Intelligent capacity expansion method and system for expressway operation period

Abstract

The invention relates to the technical field of smart traffic, in particular to a smart capacity expansion method and system for an expressway operation period, and the method comprises the steps: collecting real-time traffic flow and speed data, and obtaining traffic jam information through calculating the weighted average flow and speed; analyzing congestion data between adjacent radars according to the traffic congestion information, and combining continuous congestion points to obtain a congestion section; according to the congestion section, through a demand prediction algorithm and a traffic capacity prediction algorithm, calculating a current VC ratio and a future VC ratio; and according to a calculation result, carrying out prediction analysis on a congestion trend, and according to a prediction analysis result, adjusting traffic control measures in real time. According to the invention, the problems of response lag and insufficient data accuracy when a traditional traffic management system deals with real-time traffic changes of the expressway are effectively solved, accurate prediction and dynamic management of traffic congestion are realized, and the traffic congestion problem of the expressway is effectively relieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to an intelligent capacity expansion method and system for expressways during operation. Background Art

[0002] With the continuous growth of the economy and the improvement of residents' income level, the number of cars has continued to increase, resulting in a sharp increase in highway traffic. Especially in certain time periods and sections, the actual traffic has gradually exceeded the designed traffic, making highway congestion gradually change from an accidental phenomenon to the norm.

[0003] Traditional traffic management systems usually have problems such as delayed response, insufficient data accuracy, and inability to dynamically adjust control measures in real time when dealing with real-time traffic changes on highways. Existing technologies mainly rely on fixed sensors or historical data for analysis, and lack accurate capture and analysis of real-time traffic flow and speed, resulting in delays in congestion detection and lags in control measures.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a smart capacity expansion method and system for highway operation, thereby effectively solving the problems pointed out in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A smart capacity expansion method for expressway operation period, comprising:

[0008] Collect real-time traffic flow and speed data, and obtain traffic congestion information by calculating weighted average flow and speed;

[0009] Analyzing the congestion data between adjacent radars according to the traffic congestion information, and merging continuous congestion points to obtain a congestion section;

[0010] According to the congested section, the current VC ratio and the future VC ratio are calculated by using a demand prediction algorithm and a traffic capacity prediction algorithm, wherein the VC ratio is a ratio of demand flow to traffic capacity;

[0011] Based on the calculation results, the congestion trend is predicted and analyzed, and traffic control measures are adjusted in real time based on the prediction and analysis results.

[0012] Furthermore, the weighted average flow and speed are calculated as follows:

[0013]

[0014] Among them, V avg is the average speed per minute for 15 minutes, V i is the speed per minute, N s，i is the number of cars, recorded every i minutes, N m，i is the number of trains, recorded every i minutes, N l，i is the number of large vehicles, recorded every i minutes, F is the total flow, and is the historical mean of speed and flow, Z V and Z F is the deviation of speed and flow from historical data, D is the weighted average index of congestion detection, and α and β are weight factors.

[0015] Furthermore, the demand forecasting algorithm adopts a loop memory network, including:

[0016] i t =σ(W xi x t +W hi h t-1 +b i )

[0017] f t =σ(W xf x t +W hf h t-1 +b f )

[0018] o t =σ(W xo x t +W ho h t-1 +b o )

[0019] g t =tanh(W xc x t +W hc h t-1 +b c )

[0020] c t =f t ⊙c t-1 +i t ⊙g t

[0021] h t =o t ⊙tanh(c t )

[0022] Among them, i t , f t , o t are the activation values ​​of the input gate, forget gate, and output gate respectively; g t is the unit update vector; c t is the unit state; h t is the output hidden state.

[0023] Furthermore, the demand forecasting model constructed by the demand forecasting algorithm includes:

[0024] The input layer fuses and cleans real-time traffic flow and speed data from radar devices along the highway;

[0025] Hidden layer, uses long short-term memory network to process time series data, capture the time dependency and long-term trend of traffic flow and speed, and generate hidden states;

[0026] The output layer generates the future traffic demand forecast value according to the hidden state of the long short-term memory network, and adjusts the model parameters through loss calculation and Adam optimization algorithm;

[0027] The prediction and warning layer divides the traffic congestion level according to the traffic demand prediction value and outputs warning information.

[0028] Furthermore, the traffic capacity prediction algorithm adopts an integrated decision tree model, including:

[0029] Collect data from radars and sensors along the highway and train multiple decision tree models, each of which learns based on different features and data subsets;

[0030] Using historical data sets to train each decision tree model, and adjusting the parameter model through cross-validation to optimize the performance of each decision tree model;

[0031] New sample prediction is performed on each of the trained decision tree models, and the prediction results of each of the decision tree models are fused using a weighted average method.

[0032] Furthermore, according to the traffic congestion information, the congestion data between adjacent radars are analyzed, and continuous congestion points are merged to obtain a congestion section, including:

[0033] Obtain and record the geographic coordinate information of radar devices along the highway and determine the spatial position relationship between adjacent radar devices;

[0034] Based on the spatial position relationship, calculating the actual distance between the adjacent radar devices;

[0035] According to the traffic congestion information and the actual distance, the congestion states of the adjacent radar devices are compared, and when the congestion states of the two adjacent radar devices continue and the distances are within a preset range, it is determined that there is spatial continuity;

[0036] The congestion points of the adjacent radar devices that are determined to have spatially continuous rows are merged into continuous congestion points to obtain a congestion section.

[0037] Furthermore, traffic control measures are adjusted in real time, including:

[0038] Dynamically adjust the use strategy of emergency lanes and main lanes, and optimize the opening and closing status of ramps;

[0039] Implement gradient speed limit measures and dynamically adjust vehicle speeds based on road conditions and real-time traffic data;

[0040] Through the road information board, we can provide improved traffic information and driving suggestions, and guide drivers to change their driving routes.

[0041] Furthermore, the method also includes: combining traffic data of special traffic events, analyzing the impact of the special traffic events on traffic congestion, and designing targeted solutions.

[0042] A smart capacity expansion system for expressway operation period, the system comprising:

[0043] The congestion information acquisition module collects real-time traffic flow and speed data, and obtains traffic congestion information by calculating the weighted average flow and speed;

[0044] A congestion section acquisition module, which analyzes the congestion data between adjacent radars according to the traffic congestion information, and merges continuous congestion points to obtain a congestion section;

[0045] A VC ratio calculation module, which calculates the current VC ratio and the future VC ratio according to the congested section by using a demand prediction algorithm and a traffic capacity prediction algorithm, wherein the VC ratio is a ratio of demand flow to traffic capacity;

[0046] The control measures adjustment module predicts and analyzes the congestion trend based on the calculation results, and adjusts the traffic control measures in real time based on the prediction and analysis results.

[0047] Furthermore, the congested section acquisition module includes:

[0048] A spatial relationship determination unit, which obtains and records the geographic coordinate information of radar devices along the expressway and determines the spatial position relationship of adjacent radar devices;

[0049] an actual distance calculation unit, which calculates the actual distance between the adjacent radar devices based on the spatial position relationship;

[0050] a spatial continuity judgment unit, which compares the congestion status of the adjacent radar devices according to the traffic congestion information and the actual distance, and judges that there is spatial continuity when the congestion status of two adjacent radar devices continues and the distance is within a preset range;

[0051] The congestion section acquisition unit merges the congestion points of the adjacent radar devices that are determined to have spatially continuous rows into continuous congestion points to obtain a congestion section.

[0052] The technical solution of the present invention can achieve the following technical effects:

[0053] The present invention effectively solves the problems of delayed response and insufficient data accuracy of traditional traffic management systems when dealing with real-time traffic changes on highways, realizes accurate prediction and dynamic management of traffic congestion, and effectively alleviates the problem of highway traffic congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 It is a flowchart of a smart capacity expansion method for expressway operation period;

[0056] Figure 2 This is a flowchart of the integrated decision tree model;

[0057] Figure 3 To obtain the flow chart of the congested section. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0060] Embodiment 1

[0061] like Figure 1 As shown, the present invention provides a smart capacity expansion method for expressway operation period, the method comprising:

[0062] S1: Collect real-time traffic flow and speed data, and obtain traffic congestion information by calculating weighted average flow and speed;

[0063] Specifically, first, multiple radar devices and sensors deployed along the highway are used to collect traffic flow and speed data on each lane in real time. These data include the number of vehicles monitored by the radar every minute, the type of vehicle (such as small cars, medium cars, and large cars), and the driving speed; the collected real-time traffic data will be initially cleaned to remove outliers and noise to ensure the accuracy and reliability of the data, and the cleaned data will be used for subsequent calculations and analysis; for the cleaned data, the system will calculate the weighted average of traffic flow and speed. Through these calculations, more accurate traffic flow and speed information can be obtained, thereby better reflecting the current traffic conditions. Based on the calculated weighted average flow and speed, the changing trend of traffic flow and speed can be further analyzed.

[0064] S2: According to the traffic congestion information, the congestion data between adjacent radars are analyzed and continuous congestion points are merged to obtain the congestion section;

[0065] Specifically, by analyzing the congestion data of adjacent radar devices, it is determined whether there is spatial continuity, and continuous congestion points are merged into congestion sections. In this way, larger continuous congestion areas on the highway can be accurately identified to avoid mistaking isolated congestion points for independent congestion events. The merged congestion sections can more comprehensively reflect the actual traffic conditions, help improve the accuracy of congestion trend predictions, and provide a reliable basis for real-time adjustment of traffic control measures, thereby more effectively alleviating traffic congestion and optimizing the overall traffic efficiency of the highway.

[0066] S3: According to the congested section, the current VC ratio and the future VC ratio are calculated by the demand prediction algorithm and the capacity prediction algorithm. The VC ratio is the ratio of the demand flow to the capacity.

[0067] Specifically, first, based on the identified congested sections, current and historical traffic flow data are collected. These data include vehicle type, flow, speed, and other relevant traffic parameters in order to fully understand the traffic conditions. Next, the demand forecasting algorithm is applied to analyze the current and historical traffic flow data, identify traffic patterns and trends, and predict the demand flow in a certain period of time in the future. The demand forecasting algorithm can consider a variety of influencing factors, such as seasonal changes and special events (such as holidays, accidents, etc.), thereby improving the accuracy of the forecast. At the same time, the capacity forecasting algorithm is applied to calculate the current capacity based on the current road conditions and real-time traffic data. The capacity forecasting algorithm comprehensively considers the number of lanes, road width, traffic control measures (such as signal The maximum capacity of the road under specific conditions is estimated by taking into account factors such as traffic lights, speed limit measures, and weather conditions. Then, by comparing the current demand flow with the capacity, the current VC ratio (V / C), that is, the ratio of demand flow to capacity, is calculated. The calculation of the VC ratio can reflect the matching situation of current traffic demand and road capacity. A high VC ratio indicates that the demand flow is close to or exceeds the road capacity, indicating potential congestion risks. The predicted future demand flow and future capacity are used to calculate the VC ratio for future time periods. In this step, the future traffic conditions are estimated by combining the demand forecasting algorithm with the capacity forecasting algorithm. By comparing the current and future VC ratios, the congestion trend is analyzed to determine whether the traffic conditions will improve or deteriorate.

[0068] S4: Based on the calculation results, predict and analyze the congestion trend, and adjust traffic control measures in real time based on the prediction and analysis results.

[0069] The present invention effectively solves the problems of delayed response and insufficient data accuracy of traditional traffic management systems when dealing with real-time traffic changes on highways, realizes accurate prediction and dynamic management of traffic congestion, and effectively alleviates the problem of highway traffic congestion.

[0070] As a preferred embodiment of the above, the weighted average flow and speed are calculated using the formula:

[0071]

[0072] Among them, V avg is the average speed per minute for 15 minutes, V i is the speed per minute, N s，i is the number of cars, recorded every i minutes, N m，i is the number of trains, recorded every i minutes, N l，i is the number of large vehicles, recorded every i minutes, F is the total flow, and is the historical mean of speed and flow, Z V and ZF is the deviation of speed and flow from historical data, D is the weighted average index of congestion detection, and α and β are weight factors.

[0073] Specifically, by calculating the average speed per minute within 15 minutes, the speed fluctuation in a short period of time can be smoothed out to obtain a more stable and representative speed index, which helps to more accurately reflect the overall driving speed of the vehicle and avoid misjudgment caused by instantaneous changes; by accumulating the number of all vehicles within 15 minutes, the total flow is obtained. The purpose of this is to comprehensively reflect the traffic flow of various types of vehicles and provide a complete data basis for further traffic analysis; by calculating the deviation of the current speed and flow relative to the historical data, the abnormal situation of the current traffic condition can be effectively detected. The deviation value Z V and Z F It reflects the degree of difference between real-time data and historical averages. Standardization makes data of different dimensions comparable. By taking weighted average of speed deviation and flow deviation, the impact of these two factors on traffic congestion can be comprehensively considered. The weight factors α and β can be adjusted according to actual needs to adapt to different traffic management goals. For example, in some cases, flow may be more important than speed, and vice versa. The weighted average indicator D makes congestion detection more flexible and accurate.

[0074] As a preferred embodiment of the above, the demand forecasting algorithm adopts a loop memory network, including:

[0075] i t =σ(W xi x t +W hi h t-1 +b i )

[0076] f t =σ(W xf x t +W hf h t-1 +b f )

[0077] o t =σ(W xo x t +W ho h t-1 +b o )

[0078] g t =tanh(W xc x t +W hc h t-1 +b c )

[0079] c t =f t ⊙c t-1 +i t ⊙g t

[0080] h t =o t ⊙tanh(c t )

[0081] Among them, i t , f t , o t are the activation values ​​of the input gate, forget gate, and output gate respectively; g t is the unit update vector; c t is the unit state; h t is the output hidden state.

[0082] Specifically, the recurrent neural network (RNN) is a deep learning model that processes sequence data. In order to effectively predict traffic demand, the present invention uses a long short-term memory network (LSTM), a special RNN, to capture the time dependency and long-term trend in traffic flow data. LSTM solves the long-term dependency problem in traditional RNN through its unique gate mechanism (input gate, forget gate and output gate). The input gate determines the current input x t And the hidden state h at the previous moment t-1 The impact on the current cell state, through the activation value i of the input gate t , the LSTM unit selectively updates the current input information to ensure that important input information is passed to the unit state; the forget gate determines the unit state c at the previous moment t-1 How many are retained to the current moment, the activation value f of the forget gate t Controlling the forgetting ratio of information helps the network discard unnecessary old information and retain important historical information, thereby optimizing the accuracy of prediction; the output gate controls the current unit state c t How much influence does it have on the current hidden state h t , the activation value o of the output gate t The output gate determines the amount of information output, thus affecting the prediction of the next moment. Through the output gate, LSTM can selectively transmit information at the output moment, making the model more flexible and accurate. The unit update vector is updated by the current input x t and the previous hidden state h t-1 Perform nonlinear transformation to generate new candidate information g tThis process ensures the validity and relevance of the input information; the cell state is the core of LSTM, which combines the cell state of the previous moment and the current input information through the forget gate f t and input gate i t Adjustment, update the current unit state c t , thereby retaining important historical information and introducing new input information. This mechanism ensures the balance between long-term retention and short-term update of information; the hidden state is the output of the LSTM unit, which combines the information of the unit state and the output gate to predict the output at the next moment. The hidden state h t It not only contains historical information but also reflects the impact of current input, and is the basis for subsequent predictions. Through hidden states, LSTM can capture the dynamic changes and long-term trends of traffic flow, providing an accurate basis for demand forecasting.

[0083] As a preferred embodiment of the above, the demand forecasting model constructed by the demand forecasting algorithm includes:

[0084] The input layer fuses and cleans real-time traffic flow and speed data from radar devices along the highway;

[0085] Hidden layer, uses long short-term memory network to process time series data, capture the time dependency and long-term trend of traffic flow and speed, and generate hidden states;

[0086] The output layer generates the future traffic demand forecast value based on the hidden state of the long short-term memory network and adjusts the model parameters through loss calculation and Adam optimization algorithm;

[0087] The prediction and warning layer divides traffic congestion levels according to traffic demand forecast values ​​and outputs warning information.

[0088] Specifically, the main task of the input layer is to fuse and clean the real-time traffic flow and speed data from radar devices along the highway. These data include the number and type of vehicles per minute (such as small cars, medium cars, and large cars) and vehicle speeds. Through data fusion, the data from different radar devices are integrated to form a unified data set. Through data cleaning, outliers and noise are removed to ensure the accuracy and reliability of the input data. The hidden layer uses a long short-term memory network (LSTM) to process time series data. The LSTM network can capture the time dependency and long-term trend of traffic flow and speed. Through its input gate, forget gate, and output gate mechanism, it can effectively handle long-term dependency problems. The output of the hidden layer is a hidden state, which contains the current moment and The information of all previous moments is a comprehensive representation of traffic flow and speed; the output layer generates future traffic demand forecast values ​​based on the hidden state of the LSTM network, and the hidden state is input to the fully connected layer to generate a forecast result, which is compared with the actual traffic data, and the error is calculated through the loss function. Commonly used loss functions include mean square error (MSE), etc. In order to optimize the model parameters, the Adam optimization algorithm is used to adjust the weights and bias of the network through back propagation, thereby minimizing the loss function and improving the prediction accuracy; the prediction and warning layer divides the traffic congestion level according to the traffic demand forecast value, and compares the forecast value with the preset congestion threshold to divide the traffic conditions into different levels, such as smooth, light congestion, moderate congestion and severe congestion.

[0089] As a preferred embodiment of the above, Figure 2 As shown in the figure, the traffic capacity prediction algorithm adopts an integrated decision tree model, including:

[0090] A10: Collect data from radars and sensors along the highway and train multiple decision tree models. Each decision tree model learns based on different features and data subsets.

[0091] A20: Use historical data sets to train each decision tree model, and adjust the parameter model through cross-validation to optimize the performance of each decision tree model;

[0092] A30: Predict new samples for each trained decision tree model, and use the weighted average method to fuse the prediction results of each decision tree model.

[0093] Specifically, first, real-time and historical traffic data are collected from radars and sensors deployed along the highway. These data include features such as traffic flow, vehicle speed, vehicle type, weather conditions, and road conditions. These feature data provide rich input information for the training of the decision tree model; next, the collected data is divided into multiple subsets, each of which contains different features and samples. For each data subset, a decision tree model is trained. Each decision tree model learns according to its specific data subset to form an independent judgment on the traffic conditions. The purpose of this step is to enable each decision tree model to capture different traffic patterns and laws through diversified data subsets and features, thereby improving the accuracy and robustness of the overall prediction; then, each decision tree model is trained using historical data sets, and the model parameters are adjusted through cross-validation. Cross-validation can effectively prevent model overfitting and improve the generalization ability of the model. The performance of each decision tree model is optimized by adjusting parameters such as the depth of the tree and the minimum number of samples for splitting nodes. After the model optimization is completed, new sample predictions are made for each trained decision tree model, and new traffic data is input into each decision tree model to obtain multiple prediction results. In order to obtain the final prediction result, the prediction results of each decision tree model are fused by the weighted average method. The weighted average can assign weights to each model according to its performance in cross-validation, making the comprehensive prediction result more accurate and reliable.

[0094] As a preferred embodiment of the above, Figure 3 As shown, step S2, according to the traffic congestion information, analyzes the congestion data between adjacent radars, and merges continuous congestion points to obtain a congestion section, including:

[0095] S21: Acquire and record geographic coordinate information of radar devices along the expressway, and determine the spatial position relationship between adjacent radar devices;

[0096] S22: Calculating the actual distance between adjacent radar devices based on the spatial position relationship;

[0097] S23: comparing the congestion states of adjacent radar devices according to the traffic congestion information and the actual distance, and determining that there is spatial continuity when the congestion states of two adjacent radar devices persist and the distances are within a preset range;

[0098] S24: merging the congestion points of the adjacent radar devices that are determined to have spatially continuous rows into continuous congestion points to obtain a congestion section.

[0099] Specifically, first, the geographic coordinate information of the radar devices along the highway is obtained and recorded. These coordinate information include the latitude and longitude of each radar device, ensuring that the specific location of each radar device on the highway can be accurately located; next, based on the acquired geographic coordinate information, the spatial position relationship of adjacent radar devices is determined, and by comparing the coordinates of each radar device, which radar devices are geographically adjacent, the main purpose of determining the adjacent relationship is to provide basic data for subsequent distance calculation and congestion status analysis. Then, the distance is calculated using the geographic coordinates, and the geodetic distance formula or other suitable calculation methods can be used to ensure the accuracy of the distance calculation. The calculation of the actual distance helps to determine the degree of spatial correlation between radar devices; according to the calculated actual distance and traffic congestion information, the congestion status of adjacent radar devices is compared. When the congestion status of two adjacent radar devices continues and the distance is within a preset range, they are determined to have spatial continuity. The preset range is a threshold set according to the actual situation, which is used to distinguish between independent congestion points and continuous congestion points; finally, the congestion points of adjacent radar devices that are determined to have spatial continuity are merged into a continuous congestion point, thereby obtaining a congestion section. The merged congestion section can more comprehensively reflect the actual traffic congestion situation.

[0100] As a preferred embodiment of the above, the traffic control measures are adjusted in real time, including:

[0101] Dynamically adjust the use strategy of emergency lanes and main lanes, and optimize the opening and closing status of ramps;

[0102] Implement gradient speed limit measures and dynamically adjust vehicle speeds based on road conditions and real-time traffic data;

[0103] Through the road information board, we can provide improved traffic information and driving suggestions, and guide drivers to change their driving routes.

[0104] Specifically, according to real-time traffic data and road conditions, the system can dynamically adjust the use strategy of emergency lanes and main lanes. When traffic congestion is serious, the emergency lanes are temporarily opened for ordinary vehicles to pass to increase the road's traffic capacity. At the same time, the opening and closing status of ramps is optimized to control the vehicle flow entering the congested section. For example, some ramp entrances are closed during peak hours to prevent more vehicles from entering the already congested main road. By analyzing real-time traffic data and road conditions, the system can implement gradient speed limit measures and set different speed limit values ​​in different sections to smooth traffic flow and reduce congestion caused by sudden deceleration or acceleration of vehicles. Dynamic adjustment of speed limit measures can better adapt to current traffic conditions and improve road traffic efficiency and safety. Information boards installed along the highway can display traffic information and driving suggestions in real time. According to the current traffic conditions, the system provides the driver with the latest traffic information through the information boards, such as the current road congestion, recommended detour routes and real-time speed limit information. Through this information, the driver is guided to choose the best driving route and avoid congested sections, thereby improving the efficiency of the overall traffic flow.

[0105] As a preferred embodiment of the above embodiment, the method further includes: combining traffic data of special traffic events, analyzing the impact of special traffic events on traffic congestion, and designing targeted solutions.

[0106] Specifically, first, collect and record data on special traffic events, including traffic accidents, road construction, severe weather, and major events; next, analyze the data on special traffic events, evaluate the impact of these events on traffic flow, vehicle speed, and congestion, and identify the scope and severity of congestion through data analysis technology to provide a basis for designing solutions; then, design targeted traffic control and diversion measures based on the analysis results. For example, after a traffic accident occurs, quickly clean up the accident site, set up temporary traffic guidance signs, and guide vehicles to detour; in severe weather conditions, issue early warning information, implement speed limit measures, and increase patrols and emergency rescue forces; during major events, reasonably arrange traffic flow and driving routes, and adjust traffic control measures through real-time monitoring.

[0107] Embodiment 2

[0108] Based on the same inventive concept as the intelligent capacity expansion method for the highway operation period in the aforementioned embodiment, the present invention further provides an intelligent capacity expansion system for the highway operation period, the system comprising:

[0109] The congestion information acquisition module collects real-time traffic flow and speed data, and obtains traffic congestion information by calculating the weighted average flow and speed;

[0110] The congestion section acquisition module analyzes the congestion data between adjacent radars based on traffic congestion information and merges continuous congestion points to obtain the congestion section;

[0111] The VC ratio calculation module calculates the current VC ratio and the future VC ratio according to the congested section through the demand prediction algorithm and the traffic capacity prediction algorithm. The VC ratio is the ratio of the demand flow to the traffic capacity.

[0112] The control measures adjustment module predicts and analyzes the congestion trend based on the calculation results, and adjusts the traffic control measures in real time based on the prediction and analysis results.

[0113] The above-mentioned capacity expansion system in the present invention can effectively realize the intelligent capacity expansion method for the operation period of the highway. The technical effects that can be achieved are as described in the above-mentioned embodiments and will not be repeated here.

[0114] As a preferred embodiment of the above, the congestion section acquisition module includes:

[0115] A spatial relationship determination unit, which obtains and records the geographic coordinate information of radar devices along the expressway and determines the spatial position relationship of adjacent radar devices;

[0116] an actual distance calculation unit, which calculates the actual distance between adjacent radar devices based on the spatial position relationship;

[0117] A spatial continuity judgment unit, which compares the congestion status of adjacent radar devices according to the traffic congestion information and the actual distance, and judges that there is spatial continuity when the congestion status of two adjacent radar devices continues and the distance is within a preset range;

[0118] The congestion section acquisition unit merges the congestion points of the adjacent radar devices that are determined to have spatially continuous rows into continuous congestion points to obtain a congestion section.

[0119] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0120] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A smart capacity expansion method for expressway operation, characterized in that: include: Collect real-time traffic flow and speed data, and obtain traffic congestion information by calculating weighted average flow and speed; Analyzing the congestion data between adjacent radars according to the traffic congestion information, and merging continuous congestion points to obtain a congestion section; According to the congested section, the current VC ratio and the future VC ratio are calculated by using a demand prediction algorithm and a traffic capacity prediction algorithm, wherein the VC ratio is a ratio of demand flow to traffic capacity; Based on the calculation results, the congestion trend is predicted and analyzed, and traffic control measures are adjusted in real time based on the prediction and analysis results.

2. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: Calculate the weighted average flow and speed with the formula: Among them, V avg is the average speed per minute for 15 minutes, V i is the speed per minute, N s,i is the number of cars, recorded every i minutes, N m,i is the number of trains, recorded every i minutes, N l,i is the number of large vehicles, recorded every i minutes, F is the total flow, and is the historical mean of speed and flow, Z V and Z F is the deviation of speed and flow from historical data, D is the weighted average index of congestion detection, and α and β are weight factors.

3. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: The demand forecasting algorithm adopts a loop memory network, including: i t =σ(W xi x t +W hi h t-1 +b i ) f t =σ(W xf x t +W hf h t-1 +b f ) o t =σ(Wx xo x t +W ho h t-1 +b o ) g t =tanh(W xc x t +W hc h t-1 +b c ) c t =f t ⊙c t-1 +i t ⊙g t h t =o t ⊙tanh(c t ) Among them, i t , f t , o t are the activation values ​​of the input gate, forget gate, and output gate respectively; g t is the unit update vector; c t is the unit state; h t is the output hidden state.

4. The intelligent capacity expansion method for expressway operation period according to claim 3 is characterized in that: The demand forecasting model constructed by the demand forecasting algorithm includes: The input layer fuses and cleans real-time traffic flow and speed data from radar devices along the highway; Hidden layer, uses long short-term memory network to process time series data, capture the time dependency and long-term trend of traffic flow and speed, and generate hidden states; The output layer generates the future traffic demand forecast value according to the hidden state of the long short-term memory network, and adjusts the model parameters through loss calculation and Adam optimization algorithm; The prediction and warning layer divides the traffic congestion level according to the traffic demand prediction value and outputs warning information.

5. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: The traffic capacity prediction algorithm adopts an integrated decision tree model, including: Collect data from radars and sensors along the highway and train multiple decision tree models, each of which learns based on different features and data subsets; Using historical data sets to train each decision tree model, and adjusting the parameter model through cross-validation to optimize the performance of each decision tree model; New sample prediction is performed on each of the trained decision tree models, and the prediction results of each of the decision tree models are fused using a weighted average method.

6. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: According to the traffic congestion information, the congestion data between adjacent radars are analyzed, and continuous congestion points are merged to obtain a congestion section, including: Obtain and record the geographic coordinate information of radar devices along the highway and determine the spatial position relationship between adjacent radar devices; Based on the spatial position relationship, calculating the actual distance between the adjacent radar devices; According to the traffic congestion information and the actual distance, the congestion states of the adjacent radar devices are compared, and when the congestion states of the two adjacent radar devices continue and the distances are within a preset range, it is determined that there is spatial continuity; The congestion points of the adjacent radar devices that are determined to have spatially continuous rows are merged into continuous congestion points to obtain a congestion section.

7. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: Adjust traffic control measures in real time, including: Dynamically adjust the use strategy of emergency lanes and main lanes, and optimize the opening and closing status of ramps; Implement gradient speed limit measures and dynamically adjust vehicle speeds based on road conditions and real-time traffic data; Through the road information board, we can provide improved traffic information and driving suggestions, and guide drivers to change their driving routes.

8. The intelligent capacity expansion method for expressway operation period according to claim 1 is characterized in that: The method further includes: combining traffic data of special traffic events, analyzing the impact of the special traffic events on traffic congestion, and designing targeted solutions.

9. A smart capacity expansion system for highway operation, characterized in that: The system comprises: The congestion information acquisition module collects real-time traffic flow and speed data, and obtains traffic congestion information by calculating the weighted average flow and speed; A congestion section acquisition module, which analyzes the congestion data between adjacent radars according to the traffic congestion information, and merges continuous congestion points to obtain a congestion section; A VC ratio calculation module, which calculates the current VC ratio and the future VC ratio according to the congested section by using a demand prediction algorithm and a traffic capacity prediction algorithm, wherein the VC ratio is a ratio of demand flow to traffic capacity; The control measures adjustment module predicts and analyzes the congestion trend based on the calculation results, and adjusts the traffic control measures in real time based on the prediction and analysis results.

10. The intelligent capacity expansion system for expressway operation period according to claim 9 is characterized in that: The congested section acquisition module includes: A spatial relationship determination unit, which obtains and records the geographic coordinate information of radar devices along the expressway and determines the spatial position relationship of adjacent radar devices; an actual distance calculation unit, which calculates the actual distance between the adjacent radar devices based on the spatial position relationship; a spatial continuity judgment unit, which compares the congestion status of the adjacent radar devices according to the traffic congestion information and the actual distance, and judges that there is spatial continuity when the congestion status of two adjacent radar devices continues and the distance is within a preset range; The congestion section acquisition unit merges the congestion points of the adjacent radar devices that are determined to have spatially continuous rows into continuous congestion points to obtain a congestion section.

Images