Traffic planning method

Abstract

The invention discloses a traffic planning method, which realizes real-time monitoring, regulation and optimization of a traffic system by collecting, analyzing and processing real-time traffic data and combining an advanced traffic management algorithm. According to the invention, by using a data processing algorithm, the real-time data of the traffic system can be analyzed according to the timely acquired data, and the real-time monitoring and response to the traffic condition can be realized; through deep analysis and mining of big data, traffic demands and road condition changes can be accurately evaluated, and accurate decision support is provided for traffic planning and management. Flexible traffic management strategies and regulation and control means are adopted, and targeted traffic management and adjustment can be realized according to traffic characteristics of different time periods and regions; multiple factors such as traffic flow, road conditions, traffic facilities and the like are comprehensively considered, and the operation efficiency and safety of the traffic system are comprehensively optimized.

Description

Technical Field

[0001] The invention relates to the technical field of traffic planning, and in particular to a traffic planning method. Background Art

[0002] With the continuous advancement of technologies such as the Internet of Things and the continuous increase in data volume, cloud-based IoT solutions face the challenge of meeting growing demand. Therefore, more and more companies are turning their attention to edge computing and see it as an extension and expansion of cloud computing to increase the speed of data analysis and support companies to make decisions faster and more accurately. Although a few years ago, most people generally believed that deploying the Internet of Things to the cloud was the best choice, which brought a convenient user experience to individual users, building enterprise-level IoT solutions still requires a hybrid solution of cloud computing and edge computing. Compared with relying solely on cloud solutions, hybrid solutions that integrate edge computing can reduce latency, improve scalability, increase information access, and make business development more flexible. With the popularization of autonomous driving technologies such as assisted driving and unmanned driving, accurate identification of road conditions has become crucial. Therefore, it is of great significance to apply edge computing and cloud computing collaborative technologies to traffic sign recognition and control;

[0003] For this purpose, a variable information sign control method based on edge-cloud collaboration is proposed (patent number: CN111243306B patent name: A variable information sign control method based on edge-cloud collaboration);

[0004] However, in actual use, variable information signs need to be adaptively changed based on actual conditions. For example, at a traffic light at an intersection, if adaptive changes are not made based on the actual traffic conditions at the intersection, the traffic burden will be increased.

[0005] As urbanization continues to accelerate, traffic congestion and frequent traffic accidents have become important challenges in urban management. Traditional traffic planning methods often lack flexibility and real-time performance, and are unable to cope with the dynamic changes in traffic demand, resulting in low efficiency of traffic systems and widespread traffic congestion. Therefore, there is an urgent need for a new type of traffic planning method that can make full use of modern information technology to achieve intelligent management and optimization of urban traffic systems. Summary of the invention

[0006] The present invention aims to provide a traffic planning method to solve the deficiencies of traditional methods in terms of real-time performance and flexibility, thereby optimizing the operation of the traffic system and solving the problem of traffic congestion.

[0007] To achieve the above object, the present invention provides the following technical solution: a traffic planning method, the method comprising obtaining the current intersection saturation of the current intersection for monitoring, extracting features of the current intersection saturation, and obtaining the current intersection saturation features, including:

[0008] Mark the current intersection saturation to obtain the current intersection saturation marking information;

[0009] Performing data conversion on the current intersection saturation mark information to obtain current intersection saturation mark data;

[0010] Perform feature screening on the current intersection saturation mark data to obtain the current intersection saturation mark features;

[0011] The real-time vehicle saturation is calculated using the following formula and the current intersection is regulated:

[0012] Among them, the V ij (t) is the real-time traffic flow, which means the traffic flow from direction i to direction j at time t;

[0013] W ij (t) is the vehicle waiting time, which means the average waiting time of vehicles from direction i to direction j at time t; C ij The current intersection capacity refers to the capacity from direction i to direction j at time t; T ij (t) The duration of the traffic light indicates the duration of the traffic light from direction i to direction j at time t; α1, β1, and γ1 are the parameters of the model, which are obtained through data analysis and affect the weights of various factors.

[0014] Preferably, according to the saturation mark feature of the current intersection, the saturation status of the next intersection is queried, and the saturation feature of the next intersection is extracted, and the saturation feature of the next intersection is obtained, and the saturation feature of the next intersection is uploaded to the traffic planning platform, and the green wave section coordination plan of the next intersection is constructed through the traffic planning platform.

[0015] It includes converting the saturation information of the next intersection into data to obtain control data;

[0016] The control data is written into the green wave section collaborative control model to obtain a collaborative control file;

[0017] Upload the collaborative control file to the traffic planning platform and obtain the control instructions from the platform;

[0018] Adjust the traffic lights at the next intersection according to the control instructions;

[0019] The green wave section collaborative control model is: Among them, Vij (t) is the real-time traffic flow at the adjacent intersection; W ij (t) is the waiting time of vehicles at adjacent intersections; G ij (t) is the current green wave setting parameter; Δt ij is the optimal time difference, indicating the optimal time interval for cooperative control between adjacent intersections.

[0020] Preferably, the congestion level of the road section is evaluated according to the vehicle waiting time;

[0021] When the vehicle waiting time is less than 45 seconds, it is a smooth road section;

[0022] When the waiting time of vehicles is between 45 and 70 seconds, it is considered a slow-moving section;

[0023] When the waiting time of vehicles is between 70 and 120, it is a lightly congested section;

[0024] When the vehicle waiting time is greater than 120 seconds, it is considered a congested road section.

[0025] Preferably, based on the congestion level, the characteristics of the intersection are marked to obtain intersection congestion characteristics;

[0026] According to the congestion characteristics, query historical congestion data and current regional traffic flow distribution, generate warning information, and upload the warning information to the traffic planning platform, build a congestion evacuation plan and a low-flow road guidance strategy based on the traffic planning platform, and issue the congestion evacuation plan and the low-flow road guidance strategy to the traffic lights at each intersection and perform adjustments, including;

[0027] Convert the warning information into data to obtain warning data;

[0028] Writing warning data into a macroscopic dynamic traffic monitoring model to obtain a warning file;

[0029] Uploading the warning file to the traffic planning platform to obtain planning instructions;

[0030] Control the traffic lights at each intersection according to planning instructions;

[0031] The macroscopic dynamic flow monitoring model is: Among them, D ij (t) is the congestion level of the road section, which means the congestion level of the road section ij at time t; F ij (t) is the traffic flow distribution, which indicates the traffic flow distribution of road section ij at time t; H ij is the historical congestion data, indicating the historical congestion situation of road section ij; S ij (t) is the congestion evacuation plan and low-flow road guidance strategy, which represents the evacuation plan and guidance strategy for road section ij at time t.

[0032] Preferably, according to the number of times the congestion relief plan and the low-flow road guidance strategy are regulated at the same intersection in a natural month and historical data, a signal light timing plan for the same intersection in a future time period is predicted;

[0033] When the number of controls at the same intersection is less than 10 times in a natural month, the control is evaluated as sporadic control;

[0034] When the number of controls at the same intersection is between 10 and 30 times within a natural month, the control evaluation is routine control;

[0035] When the number of controls at the same intersection exceeds 30 times within a calendar month, the control evaluation will be key control.

[0036] Preferably, predictive traffic control is performed on the target intersection according to the key control, and the historical peak-time traffic flow data of the target intersection is collected. When the traffic flow during the natural day reaches the lower limit of the historical peak-time traffic flow threshold of the key control intersection, predictive traffic control is triggered;

[0037] The collected current real-time traffic flow, when the traffic flow reaches the lower limit of the historical peak traffic flow threshold of the key control intersection, generates key control information, uploads the key control information to the traffic planning platform, and constructs the signal light timing plan for the future period based on the traffic planning platform, including:

[0038] Carry out data conversion on key control information to obtain key control data;

[0039] Write the key control data into the predictive traffic control model to obtain the predictive traffic control file; upload the predictive traffic control file to the traffic planning platform to obtain the signal light timing plan;

[0040] The current signal light executes the signal light timing plan to conduct advance road section control;

[0041] The predictive traffic control model is:

[0042] Where V peak is the traffic flow data during the historical peak period, indicating the traffic flow during the historical peak period at time t;

[0043] T now is the current date and time;

[0044] T future For predicted future moments;

[0045] P signal (T future ) is the signal light timing scheme based on prediction, indicating that at the future time Tfuture , adjusted the traffic light timing plan.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention realizes the functions of real-time monitoring of intersection saturation, feature extraction and traffic flow control by setting up an intelligent traffic planning platform and a signal light for executing planning instructions, thereby effectively reducing intersection congestion and improving traffic fluency;

[0048] 2. Compared with the traditional traffic planning method, the present invention realizes the coordinated control between adjacent intersections through the green wave section coordinated control model, effectively improves the traffic efficiency, reduces the waiting time of vehicles, and optimizes the road traffic capacity;

[0049] 3. The present invention evaluates the degree of road congestion according to the waiting time of vehicles, and formulates corresponding congestion evacuation plans and low-flow road guidance strategies for different degrees of congestion, so as to deal with traffic congestion more accurately and improve road traffic efficiency;

[0050] 4. Through the predictive traffic control model, the present invention can predict the traffic conditions in future time periods and adjust the signal light timing plan based on historical traffic flow data and current real-time traffic flow conditions, making traffic control more flexible and efficient, and effectively alleviating traffic congestion problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a schematic diagram of the current intersection saturation monitoring and control process of the present invention;

[0052] Figure 2 This is a schematic diagram of the green wave section cooperative control model for the next intersection of the present invention;

[0053] Figure 3 This is a schematic diagram of the vehicle waiting time and road congestion evaluation of the present invention;

[0054] Figure 4 This is a schematic diagram of the predictive traffic control process of the present invention. DETAILED DESCRIPTION

[0055] 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.

[0056] See also Figures 1 to 4The present invention provides a technical solution: a traffic planning method, including a traffic planning platform and a signal light for executing planning instructions, characterized in that: the method includes obtaining the current intersection saturation of the current intersection for monitoring, and extracting features of the current intersection saturation, and obtaining the current intersection saturation features, including:

[0057] Mark the current intersection saturation to obtain the current intersection saturation marking information;

[0058] Performing data conversion on the current intersection saturation mark information to obtain current intersection saturation mark data;

[0059] Perform feature screening on the current intersection saturation mark data to obtain the current intersection saturation mark features;

[0060] The real-time vehicle saturation is calculated using the following formula and the current intersection is regulated:

[0061] Among them, the V ij (t) is the real-time traffic flow, which means the traffic flow from direction i to direction j at time t;

[0062] W ij (t) is the vehicle waiting time, which means the average waiting time of vehicles from direction i to direction j at time t;

[0063] C ij The current intersection capacity refers to the capacity from direction i to direction j at time t;

[0064] T ij (t) The signal light duration refers to the signal light duration from direction i to direction j at time t;

[0065] α1, β1, γ1 are the parameters of the model, obtained through data analysis, affecting the weight of each factor; square root is used is to mitigate the undue impact of higher latency, since the effect of increased latency on signal duration grows more slowly after a certain point;

[0066] Using the logarithm log(1+C ij ) is to deal with the nonlinear effects of capacity and ensure that even very high values ​​of capacity do not lead to an infinite increase in signal duration.

[0067] α1 (real-time traffic flow weight coefficient): It represents the real-time traffic flow V ij (t) for T ij(t) is directly affected. Traffic flow is usually data collected in real time by traffic monitoring systems through cameras, induction coils or other sensor devices at intersections. The coefficient α1 is obtained through historical data analysis, for example, by observing the relationship between traffic flow and traffic light operating efficiency in different time periods, and using statistical methods such as regression analysis to determine it.

[0068] β1 (influence of vehicle waiting time): coefficient β1 and vehicle waiting time W ij The square root of (t) is used to mitigate the impact of long waiting time. This parameter is obtained by analyzing past traffic congestion data. The specific method includes analyzing the traffic flow and traffic diversion effect under different waiting times.

[0069] γ1 (capacity adjustment coefficient): Coefficient γ1 and intersection capacity C ij This logarithmic form is used to handle nonlinear effects and ensure that the capacity does not cause the objective function T to exceed the value even at high values. ij (t) value increases infinitely. This parameter is determined by studying the relationship between the capacity and traffic efficiency of different intersections and using data analysis methods to determine its specific value;

[0070] In the actual data analysis process, it is necessary to collect a large amount of real-time and historical traffic data, including but not limited to vehicle flow, waiting time, traffic capacity, etc. These data can be processed through data mining, machine learning algorithms, or optimization algorithms to identify the best parameter values ​​so that the model can accurately reflect the traffic conditions and provide effective control strategies.

[0071] 1. Data collection method:

[0072] In order to obtain the data needed to monitor the current intersection saturation at the current intersection, a variety of advanced data collection technologies are used. First, a network of high-definition cameras and sensors are deployed near the intersection to monitor the vehicles passing through the intersection in real time. These devices can capture key information such as vehicle flow, vehicle waiting time, and traffic signal status. In addition, a modern traffic management system is used to obtain real-time intersection data through the traffic monitoring center and integrate data with other traffic monitoring equipment to ensure the comprehensiveness and accuracy of the data.

[0073] 2. Information marking and data conversion process:

[0074] Once the raw data is acquired, it will be subjected to information labeling, data conversion, and feature screening to extract key intersection saturation features. First, machine learning and computer vision techniques are used to label traffic information, and data such as vehicle flow and waiting time are classified and labeled for subsequent processing. Then, data processing algorithms are used to convert the labeled information into numerical data that can be used for calculations. Next, feature extraction algorithms are used to screen the converted data and select features related to intersection saturation, such as peak vehicle flow and average waiting time, for subsequent analysis and regulation.

[0075] 3. Feature selection and calculation formula:

[0076] In the feature selection stage, a method based on statistical analysis and machine learning was used to select important features related to intersection saturation from a large number of features. Multiple factors such as traffic volume, vehicle waiting time, and intersection capacity were taken into consideration, and optimized and adjusted in combination with actual conditions to ensure that the selected features can accurately reflect the actual traffic conditions at the intersection. Based on the selected features, a formula for calculating real-time vehicle saturation was designed to evaluate the current traffic conditions at the intersection and regulate them. This formula comprehensively considers factors such as traffic volume, waiting time, and capacity, providing an effective decision-making basis for the traffic planning platform.

[0077] 4. Real-time control mechanism:

[0078] Based on the calculated real-time vehicle saturation, a series of real-time control measures are taken to optimize the traffic flow at the current intersection. These measures include adjusting the duration of traffic lights, optimizing traffic flow guidance strategies, etc., aiming to effectively reduce traffic congestion and improve traffic efficiency. Through the traffic planning platform, the traffic lights at intersections can be remotely controlled, and traffic congestion can be responded to in a timely manner to ensure smooth and safe traffic at intersections.

[0079] According to the saturation mark feature of the current intersection, the saturation status of the next intersection is queried, and the saturation feature of the next intersection is extracted, and the saturation feature of the next intersection is obtained, and the saturation feature of the next intersection is uploaded to the traffic planning platform, and the green wave section coordination plan of the next intersection is constructed through the traffic planning platform.

[0080] It includes converting the saturation information of the next intersection into data to obtain control data;

[0081] The control data is written into the green wave section collaborative control model to obtain a collaborative control file;

[0082] Upload the collaborative control file to the traffic planning platform and obtain the control instructions from the platform;

[0083] Adjust the traffic lights at the next intersection according to the control instructions;

[0084] The green wave section collaborative control model is:

[0085]

[0086] Among them, V ij (t) is the real-time traffic flow at the adjacent intersection; W ij (t) is the waiting time of vehicles at adjacent intersections; G ij (t) is the current green wave setting parameter; Δt ij is the optimal time difference, indicating the optimal time interval for cooperative control between adjacent intersections; α2, b, c are the parameters of the model, obtained through field data and model fitting, which reflect the influence of different factors on the time difference;

[0087] Using Logical Functions This is to capture the nonlinear response of traffic flow to time difference and ensure that the increase of traffic flow will lead to a reasonable increase of time difference within a certain range;

[0088] Use log(1+W ij (t)) can handle the nonlinear effect of waiting time and avoid the excessive impact of very long waiting time on the time difference;

[0089] Direct linear term G ij (t) reflects the direct relationship between the green wave setting parameters and the time difference;

[0090] In traffic engineering, the green wave setting parameter G ij (t) usually involves multiple factors, including but not limited to signal timing plans, traffic flow data, historical congestion patterns, etc. The purpose of setting this parameter is to optimize a series of traffic lights so that vehicles can pass through multiple consecutive intersections during the green light period, thereby reducing parking waiting time and improving road traffic efficiency. Here are several common ways to define and obtain this parameter:

[0091] 1. Traffic signal control system

[0092] Modern traffic control systems automatically adjust green waves based on real-time or scheduled traffic flow data. These systems typically include:

[0093] Real-time monitoring equipment, such as cameras and induction loops, provides real-time traffic flow and vehicle waiting time data.

[0094] Signal control algorithms, such as adaptive signal control systems (e.g., SCATS, SCOOT, etc.), automatically adjust the duration and sequence of traffic lights based on real-time data;

[0095] Among them, α2 (adjustment coefficient of real-time traffic flow at adjacent intersections): This coefficient takes into account the real-time traffic flow V ij (t) has an impact on the optimal time difference. It is usually obtained through time series analysis of traffic flow data, such as fitting the relationship between flow and time difference using real-time monitoring data. The logistic function in the denominator It is used to limit the impact of traffic flow within a certain range to avoid excessive impact caused by extreme traffic flow values;

[0096] b (Logarithmic influence coefficient of vehicle waiting time): This parameter indicates the vehicle waiting time W ij The effect of the change of (t) on the time difference. The logarithmic form is used to alleviate the nonlinear growth effect caused by the long waiting time, so that the growth trend slows down under longer waiting time. It is obtained by long-term traffic condition monitoring and statistically analyzing the relationship between vehicle waiting time and the green wave time difference that needs to be adjusted;

[0097] c (the influence coefficient of the current green wave setting parameters): G ij (t) represents the parameter of the current green wave setting, which directly affects the optimal time difference. This parameter is based on past green wave setting experience and simulation studies, considering how the green wave setting can best adapt to the current and predicted traffic flow conditions.

[0098] 2. Traffic Engineering Simulation

[0099] Use traffic simulation software (such as VISSIM, Synchro, etc.) to simulate different traffic scenarios and signal settings to predict the optimal green wave settings. These simulations take into account various variables such as traffic flow forecasts, special events (such as sports events), etc.

[0100] 3. Experience and historical data

[0101] In the absence of complex systems or simulation capabilities, green wave settings can also be set manually based on historical data and the experience of traffic engineers. For example, different signal timing schemes can be set by analyzing the difference in traffic volume during peak and non-peak periods.

[0102] You can also customize the parameter G ij (t)=k·(1-parking rate), where k is an adjustment factor and the parking rate refers to the proportion of vehicles that need to park at the intersection, which is obtained through observation or simulation.

[0103] Next intersection saturation query and collaborative control solution:

[0104] In order to achieve coordinated control and optimization of traffic flow, the saturation status of the next intersection is queried according to the saturation mark feature of the current intersection, and feature extraction is performed to obtain the saturation feature of the next intersection. The specific steps are as follows:

[0105] Data query and feature extraction:

[0106] First, the traffic data of the next intersection is queried in real time through the traffic planning platform, including key information such as vehicle flow and vehicle waiting time. Then, data processing technology is used to extract features from these data and filter out features related to the saturation of the next intersection, such as peak vehicle flow and average waiting time, to provide a basis for subsequent collaborative control.

[0107] Construction of collaborative control scheme for green wave sections:

[0108] Based on the acquired saturation characteristics of the next intersection, a green wave section collaborative control scheme for the next intersection is constructed through the traffic planning platform. The specific steps are as follows:

[0109] Data conversion and control data acquisition: The saturation information of the next intersection is converted into data to obtain control data. These data include key indicators such as vehicle flow and vehicle waiting time, which are used for subsequent collaborative control model calculations.

[0110] Cooperative control model calculation: The control data is input into the green wave section cooperative control model to calculate the cooperative control scheme. The model takes into account the real-time traffic flow of adjacent intersections, vehicle waiting time and current green wave setting parameters to determine the optimal time difference, that is, the optimal time interval for cooperative control between adjacent intersections.

[0111]

[0112] Among them, V ij (t) is the real-time traffic flow at adjacent intersections;

[0113] W ij (t) is the waiting time of vehicles at adjacent intersections;

[0114] G ij (t) Set parameters for the current green wave;

[0115] Δt ij is the optimal time difference, indicating the optimal time interval for cooperative control between adjacent intersections;

[0116] Collaborative control file generation and upload: The calculated collaborative control scheme is converted into a collaborative control file and uploaded to the traffic planning platform to obtain the platform's control instructions.

[0117] Traffic light control execution: According to the control instructions issued by the traffic planning platform, the traffic lights at the next intersection are controlled to achieve coordinated optimization of traffic flow.

[0118] Through the above steps, coordinated control of traffic flow can be achieved, the traffic efficiency and transportation capacity of intersections can be improved, and strong support can be provided for urban traffic planning and management.

[0119] evaluating the congestion level of the road section according to the vehicle waiting time;

[0120] When the vehicle waiting time is less than 45 seconds, it is a smooth road section;

[0121] When the waiting time of vehicles is between 45 and 70 seconds, it is considered a slow-moving section;

[0122] When the waiting time of vehicles is between 70 and 120, it is a lightly congested section;

[0123] When the waiting time of vehicles is more than 120 seconds, it is a congested section;

[0124] Vehicle waiting time assessment and road congestion level classification:

[0125] This method evaluates and divides the congestion level of a road section according to the waiting time of vehicles on the road section, so as to carry out traffic planning and management more accurately. The specific steps are as follows:

[0126] Vehicle waiting time assessment:

[0127] The waiting time data of vehicles on the road section is obtained in real time through monitoring equipment or traffic cameras, and then the congestion level is divided according to the different ranges of vehicle waiting time. The specific division is as follows:

[0128] Smooth road section (Wait < 45 seconds): When the vehicle waiting time is less than 45 seconds, the road section is considered to be smooth and vehicles can pass smoothly.

[0129] Slow-moving road section (45 seconds ≤ Wait < 70 seconds): When the vehicle waiting time is between 45 and 70 seconds, it means that there is a certain degree of congestion on the road section, and the vehicle speed is slower, but it can still pass smoothly.

[0130] Mildly congested road section (70 seconds ≤ Wait < 120 seconds): When the vehicle waiting time is between 70 and 120 seconds, it indicates that the road section has been slightly congested and the vehicle speed has slowed down significantly, but it can still pass normally.

[0131] Congested road section (Wait ≥ 120 seconds): When the waiting time of a vehicle exceeds 120 seconds, it means that the road section is seriously congested, the vehicle speed is very slow, and the traffic flow is seriously affected.

[0132] Road congestion level classification:

[0133] Based on the evaluation results of vehicle waiting time, the congestion degree of road sections is divided to provide a reference for traffic planning and management. The specific divisions are as follows:

[0134] Excellent road sections: including smooth sections and slow-moving sections, where vehicles can pass smoothly, congestion levels are low, and traffic efficiency is high.

[0135] General road sections: including lightly congested sections, where vehicle speeds are reduced but traffic flows are still relatively smooth.

[0136] Congested road sections: Road sections where the vehicle waiting time exceeds 120 seconds are considered congested sections. Traffic flow is seriously affected and corresponding measures need to be taken to guide and manage the traffic.

[0137] By evaluating vehicle waiting times and dividing the degree of congestion on road sections, we can more accurately understand the traffic conditions on the road sections, formulate targeted traffic management strategies, improve transportation efficiency, and optimize the urban transportation network.

[0138] Based on the congestion degree, marking the characteristics of the intersection to obtain intersection congestion characteristics;

[0139] According to the congestion characteristics, query historical congestion data and current regional traffic flow distribution, generate warning information, and upload the warning information to the traffic planning platform, build a congestion evacuation plan and a low-flow road guidance strategy based on the traffic planning platform, and issue the congestion evacuation plan and the low-flow road guidance strategy to the traffic lights at each intersection and perform adjustments, including;

[0140] Convert the warning information into data to obtain warning data;

[0141] Writing warning data into a macroscopic dynamic traffic monitoring model to obtain a warning file;

[0142] Uploading the warning file to the traffic planning platform to obtain planning instructions;

[0143] Control the traffic lights at each intersection according to planning instructions;

[0144] The macroscopic dynamic flow monitoring model is:

[0145] Among them, D ij (t) is the congestion level of the road section, which means the congestion level of the road section ij at time t;

[0146] F ij (t) is the traffic flow distribution, which indicates the traffic flow distribution of road section ij at time t;

[0147] Hij is the historical congestion data, indicating the historical congestion situation of road section ij;

[0148] S ij (t) is the congestion evacuation plan and low-flow road guidance strategy, which represents the evacuation plan and guidance strategy for road section ij at time t;

[0149] λ,μ,ν are the coefficients of the model, which need to be determined through data analysis. These coefficients reflect the contribution of different variables to the formation of the strategy;

[0150] Using log(1+D ij (t)) is to adjust the impact of congestion level to prevent excessive impact on the strategy during extreme congestion;

[0151] F ij (t) The linear term directly indicates the impact of traffic flow distribution on the strategy;

[0152] Exponential decay It can indicate the impact of historical congestion data on the current strategy and make adjustments based on historical data, where the more severe the historical congestion, the lower the impact weight;

[0153] Practical application and parameter setting,

[0154] Intersection congestion feature marking and traffic planning:

[0155] This method marks the intersection features according to the degree of congestion at the intersection, and generates warning information and performs traffic planning based on historical congestion data and regional traffic flow distribution to optimize traffic flow and improve intersection efficiency. The specific steps are as follows:

[0156] Intersection congestion feature markers: λ,μ,ν are set based on field data through regression analysis, machine learning or other statistical methods.

[0157] The congestion level of the intersection is evaluated and the evaluation results are marked as the characteristics of the intersection. The congestion level can be marked in multiple levels, such as light congestion, moderate congestion, severe congestion, etc., to facilitate subsequent traffic planning and management.

[0158] Generates the warning message:

[0159] The congestion feature marks of the intersection are combined with the historical congestion data and the current regional traffic flow distribution to generate corresponding warning information. The warning information may include the current congestion level of the intersection, the expected congestion duration, and the recommended traffic diversion strategy.

[0160] Traffic planning and evacuation plan:

[0161] Based on the generated warning information, the traffic planning platform can formulate corresponding traffic evacuation plans and low-flow road guidance strategies. The evacuation plan can include measures such as adjusting the timing of intersection lights and optimizing traffic routes to alleviate congestion and improve intersection traffic efficiency.

[0162] Implementation of control: The traffic planning scheme is issued to the signal light control system of each intersection to implement corresponding control measures. Control includes adjusting the timing of signal lights according to warning information, guiding traffic flow to low-volume roads, etc., so as to achieve timely guidance and management of traffic congestion.

[0163] Macro-dynamic traffic monitoring model: The macro-dynamic traffic monitoring model is used to monitor and analyze traffic flow distribution and road congestion in real time. The model combines multiple factors such as road congestion, traffic flow distribution and historical congestion data to provide scientific basis and data support for traffic planning.

[0164] According to the number of control times of the congestion relief plan and the low-flow road guidance strategy at the same intersection in a natural month and historical data, predict the signal light timing plan for the same intersection in the future time period;

[0165] When the number of controls at the same intersection is less than 10 times in a natural month, the control is evaluated as sporadic control;

[0166] When the number of controls at the same intersection is between 10 and 30 times within a natural month, the control evaluation is routine control;

[0167] When the number of controls at the same intersection exceeds 30 times within a calendar month, the control evaluation will be key control.

[0168] Intersection control evaluation and signal timing prediction:

[0169] This method predicts the traffic light timing plan for the same intersection in the future time period based on the number of control times of the congestion evacuation plan and low-flow road guidance strategy at the same intersection in a natural month and historical data, so as to achieve more accurate and effective traffic control. The specific steps are as follows:

[0170] Intersection control evaluation: The control situation of the intersection is evaluated based on the number of controls at the same intersection within a natural month. Intersections with less than 10 controls are evaluated as occasional controls, intersections with 10-30 controls are evaluated as routine controls, and intersections with more than 30 controls are evaluated as key controls. These evaluations will be used for subsequent signal light timing predictions.

[0171] Historical data analysis: Analyze historical control data, including the number of controls at different intersections, control methods, control effects, etc. By analyzing historical data, we can understand the traffic conditions and control needs of each intersection, and provide a reference for future signal timing predictions.

[0172] Traffic light timing prediction: Based on the evaluation of intersection control and historical data analysis results, data mining and machine learning techniques are used to predict the traffic light timing plan for the same intersection in the future time period. The prediction model will consider multiple factors such as intersection control evaluation, historical control data, current traffic conditions, etc. to provide a more accurate and reliable traffic light timing plan.

[0173] Implement control plan: According to the signal light timing prediction results, formulate corresponding traffic control plan and issue it to the signal light control system of each intersection. The control plan includes measures such as signal light timing adjustment and traffic route optimization to adapt to future traffic needs and improve traffic efficiency at intersections.

[0174] Real-time monitoring and adjustment: Real-time monitoring and adjustment of the predicted signal light timing plan, and timely feedback and adjustment based on actual traffic conditions. Through real-time monitoring and adjustment, the effectiveness and flexibility of the traffic control plan are guaranteed to cope with sudden traffic incidents and changes in road conditions.

[0175] According to the key control, predictive traffic control is performed on the target intersection, and the historical peak-hour traffic flow data of the target intersection is collected. When the traffic flow during the natural day reaches the lower limit of the historical peak-hour traffic flow threshold of the key control intersection, the predictive traffic control is triggered;

[0176] The collected current real-time traffic flow, when the traffic flow reaches the lower limit of the historical peak traffic flow threshold of the key control intersection, generates key control information, uploads the key control information to the traffic planning platform, and constructs the signal light timing plan for the future period based on the traffic planning platform, including:

[0177] Carry out data conversion on key control information to obtain key control data;

[0178] Write the key control data into the predictive traffic control model to obtain the predictive traffic control file; upload the predictive traffic control file to the traffic planning platform to obtain the signal light timing plan;

[0179] The current signal light executes the signal light timing plan to conduct advance road section control;

[0180] The predictive traffic control model is:

[0181] Where V peak It is the traffic flow data during the historical peak period, indicating the traffic flow during the historical peak period at time t.

[0182] T now Is the current date and time.

[0183] Tfuture To predict the future moment.

[0184] P signal (T future ) is the signal light timing scheme based on prediction, indicating that at the future time T future , adjusted the traffic light timing plan.

[0185] α3 and β2 are coefficients of the model and need to be determined through actual data. In practical applications, parameters α3 and β2 are set through historical data and statistical analysis methods (such as regression analysis).

[0186] α3 represents the maximum impact of traffic flow on signal timing when there are no other changes.

[0187] β2 is a decay factor that represents the effect of time distance, that is, the influence of historical data gradually weakens over time.

[0188] Implementation of predictive traffic control scheme:

[0189] According to the key control, predictive traffic control is carried out at the target intersection to cope with the changes in future traffic flow and fluctuations in road conditions, and effective measures are taken in advance to manage and control traffic. The specific steps are as follows:

[0190] Historical peak-hour traffic flow data collection:

[0191] Collect historical peak-hour traffic flow data for target intersections, including traffic flow conditions in different time periods, peak times, etc. By analyzing historical data, determine the lower limit of the historical peak-hour traffic flow threshold for key control intersections.

[0192] Triggering predictive traffic control:

[0193] Monitor the current real-time traffic flow. When the traffic flow reaches the lower limit of the historical peak traffic flow threshold at the key control intersection, predictive traffic control is triggered. The determination of this threshold is to warn of traffic congestion in advance and provide sufficient preparation time for traffic management departments.

[0194] Focus on regulating information uploading and data conversion:

[0195] Generate key control information and upload it to the traffic planning platform. Then, perform data conversion on the key control information to facilitate the establishment and analysis of the subsequent predictive traffic control model.

[0196] Establishment of predictive traffic control model:

[0197] Using historical peak-hour traffic flow data and current real-time traffic flow data, a predictive traffic control model is established. The model will take into account historical data, current traffic flow conditions and other influencing factors, predict the traffic flow conditions at the target intersection in the future period, and provide corresponding signal light timing solutions.

[0198] Implementation of the signal light timing scheme:

[0199] The signal light timing scheme generated by the predictive traffic control model is uploaded to the traffic planning platform and executed by the current signal light. According to the predicted traffic flow, the signal light timing is adjusted in advance and the road section is controlled in advance to ensure smooth traffic and efficient traffic at the intersection.

[0200] Example 1: Urban Traffic Congestion Optimization

[0201] Background:

[0202] Traffic congestion has always been a serious problem in modern urban traffic management. To address this challenge, a traffic planning method based on real-time data is proposed, which aims to optimize urban traffic flow and reduce congestion.

[0203] Technical purpose:

[0204] This embodiment aims to utilize real-time traffic data and intelligent control algorithms to achieve dynamic control of urban traffic flow, thereby improving traffic efficiency and reducing the occurrence of congestion.

[0205] Innovation points and application scenarios:

[0206] Use advanced sensor technology to monitor traffic conditions on various roads in the city in real time, and achieve accurate data collection and analysis.

[0207] Based on big data and artificial intelligence algorithms, traffic light timing plans are dynamically optimized to quickly resolve traffic congestion.

[0208] Application effects and advantages:

[0209] By implementing the traffic planning method of the present invention, a field experiment was conducted in City A to compare the traditional fixed-period timing scheme and the dynamic control scheme. The results show that the method can significantly reduce the average waiting time of vehicles and reduce the traffic congestion index.

[0210] index Traditional methods Method of the present invention Average waiting time 120 seconds 60 seconds Congestion Index 0.8 0.4

[0211] Example 2: Smart City Traffic Management

[0212] Background:

[0213] With the continuous acceleration of urbanization, urban traffic management has become increasingly complex. In order to improve the intelligence level of urban traffic, a smart city traffic management solution is proposed.

[0214] Technical purpose:

[0215] This embodiment aims to use Internet of Things technology and data analysis methods to build a smart city traffic management system to achieve real-time monitoring and regulation of urban traffic flow.

[0216] Innovation points and application scenarios:

[0217] Based on IoT sensors and Internet technology, remote monitoring and data collection of various traffic nodes in the city can be realized.

[0218] Through data analysis and machine learning algorithms, traffic flow can be predicted and dynamically controlled, thereby improving the operating efficiency and safety of the transportation system.

[0219] Application effects and advantages:

[0220] An implementation experiment was conducted in City B to compare the smart traffic management system with the traditional traffic monitoring system. The results showed that the system was able to detect traffic anomalies in a timely manner and take effective measures to adjust them, greatly improving the efficiency and reliability of traffic operations.

[0221] index Traditional systems Smart system Fault detection time 30 minutes 5 minutes Mean time to recovery 60 minutes 15 minutes

[0222] Example 3: Collaborative Optimization of Cross-City Traffic

[0223] Background:

[0224] Inter-city traffic coordination is the key to improving the overall efficiency of regional traffic. In order to achieve inter-city traffic coordination optimization, a traffic planning method based on joint regulation is proposed.

[0225] Technical purpose:

[0226] This embodiment aims to achieve optimal allocation and coordinated regulation of traffic flows between different cities through the sharing of cross-city traffic data and the application of a joint regulation algorithm.

[0227] Innovation points and application scenarios:

[0228] Utilize the regional traffic big data platform to integrate and share traffic data from different cities.

[0229] Through the joint control algorithm, the cross-city traffic light timing schemes can be optimized and coordinated to reduce cross-city traffic congestion and delays.

[0230] Application effects and advantages:

[0231] A joint regulation experiment was conducted in City C and City D. Compared with the individual regulation schemes, the joint regulation scheme significantly reduced the degree of cross-city traffic congestion and improved traffic efficiency.

[0232] index Individual control scheme Joint regulatory program Average delay time 90 seconds 45 seconds Cross-city congestion index 0.6 0.3

[0233] 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 traffic planning method, comprising a traffic planning platform and a traffic light for executing planning instructions, characterized in that: The method includes acquiring the saturation of the current intersection of the current intersection, performing feature extraction on the saturation of the current intersection, and obtaining the saturation feature of the current intersection, including: Mark the current intersection saturation to obtain the current intersection saturation marking information; Performing data conversion on the current intersection saturation mark information to obtain current intersection saturation mark data; Perform feature screening on the current intersection saturation mark data to obtain the current intersection saturation mark features; The real-time vehicle saturation is calculated using the following formula and the current intersection is regulated: Among them, the V ij (t) is the real-time traffic flow, which means the traffic flow from direction i to direction j at time t; W ij (t) is the vehicle waiting time, which means the average waiting time of vehicles from direction i to direction j at time t; C ij The current intersection capacity refers to the capacity from direction i to direction j at time t; T ij (t) The duration of the signal light refers to the duration of the signal light from direction i to direction j at time t; α1, β1, and γ1 are parameters of the model, obtained through data analysis, and affect the weights of various factors.

2. A traffic planning method according to claim 1, characterized in that: According to the saturation mark feature of the current intersection, the saturation status of the next intersection is queried, and the saturation feature of the next intersection is extracted, and the saturation feature of the next intersection is obtained, and the saturation feature of the next intersection is uploaded to the traffic planning platform, and the green wave section coordination plan of the next intersection is constructed through the traffic planning platform. It includes converting the saturation information of the next intersection into data to obtain control data; The control data is written into the green wave section collaborative control model to obtain a collaborative control file; Upload the collaborative control file to the traffic planning platform and obtain the control instructions from the platform; Adjust the traffic lights at the next intersection according to the control instructions; The green wave section collaborative control model is: Among them, V ij (t) is the real-time traffic flow at adjacent intersections; W ij (t) is the waiting time of vehicles at adjacent intersections; G ij (t) Set parameters for the current green wave; Δt ij is the optimal time difference, indicating the optimal time interval for cooperative control between adjacent intersections.

3. A traffic planning method according to claim 2, characterized in that: evaluating the congestion level of the road section according to the vehicle waiting time; When the vehicle waiting time is less than 45 seconds, it is a smooth road section; When the waiting time of vehicles is between 45 and 70 seconds, it is considered a slow-moving section; When the waiting time of vehicles is between 70 and 120, it is a lightly congested section; When the vehicle waiting time is greater than 120 seconds, it is considered a congested road section.

4. A traffic planning method according to claim 3, characterized in that: Based on the congestion degree, marking the characteristics of the intersection to obtain intersection congestion characteristics; According to the congestion characteristics, query historical congestion data and current regional traffic flow distribution, generate warning information, and upload the warning information to the traffic planning platform, build a congestion evacuation plan and a low-flow road guidance strategy based on the traffic planning platform, and issue the congestion evacuation plan and the low-flow road guidance strategy to the traffic lights at each intersection and perform adjustments, including; Convert the warning information into data to obtain warning data; Writing warning data into a macroscopic dynamic traffic monitoring model to obtain a warning file; Uploading the warning file to the traffic planning platform to obtain planning instructions; Control the traffic lights at each intersection according to planning instructions; The macroscopic dynamic flow monitoring model is: Among them, D ij (t) is the congestion level of the road section, which means the congestion level of the road section ij at time t; F ij (t) is the traffic flow distribution, which indicates the traffic flow distribution of road section ij at time t; H ij is the historical congestion data, indicating the historical congestion situation of road section ij; S ij (t) is the congestion evacuation plan and low-flow road guidance strategy, which represents the evacuation plan and guidance strategy for road section ij at time t.

5. A traffic planning method according to claim 4, characterized in that: According to the number of control times of the congestion relief plan and the low-flow road guidance strategy at the same intersection in a natural month and historical data, predict the signal light timing plan for the same intersection in the future time period; When the number of controls at the same intersection is less than 10 times in a natural month, the control is evaluated as sporadic control; When the number of controls at the same intersection is between 10 and 30 times within a natural month, the control evaluation is routine control; When the number of controls at the same intersection exceeds 30 times within a calendar month, the control evaluation will be key control.

6. A traffic planning method according to claim 5, characterized in that: According to the key control, predictive traffic control is performed on the target intersection, and the historical peak-hour traffic flow data of the target intersection is collected. When the traffic flow during the natural day reaches the lower limit of the historical peak-hour traffic flow threshold of the key control intersection, the predictive traffic control is triggered; The collected current real-time traffic flow, when the traffic flow reaches the lower limit of the historical peak traffic flow threshold of the key control intersection, generates key control information, uploads the key control information to the traffic planning platform, and constructs the signal light timing plan for the future period based on the traffic planning platform, including: Carry out data conversion on key control information to obtain key control data; Write the key control data into the predictive traffic control model to obtain the predictive traffic control file; upload the predictive traffic control file to the traffic planning platform to obtain the signal light timing plan; The current signal light executes the signal light timing plan to conduct advance road section control; The predictive traffic control model is: Where V peak is the traffic flow data during the historical peak period, indicating the traffic flow during the historical peak period at time t; T now is the current date and time; T future For predicted future moments; P signal (T future ) is the signal light timing scheme based on prediction, indicating that at the future time T future , adjusted the traffic light timing plan.

Images