

[0150] The front-end and back-end two-way data communication module establishes a mechanism between the user interface and the server, which enables the front-end and back-end to transmit data to each other in real time. It includes the following components: Long polling technology: Long polling technology is a method of simulating real-time communication. The client sends a request to the server, and the server keeps the connection open until there is data to send to the client or timeout, so as to achieve two-way communication; front-end implementation: by using JavaScript to send long polling requests to the server and process the data received from the server; back-end implementation: write code in Python to process the request sent by the front-end, and send data to the front-end as needed. At the same time, set an appropriate timeout on the server and send a response to the client when there is new data; message format and protocol: when transmitting data between the front-end and back-end, it is necessary to define a unified message format and communication protocol to ensure that both parties can correctly parse and process data. This module uses JSON format data.

[0151] Step 4: Based on the data of the real-time linkage global dynamic traffic scene interface and the front-end and back-end two-way data communication modules, a real-time model of typical traffic scenes in the urban road network is constructed for large-scale complex traffic network scenes: road construction scenes, isolated ramp bottleneck section scenes, such as Figure 4 , Figure 5 As shown, and a fast analysis calculation module;

[0152] The fast calculation and analysis module analyzes the microscopic traffic flow individual interaction model and calculates the individual status, individual behavior rules, individual interaction rule functions and overall system behavior functions:

[0153] s i (t) = (x i (t),v i (t),a i (t),l i (t))

[0154] a i (t) = min(a max ,max(0,f desired (v i (t), Δx i (t))+f safe (Δv i(t),Δx i (t))+f lane (l i (t),l desired (i))))

[0155]

[0156] Step 5: Based on the data communication module, a visual intelligent transportation front-end large screen is designed and developed to realize microscopic 3D simulation demonstration of road construction scenes and isolated ramp bottleneck sections, such as Figure 6 , 7 At the same time, the microscopic traffic evaluation index calculation system based on the principal component fuzzy fusion algorithm outputs the simulation analysis results, such as Figure 8 ;

[0157] In step 6, while the simulation is in progress, key data is stored in the background for multiple review by management and control decision makers, facilitating a more comprehensive analysis of the micro-behavior of urban travelers.

[0158] To sum up, this embodiment can more comprehensively consider various traffic scenarios and factors through the macro-micro integrated model, improve the accuracy of traffic flow simulation, and reflect the real traffic conditions. At the same time, through multi-dimensional data calibration and micro traffic flow individual interaction model, it can effectively respond to various complex traffic scenarios, such as bad weather, sudden accidents, etc., provide more flexible and real-time simulation support, help traffic management departments formulate emergency response measures, and provide a scientific basis for traffic management.

[0159] Embodiment 2:

[0160] The present invention provides a microscopic three-dimensional simulation system for macro-micro integration of a composite transportation network, which is used to implement the above-mentioned microscopic three-dimensional method for macro-micro integration of a composite transportation network, comprising:

[0161] Build a module to establish a real-time linkage global dynamic traffic scene interface. Based on the high-precision GIS map information platform API, develop a high-precision editable three-dimensional traffic scene twin module. Through real-time online editing of weather and road property models, add the time change dimension on the basis of three-dimensional space to build a full traffic four-dimensional scene model platform;

[0162] The multi-dimensional calibration module is used to construct control unit agents, traveler agents, road sensor agents and communication unit agents. Through the collaborative information collection and transmission of multiple agents, the multi-source micro travel data actually collected is processed and gradient updated. The multi-dimensional fusion algorithm matrix of traffic scene parameters is used to realize the multi-dimensional calibration of micro simulation model parameters.

[0163] The communication rendering module is used to develop a two-way integrated data communication module for the front and back ends based on the mature API of the micro-simulation platform, so as to achieve real-time rendering of the vehicle trajectory and micro-simulation results in the three-dimensional scene while realizing the evolution of the macro-simulation;

[0164] The analysis and response module is used to build a real-time model and a rapid analysis and calculation model for typical traffic scenarios in urban road networks for large-scale complex traffic network scenarios, so as to achieve timely response of management and control strategies in the micro dimension;

[0165] Visualization module, which is used to build a microscopic traffic evaluation index calculation system based on the main component fuzzy fusion algorithm based on the front-end and back-end two-way integrated communication rendering module, and realize the visualization of simulation analysis results on the screen;

[0166] The archiving module is used to store key data in the background while the simulation is in progress, for follow-up analysis.

[0167] Specifically, the above-mentioned building module, multi-dimensional calibration module, communication rendering module, analysis and response module, visualization module and archiving module can be embedded in a computer processing system. The computer can call the above-mentioned building module, multi-dimensional calibration module, communication rendering module, analysis and response module, visualization module and archiving module according to the above-mentioned microscopic three-dimensional simulation method for macro-micro integration of composite transportation network, and can perform operations according to the specific steps given in the microscopic three-dimensional simulation method for macro-micro integration of composite transportation network.

[0168] It should be noted that it should be understood that the division of the various modules of the above system is only the division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated, and these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the construction module can be a separately established processing element, or it can be integrated in a chip of the above-mentioned device. In addition, it can also be stored in the memory of the above-mentioned device in the form of program code, and a processing element of the above-mentioned device calls and executes the functions of the above signal processing module. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit in the processor element or the instructions in the form of software.

[0169] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0170] Embodiment three:

[0171] The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor. When the processor loads and executes the computer program, the microscopic three-dimensional simulation method of the macro-micro integrated composite transportation network in Example 1 is adopted.

[0172] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses.

[0173] Furthermore, the processor may adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and the present application does not impose any restrictions on this.

[0174] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0175] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a centered element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are only for illustrative purposes and are not intended to be the only implementation method.

[0176] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0177] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

Claims

1. A microscopic three-dimensional simulation method for macro-micro integration of composite transportation networks, characterized in that: The following steps are involved: Build a real-time linkage global dynamic traffic scene: Establish a real-time linkage global dynamic traffic scene interface, develop a high-precision editable three-dimensional traffic scene twin module based on the high-precision GIS map information platform API, and build a full-traffic four-dimensional scene model platform by adding the time change dimension on the basis of three-dimensional space through real-time online editing of weather and road property models; Multi-source traffic agent parameter fusion calibration: construct control unit agent, traveler agent, road sensor agent and communication unit agent, and through the collaborative information collection and transmission of multiple agents, process the actually collected multi-source micro travel data, perform gradient update, and realize multi-dimensional calibration of micro simulation model parameters through the multi-dimensional fusion algorithm matrix of traffic scene parameters. Front-end and back-end two-way integrated communication rendering module: Based on the mature API of the micro-simulation platform, a front-end and back-end two-way integrated data communication module is developed to achieve macro-simulation evolution while simultaneously rendering the vehicle trajectory and micro-simulation results in real time in the three-dimensional scene; Reproduction, analysis and response of complex traffic network scenarios: For large-scale complex traffic network scenarios, a real-time model of typical traffic scenarios in urban road networks and a rapid analysis and calculation model are constructed to achieve timely response of control strategies in the micro dimension; Development of dynamic visualization demonstration interaction and evaluation system: Based on the front-end and back-end two-way integrated communication rendering module, a microscopic traffic evaluation index calculation system is constructed based on the principal component fuzzy fusion algorithm to realize the visualization of simulation analysis results on the screen; Full-view simulation video stream archiving and revisit analysis: While the simulation is in progress, key data is stored in the background for revisit analysis.

2. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 1 is characterized in that: Real-time linkage of the global dynamic traffic scene interface, adding the time change dimension on the basis of three-dimensional space, and building a four-dimensional traffic scene model platform, as follows: (11) Linking with the GIS map platform API, high-precision real-time conversion and positioning of spatial geographic location information of road property models around road scenes, where the road property models include roadside greening facilities, buildings, signs and markings, and roadside equipment; (12) By specifying the model size, border, color and multi-dimensional information through parameters, the full traffic scene model itself is digitally and dynamically defined; (13) The interface supports seamless integration of real-time data streams and dynamic updates of the entire scene model module. Examples of dynamic scene updates include real-time changes to road alignment, real-time changes to road signs, and real-time addition or deletion of static obstacles.

3. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 2 is characterized by: The multi-source traffic agent parameter fusion calibration specifically includes the following: (21) Constructing a signal control unit agent, a traveler agent, a road sensor agent, and a communication unit agent; the signal control unit agent is used to control traffic signals, the traveler agent is used to make intelligent predictions of traffic conditions based on historical traffic data, the road sensor agent is used to make intelligent and accurate collaborative perceptions of road traffic volume, and the communication unit agent is used to perform preliminary processing and transmission of traffic flow data; (22) Through the collaborative information collection and transmission of multiple intelligent agents, the collected information is subjected to the improved centralized learning-distributed execution deep multi-source intelligent agent algorithm: θ=(θ1,…,θ i ) represents the parameters of the signal control unit agent, the traveler agent, the road sensor agent, and the communication unit agent. The gradient is updated through the strategy optimization function and gradient function of each traffic agent. The optimization function and its gradient function are: Where: i Represents the discount factor of each agent, which represents the degree of influence of each agent on the traffic conditions, usually between 0 and 1; r i,t Represents the reward function of each agent for improving traffic conditions at time t; β i The correction bias parameter representing the impact of each agent on the traffic conditions; π i Represents the strategy of each agent; t i represents the state value currently collected by each traffic agent; α=(t1,…,t i ) represents the state value vector collected by each traffic agent through centralized learning; a i,n represents the nth control strategy action executed by each traffic agent; Q i π Represents the traffic impact-control strategy action function under the improved centralized learning-distributed execution of each traffic agent; (23) Multi-dimensional calibration refers to the three dimensions of scene, vehicle type, and traffic flow parameters. The underlying layer selects various traffic conditions in complex urban traffic as scenes. The above traffic scenes realize the multi-dimensional calibration of micro-simulation model parameters through the multi-dimensional fusion algorithm matrix of traffic scene parameters; There are various types of vehicles corresponding to each scenario. In complex urban traffic, there are many types of vehicles, including family cars, taxis, buses, trucks, trailers and non-motor vehicles. At the same time, different types of vehicles have different traffic parameters. Following, lane changing and speed are selected as the matrix parameters of the last layer. The three traffic scene parameters are multi-dimensionally integrated to form a set of traffic scene parameter multi-dimensional fusion algorithm matrix: Microsim p =[scene1,scene2,scene3,…,scene n ] micropra ij =[CF LC CS] Among them, Microsim p is a multidimensional microscopic parameter matrix, scene n is the dimensional parameter matrix of a typical traffic scene, micropra ij is the vehicle model dimension parameter matrix under the typical traffic scenario dimension, CF is the vehicle following parameter model library, LC is the vehicle lane changing parameter model library, and CS is the vehicle speed model library.

4. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 3 is characterized in that: Various traffic conditions in complex urban traffic are selected as scenarios, which are specifically divided into severe weather scenarios, road construction scenarios, isolated ramp bottleneck section scenarios, active intervention traffic accident risk scenarios, bottleneck section traffic flow fluctuation scenarios, high truck ratio traffic scenarios, occasional continuous ramp multiple bottleneck section scenarios, frequent continuous ramp multiple bottleneck section scenarios, large-scale sudden emergency events and accidents, and emergency lane management scenarios.

5. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 3 is characterized by: The front-end and back-end bidirectional integrated communication rendering module uses JSON format data, which specifically includes the following: Long polling technology used to simulate real-time communication: the client sends a request to the server, and the server keeps the connection open until there is data to send to the client or it times out, achieving two-way communication; Front-end implementation: Use JavaScript to send long polling requests to the server and process the data received from the server; Backend implementation: Write code in Python to process requests from the frontend and send data to the frontend as needed. At the same time, set a timeout on the server and send a response to the client when there is new data. Message format and protocol: When transmitting data between the front-end and back-end, define a unified message format and communication protocol to ensure that the front-end and back-end can correctly parse and process the data.

6. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 5 is characterized in that: Construct a real-time model and a rapid analysis and calculation model for typical traffic scenarios in urban road networks, including the following: (41) Call the digital definition interface of the full traffic scene model, link the GIS map platform to locate the scene construction content, and start updating the road three-dimensional modeling in real time; (42) Use information collection equipment to monitor road conditions in real time, receive multi-source micro-travel data through the front-end and back-end two-way integrated communication rendering module, including road event type, event occurrence time, event generation space coordinates, and traffic flow scene information around the event occurrence point, and perform real-time rendering simulation of the traffic flow module; (43) Form a fast computational analysis model for micro-simulation to predict simulation results for management and control decisions.

7. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 6 is characterized in that: The fast calculation and analysis model for microscopic simulation mainly includes traffic flow simulation at the microscopic level. In order to reflect the interaction between vehicles and the diversity of driving behaviors, a microscopic traffic flow individual interaction model is constructed. Each vehicle is regarded as an independent intelligent agent and simulated according to its individual characteristics and behavior rules. The details are as follows: (431) Individual state definition: Each vehicle is defined as an individual, and its state includes information such as position, speed, acceleration, lane position, etc. The state formula is as follows: s i (t)=(x i (t),v i (t),a i (t),l i (t)) Where: x i (t) represents the position of vehicle i at time t, v i (t) represents speed, a i (t) represents acceleration, l i (t) represents the lane position; (432) Individual behavior rules: define the situations in which the vehicle follows the speed of the vehicle in front, maintains a safe following distance, and limits the maximum acceleration. In addition, lane position information is introduced into the behavior rule function and used to adjust the vehicle's acceleration. The formula of the individual behavior rule is as follows: a i (t) = min(a max ,max(0,f desired (v i (t),Δx,(t))+f safe (Δv i (t),Δx i (t))+f lane (l i (t), l desired (i)))) Where a max is the maximum acceleration limit of the vehicle; Δv i (t) is the speed difference between vehicle i and the preceding vehicle at time t; Δx i (t) is the distance difference between vehicle i and the preceding vehicle at time t; f desired is a function of the vehicle's drag speed, which depends on the relative speed of the vehicle and the preceding vehicle, f safe It is a function of the safe following distance maintained by the vehicle, which depends on the distance and speed difference between the vehicle and the vehicle ahead; l i (t) is the lane position of vehicle i at time t; l desired (i) is the desired lane position of vehicle i; f lane is the lane position adjustment function that adjusts the vehicle acceleration, depending on the The difference between the preceding lane and the desired lane position, using cubic spline interpolation. (433) Define individual interaction rules. The interaction rules defined here adjust the speed according to the vehicle spacing, speed difference, acceleration difference and prediction of other vehicle behaviors to avoid collision or maintain an appropriate spacing. The formula of the individual interaction rule function is as follows: In the formula, v desired is the desired speed function of the vehicle, which is adjusted according to the distance to the vehicle ahead and the acceleration of the current vehicle; v safe is the speed function when maintaining a safe distance, which is adjusted according to the distance to the vehicle in front; T is a time constant used to control the safe distance between vehicles; T p is a prediction time, used to predict the speed of other vehicles; α, β, γ, δ, η, θ are adjustment parameters with a value range of [0, 1], used to control the impact of different factors on vehicle speed; (434) Define the overall system behavior function of the traffic flow, which is specifically expressed as: Where: ρ(t) is the density of traffic flow at time t, which can be calculated based on the number of vehicles and road length; represents the acceleration change rate of vehicle i, represents the rate of change of traffic flow density; ω1, ω2, ω3, ω4, ω5 are the weights of various factors, which are used to adjust the impact of each factor on the overall behavior; other factors include weather factors and road control.

8. The microscopic three-dimensional simulation method for composite transportation network macro-micro integration according to claim 1 is characterized in that: The microscopic traffic evaluation index calculation system is constructed based on the principal component fuzzy fusion algorithm, which includes the following: (51) Specify the traffic evaluation object system: For complex urban traffic scenarios, select vehicle delay time, number of stops, travel speed, travel time and fuel consumption indicators to construct an evaluation index set, P = {u1,u2,…,u i },u i = {u i1 ,u i2 ,…,u ij }(i=1,2,…n)(j=1,2,…m) is a single set of various traffic evaluation indicators; (52) The traffic index standardization processing method is expressed as: Where: is the average of the set of i traffic evaluation indicators, s i is the standard deviation of the set of i-item traffic evaluation indicators; α ij is u ij Indicator relative to the overall u i The scaling index of the indicator set is between 0 and 1; β i It is the indicator factor of each traffic indicator, representing the degree of indication of each traffic indicator on the traffic conditions; ε i is the dimensional conversion parameter of each traffic index, representing the sensitivity of the dimensional indication of traffic conditions; the correction bias parameter δ i Correction dimension conversion parameters; (53) Principal component analysis of traffic evaluation index data: Analyze the index set, extract the traffic evaluation index component load matrix, score matrix and weight coefficient, and obtain the traffic evaluation index weight vector C = {c1, c2, …, c i }, representing vehicle delay time Vt, number of stops Vs, travel speed Vv, travel time VT and fuel consumption Ve; (54) Determine the fuzzy evaluation level of traffic indicators and establish the fuzzy evaluation matrix R of traffic indicators: Determine the fuzzy evaluation level of traffic indicators as needed into i layers: V = {v1, v2, …, v i }, from high to low; According to the fuzzy evaluation level of traffic indicators, the evaluation indicator set P is fuzzy judged, and the traffic indicator fuzzy evaluation matrix R is established according to the frequency to complete the fuzzy evaluation of indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption: Where: γ i,i is the discount factor of the i-th traffic index to the i-th level fuzzy evaluation level, usually between 0 and 1; θ i,i is the fuzzy weight of the i-th level fuzzy evaluation level for the i-th traffic index pair; (55) Synthesize the evaluation result of the principal component fuzzy fusion algorithm: The traffic evaluation index weight vector C and the traffic index fuzzy evaluation matrix R are synthesized to obtain the evaluation result S of the principal component fuzzy fusion algorithm: The evaluation result S is evaluated to complete the evaluation of traffic indicators such as vehicle delay time, number of stops, travel speed, travel time and fuel consumption.

9. A microscopic three-dimensional simulation system for macro-micro integration of a composite transportation network, used to implement the microscopic three-dimensional method for macro-micro integration of a composite transportation network as claimed in any one of claims 1 to 8, characterized in that: include: Build a module to establish a real-time linkage global dynamic traffic scene interface. Based on the high-precision GIS map information platform API, develop a high-precision editable three-dimensional traffic scene twin module. Through real-time online editing of weather and road property models, add the time change dimension on the basis of three-dimensional space to build a full traffic four-dimensional scene model platform; The multi-dimensional calibration module is used to construct control unit agents, traveler agents, road sensor agents and communication unit agents. Through the collaborative information collection and transmission of multiple agents, the multi-source micro travel data actually collected is processed and gradient updated. The multi-dimensional fusion algorithm matrix of traffic scene parameters is used to realize the multi-dimensional calibration of micro simulation model parameters. The communication rendering module is used to develop a two-way integrated data communication module for the front and back ends based on the mature API of the micro-simulation platform, so as to achieve real-time rendering of the vehicle trajectory and micro-simulation results in the three-dimensional scene while realizing the evolution of the macro-simulation; The analysis and response module is used to build a real-time model and a rapid analysis and calculation model for typical traffic scenarios in urban road networks for large-scale complex traffic network scenarios, so as to achieve timely response of management and control strategies in the micro dimension; Visualization module, which is used to build a microscopic traffic evaluation index calculation system based on the main component fuzzy fusion algorithm based on the front-end and back-end two-way integrated communication rendering module, and realize the visualization of simulation analysis results on the screen; The archiving module is used to store key data in the background while the simulation is in progress, for follow-up analysis.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the microscopic three-dimensional simulation method for macro-micro integration of a composite transportation network as described in any one of claims 1 to 8 is adopted.

Images