Real-time online collaborative reasoning system and method for urban multi-modal transportation system

Abstract

The application discloses a real-time online collaborative deduction system and method of an urban multi-mode traffic system, and belongs to the technical field of intelligent traffic. In order to solve the problem of real-time online collaborative evolution simulation, the application determines the initial travel demand of the whole network through a multi-mode network model; a multi-mode travel demand distribution module distributes the multi-mode travel proportion based on generalized travel cost, and distributes road travel demand and rail travel demand; a road simulation deduction module outputs the road traffic operation situation; a rail simulation deduction module outputs the rail traffic operation situation; a multi-mode interactive fusion module performs coupling deviation calculation; a dynamic calibration module dynamically checks the calculation results of the road simulation deduction module, the rail simulation deduction module and the multi-mode interactive fusion module, and obtains the whole-network multi-mode operation situation which is qualified in checking. A global optimization and output module is used for receiving the whole-network multi-mode operation situation which is qualified in checking, and outputting the accurate restoration and prediction results of the multi-mode whole-network state.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation technology, specifically relating to a real-time online collaborative simulation system and method for urban multimodal transportation systems. Background Technology

[0002] Traditional traffic simulation primarily focuses on single modes, such as road traffic flow simulation or rail transit passenger flow simulation, and has reached a relatively mature stage. However, real-world transportation systems are complex mega-systems where multiple modes, including roads, rail, and pedestrians, interact dynamically. Multimodal traffic simulation, as a cutting-edge field, aims to integrate interdisciplinary models to reproduce and deduce the macro, meso, and micro dynamic processes of all modes of travel in a digital twin manner. It has irreplaceable value for comprehensive traffic management, travel service optimization, and large-scale hub planning.

[0003] Achieving high-fidelity multi-mode real-time online simulation faces two key technical challenges: multi-mode dynamic interactive simulation and real-time data assimilation across the entire system. Regarding dynamic interaction, the core challenge lies in accurately depicting the micro-behavioral feedback and macro-state coupling of different transportation modes (such as motor vehicles, pedestrians, and trains) in shared spaces (such as at-grade intersections) or connecting nodes (such as integrated transportation hubs). In terms of data assimilation, the difficulty lies in integrating multi-source heterogeneous real-time detection data (such as checkpoint traffic flow and track AFC / ATS data), continuously calibrating the simulation state of each subsystem and interactive interface, and ensuring the synchronization and consistency of the virtual system and the physical system under time-varying conditions.

[0004] The existing technology has the following problems:

[0005] Interactive simulation is static and rigid: Existing solutions rely on preset static rules to process interactions between modes, which cannot reflect the complex behaviors of conflict, avoidance, and coordination that evolve dynamically among multi-modal traffic flows in the real world as conditions such as density, speed, and signal phase change. This leads to distorted simulation results in key areas such as hubs and intersections.

[0006] The system suffers from low coupling and a lack of closed-loop feedback: each mode simulator often operates independently or only performs unidirectional data transfer (such as using track timetables as a fixed input for road simulation), failing to construct a closed loop for bidirectional real-time influence and state calibration between the road and track. When a subsystem experiences state deviation due to disturbances (such as traffic accidents or train delays), the deviation cannot be transmitted dynamically and in real time to affect other subsystems and transfer behavior between them, thus disrupting the overall integrity of the system.

[0007] Large-scale online synchronization and calibration are difficult to achieve: Existing solutions lack a unified framework to integrate multi-source real-time data for collaborative calibration of various modes and interaction interfaces. Road detection data cannot be used to correct rail passenger flow origin-destination (OD), and rail data cannot be used to optimize road allocation. This makes it difficult for the simulation system to maintain high-precision synchronization with the full state of a real large-scale transportation system online, limiting its application in real-time decision support. Summary of the Invention

[0008] This invention aims to address the technical problem of realistically reflecting multi-modal dynamic interactions and utilizing multi-source data to achieve real-time online collaborative evolution simulation of large-scale road and rail networks. It proposes a real-time online collaborative simulation system and method for urban multi-modal transportation systems.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A real-time online collaborative simulation system for urban multimodal transportation systems includes a multimodal network model construction module, a multimodal travel demand allocation module, a road simulation module, a rail simulation module, a multimodal interactive fusion module, a dynamic calibration module, and a global optimization and output module.

[0011] The multi-mode network model construction module is used to integrate traffic network data, traffic control data, traffic operation data, and multiple types of detection data to determine the initial travel demand of the entire network.

[0012] The multi-modal travel demand allocation module is used to allocate the proportion of multi-modal travel based on the initial travel demand of the entire network and the generalized travel cost, and allocate it into road travel demand and rail travel demand.

[0013] The road simulation module is used to receive road travel demand, perform path flow adjustment based on dynamic vehicle OD estimation, perform real-time simulation of mixed road traffic flow, and output the road traffic operation status.

[0014] The track simulation module is used to receive rail travel demand, perform passenger flow regulation based on dynamic passenger flow OD estimation, conduct real-time online simulation of rail transit, and output the rail transit operation status.

[0015] The multi-mode interactive fusion module performs coupling deviation calculation on the traffic flow of hub nodes output by the road simulation module and the passenger flow of hub nodes output by the track simulation module.

[0016] The dynamic calibration module dynamically verifies the calculation results of the road simulation and extrapolation module, the track simulation and extrapolation module, and the multi-mode interactive fusion module to obtain a qualified multi-mode operation status of the entire network.

[0017] The global optimization and output module is used to receive the verified multi-mode operation status of the entire network and output the accurate restoration and prediction results of the multi-mode full network status.

[0018] Furthermore, the multi-mode network model construction module is connected to the multi-mode travel demand allocation module, which is connected to the road simulation and deduction module, the track simulation and deduction module, and the dynamic calibration module, respectively. The road simulation and deduction module and the track simulation and deduction module are connected to the multi-mode interaction and fusion module and the dynamic calibration module, respectively. The multi-mode interaction and fusion module is connected to the dynamic calibration module, and the dynamic calibration module is connected to the global optimization and output module.

[0019] Furthermore, the multi-mode network model construction module sets road intersections and rail stations as road nodes and rail nodes with mutual conversion functions, sets road segments and rail intervals as edges with mode attributes, and connects road nodes and rail nodes through transfer edges to obtain a multi-mode network model.

[0020] Furthermore, the multimodal travel demand allocation module calculates the mode share rate based on the generalized travel cost, performs a convergence judgment on the obtained mode share rate, and multiplies the converged mode share rate with the total network travel demand to obtain the updated road travel demand and rail travel demand.

[0021] Furthermore, the road simulation and dynamic calibration modules employ an adjustment formula based on the accumulation of path flow deviations to estimate dynamic traffic flow. The calculation formula is as follows:

[0022] (1)

[0023] in, For the first Path within each calibration cycle The amount of traffic adjustment on the screen. The step size factor has a range of values. ; The path of the previous cycle The generalized cost of travel; For path indexing, This represents the sum of the costs of all paths; The total road network demand for the current period is estimated from road inspection data; The path of the previous cycle Simulated traffic flow; For the previous two cycle paths Simulated traffic flow; The inertia coefficient has a range of values. .

[0024] Furthermore, the dynamic calibration module and track simulation module use a correction formula based on the deviation between the arrival sequence and the train's full load rate to estimate dynamic passenger flow. The calculation formula is as follows:

[0025] (2)

[0026] in, For the first The passenger flow adjustment amount for od pairs within a calibration cycle; o is the starting point, and d is the ending point; The learning rate has a range of values. ; For the first The real-time AFC entry matching probability of od pairs within each calibration cycle is calculated by the time-series correlation between the entry gate and the exit gate. This is the train's maximum passenger capacity; For the first The sum of all od pairs originating from the starting point o within a calibration cycle; For the first Simulated passenger flow of od pairs within 1 calibration cycle; This is the section correction factor, with a range of values. ; For the first Measured passenger flow at key sections within each calibration cycle; Let be the 0-1 variable corresponding to the i-th OD pair; For the first Simulated passenger flow for the i-th OD pair within -1 calibration cycles; It is a very small positive number, used to prevent division by zero; This is the sensitivity coefficient.

[0027] Furthermore, the method for coupling verification of the multi-mode interaction fusion module is based on the transfer travel ratio matrix. The passenger flow of hub nodes output by the track simulation and deduction module is converted into the expected road traffic flow. Then, the deviation between the expected road traffic flow and the hub node traffic flow output by the road simulation and deduction module is calculated, and the maximum deviation is selected as the comprehensive coupling deviation.

[0028] Furthermore, the dynamic calibration module judges the comprehensive coupling deviation. If the comprehensive coupling deviation is less than the preset coupling deviation threshold, it is determined that the multi-mode travel demand allocation is reasonable and outputs the multi-mode operation status of the entire network; otherwise, it is determined that the multi-mode travel demand allocation is unreasonable and the deviation information is fed back to the multi-mode travel demand allocation module for reallocation.

[0029] A simulation method for a real-time online collaborative simulation system for urban multimodal transportation systems includes the following steps:

[0030] S1. The multi-modal network model construction module integrates traffic network data, traffic control data, traffic operation data, and various types of detection data to establish a multi-modal network model and output the initial travel demand of the entire network.

[0031] S2. The multi-modal travel demand allocation module allocates the proportion of multi-modal travel based on the initial travel demand of the entire network obtained in step S1 and the generalized travel cost, dividing it into road travel demand and rail travel demand.

[0032] S3. The road simulation module and the rail simulation module receive road travel demand and rail travel demand respectively, run road simulation and rail simulation independently and concurrently, and output dynamic traffic flow and dynamic passenger flow respectively;

[0033] S4. The multi-mode interactive fusion module performs coupling deviation calculation on the dynamic vehicle flow and dynamic passenger flow obtained in step S3;

[0034] S5. The dynamic calibration module dynamically judges the road simulation results, track simulation results, and coupling deviation. If the multi-mode travel demand allocation is deemed reasonable, it outputs the overall network multi-mode operation status. If the multi-mode travel demand allocation is deemed unreasonable, it returns to step S2 to re-perform the whole system simulation until the requirements are met.

[0035] The beneficial effects of this invention are:

[0036] This invention discloses a real-time online collaborative simulation system for urban multimodal transportation systems. It enables bidirectional dynamic interaction and collaborative simulation of road mixed traffic flow and rail transit passenger flow in terms of both micro-behavior and macro-state, transcending static rules to simulate their mutually adaptive evolutionary process. The multi-source data closed-loop dynamic calibration mechanism designed in this invention simultaneously and in a closed loop feeds real-time detection data back to the road and rail simulation subsystems and their interactive interfaces, driving adaptive adjustments to parameters of each subsystem (such as OD matrix and path flow) to ensure the simulation accuracy of the entire system. The efficient unified computing architecture constructed in this invention supports the fusion modeling and parallel computing of large-scale road and rail networks, ultimately outputting a high-fidelity real-time state of the entire network after collaborative evolution and dynamic calibration, and providing a reliable sandbox environment for multimodal travel structure optimization and management strategy evaluation.

[0037] This invention discloses a real-time online collaborative simulation system for urban multimodal transportation systems. It constructs a unified digital twin model by abstracting road intersections and rail stations as interchangeable nodes, road segments and rail intervals as edges with mode attributes, and connecting these two types of nodes through transfer edges. This unified representation method aligns the road and rail networks at the data and topology layers, eliminating the obstacle of direct data interaction caused by independent network models in traditional methods. The unified modeling of this invention allows for the allocation of multimodal travel demand, transfer behavior simulation, and coupling verification to be completed within the same data framework, significantly reducing data redundancy and model conversion errors in multimodal integrated simulation.

[0038] This invention discloses a real-time online collaborative simulation system for urban multimodal transportation systems, employing a "pre-allocation" step: based on generalized travel costs, the initial travel demand across the entire network is allocated into road travel demand and rail travel demand, driving the road and rail simulation engines respectively. This allocation method utilizes the congestion sensitivity index in the generalized cost function and a threshold-based transfer penalty model (piecewise quadratic function), making the initial allocation closer to actual travel choices. Subsequently, road and rail simulations run independently, and through first and second-level self-calibration, the simulation results are compared with real-time detection data (road checkpoint flow, rail AFC data), dynamically adjusting path flow and passenger OD. This self-calibration mechanism can control the accuracy of single-mode simulations within a preset tolerance.

[0039] This invention discloses a real-time online collaborative simulation system for urban multimodal transportation systems. At key transportation hubs, this system utilizes a multimodal interaction and fusion module to receive road traffic flow (cars, taxis, buses) and rail passenger flow. Based on a pre-defined transfer travel ratio matrix, it converts rail passenger flow into expected road traffic flow and calculates the comprehensive coupling deviation. If the deviation exceeds a preset threshold, it triggers a multimodal travel demand allocation module to readjust the travel ratio between roads and rail.

[0040] The real-time online collaborative simulation system for urban multimodal transportation systems described in this invention employs a threshold-based conditional calculation model for transfer time in the generalized travel cost: when the number of transfers or waiting time exceeds a preset threshold, the cost increases according to a quadratic function. This model differs from traditional fixed or linear penalties and better reflects the actual behavior of travelers in multimodal transportation—slight transfer delays are tolerable, but long waits or multiple transfers will cause significant aversion.

[0041] The present invention discloses a real-time online collaborative simulation system for urban multimodal transportation systems. The final output of the multimodal network status includes: road segment traffic flow, travel speed, and congestion index; passenger numbers entering and exiting rail stations, platform congestion, and train occupancy rates; hub transfer volume and transfer delays, etc. This system can support traffic management departments in real-time assessment of the impact of traffic restrictions, public transport scheduling, and other strategies on the multimodal network, improving the scientific rigor and timeliness of decision support. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the structure of a real-time online collaborative simulation system for an urban multimodal transportation system according to the present invention;

[0043] Figure 2 This is a flowchart of a real-time online collaborative simulation method for an urban multimodal transportation system as described in this invention;

[0044] Figure 3 This is a structural block diagram of a real-time online collaborative simulation method for an urban multimodal transportation system according to the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described specific embodiments are merely a part of the embodiments of the invention, and not all of them. The components of the specific embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations, and the invention may also have other embodiments.

[0046] Therefore, the following detailed description of specific embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected specific embodiments of the invention. All other specific embodiments obtained by those skilled in the art based on these specific embodiments without inventive effort are within the scope of protection of this invention.

[0047] To further understand the invention's content, features, and effects, the following specific embodiments are provided, along with accompanying drawings. Figure 1 - Appendix Figure 3 Detailed explanation is as follows:

[0048] Example 1:

[0049] A real-time online collaborative simulation system for urban multimodal transportation systems includes a multimodal network model construction module, a multimodal travel demand allocation module, a road simulation module, a rail simulation module, a multimodal interactive fusion module, a dynamic calibration module, and a global optimization and output module.

[0050] The multi-mode network model construction module is used to integrate traffic network data, traffic control data, traffic operation data and multiple types of detection data to determine the initial travel demand of the entire network;

[0051] The multi-modal travel demand allocation module is used to allocate the proportion of multi-modal travel based on the initial travel demand of the entire network and the generalized travel cost, and allocate it into road travel demand and rail travel demand.

[0052] The road simulation module is used to receive road travel demand, perform path flow adjustment based on dynamic vehicle OD estimation, perform real-time simulation of mixed road traffic flow, and output the road traffic operation status.

[0053] The track simulation and deduction module is used to receive rail travel demand, perform passenger flow regulation based on dynamic passenger flow OD estimation, conduct real-time online deduction of rail transit, and output the rail transit operation status.

[0054] The multi-mode interactive fusion module performs coupling deviation calculation on the traffic flow of hub nodes output by the road simulation module and the passenger flow of hub nodes output by the track simulation module.

[0055] The dynamic calibration module dynamically verifies the calculation results of the road simulation and extrapolation module, the track simulation and extrapolation module, and the multi-mode interactive fusion module to obtain a qualified multi-mode operation status of the entire network.

[0056] The global optimization and output module is used to receive the verified multi-mode operation status of the entire network and output the accurate restoration and prediction results of the multi-mode full network status.

[0057] Furthermore, the multi-mode network model construction module is connected to the multi-mode travel demand allocation module, which is connected to the road simulation and deduction module, the track simulation and deduction module, and the dynamic calibration module, respectively. The road simulation and deduction module and the track simulation and deduction module are connected to the multi-mode interaction and fusion module and the dynamic calibration module, respectively. The multi-mode interaction and fusion module is connected to the dynamic calibration module, and the dynamic calibration module is connected to the global optimization and output module.

[0058] Furthermore, the multi-mode network model construction module sets road intersections and rail stations as road nodes and rail nodes with mutual conversion functions, sets road segments and rail intervals as edges with mode attributes, and connects road nodes and rail nodes through transfer edges to obtain a multi-mode network model.

[0059] Furthermore, the multimodal travel demand allocation module calculates the mode share rate based on the generalized travel cost, performs a convergence judgment on the obtained mode share rate, and multiplies the converged mode share rate with the total network travel demand to obtain the updated road travel demand and rail travel demand.

[0060] Furthermore, the road simulation and dynamic calibration modules employ an adjustment formula based on the accumulation of path flow deviations to estimate dynamic traffic flow. The calculation formula is as follows:

[0061] (1)

[0062] in, For the first Path within each calibration cycle The amount of traffic adjustment on the screen. The step size factor has a range of values. ; The path of the previous cycle The generalized cost of travel; For path indexing, This represents the sum of the costs of all paths; The total road network demand for the current period is estimated from road inspection data; The path of the previous cycle Simulated traffic flow; For the previous two cycle paths Simulated traffic flow; The inertia coefficient has a range of values. .

[0063] Furthermore, mixed traffic flow on roads includes cars, taxis, buses, pedestrians, and non-motorized vehicles.

[0064] Furthermore, the dynamic calibration module and track simulation module use a correction formula based on the deviation between the arrival sequence and the train's full load rate to estimate dynamic passenger flow. The calculation formula is as follows:

[0065] (2)

[0066] in, For the first The passenger flow adjustment amount for od pairs within a calibration cycle; o is the starting point, and d is the ending point; The learning rate has a range of values. ; For the first The real-time AFC entry matching probability of od pairs within each calibration cycle is calculated by the time-series correlation between the entry gate and the exit gate. This is the train's maximum passenger capacity; For the first The sum of all od pairs originating from the starting point o within a calibration cycle; For the first Simulated passenger flow of od pairs within 1 calibration cycle; This is the section correction factor, with a range of values. ; For the first Measured passenger flow at key sections within each calibration cycle; Let be the 0-1 variable corresponding to the i-th OD pair; For the first Simulated passenger flow for the i-th OD pair within -1 calibration cycles; It is a very small positive number, used to prevent division by zero; This is the sensitivity coefficient.

[0067] Furthermore, the real-time online simulation of rail transit includes simulations of the entire process of passengers entering the station hall, turning off the gates, walking in the station, using escalators and stairs, and getting on and off the train, as well as detailed simulations of the entire process of the train from the starting station, stopping at stations, to the terminal station.

[0068] Furthermore, the method for coupling verification of the multi-mode interaction fusion module is based on the transfer travel ratio matrix. The passenger flow of hub nodes output by the track simulation and deduction module is converted into the expected road traffic flow. Then, the deviation between the expected road traffic flow and the hub node traffic flow output by the road simulation and deduction module is calculated, and the maximum deviation is selected as the comprehensive coupling deviation.

[0069] Furthermore, the dynamic calibration module judges the comprehensive coupling deviation. If the comprehensive coupling deviation is less than the preset coupling deviation threshold, it is determined that the multi-mode travel demand allocation is reasonable and outputs the multi-mode operation status of the entire network; otherwise, it is determined that the multi-mode travel demand allocation is unreasonable and the deviation information is fed back to the multi-mode travel demand allocation module for reallocation.

[0070] Furthermore, the dynamic calibration module performs a three-level progressive calibration:

[0071] First-level road simulation self-calibration: The road segment flow output by the road simulation deduction module is compared with the road detection data (checkpoint flow, road segment flow). If the accuracy is inconsistent, dynamic vehicle OD estimation is performed, and the path flow is updated using formula (1). The deduction is repeated until the road simulation accuracy meets the requirements.

[0072] Second-level track simulation self-calibration: The platform congestion and train load factor output by the track simulation deduction module are compared with the track detection data (video surveillance, AFC data). If the accuracy is inconsistent, dynamic passenger flow OD estimation is performed, and the passenger flow OD matrix is ​​updated using formula (2). The simulation is repeated until the track simulation accuracy meets the requirements.

[0073] The third-level hub transfer coupling calibration compares the expected road traffic flow of the hub node output by the multi-mode interaction fusion module with the actual road traffic flow. If the deviation exceeds the preset threshold, the multi-mode travel demand allocation module is triggered. The generalized cost iterative allocation formula is used to readjust the ratio of road travel demand to rail travel demand, and the road simulation and rail simulation modules are re-driven until the hub node coupling accuracy meets the requirements.

[0074] Example 2:

[0075] A simulation method for a real-time online collaborative simulation system for an urban multimodal transportation system, as described in Embodiment 1, includes the following steps:

[0076] S1. The multi-modal network model construction module integrates traffic network data, traffic control data, traffic operation data, and various types of detection data to establish a multi-modal network model and output the initial travel demand of the entire network.

[0077] S2. The multi-modal travel demand allocation module allocates the proportion of multi-modal travel based on the initial travel demand of the entire network obtained in step S1 and the generalized travel cost, dividing it into road travel demand and rail travel demand.

[0078] Furthermore, the multimodal travel demand allocation module adopts a generalized cost iterative allocation formula that considers transfer impedance. The generalized travel cost comprehensively considers the time and economic costs of individual travel behavior, including the travel costs of different modes of travel, the transfer time cost during the mode-of-travel process, and the waiting time cost. The transfer time cost and the waiting time cost adopt a threshold-based conditional calculation model, that is, when the number of transfers or the waiting time exceeds a preset threshold, the cost increases non-linearly. Impedance functions are constructed for sub-network segments and transfer segments respectively to estimate time costs.

[0079] Furthermore, the specific implementation method of step S2 includes the following steps:

[0080] S2.1. Calculate the generalized travel cost using the following formula:

[0081] (3)

[0082] in, For the first Pattern in the next iteration Average generalized cost (of roads or tracks), in yuan or normalized value per minute; For pattern Average travel time For pattern The average monetary cost, including ticket price, fuel surcharge, parking fee, etc., is expressed in yuan. For pattern Demand volume to be met, in person-times / hour or veh / hour; For pattern The system throughput capacity, in units of people / hour or veh / hour; As a crowding sensitivity index, This causes congestion costs to increase non-linearly with demand; For pattern Average number of transfers, in times; For transfer sensitivity coefficient, Control the saturation speed of transfer penalties; , , , These are weighting coefficients, determined through the SP survey;

[0083] It is obtained by weighted summation of the impedance of the sub-network segment and the impedance of the transfer segment:

[0084]

[0085] in, For pattern The set of subnet segments, For road section The impedance function (in BPR function form); For pattern A collection of transfer routes, Transfer section The transfer time is calculated using a threshold-based conditional calculation model;

[0086] S2.2. Update Mode Distributive Ratio:

[0087] (4)

[0088] in, For the first Pattern in the next iteration Demand sharing rate, dimensionless. ; These are the scaling parameters for the Logit model. ; For pattern index;

[0089] S2.3. Define convergence criterion:

[0090] (5)

[0091] in, The preset convergence threshold ranges from 0.001 to 0.01. This represents the maximum number of iterations, ranging from 10 to 30.

[0092] S2.4. Demand Allocation: The converged mode sharing ratio Multiplying the total network demand by the updated road travel demand and rail travel demand yields the updated road travel demand and rail travel demand, which are then sent to the road simulation module and rail simulation module, respectively.

[0093] S3. The road simulation module and the rail simulation module receive road travel demand and rail travel demand respectively, run road simulation and rail simulation independently and concurrently, and output dynamic traffic flow and dynamic passenger flow respectively;

[0094] Furthermore, the specific implementation method of step S3 includes the following steps:

[0095] S3.1. Road Simulation: Assign road travel demand to the road network, deduce traffic status, and if the deviation from the real-time detection data exceeds a threshold, execute formula (1) to update the path flow. Repeat the deduction until the road simulation accuracy meets the requirements, and output the road operation status and traffic flow of hub nodes. ;

[0096] S3.2. Track Simulation: Based on the rail travel demand, generate individual trains and passengers, and perform "passenger-train" trajectory matching simulation. If the platform congestion, train occupancy rate and the deviation from the detection data exceed the threshold, execute formula (2) to update the passenger flow OD. Repeat the simulation until the track simulation accuracy meets the requirements, and output the track operation status and the passenger flow of the hub nodes. ;

[0097] S4. The multi-mode interactive fusion module performs coupling deviation calculation on the dynamic vehicle flow and dynamic passenger flow obtained in step S3;

[0098] Furthermore, the specific implementation method of step S4 includes the following steps:

[0099] S4.1. Receive the output of the road simulation and deduction module to obtain the time period Internal arrival hub Road traffic volume :

[0100]

[0101] Including car traffic Taxi traffic Bus traffic ;

[0102] S4.2. Receive the output of the orbit simulation and extrapolation module to obtain the time period. Internal hub Passenger flow in and out of the station :

[0103]

[0104] in, To increase passenger flow at the station, This refers to the number of passengers exiting the station, expressed in person-hours.

[0105] S4.3. Based on the preset transfer travel ratio matrix ,in Indicates from orbital mode (Entering or exiting the station) Switch to road mode The ratio of (cars, taxis, buses, and pedestrians) converts rail passenger flow into expected road traffic flow. Considering only the transfers between cars, taxis, and buses, the walking portion does not generate road traffic:

[0106]

[0107] Right now:

[0108]

[0109]

[0110]

[0111] Among them, the transfer travel ratio matrix Pre-defined based on traffic survey data or historical patterns, and meeting the following requirements: S4.4. Calculate the relative deviation between the expected road traffic volume and the actual road traffic volume:

[0112]

[0113]

[0114] ;

[0115] S4.5. Define the integrated coupling deviation as follows:

[0116] ;

[0117] S5. The dynamic calibration module dynamically judges the road simulation results, track simulation results, and coupling deviation. If the multi-mode travel demand allocation is deemed reasonable, it outputs the multi-mode operation status of the entire network to the global optimization and output module. If the multi-mode travel demand allocation is deemed unreasonable, it returns to step S2 to re-perform the whole system simulation until the reasonable requirements are met.

[0118] Furthermore, the specific implementation method of step S5 is as follows: , If the preset coupling deviation threshold is set to 0 to 0.1, the multi-mode coupling accuracy is determined to meet the requirements; otherwise, the multi-mode travel demand allocation is determined to be unreasonable, and the deviation information is fed back to the multi-mode travel demand allocation module.

[0119] Furthermore, this invention first integrates traffic network data, traffic control data, traffic operation data, and multi-source detection data to construct a unified road and rail network model—abstracting road intersections and rail stations as mutually convertible nodes, achieving topological connections between modes through transfer edges, and determining the initial travel demand of the entire network based on this model. Subsequently, the system adopts an architecture of "allocation first, deduction later, and coupling later": based on generalized travel costs, the demand of the entire network is allocated into road travel demand and rail travel demand; the road simulation deduction module and the rail simulation deduction module run in parallel and independently, each performing internal self-calibration through dynamic vehicle OD estimation (path flow adjustment) and dynamic passenger flow OD estimation (passenger flow adjustment) to ensure that the accuracy of single-mode simulation is consistent with real-time detection data.

[0120] Furthermore, at the hub nodes, the multi-modal interaction fusion module receives hub traffic flow (cars, taxis, buses) from road simulation and passenger flow entering and exiting the station from rail simulation. Based on a pre-calibrated transfer travel ratio matrix (including walking), it converts the rail passenger flow into the expected road traffic flow and performs coupling deviation calculation. If the deviation exceeds a preset threshold, the initial multi-modal travel allocation is deemed unreasonable. The dynamic calibration module triggers the multi-modal travel demand allocation module to readjust the travel ratio between roads and rail with generalized travel cost as the optimization objective, forming a complete closed-loop feedback mechanism of "allocation—deduction—self-calibration—coupling verification—reallocation".

[0121] Furthermore, the final output is a high-precision full-network traffic state reconstruction and short-term future prediction results after multi-level calibration, realizing the organic synergy between road and rail transit simulation, and significantly improving the accuracy and efficiency of urban traffic system state perception, assessment and decision support.

[0122] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0123] Although this application has been described above with reference to specific embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of this application. In particular, as long as there is no structural conflict, the features in the specific embodiments disclosed in this application can be combined with each other in any way. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, this application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A real-time online collaborative simulation system for urban multimodal transportation systems, characterized in that, It includes a multi-mode network model construction module, a multi-mode travel demand allocation module, a road simulation and extrapolation module, a track simulation and extrapolation module, a multi-mode interactive fusion module, a dynamic calibration module, and a global optimization and output module; The multi-mode network model construction module is used to integrate traffic network data, traffic control data, traffic operation data and multiple types of detection data to determine the initial travel demand of the entire network; The multi-modal travel demand allocation module is used to allocate the proportion of multi-modal travel based on the initial travel demand of the entire network and the generalized travel cost, and allocate it into road travel demand and rail travel demand. The road simulation module is used to receive road travel demand, perform path flow adjustment based on dynamic vehicle OD estimation, perform real-time simulation of mixed road traffic flow, and output the road traffic operation status. The track simulation and deduction module is used to receive rail travel demand, perform passenger flow regulation based on dynamic passenger flow OD estimation, conduct real-time online deduction of rail transit, and output the rail transit operation status. The multi-mode interactive fusion module performs coupling deviation calculation on the traffic flow of hub nodes output by the road simulation module and the passenger flow of hub nodes output by the track simulation module. The dynamic calibration module dynamically verifies the calculation results of the road simulation and extrapolation module, the track simulation and extrapolation module, and the multi-mode interactive fusion module to obtain a qualified multi-mode operation status of the entire network. The global optimization and output module is used to receive the verified multi-mode operation status of the entire network and output the accurate restoration and prediction results of the multi-mode full network status.

2. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 1, characterized in that, The multi-mode network model construction module is connected to the multi-mode travel demand allocation module. The multi-mode travel demand allocation module is connected to the road simulation and deduction module, the track simulation and deduction module, and the dynamic calibration module. The road simulation and deduction module and the track simulation and deduction module are connected to the multi-mode interaction and fusion module and the dynamic calibration module, respectively. The multi-mode interaction and fusion module is connected to the dynamic calibration module. The dynamic calibration module is connected to the global optimization and output module.

3. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 2, characterized in that, The multi-mode network model construction module sets road intersections and rail stations as road nodes and rail nodes with mutual conversion functions, sets road segments and rail intervals as edges with mode attributes, and connects road nodes and rail nodes through transfer edges to obtain a multi-mode network model.

4. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 3, characterized in that, The multimodal travel demand allocation module calculates the mode share rate based on the generalized travel cost, performs a convergence judgment on the obtained mode share rate, and multiplies the converged mode share rate with the total network travel demand to obtain the updated road travel demand and rail travel demand.

5. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 4, characterized in that, The road simulation and dynamic calibration modules use an adjustment formula based on the accumulation of path flow deviation to estimate dynamic traffic flow. The calculation formula is as follows: (1) in, For the first Path within each calibration cycle The amount of traffic adjustment on the screen. The step size factor has a range of values. ; The path of the previous cycle The generalized cost of travel; For path indexing, This represents the sum of the costs of all paths; The total road network demand for the current period is estimated from road inspection data; The path of the previous cycle Simulated traffic flow; For the previous two cycle paths Simulated traffic flow; The inertia coefficient has a range of values. .

6. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 5, characterized in that, The dynamic calibration module and track simulation module use a correction formula based on the deviation between the arrival sequence and the train's full load rate to estimate dynamic passenger flow. The calculation formula is as follows: (2) in, For the first The passenger flow adjustment amount for od pairs within a calibration cycle; o is the starting point, and d is the ending point; The learning rate has a range of values. ; For the first The real-time AFC entry matching probability of od pairs within each calibration cycle is calculated by the time-series correlation between the entry gate and the exit gate. This is the train's maximum passenger capacity; For the first The sum of all od pairs originating from the starting point o within a calibration cycle; For the first Simulated passenger flow of od pairs within 1 calibration cycle; This is the section correction factor, with a range of values. ; For the first Measured passenger flow at key sections within each calibration cycle; Let be the 0-1 variable corresponding to the i-th OD pair; For the first Simulated passenger flow for the i-th OD pair within -1 calibration cycles; It is a very small positive number, used to prevent division by zero; This is the sensitivity coefficient.

7. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 6, characterized in that, The method for coupling verification of the multi-mode interaction fusion module is based on the transfer travel ratio matrix. The passenger flow of hub nodes output by the track simulation and deduction module is converted into the expected road traffic flow. Then, the deviation between the expected road traffic flow and the hub node traffic flow output by the road simulation and deduction module is calculated, and the maximum deviation is selected as the comprehensive coupling deviation.

8. The real-time online collaborative simulation system for urban multimodal transportation systems according to claim 7, characterized in that, The dynamic calibration module judges the overall coupling deviation. If the overall coupling deviation is less than the preset coupling deviation threshold, it determines that the multi-mode travel demand allocation is reasonable and outputs the multi-mode operation status of the entire network; otherwise, it determines that the multi-mode travel demand allocation is unreasonable and feeds back the deviation information to the multi-mode travel demand allocation module for reallocation.

9. A deduction method for a real-time online collaborative deduction system for an urban multimodal transportation system as described in any one of claims 1-8, characterized in that, Includes the following steps: S1. The multi-modal network model construction module integrates traffic network data, traffic control data, traffic operation data, and various types of detection data to establish a multi-modal network model and output the initial travel demand of the entire network. S2. The multi-modal travel demand allocation module allocates the proportion of multi-modal travel based on the initial travel demand of the entire network obtained in step S1 and the generalized travel cost, dividing it into road travel demand and rail travel demand. S3. The road simulation module and the rail simulation module receive road travel demand and rail travel demand respectively, run road simulation and rail simulation independently and concurrently, and output dynamic traffic flow and dynamic passenger flow respectively; S4. The multi-mode interactive fusion module performs coupling deviation calculation on the dynamic vehicle flow and dynamic passenger flow obtained in step S3; S5. The dynamic calibration module dynamically judges the road simulation results, track simulation results, and coupling deviation. If the multi-modal travel demand allocation is deemed reasonable, it outputs the multi-modal operation status of the entire network to the global optimization and output module. If the multi-modal travel demand allocation is deemed unreasonable, it returns to step S2 to re-perform the full system simulation until the reasonable requirements are met.

Images