A dynamic scheduling method and system for intelligent transportation network

Abstract

The present invention discloses a method and system for dynamic scheduling of intelligent transportation networks, which relates to the technical field of transportation networks. Based on the flow conservation equation, an objective function is established and minimized, which can effectively identify the impact of temporary occupation of buses on non-motor vehicle traffic, accurately calculate the flow efficiency L of non-motor vehicles, so as to improve the accuracy of traffic flow prediction in the later stage; through comparative analysis of predicted traffic and actual traffic, the system can intelligently judge the congestion status of the current non-motor vehicle lanes, and quickly issue scheduling optimization instructions, which not only helps to alleviate traffic congestion, but also reduces the delay time of non-motor vehicles. In addition, the scheduling optimization model constructed in combination with PyTorch technology makes the calculation of the scheduling evaluation index Dgzs more accurate, providing a scientific basis for the bus dispatching plan. According to the numerical value of the evaluation index, the system can intelligently select the corresponding bus scheduling strategy, and this process further realizes the efficient allocation of transportation resources.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic networks, and in particular to a method and system for dynamic scheduling of intelligent traffic networks. Background Art

[0002] Intelligent Transportation System (ITS) is an important part of modern traffic management and transportation. It aims to improve traffic operation efficiency and safety by using advanced information technology and communication technology, and focuses on optimizing traffic flow and resource allocation through data monitoring and analysis. In this field, the dispatching and management of buses is particularly important, and related research and applications are constantly deepening to optimize traffic flow and enhance travel experience. However, in actual operation, buses may temporarily stop on non-motorized vehicle lanes during the process of picking up and dropping off passengers due to the lack of dedicated bus stops or stops. This behavior not only affects the traffic efficiency of non-motorized vehicles, but may also cause traffic flow disorder, further exacerbating road congestion.

[0003] When analyzing the impact of temporary bus parking on non-motorized vehicle traffic, the current shortcomings are mainly reflected in the lack of a systematic monitoring and analysis mechanism. Existing methods often focus on monitoring the traffic flow of motor vehicle lanes in the traffic network and analyzing pedestrians on the road. They may cause delays and congestion in non-motorized vehicle traffic caused by temporary bus stops. Existing methods often ignore some "blocking points" that suddenly appear in traffic flow, that is, drastic changes in the relationship between flow and density. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for dynamic scheduling of an intelligent transportation network, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for dynamic scheduling of intelligent transportation networks, comprising the following steps:

[0006] S1. Deploy several groups of monitoring equipment to monitor the relevant vehicle flow data in the non-motorized vehicle lanes, and set up flow meters at bus stops to monitor the relevant flow data in real time, generate a traffic network data set after pre-processing, and upload the traffic network data set to the cloud platform;

[0007] S2. Establishing the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ;

[0008] S3: Based on the predicted traffic flow value obtained in S2 and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued;

[0009] S4. After receiving the dispatch optimization instruction, analyze the impact of the temporary stop of the bus in the non-motorized vehicle lane on the travel delay of the non-motorized vehicle to obtain the delay time Ywsc, and calculate the service demand ratio Fxb of the current bus station to the bus based on the relevant passenger flow data in the bus station, and combine the dispatch optimization model constructed by PyTorch technology to obtain the dispatch evaluation index Dgzs;

[0010] S5. Pre-set the evaluation threshold W and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

[0011] Preferably, the specific steps of S1 include:

[0012] S11. Deploy several groups of monitoring equipment in the non-motorized vehicle lane to monitor the relevant vehicle flow data in the non-motorized vehicle lane, wherein the relevant vehicle flow data includes the actual non-motorized vehicle flow Scz in each time period, the density value md in the non-motorized vehicle lane in each time period, and the maximum density value md in the non-motorized vehicle lane. max , the length of time that non-motor vehicles stay due to temporary road occupation by buses T stop , the speed of buses when temporarily occupying the road Gcsd and the average speed of non-motor vehicles ; and set up a flow meter at the bus station to monitor the relevant flow data in the bus station in real time, where the relevant flow data includes the number of passengers waiting in the bus station RLL;

[0013] S12, preprocessing the relevant vehicle flow data and the relevant pedestrian flow data to identify and remove noise duplicate values ​​and abnormal values ​​in the relevant vehicle flow data and the relevant pedestrian flow data, and processing the missing values ​​therein by interpolation method, and then converting data of different scales into the same scale according to the dimensionless processing technology;

[0014] S13. The pre-processed relevant vehicle flow data and relevant passenger flow data are aggregated to generate a traffic network data set, and the traffic network data set is uploaded to the cloud platform in real time through streaming tools using Internet technology. After the data is uploaded, automatic backup and archiving operations are performed, and services are provided for subsequent analysis.

[0015] Preferably, the specific steps of S2 include:

[0016] S21. Analyze the relationship between traffic flow and density through the flow conservation equation to preliminarily construct the non-motor vehicle flow prediction value , the specific contents are as follows:

[0017] ;

[0018] Where L represents the flow efficiency of non-motor vehicles, md represents the density value in the non-motor vehicle lane, and md max It is expressed as the maximum density value in the non-motorized vehicle lane;

[0019] S22. Analyze the predicted value of non-motor vehicle flow The difference between the actual non-motor vehicle flow value Scz is used to establish the objective function , which can be obtained by:

[0020] ;

[0021] The significance of this formula is to measure the gap between the predicted value and the actual value, where n represents the monitoring period, i=1, 2, 3, ..., n, Scz i It is represented as the actual non-motor vehicle flow in the i-th period, It is expressed as the predicted value of non-motor vehicle flow in the i-th period.

[0022] Preferably, S2 also includes:

[0023] S23. Solve the objective function by using the least squares method , the partial derivative of the flow efficiency L of non-motor vehicles is calculated to convert the objective function After minimization operation, the numerical value of the flow efficiency L of non-motor vehicles is finally determined.

[0024] Preferably, S2 also includes:

[0025] S24: Substitute the non-motor vehicle flow efficiency L finally determined in S23 into the non-motor vehicle flow prediction value The formula involved is used to recalculate the final non-motor vehicle flow forecast value .

[0026] Preferably, the specific steps of S3 include:

[0027] S31, based on the final non-motor vehicle flow prediction value , and combined with the actual non-motor vehicle flow value Scz during the corresponding monitoring period, it is estimated whether there is congestion in the current non-motor vehicle lane. The specific judgment content is as follows:

[0028] S311, when the final non-motor vehicle flow prediction value ≥ the actual non-motor vehicle flow value Scz, it is estimated that there is a congestion risk in the current non-motor vehicle lane, and the result is transmitted to S313 for further analysis;

[0029] S312, when the final non-motor vehicle flow prediction value When the actual non-motor vehicle flow value Scz is less than the actual non-motor vehicle flow value Scz, it is estimated that there is no congestion risk in the current non-motor vehicle lane;

[0030] S313: After receiving the result in S311, the non-motor vehicle flow prediction value in the corresponding monitoring period is calculated. The density value md in the corresponding non-motorized vehicle lane is extracted, and the density value md in the non-motorized vehicle lane is compared with the maximum density value md in the non-motorized vehicle lane. max of times to determine whether shock wave occurs, specifically:

[0031] S3131. If the density value md in the non-motorized vehicle lane exceeds the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is a shock wave phenomenon in the bus stop area in the current non-motorized vehicle lane;

[0032] S3132: If the density value md in the non-motorized vehicle lane does not exceed the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is no shock wave phenomenon in the bus stop area in the current non-motor vehicle lane;

[0033] S314, analyzing the shock wave phenomenon to calculate the shock wave propagation velocity Jcsd, which is specifically obtained by the following formula:

[0034] ;

[0035] In the formula, Scz down It is expressed as the non-motor vehicle flow value after the bus temporarily occupies the road, Scz up It is expressed as the non-motor vehicle flow value before the bus temporarily occupies the road, md down It is expressed as the density value after the bus temporarily occupies the road, md up It is expressed as the density value before the bus temporarily occupies the road;

[0036] S315. Based on the value of the shock wave propagation velocity Jcsd, when the shock wave propagation velocity Jcsd is a positive value, it means that the temporary occupation of the bus causes the shock wave appearing in the current non-motorized vehicle lane to propagate downstream; when the shock wave propagation velocity Jcsd is a negative value, it means that the temporary occupation of the bus causes the shock wave appearing in the current non-motorized vehicle lane to propagate upstream, and it is estimated that there is congestion in the current non-motorized vehicle lane, and a scheduling optimization instruction is issued to the outside.

[0037] Preferably, the specific steps of S4 include:

[0038] S41. After receiving the dispatch optimization instruction, the influence of the temporary stop of the bus in the non-motor vehicle lane on the travel delay of the non-motor vehicle is analyzed to obtain the delay time Ywsc, which is obtained specifically in the following way:

[0039] ;

[0040] Where, T stop It indicates the length of time that non-motor vehicles stay on the road due to temporary occupation of the bus. It is represented by the average speed of non-motor vehicles, Gcsd is represented by the speed of buses when they temporarily occupy the road, and Lr is represented by the length of the occupied road;

[0041] S42, based on the relevant passenger flow data in the bus station, calculate the service demand ratio Fxb of the current bus station to the bus, which is obtained in the following way:

[0042] ;

[0043] Where RLL is the number of passengers waiting at the bus stop.

[0044] Preferably, S4 also includes:

[0045] S43. Use convolutional neural network technology and combine it with PyTorch technology to build a scheduling optimization model. Input the delay time Ywsc and service demand ratio Fxb into the scheduling optimization model, and after dimensionless processing, obtain the scheduling evaluation index Dgzs by fitting:

[0046] ;

[0047] In the formula, and They are all weight values, and U represents the correction constant.

[0048] Preferably, the specific steps of S5 include:

[0049] S51. By comparing and analyzing the dispatch evaluation index Dgzs with the evaluation threshold W, the current bus dispatching plan is intelligently selected. The specific contents are as follows:

[0050] If the dispatch evaluation index Dgzs falls within the evaluation threshold W, the departure time of the bus will be adjusted to shorten the departure time to 80% of the original time, and the latest departure information and recommended routes will be released to passengers in a timely manner through mobile phone applications or platform display screens;

[0051] If the dispatch evaluation index Dgzs does not fall within the evaluation threshold W, the bus departure time will continue to be executed according to the current plan.

[0052] An intelligent transportation network dynamic dispatching system, comprising a collection subsystem, a difference analysis subsystem, an instruction issuing subsystem, an optimization analysis subsystem and a dispatching subsystem;

[0053] The acquisition subsystem monitors the relevant vehicle flow data in the non-motorized vehicle lane by deploying several groups of monitoring equipment, and sets up flow meters at bus stops to monitor the relevant flow data in real time, and generates a traffic network data set after pre-processing, and uploads the traffic network data set to the cloud platform;

[0054] The difference analysis subsystem sets up the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ;

[0055] The instruction issuing subsystem is based on the predicted traffic flow value and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued;

[0056] The optimization analysis subsystem is used to analyze the impact of temporary bus stops on non-motorized vehicle lanes on non-motorized vehicle travel delays after receiving a dispatch optimization instruction, so as to obtain the delay duration Ywsc, and calculate the current bus station to bus service demand ratio Fxb based on relevant passenger flow data in the bus station, and fit the dispatch optimization model constructed using PyTorch technology to obtain the dispatch evaluation index Dgzs;

[0057] The dispatch subsystem is used to pre-set the evaluation threshold W, and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

[0058] The present invention provides a method and system for dynamic scheduling of intelligent transportation networks, which have the following beneficial effects:

[0059] (1) This method obtains vehicle and passenger flow data in real time by deploying monitoring equipment to ensure the accuracy and timeliness of information; secondly, based on the flow conservation equation, the objective function is established and minimized, which can effectively identify the impact of temporary bus occupation on non-motor vehicle flow and accurately calculate the flow efficiency L of non-motor vehicles to improve the accuracy of subsequent vehicle flow prediction; through the comparative analysis of predicted flow and actual flow, the system can intelligently judge the current congestion status of non-motor vehicle lanes and quickly issue dispatch optimization instructions, which not only helps to alleviate traffic congestion, but also reduces the delay time of non-motor vehicles and improves the travel experience. In addition, the dispatch optimization model built with PyTorch technology makes the calculation of the dispatch evaluation index Dgzs more accurate, providing a scientific basis for the bus dispatch plan. When the evaluation index is compared with the preset threshold, the system can intelligently select the corresponding bus dispatch strategy. This process further realizes the efficient allocation of traffic resources, reduces operating costs, and improves the response speed and quality of bus services. In short, this method fully considers the problem of congestion in the corresponding lanes caused by temporary bus stops, thus laying a foundation for improving the management efficiency of traffic flow.

[0060] (2) The application of the flow conservation equation in step S2 of the present invention not only provides a scientific basis for the prediction of non-motor vehicle flow, but also realizes the in-depth analysis between traffic flow and density. This process makes the understanding of traffic status more accurate by establishing the relationship between non-motor vehicle flow efficiency L and density and maximum density. The construction of this mathematical model helps to accurately predict the non-motor vehicle flow in different time periods and provides data support for urban traffic management. Through the difference analysis between the predicted value and the actual value in S22, the established objective function becomes an important tool for measuring the performance of the model. This objective function can quantify the prediction accuracy, promote the optimization and iteration of the model, and ensure that the accuracy of the prediction can be continuously improved in practical applications. This data-driven feedback mechanism makes traffic flow prediction not only a static analysis, but also an important strategy for dynamically adapting to changes in urban traffic. In addition, the combination of flow prediction and actual monitoring data forms a closed-loop feedback system. Through the continuously updated real-time data, traffic managers can quickly identify the deviation of flow prediction, so as to adjust the accuracy of the prediction results in time. In short, the present invention not only enhances the scientificity and practicality of traffic management through accurate flow analysis and efficient objective function establishment, but also lays a solid foundation for the construction of future intelligent transportation systems.

[0061] (3) In step S3, by comparing the predicted value of non-motorized vehicle flow with the actual flow value Scz, the system can effectively identify the congestion risk in the non-motorized vehicle lane. The innovation of this process lies in real-time monitoring of traffic conditions and providing accurate congestion judgment through data-driven methods. This intelligent monitoring mechanism can significantly improve the response speed of traffic management and take timely measures to alleviate congestion. Specifically, when the predicted value of non-motorized vehicle flow is greater than or equal to the actual flow value, the system will actively identify potential congestion risks and transmit the information to subsequent analysis links. This pre-warning mechanism enables traffic managers to formulate response strategies in advance to avoid further deterioration of congestion. By comparing the density value with the maximum density value, the system can not only determine the occurrence of shock wave phenomenon, but also provide in-depth analysis in a dynamic traffic environment and provide data support for traffic flow optimization. Furthermore, by calculating the shock wave propagation speed Jcsd, the system can gain insight into the dynamic evolution of congestion conditions. This dynamic feedback mechanism provides a basis for real-time scheduling and resource allocation, ensuring the efficient operation of traffic flow. In addition, the creativity of this method lies in its universality and flexibility, which can adapt to different cities and their specific traffic conditions.

[0062] (4) By combining the quantification of the delay time Ywsc and the service demand ratio Fxb in step S4, the intelligent transportation network can further realize dynamic feedback and real-time scheduling to form a closed-loop management. By analyzing the mutual influence between buses and non-motorized vehicle flows, the system can not only adjust the bus departure interval in real time, but also optimize the traffic strategy of non-motorized vehicle lanes during peak hours, thereby effectively alleviating traffic congestion. In addition, using the service demand ratio Fxb, the system can arrange additional bus resources in advance when passenger flow increases suddenly, further improving the service level of public transportation. In general, this method not only improves the operating efficiency of traffic flow through precise data analysis and dynamic adjustment, but also provides innovative solutions for urban traffic management, helping to build a more intelligent and efficient transportation network. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic diagram of a flow chart of a dynamic scheduling method for an intelligent transportation network according to the present invention;

[0064] Figure 2 This is a block diagram of an intelligent transportation network dynamic scheduling system of the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Example 1

[0067] See also Figure 1 The present invention provides a method for dynamic scheduling of intelligent transportation networks, comprising the following steps:

[0068] S1. Deploy several groups of monitoring equipment to monitor the relevant vehicle flow data in the non-motorized vehicle lanes, and set up flow meters at bus stops to monitor the relevant flow data in real time, generate a traffic network data set after pre-processing, and upload the traffic network data set to the cloud platform;

[0069] S2. Establishing the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ;

[0070] S3: Based on the predicted traffic flow value obtained in S2 and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued;

[0071] S4. After receiving the dispatch optimization instruction, analyze the impact of the temporary stop of the bus in the non-motorized vehicle lane on the travel delay of the non-motorized vehicle to obtain the delay time Ywsc, and calculate the service demand ratio Fxb of the current bus station to the bus based on the relevant passenger flow data in the bus station, and combine the dispatch optimization model constructed by PyTorch technology to obtain the dispatch evaluation index Dgzs;

[0072] S5. Pre-set the evaluation threshold W and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

[0073] In this embodiment, by deploying monitoring equipment and pedestrian flow meters, it is possible to further collect real-time data on vehicle and pedestrian flow in non-motorized vehicle lanes and bus stops. This real-time monitoring capability provides an accurate basis for subsequent data analysis and decision-making, ensuring the scientificity and rationality of the scheduling plan. The objective function established based on the flow conservation equation can further accurately analyze the impact of temporary bus occupation of the lane on non-motorized vehicle flow. By minimizing the difference between the predicted flow and the actual flow, the flow efficiency L of non-motorized vehicles can be further determined. It can be regarded as the number of vehicles that can pass through at a unit density in a unit time. By determining the flow efficiency L of non-motorized vehicles, the final non-motorized vehicle flow prediction value can be obtained. The system can grasp the congestion in the non-motorized vehicle lane and issue dispatch optimization instructions immediately after identifying the congestion in the non-motorized vehicle lane. This rapid response mechanism can further reduce the frequency of traffic jams and reduce the delay of non-motorized vehicles, thereby improving the overall traffic mobility. By analyzing the impact of temporary bus stops on non-motorized vehicle delays and calculating the service demand ratio based on passenger flow data, it can more accurately match bus services with passenger flow demand. This dynamic adjustment can reasonably allocate bus resources during peak hours to ensure passengers' travel experience. The dispatch optimization model built with PyTorch technology makes the calculation of the dispatch evaluation index more scientific and intelligent. By setting and analyzing the evaluation threshold, the current bus dispatch plan can be effectively judged, the flexibility and adaptability of dispatch can be improved, and the changing traffic needs can be met. By optimizing the bus dispatch strategy, not only can the traffic efficiency be improved, but also the congestion in non-motorized vehicle lanes can be further reduced, thus making a positive contribution to environmental protection and sustainable development. In short, this method provides an efficient solution for intelligent traffic management, which can consider the overall traffic mobility while improving the quality of public transportation services, laying the foundation for the intelligence and modernization of urban traffic management.

[0074] Example 2

[0075] Please refer to Figure 1 , specifically: S1 specific steps include:

[0076] S11. Deploy several groups of monitoring equipment in the non-motorized vehicle lane to monitor the relevant vehicle flow data in the non-motorized vehicle lane, wherein the relevant vehicle flow data includes the actual non-motorized vehicle flow Scz in each time period, the density value md in the non-motorized vehicle lane in each time period, and the maximum density value md in the non-motorized vehicle lane. max , the length of time that non-motor vehicles stay due to temporary road occupation by buses T stop , the speed of buses when temporarily occupying the road Gcsd and the average speed of non-motor vehicles ; and set up a flow meter at the bus station to monitor the relevant flow data in the bus station in real time, where the relevant flow data includes the number of passengers waiting in the bus station RLL;

[0077] The density value md in the non-motorized vehicle lane can be monitored and obtained by a ground sensor or a video monitoring system, wherein the video monitoring system analyzes the video stream through image processing technology to calculate the number of vehicles within a unit length;

[0078] The duration of non-motor vehicles staying due to temporary road occupation by buses T stop The time a vehicle spends in a particular area can be monitored through ground sensors;

[0079] The speed of the bus when temporarily occupying the road Gcsd can be monitored in real time by the GPS device on the bus, or the real-time speed data can be obtained by using the on-board sensor;

[0080] Average speed of non-motor vehicles It can be monitored and obtained through speed radar or ground sensors.

[0081] S12, preprocessing the relevant vehicle flow data and the relevant pedestrian flow data to identify and remove noise duplicate values ​​and abnormal values ​​in the relevant vehicle flow data and the relevant pedestrian flow data, and processing the missing values ​​therein by interpolation method, and then converting data of different scales into the same scale according to the dimensionless processing technology;

[0082] S13. Generate a traffic network data set by aggregating the pre-processed relevant vehicle flow data and relevant pedestrian flow data, and use Internet technology to upload the traffic network data set to the cloud platform in real time through streaming tools (such as Kafka, AWS Kinesis, Google Pub / Sub). After the data is uploaded, perform automated backup and archiving operations, and provide services for subsequent analysis.

[0083] In this embodiment, step S1 of the method realizes multi-dimensional monitoring of traffic flow and pedestrian flow by collecting real-time data of non-motorized vehicle lanes through multiple sets of monitoring equipment. This accurate data acquisition not only improves the scientific nature of traffic management, but also provides a solid foundation for subsequent analysis. Specifically, various types of data monitored in step S11, such as actual non-motorized vehicle flow Scz, density value md, maximum density value md maxAs well as the length of stay Tstop caused by temporary road occupation by buses, it can help traffic managers to fully understand the traffic conditions. The monitoring of this data helps to accurately assess the driving conditions of non-motor vehicles and the extent to which they are affected by temporary road occupation by buses, providing a key basis for future scheduling decisions. In step S12, the high quality and reliability of the data are ensured by eliminating noise, identifying and processing outliers for the relevant data. At the same time, the interpolation method is used to process missing values ​​and dimensionless processing technology, so that data of different scales can be analyzed uniformly. This process of data cleaning and standardization improves the accuracy and effectiveness of subsequent analysis as much as possible, laying a good foundation for the establishment of intelligent scheduling models. Step S13 uses streaming technology to achieve real-time data upload, which not only improves the efficiency of data transmission, but also ensures that the data is automatically backed up and archived after uploading. This process not only reduces the risk of data loss, but also provides continuous support for subsequent real-time analysis and decision-making. In addition, through the application of Internet technology, the rapid update and sharing of traffic data is realized, so that all relevant parties can obtain the latest information in a timely manner, thereby improving the responsiveness and management efficiency of the entire transportation system. In summary, this method further improves the efficiency and management level of the intelligent transportation network through multi-faceted monitoring, data cleaning and real-time uploading, and provides an innovative solution for future traffic scheduling and optimization.

[0084] Example 3

[0085] Please refer to Figure 1 , specifically: S2 specific steps include:

[0086] S21. Analyze the relationship between traffic flow and density through the flow conservation equation to preliminarily construct the non-motor vehicle flow prediction value , the specific contents are as follows:

[0087] ;

[0088] Where L represents the flow efficiency of non-motor vehicles, md represents the density value in the non-motor vehicle lane, and md max It is expressed as the maximum density value in the non-motorized vehicle lane;

[0089] S22. Analyze the predicted value of non-motor vehicle flow The difference between the actual non-motor vehicle flow value Scz is used to establish the objective function , which can be obtained by:

[0090] ;

[0091] The significance of this formula is to measure the gap between the predicted value and the actual value, where n represents the monitoring period, i=1, 2, 3, ..., n, Sczi It is represented as the actual non-motor vehicle flow in the i-th period, It is expressed as the predicted value of non-motor vehicle flow in the i-th period.

[0092] It should be noted that the flow conservation equation is used to describe the relationship between the change in traffic flow and the number of vehicles in a specific area; its expression is: ;in, Expressed as the number of vehicles at position x and time t; It is expressed as the flow rate at position x and time t (the number of vehicles passing a point per unit time); represents the partial derivative with respect to time t, which describes the rate at which the number of vehicles changes over time; represents the partial derivative with respect to position x, which describes the rate at which vehicle flow changes with position; is the partial derivative operator, which represents the rate of change of a function with respect to some variable.

[0093] In this embodiment, in step S2, the method sets up an objective function to analyze the difference between the predicted value of non-motor vehicle flow and the actual non-motor vehicle flow value Scz. This process provides a quantitative indicator for the optimization of the model. The design of the objective function enables traffic managers to clearly evaluate the prediction accuracy and then adjust the scheduling strategy in a targeted manner. This difference analysis can help identify the source of the prediction error and provide important feedback for subsequent data processing and model iteration, thereby improving the robustness and accuracy of the prediction model. Further, by setting the monitoring cycle, the system can analyze the flow change trend in different time periods, which not only provides a dynamic perspective for the monitoring of real-time traffic conditions, but also provides historical data support for future traffic flow prediction, forming a benign feedback loop. In addition, through continuous flow and density analysis, the traffic management system can achieve adaptive scheduling, respond to changes in non-motor vehicle flow in a timely manner, effectively alleviate congestion, and optimize traffic efficiency. In summary, the method constructs a data-driven non-motor vehicle flow prediction and analysis method through the combination of flow conservation equation and objective function, which not only improves the accuracy of flow prediction, but also provides important technical support for the intelligentization of urban traffic management, and helps to build a more efficient and sustainable urban transportation network.

[0094] Example 4

[0095] Please refer to Figure 1 , specifically:

[0096] S2 also includes: S2 also includes:

[0097] S23. Solve the objective function by using the least squares method , the partial derivative of the flow efficiency L of non-motor vehicles is calculated to convert the objective function After minimization operation, the numerical value of the flow efficiency L of non-motor vehicles is finally determined.

[0098] Among them, the steps of using the least squares method to solve are: first calculate the error value, that is, , each observation point has an error value; then, using the objective function, in fact, is to calculate the error square sum; then find the partial derivative of the objective function, in order to minimize the error square sum, we take the derivative of the non-motor vehicle flow efficiency L, and set the partial derivative to 0, to solve the final non-motor vehicle flow efficiency L;

[0099] S24: Substitute the non-motor vehicle flow efficiency L finally determined in S23 into the non-motor vehicle flow prediction value The formula involved is used to recalculate the final non-motor vehicle flow forecast value .

[0100] In this embodiment, in step S2, the specific implementation of S23 and S24 provides a more accurate and scientific basis for the prediction of non-motor vehicle flow. By using the least squares method to solve the objective function, the researchers can effectively identify and optimize the flow efficiency L of non-motor vehicles. This method not only realizes the systematization and standardization of non-motor vehicle flow prediction, but also ensures the sensitivity and accuracy of the model to flow efficiency by solving partial derivatives. Combining the gradient descent method for minimization operation makes the objective function more operational and flexible in practical applications. The iterative process of the algorithm can quickly adjust the flow efficiency L of non-motor vehicles, and then optimize the prediction model in real time. This dynamic adjustment mechanism improves the reliability of traffic flow prediction and can better cope with the ever-changing traffic environment. Substituting the final determined flow efficiency L into the flow prediction formula in S24 further enhances the practicality of the model. Through this process, the researchers can update the non-motor vehicle flow prediction value in real time to ensure that the prediction results reflect the latest traffic conditions, which not only helps to improve the response speed of traffic management, but also provides a basis for decision makers to facilitate timely adjustments during peak hours or special circumstances. In addition, the innovation of this method lies in its use of data-driven feedback mechanism to form a close connection between traffic flow prediction and actual operation data. This mechanism not only improves the adaptability of the model, but also lays the foundation for the subsequent development of intelligent transportation systems. In short, through accurate flow efficiency optimization and dynamic flow prediction, it lays the foundation for the subsequent analysis of non-motorized vehicle lane congestion caused by temporary occupation of buses.

[0101] Example 5

[0102] Please refer to Figure 1 , specifically: S3 specific steps include:

[0103] S31, based on the final non-motor vehicle flow prediction value , and combined with the actual non-motor vehicle flow value Scz during the corresponding monitoring period, it is estimated whether there is congestion in the current non-motor vehicle lane. The specific judgment content is as follows:

[0104] S311, when the final non-motor vehicle flow prediction value ≥ the actual non-motor vehicle flow value Scz, it is estimated that there is a congestion risk in the current non-motor vehicle lane, and the result is transmitted to S313 for further analysis;

[0105] S312, when the final non-motor vehicle flow prediction value When the actual non-motor vehicle flow value Scz is less than the actual non-motor vehicle flow value Scz, it is estimated that there is no congestion risk in the current non-motor vehicle lane;

[0106] The actual non-motor vehicle flow value Scz here refers to the number of vehicles passing through a certain location within a certain period of time, such as the number of vehicles passing through per hour; the final non-motor vehicle flow prediction value It takes into account the characteristics of traffic flow, including the impact of density on flow. Under certain conditions (for example, specific traffic rules, road conditions, etc.), it reflects the maximum flow that can be achieved theoretically if the vehicle density on the road is md, that is, the flow that can be achieved theoretically at a specific density.

[0107] S313: After receiving the result in S311, the non-motor vehicle flow prediction value in the corresponding monitoring period is calculated. The density value md in the corresponding non-motorized vehicle lane is extracted, and the density value md in the non-motorized vehicle lane is compared with the maximum density value md in the non-motorized vehicle lane. max of times to determine whether shock wave occurs, specifically:

[0108] S3131. If the density value md in the non-motorized vehicle lane exceeds the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is a shock wave phenomenon in the bus stop area in the current non-motorized vehicle lane;

[0109] S3132: If the density value md in the non-motorized vehicle lane does not exceed the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is no shock wave phenomenon in the bus stop area in the current non-motor vehicle lane;

[0110] S314, analyzing the shock wave phenomenon to calculate the shock wave propagation velocity Jcsd, which is specifically obtained by the following formula:

[0111] ;

[0112] In the formula, Scz down It is expressed as the non-motor vehicle flow value after the bus temporarily occupies the road, Scz up It is expressed as the non-motor vehicle flow value before the bus temporarily occupies the road, md down It is expressed as the density value after the bus temporarily occupies the road, md up It is expressed as the density value before the bus temporarily occupies the road;

[0113] S315, based on the value of the shock wave propagation velocity Jcsd, when the shock wave propagation velocity Jcsd is a positive value, it means that the temporary occupation of the bus causes the shock wave in the current non-motorized vehicle lane to propagate downstream, which may be because the subsequent traffic volume begins to increase and the congestion is gradually alleviated;

[0114] When the shock wave propagation speed Jcsd is negative, it means that the temporary occupation of the bus causes the shock wave in the current non-motorized vehicle lane to propagate upstream, indicating that congestion is spreading, and it is estimated that there is congestion in the current non-motorized vehicle lane, and at the same time, a scheduling optimization instruction is issued to the outside.

[0115] It should be noted that temporary occupation of the road by buses will lead to a reduction in the effective width of non-motorized vehicle lanes, thereby increasing traffic density in local areas. If the density increases to a certain threshold, it will cause traffic flow to reach saturation or even congestion, thus producing a shock wave phenomenon.

[0116] In this embodiment, based on step S2, the predicted value of non-motor vehicle flow is accurately determined, so that the congestion of non-motor vehicle lanes can be judged in the comparative analysis of the predicted value of non-motor vehicle flow and the actual flow value, thereby effectively alleviating traffic pressure. This early identification mechanism can reduce traffic accidents and delays caused by congestion in practical applications and improve traffic fluency. In S313, the occurrence of shock wave phenomenon can be further accurately judged by the density value md. This density-based judgment method provides a quantitative analysis basis for traffic flow status, allowing traffic managers to make scientific decisions based on real-time data. For example, if a shock wave phenomenon is found, the traffic management system can immediately adjust the timing of signal lights, optimize bus departure intervals, or guide non-motor vehicles to choose other routes to reduce traffic pressure; further, S314 and S315 provide quantitative indicators for analyzing congestion conditions by calculating the shock wave propagation speed Jcsd. When the shock wave speed is positive, it means that congestion is being alleviated; otherwise, it means that congestion is intensifying. The introduction of this dynamic monitoring mechanism makes traffic management no longer rely on single data, but can make real-time adjustments and responses to ensure the efficient operation of the traffic system. In addition, the creative effect of this method is also reflected in its adaptability and scalability. By continuously collecting and analyzing real-time data, the system can learn and optimize its own judgment logic, gradually form an intelligent scheduling model, and can adapt to changes in different time periods and traffic conditions. This flexible scheduling capability not only provides new ideas for urban traffic management, but also lays a solid foundation for the development of future intelligent transportation systems. In short, through data-driven congestion judgment and shock wave analysis, the present invention not only improves the judgment of the flow efficiency of non-motorized vehicle lanes, but also provides a scientific and effective solution for intelligent traffic scheduling, thereby promoting the intelligentization and modernization of urban traffic management.

[0117] Example 6

[0118] Please refer to Figure 1 , specifically: S4 specific steps include:

[0119] S41. After receiving the dispatch optimization instruction, the influence of the temporary stop of the bus in the non-motor vehicle lane on the travel delay of the non-motor vehicle is analyzed to obtain the delay time Ywsc, which is obtained specifically in the following way:

[0120] ;

[0121] Where, T stop It indicates the length of time that non-motor vehicles stay on the road due to temporary occupation of the bus. It is represented by the average speed of non-motor vehicles, Gcsd is represented by the speed of buses when temporarily occupying the road, and Lr is represented by the length of the occupied road; It is expressed as the time required for non-motor vehicles to pass when there is no bus temporarily occupying the road; It is expressed as the length of time that non-motor vehicles pass when the bus temporarily occupies the road; among them, the occupied road length Lr can be determined according to the length of the bus.

[0122] S42, based on the relevant passenger flow data in the bus station, calculate the service demand ratio Fxb of the current bus station to the bus, which is obtained in the following way:

[0123] ;

[0124] In the formula, RLL represents the number of passengers waiting at the bus stop. The purpose of adding one to the denominator is to avoid division by zero errors. In calculations, if a certain quantity may be zero, direct division may lead to calculation errors. By adding one to the denominator, the safety of the division operation can be ensured.

[0125] In this embodiment, in step S4, by analyzing the temporary stop behavior of the bus in the non-motorized vehicle lane, the system can calculate the non-motorized vehicle travel delay time Ywsc caused by the bus occupying the lane. This innovative delay calculation method comprehensively considers multiple factors, including the non-motorized vehicle stop time Tstop, average speed, bus travel speed Gcsd and occupied lane length Lr, thereby providing a more comprehensive delay evaluation mechanism. This refined analysis can not only provide a basis for real-time traffic scheduling, but also help traffic managers formulate more scientific response strategies to reduce traffic congestion. Specifically, by clearly calculating the passing time of non-motorized vehicles under different conditions, the system can effectively identify the impact of occupying the lane on traffic flow. This dynamic evaluation method can be flexibly applied in the actual traffic environment, timely reflect the changes in traffic status, and provide real-time data support for traffic management. For example, when the delay time Ywsc exceeds a certain threshold, the system can quickly issue a scheduling optimization instruction to guide the bus to leave the non-motorized vehicle lane in time to reduce interference with traffic flow. In addition, the system further improves the intelligence level of dispatching by calculating the service demand ratio Fxb based on the passenger flow data in the bus station. The design of adding one to the denominator is introduced in the calculation formula to ensure that the calculation can be performed safely when the number of passengers is zero, avoiding the error of dividing by zero. This design idea demonstrates the flexibility of the system in achieving efficient data processing and safe calculation, and ensures the robustness of the entire analysis process. Combining these two innovations, the system can not only improve the dispatching efficiency of buses, but also improve the traffic efficiency of non-motor vehicles, and optimize traffic mobility as a whole. In addition, through continuous data monitoring and analysis, this intelligent traffic dispatching method can provide valuable insights for urban traffic management, promote the scientific formulation of future traffic policies, and thus improve the overall traffic environment of the city. Ultimately, the implementation of this method will help build a more efficient, convenient and safe intelligent transportation network and promote sustainable urban development.

[0126] Example 7

[0127] Please refer to Figure 1 Specifically, S4 also includes:

[0128] S43. Use convolutional neural network technology and combine it with PyTorch technology to build a scheduling optimization model. Input the delay time Ywsc and service demand ratio Fxb into the scheduling optimization model, and after dimensionless processing, obtain the scheduling evaluation index Dgzs by fitting:

[0129] ;

[0130] In the formula, and are all weight values, U represents the correction constant, where 0< ≤1,0< ≤1, and + =1, the weight value can be obtained by referring to the hierarchical analysis method;

[0131] The specific steps of S5 include:

[0132] S51. By comparing and analyzing the dispatch evaluation index Dgzs with the evaluation threshold W, the current bus dispatching plan is intelligently selected. The specific contents are as follows:

[0133] If the dispatch evaluation index Dgzs falls within the evaluation threshold W, the departure time of the bus will be adjusted to shorten it to 80% of the original time to optimize the passengers' boarding and alighting experience. The latest departure information and recommended routes will be released to passengers in a timely manner through mobile phone applications or platform displays to reduce unnecessary waiting.

[0134] If the dispatch evaluation index Dgzs does not fall within the evaluation threshold W, the bus departure time will continue to be executed according to the current plan.

[0135] It should be noted that the hierarchical analysis method is a combination of qualitative and quantitative analysis methods. It can decompose complex problems into multiple levels. By comparing the importance of factors at each level, it can help decision makers make decisions on complex problems and determine the final decision plan. In this process, the hierarchical analysis method can be used to determine the weight values ​​of these indicators.

[0136] In this embodiment, in steps S4 and S5, by applying the scheduling optimization model constructed by convolutional neural network (CNN) and PyTorch technology, the system can analyze the delay time Ywsc and service demand ratio Fxb in real time, thereby generating a scheduling evaluation index Dgzs. The innovation of this process lies in its use of deep learning technology, which enables the model to learn complex traffic patterns and passenger needs, significantly improving the intelligent level of scheduling. The calculation of the scheduling evaluation index Dgzs and the comparison with the evaluation threshold W provide a scientific basis for the departure plan of the bus. When the scheduling evaluation index Dgzs falls within the threshold range, the system will automatically adjust the departure time of the bus to 80% of the original time, thereby optimizing the passengers' boarding and alighting experience. This flexibility can not only reduce the waiting time of passengers, but also improve the capacity utilization rate of the bus and alleviate the traffic pressure during peak hours. In addition, the latest departure information and recommended routes are released in real time through mobile phone applications and platform display screens, so that passengers can obtain dynamic information in a timely manner, optimize passengers' travel options, and enhance passengers' satisfaction and convenience. This information transparency strategy effectively reduces the anxiety of passengers caused by information asymmetry and improves the utilization rate of public transportation. When the dispatch evaluation index does not fall within the evaluation threshold, the system will maintain the existing departure plan to ensure the stability and predictability of traffic management. This dual dispatch strategy not only improves the response speed of the bus system, but also effectively responds to various emergencies and ensures the continuity of public transportation services. In summary, by combining deep learning with intelligent dispatching, the system can not only respond to traffic changes in real time, but also improve passengers' travel experience, forming a more intelligent, efficient and humane public transportation network.

[0137] Example 8

[0138] Please refer to Figure 2 ,Specifically: An intelligent transportation network dynamic dispatching system, including a collection subsystem, a difference analysis subsystem, an instruction issuing subsystem, an optimization analysis subsystem and a dispatching subsystem;

[0139] The acquisition subsystem monitors the relevant vehicle flow data in the non-motorized vehicle lane by deploying several groups of monitoring equipment, and sets up flow meters at bus stops to monitor the relevant flow data in real time, and generates a traffic network data set after pre-processing, and uploads the traffic network data set to the cloud platform;

[0140] The difference analysis subsystem sets up the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ;

[0141] The instruction issuing subsystem is based on the predicted traffic flow value and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued;

[0142] The optimization analysis subsystem is used to analyze the impact of temporary bus stops on non-motorized vehicle lanes on non-motorized vehicle travel delays after receiving a dispatch optimization instruction, so as to obtain the delay duration Ywsc, and calculate the current bus station to bus service demand ratio Fxb based on relevant passenger flow data in the bus station, and fit the dispatch optimization model constructed using PyTorch technology to obtain the dispatch evaluation index Dgzs;

[0143] The dispatch subsystem is used to pre-set the evaluation threshold W, and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

[0144] 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.

Claims

1. A method for dynamic scheduling of intelligent transportation networks, characterized in that: The following steps are included: S1. Deploy several groups of monitoring equipment to monitor the relevant vehicle flow data in the non-motorized vehicle lanes, and set up flow meters at bus stops to monitor the relevant flow data in real time, generate a traffic network data set after pre-processing, and upload the traffic network data set to the cloud platform; S2. Establishing the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ; S3, based on the predicted traffic flow value obtained in S2 and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued; S4. After receiving the dispatch optimization instruction, analyze the impact of the temporary stop of the bus in the non-motorized vehicle lane on the travel delay of the non-motorized vehicle to obtain the delay time Ywsc, and calculate the service demand ratio Fxb of the current bus station to the bus based on the relevant passenger flow data in the bus station, and combine the dispatch optimization model constructed by PyTorch technology to obtain the dispatch evaluation index Dgzs; S5. Pre-set the evaluation threshold W and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

2. The method for dynamic scheduling of an intelligent transportation network according to claim 1, characterized in that: The specific steps of S1 include: S11. Deploy several groups of monitoring equipment in the non-motorized vehicle lane to monitor the relevant vehicle flow data in the non-motorized vehicle lane, wherein the relevant vehicle flow data includes the actual non-motorized vehicle flow Scz in each time period, the density value md in the non-motorized vehicle lane in each time period, and the maximum density value md in the non-motorized vehicle lane. max , the length of time that non-motor vehicles stay due to temporary road occupation by buses T stop , the speed of buses when temporarily occupying the road Gcsd and the average speed of non-motor vehicles ; and set up a flow meter at the bus station to monitor the relevant flow data in the bus station in real time, where the relevant flow data includes the number of passengers waiting in the bus station RLL; S12, preprocessing the relevant vehicle flow data and the relevant pedestrian flow data to identify and remove noise duplicate values ​​and abnormal values ​​in the relevant vehicle flow data and the relevant pedestrian flow data, and processing the missing values ​​therein by interpolation method, and then converting data of different scales into the same scale according to the dimensionless processing technology; S13. The pre-processed relevant vehicle flow data and relevant passenger flow data are aggregated to generate a traffic network data set, and the traffic network data set is uploaded to the cloud platform in real time through streaming tools using Internet technology. After the data is uploaded, automatic backup and archiving operations are performed, and services are provided for subsequent analysis.

3. The method for dynamic scheduling of an intelligent transportation network according to claim 2, characterized in that: The specific steps of S2 include: S21. Analyze the relationship between traffic flow and density through the flow conservation equation to preliminarily construct the non-motor vehicle flow prediction value , the specific contents are as follows: ； Where L represents the flow efficiency of non-motor vehicles, md represents the density value in the non-motor vehicle lane, and md max It is expressed as the maximum density value in the non-motorized vehicle lane; S22. Analyze the predicted value of non-motor vehicle flow The difference between the actual non-motor vehicle flow value Scz is used to establish the objective function , which can be obtained by: ； The significance of this formula is to measure the gap between the predicted value and the actual value, where n represents the monitoring period, i=1, 2, 3, ..., n, Scz i It is represented as the actual non-motor vehicle flow in the i-th period, It is expressed as the predicted value of non-motor vehicle flow in the i-th period.

4. The method for dynamic scheduling of an intelligent transportation network according to claim 3, characterized in that: S2 also includes: S23. Solve the objective function by using the least squares method , the partial derivative of the flow efficiency L of non-motor vehicles is calculated to convert the objective function After minimization operation, the numerical value of the flow efficiency L of non-motor vehicles is finally determined.

5. The method for dynamic scheduling of intelligent transportation network according to claim 4, characterized in that: S2 also includes: S24: Substitute the non-motor vehicle flow efficiency L finally determined in S23 into the non-motor vehicle flow prediction value The formula involved is used to recalculate the final non-motor vehicle flow forecast value .

6. The method for dynamic scheduling of an intelligent transportation network according to claim 2, characterized in that: The specific steps of S3 include: S31, based on the final non-motor vehicle flow prediction value , and combined with the actual non-motor vehicle flow value Scz during the corresponding monitoring period, it is estimated whether there is congestion in the current non-motor vehicle lane. The specific judgment content is as follows: S311, when the final non-motor vehicle flow prediction value ≥ the actual non-motor vehicle flow value Scz, it is estimated that there is a congestion risk in the current non-motor vehicle lane, and the result is transmitted to S313 for further analysis; S312, when the final non-motor vehicle flow prediction value When the actual non-motor vehicle flow value Scz is less than the actual non-motor vehicle flow value Scz, it is estimated that there is no congestion risk in the current non-motor vehicle lane; S313: After receiving the result in S311, the non-motor vehicle flow prediction value in the corresponding monitoring period is calculated. The density value md in the corresponding non-motorized vehicle lane is extracted, and the density value md in the non-motorized vehicle lane is compared with the maximum density value md in the non-motorized vehicle lane. max of times to determine whether shock wave occurs, specifically: S3131. If the density value md in the non-motorized vehicle lane exceeds the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is a shock wave phenomenon in the bus stop area in the current non-motorized vehicle lane; S3132: If the density value md in the non-motorized vehicle lane does not exceed the maximum density value md in the non-motorized vehicle lane max of times, it is judged that there is no shock wave phenomenon in the bus stop area in the current non-motor vehicle lane; S314, analyzing the shock wave phenomenon to calculate the shock wave propagation velocity Jcsd, which is specifically obtained by the following formula: ； In the formula, Scz down It is expressed as the non-motor vehicle flow value after the bus temporarily occupies the road, Scz up It is expressed as the non-motor vehicle flow value before the bus temporarily occupies the road, md down It is expressed as the density value after the bus temporarily occupies the road, md up It is expressed as the density value before the bus temporarily occupies the road; S315. Based on the value of the shock wave propagation velocity Jcsd, when the shock wave propagation velocity Jcsd is a positive value, it means that the temporary occupation of the bus causes the shock wave appearing in the current non-motorized vehicle lane to propagate downstream; when the shock wave propagation velocity Jcsd is a negative value, it means that the temporary occupation of the bus causes the shock wave appearing in the current non-motorized vehicle lane to propagate upstream, and it is estimated that there is congestion in the current non-motorized vehicle lane, and a scheduling optimization instruction is issued to the outside.

7. The method for dynamic scheduling of an intelligent transportation network according to claim 6, characterized in that: The specific steps of S4 include: S41. After receiving the dispatch optimization instruction, the influence of the temporary stop of the bus in the non-motor vehicle lane on the travel delay of the non-motor vehicle is analyzed to obtain the delay time Ywsc, which is obtained specifically in the following way: ； Where, T stop It indicates the length of time that non-motor vehicles stay on the road due to temporary occupation of the bus. It is represented by the average speed of non-motor vehicles, Gcsd is represented by the speed of buses when they temporarily occupy the road, and Lr is represented by the length of the occupied road; S42, based on the relevant passenger flow data in the bus station, calculate the service demand ratio Fxb of the current bus station to the bus, which is obtained in the following way: ； Where RLL is the number of passengers waiting at the bus stop.

8. The method for dynamic scheduling of intelligent transportation network according to claim 7, characterized in that: S4 also includes: S43. Use convolutional neural network technology and combine it with PyTorch technology to build a scheduling optimization model. Input the delay time Ywsc and service demand ratio Fxb into the scheduling optimization model, and after dimensionless processing, obtain the scheduling evaluation index Dgzs by fitting: ； In the formula, and They are all weight values, and U represents the correction constant.

9. The method for dynamic scheduling of intelligent transportation network according to claim 1, characterized in that: The specific steps of S5 include: S51. By comparing and analyzing the dispatch evaluation index Dgzs with the evaluation threshold W, the current bus dispatching plan is intelligently selected. The specific contents are as follows: If the dispatch evaluation index Dgzs falls within the evaluation threshold W, the departure time of the bus will be adjusted to shorten the departure time to 80% of the original time, and the latest departure information and recommended routes will be released to passengers in a timely manner through mobile phone applications or platform display screens; If the dispatch evaluation index Dgzs does not fall within the evaluation threshold W, the bus departure time will continue to be executed according to the current plan.

10. An intelligent transportation network dynamic scheduling system, used to implement an intelligent transportation network dynamic scheduling method according to any one of claims 1 to 9, characterized in that: It includes collection subsystem, difference analysis subsystem, instruction issuing subsystem, optimization analysis subsystem and scheduling subsystem; The acquisition subsystem monitors the relevant vehicle flow data in the non-motorized vehicle lane by deploying several groups of monitoring equipment, and sets up flow meters at bus stops to monitor the relevant flow data in real time, and generates a traffic network data set after pre-processing, and uploads the traffic network data set to the cloud platform; The difference analysis subsystem sets up the objective function based on the flow conservation equation , in order to analyze the predicted value of non-motor vehicle flow when buses are temporarily occupying the road The difference between the actual non-motor vehicle flow value Scz changes, and the objective function After minimization, the flow efficiency L of non-motor vehicles is determined, and based on the flow efficiency L of non-motor vehicles, the final non-motor vehicle flow prediction value is constructed. ; The instruction issuing subsystem is based on the predicted traffic flow value and the actual non-motor vehicle flow value Scz, to estimate whether there is congestion in the current non-motor vehicle lane. If so, a dispatch optimization instruction is issued; The optimization analysis subsystem is used to analyze the impact of temporary bus stops on non-motorized vehicle lanes on non-motorized vehicle travel delays after receiving a dispatch optimization instruction, so as to obtain the delay duration Ywsc, and calculate the current bus station to bus service demand ratio Fxb based on relevant passenger flow data in the bus station, and fit the dispatch optimization model constructed using PyTorch technology to obtain the dispatch evaluation index Dgzs; The dispatch subsystem is used to pre-set the evaluation threshold W, and compare and analyze it with the dispatch evaluation index Dgzs to intelligently select the current bus dispatch plan.

Images