Cloud side end toll station traffic flow monitoring method and system based on digital twinning

Abstract

The invention provides a cloud side end toll station traffic flow monitoring method and system based on digital twinning, and relates to the technical field of traffic control, and the method comprises the steps: collecting the actual operation data of a toll station in real time, including the vehicle arrival time, the lane use condition and the equipment state information, constructing a digital twinborn model corresponding to the entity toll station on the cloud server; an edge computing node is arranged at the edge of the entity toll station, processing and anomaly detection analysis are carried out on the actual operation data to obtain a processing result, and the processing result is uploaded to a cloud server; and in the cloud server, the processing result is predicted and analyzed to generate a traffic flow monitoring scheme of the toll station, and the traffic flow monitoring scheme comprises lane opening or closing suggestions, lane proportion adjustment and emergency response measures under special events. Through cooperation of the digital twin technology and the cloud side end, intelligent monitoring and dynamic optimization of the traffic flow of the toll station are realized, and the traffic efficiency and the management level are improved.

Description

Technical Field

[0001] The present invention relates to the field of traffic control technology, and in particular to a cloud-edge toll station vehicle flow monitoring method and system based on digital twins. Background Art

[0002] Traditional methods mostly rely on manual inspections, fixed camera monitoring and simple data statistics, which makes it difficult to achieve real-time and accurate monitoring of traffic flow at toll stations. The data update cycle is long and cannot reflect the dynamic changes of traffic flow in a timely manner, resulting in large time differences and errors between the monitoring results and the actual situation. Lack of advanced data analysis and processing capabilities, unable to automatically adjust lane configuration, signal light control, etc. according to real-time traffic flow, monitoring plans are often formulated based on experience and are difficult to adapt to changes in traffic demand at different times and under different weather conditions.

[0003] The monitoring system can usually only provide on-site monitoring images. Managers cannot remotely view the detailed operating status of the toll station in real time, let alone remotely control and adjust parameters of the toll station equipment. At the same time, there is a lack of effective feedback mechanism, and the monitoring results can often only be used for post-analysis and cannot immediately guide on-site operations.

[0004] In addition, with the improvement of information technology, traditional monitoring systems are facing more and more network security threats. Due to the lack of professional security protection measures, the system is vulnerable to hacker attacks and virus intrusions, resulting in serious consequences such as data leakage and system paralysis, which seriously affect the normal operation of toll stations and traffic safety. The hardware and software of traditional monitoring systems are often tightly coupled, making it difficult to flexibly expand or upgrade according to actual needs. At the same time, the system maintenance cost is high, and once a failure occurs, the repair cycle is long, affecting the continuity and stability of monitoring. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a cloud-edge toll station traffic flow monitoring method and system based on digital twins, which realizes comprehensive, real-time and accurate monitoring of the operating status of the toll station.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows: In a first aspect, a cloud-edge toll station traffic flow monitoring method based on digital twins, the method comprising: Collect the actual operation data of the toll station in real time, including vehicle arrival time, lane usage and equipment status information, and build a digital twin model corresponding to the physical toll station on the cloud server; Set up edge computing nodes at the edge of the physical toll station to process and analyze the actual operation data to obtain processing results, and upload the processing results to the cloud server; In the cloud server, the processing results are predicted and analyzed to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment, and emergency response measures under special events; The traffic flow monitoring plan is sent to the edge computing node and the on-site control system of the toll station. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring plan; According to the real-time adjustment of the lane configuration and signal light control of the toll station, the real-time operation status and traffic flow of the toll station are remotely monitored through the cloud platform, and the toll station is remotely controlled, including adjusting the number of lanes open and modifying the toll rules; The actual operating results after remote control are fed back to the digital twin model. The digital twin model compares and analyzes the actual operating status based on the actual operating effect feedback, and continuously optimizes the traffic flow monitoring solution through continuous interaction and iteration.

[0007] Furthermore, the actual operation data of the toll station is collected in real time, including vehicle arrival time, lane usage and equipment status information, and a digital twin model corresponding to the physical toll station is built on the cloud server, including: Place sensors and monitoring equipment at key locations of toll booths, including lane entrances, exits and toll booths, to capture real-time data on the actual operation of toll booths, including vehicle arrival times, lane usage and equipment status information; Transmit the actual operation data of the toll station to the cloud server in real time; On the cloud server, 3D modeling software is used to build a digital twin model that corresponds one-to-one to the physical toll station based on the actual operating data of the toll station, and the digital twin model reflects the operating status of the toll station in real time.

[0008] Furthermore, edge computing nodes are set up at the edge of the physical toll station to process and analyze the actual operation data to obtain processing results, and upload the processing results to the cloud server, including: The edge computing node monitors and receives the raw data streams from the toll station sensors and monitoring equipment, including vehicle arrival time, lane occupancy status, and equipment working status; Clean and format the original data stream to obtain cleaned and formatted data, and extract key features from the cleaned and formatted data, including vehicle passing speed, lane occupancy time, and equipment status change frequency; The key features are input into the preset random forest model, which evaluates each feature point to determine whether it is an anomaly. Anomalies include equipment failure, traffic congestion, and special events. For detected anomalies, the edge computing node generates an anomaly report, including the anomaly type, occurrence time and impact range; The cleaned and formatted data and anomaly detection reports are packaged into data packets, and the data packets are uploaded to the cloud server.

[0009] Furthermore, in the cloud server, the processing results are predicted and analyzed to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment, and emergency response measures under special events, including: The cloud server receives the processing results from the edge computing node, including cleaned traffic flow data and anomaly detection reports; Use time series prediction models to predict traffic flow data, identify traffic flow trends and peak hours, and obtain prediction and congestion analysis results; According to the prediction and congestion analysis results, the lane opening or closing status, lane ratio adjustment plan and emergency response measures under special events are used as optimization variables to initialize an artificial fish swarm, where each fish represents a traffic flow monitoring plan for a toll station. Define an objective function to quantitatively evaluate the comprehensive performance of each toll station's traffic flow monitoring solution under given traffic flow prediction results and congestion analysis conditions, including traffic efficiency, resource utilization, and responsiveness to special events; During the iteration process of the artificial fish school, each fish performs foraging, clustering and tailgating behaviors according to its current position and objective function value. When the preset number of iterations is reached, the final traffic flow monitoring plan for the toll station is obtained, including suggestions for lane opening or closing, lane ratio adjustment plans and emergency response measures under special events.

[0010] Furthermore, the calculation formula of the objective function is: ; in, Representation scheme The objective function value of , , , , , , , represents the weight coefficient; Indicated in the plan The total number of vehicles passing through the toll booth; Representation scheme The average speed of vehicles passing through the toll booth; Indicated in the plan the number of lanes available at the lower toll booth; Indicates the time period hours; Indicated in the plan The effective length of each lane below; Indicated in the plan The number of lanes actually opened; Indicated in the plan The operating costs required for each lane to be opened; Indicates the total number of lanes; Indicated in the plan the total operating expenses of the lower toll booth; Indicated in the plan Average response time to special events at lower toll booths; Indicated in the plan The duration of the special event; Indicated in the plan The number of lanes affected by the special event; Indicated in the plan The degree of traffic congestion caused by special events; Indicated in the plan The total number of vehicles expected to pass through the toll booth; Indicated in the plan The number of vehicles or the length of the queue before the toll booth; Indicates the maximum queue length.

[0011] Furthermore, the traffic flow monitoring solution is sent to the edge computing node and the on-site control system of the toll station. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring solution, including: The edge computing node analyzes the traffic flow monitoring solution and extracts key information, including lane opening or closing instructions, lane direction adjustment requirements, and signal light control parameters; Using traffic flow monitoring equipment, including cameras and geomagnetic sensors, edge computing nodes collect and process traffic flow data in real time to obtain the real-time status of current traffic flow; Determine the final lane configuration and signal light control plan based on the traffic flow monitoring plan and the real-time status of the current traffic flow; The edge computing node sends the control instructions of the lane configuration and signal light control scheme to the toll station on-site control system, and performs corresponding lane configuration adjustments and signal light control operations, including physically opening or closing lane gates, adjusting lane signs, and changing signal light display status.

[0012] Furthermore, the actual operation effect after remote control is fed back to the digital twin model. The digital twin model compares and analyzes the actual operation status based on the actual operation effect feedback. Through continuous interaction and iteration, the traffic flow monitoring solution is continuously optimized, including: The digital twin model simulates the actual operating status of the toll station based on the real-time feedback effect data and the actual operating data of the toll station, including the arrival distribution of vehicles, the use of lanes and the control logic of traffic lights, and generates virtual operating status data corresponding to the actual operating status; Compare the feedback effect data with the virtual operation status data generated in the digital twin model, including comparing the improvement in lane traffic efficiency, the reduction in vehicle delay time, and the change in queue length, identifying the differences and trends between the feedback effect data and the virtual operation status data, and obtaining the results of the comparative analysis; Based on the results of the comparative analysis, the digital twin model continues to interact with the physical toll station, and the traffic flow monitoring solution is continuously iterated and optimized.

[0013] Second, a cloud-edge toll station traffic flow monitoring system based on digital twins, including: Data collection module, used to capture the actual operation data of toll stations in real time, including vehicle arrival time, lane usage, equipment status, lane efficiency, vehicle delay duration and queue length traffic flow information; The edge computing processing module is used to perform preliminary sorting and abnormal identification analysis on the actual operation data obtained by the data collection module, obtain the analysis results, and upload the analysis results to the cloud server; The cloud service management module is used to build a digital twin model corresponding to the physical toll station, receive and process the data uploaded by the edge computing processing module to generate a traffic flow monitoring plan, and send the traffic flow monitoring plan to the edge computing processing module and the toll station on-site control system; Field control execution module, used to adjust the lane layout and signal light control logic of the toll station in real time according to the traffic flow monitoring plan; Remote monitoring and control module, which is used to remotely monitor the real-time operating status and traffic flow of toll stations through the cloud platform, and supports remote control operations, including adjustment of the number of lanes open and modification of toll rules; The effect feedback and optimization iteration module is used to transmit the actual operation data after remote control back to the digital twin model, and compare and analyze it with the actual operation status, so as to continuously iterate and optimize the traffic flow monitoring solution.

[0014] According to a third aspect, a computing device includes: one or more processors; The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0015] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0016] The above solution of the present invention includes at least the following beneficial effects: By collecting the actual operation data of toll stations in real time and building a digital twin model that accurately corresponds to the physical toll stations on the cloud server, real-time and accurate monitoring of the traffic flow at toll stations is achieved. The data is updated in a timely manner, and the monitoring results are highly consistent with the actual situation, providing reliable data support for traffic management.

[0017] By using edge computing nodes to process actual operation data and perform anomaly detection and analysis, combined with the prediction and analysis capabilities of cloud servers, an optimized traffic flow monitoring solution can be automatically generated. The solution is dynamically adjusted according to real-time traffic flow and special events, improving the intelligence and adaptability of monitoring.

[0018] Through the cloud platform, the real-time operating status and traffic flow of toll stations can be remotely monitored. Managers can instantly view and remotely control toll station equipment, including adjusting the number of open lanes and modifying toll rules. At the same time, the actual operating effect after remote control is fed back to the digital twin model. Through continuous interaction and iteration, the monitoring plan is continuously optimized to form a closed-loop monitoring and feedback mechanism. Advanced network security technologies such as encrypted transmission, access control, and firewalls are used in various links such as data storage, remote control, and data storage, which effectively prevent network security threats such as hacker attacks and virus intrusions, ensuring the security and reliability of the system.

[0019] The system adopts a cloud-edge-end architecture, with relatively independent hardware and software, which can be flexibly expanded or upgraded according to actual needs. At the same time, the cloud server is centrally managed, which reduces maintenance costs and improves the maintainability and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a cloud-edge toll station traffic flow monitoring method based on digital twins provided in an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of a cloud-edge toll station vehicle flow monitoring system based on digital twins provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0023] like Figure 1 As shown, an embodiment of the present invention proposes a cloud-edge toll station vehicle flow monitoring method and system based on digital twins, and the method includes the following steps: Step 11: collect the actual operation data of the toll station in real time, including vehicle arrival time, lane usage and equipment status information, and build a digital twin model corresponding to the physical toll station on the cloud server; Step 12: Set up edge computing nodes at the edge of the physical toll station to process and perform anomaly detection analysis on the actual operation data to obtain processing results, and upload the processing results to the cloud server; Step 13, in the cloud server, the processing results are predicted and analyzed to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment, and emergency response measures under special events; Step 14: Send the traffic flow monitoring plan to the edge computing node and the on-site control system of the toll station. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring plan. Step 15: According to the real-time adjustment of the lane configuration and signal light control of the toll station, the real-time operation status and traffic flow of the toll station are remotely monitored through the cloud platform, and the toll station is remotely controlled, including adjusting the number of lanes open and modifying the toll rules; Step 16: Feedback the actual operation effect after remote control to the digital twin model. The digital twin model compares and analyzes the actual operation status based on the actual operation effect feedback, and continuously optimizes the vehicle flow monitoring solution through continuous interaction and iteration.

[0024] In the embodiment of the present invention, by collecting the actual operation data of the toll station in real time and building a digital twin model that accurately corresponds to the physical toll station on the cloud server, a comprehensive and real-time simulation of the operation status of the toll station is achieved. This simulation not only improves the intuitiveness and accuracy of monitoring, but also provides a solid foundation for subsequent data analysis, prediction and decision-making. As a virtual "mirror", the digital twin model can reflect the actual situation of the toll station in real time, providing managers with a powerful visual monitoring tool.

[0025] Step 12: Set up edge computing nodes at the edge of the physical toll station to perform preliminary processing and anomaly detection analysis on the actual operation data, effectively reducing the burden on the cloud server and improving the real-time and efficiency of data processing. At the same time, the anomaly detection capability of the edge computing node can timely discover and warn potential problems, providing a strong guarantee for the stable operation of the toll station.

[0026] Step 13: The cloud server uses powerful computing power and intelligent algorithms to predict and analyze the processing results and generate an optimized traffic flow monitoring plan. This plan not only takes into account the current traffic flow and lane usage, but also predicts future traffic trends, providing scientific guidance for lane configuration and signal light control at toll stations. By implementing these plans, the traffic efficiency and vehicle traffic experience at toll stations can be improved.

[0027] Step 14: Send the traffic flow monitoring plan to the edge computing node and the on-site control system of the toll station, so as to achieve the rapid issuance and real-time adjustment of the monitoring plan. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the plan instructions, ensuring the effective implementation of the monitoring plan. This real-time adjustment capability can quickly respond to changes in traffic flow and keep the toll station running smoothly.

[0028] Step 15: By remotely monitoring the real-time operating status and traffic flow of the toll station through the cloud platform, managers can understand the operating status of the toll station anytime and anywhere, and remotely control the toll station. This remote monitoring and control capability not only improves management efficiency, but also enhances the ability to respond to emergencies. Managers can adjust the number of lanes open and modify toll rules in a timely manner according to actual conditions to ensure the stable operation of the toll station and smooth traffic.

[0029] Step 16: Feedback the actual operation effect after remote control to the digital twin model. The model compares and analyzes the actual operation effect feedback with the actual operation status, and continuously optimizes the traffic flow monitoring plan through continuous interaction and iteration. This feedback and optimization iteration mechanism can continuously improve the monitoring plan and improve the accuracy and effectiveness of monitoring. At the same time, the continuous optimization of the digital twin model also provides strong support for the long-term stable operation of the toll station.

[0030] In a preferred embodiment of the present invention, the above step 11 collects the actual operation data of the toll station in real time, including vehicle arrival time, lane usage and equipment status information, and builds a digital twin model corresponding to the physical toll station on the cloud server, which may include: Step 112, placing sensors and monitoring equipment at key locations of the toll station, including lane entrances, exits, and toll booths, to capture real-time actual operation data of the toll station, including vehicle arrival time, lane usage, and equipment status information; Step 113, transmitting the actual operation data of the toll station to the cloud server in real time; Step 114, on the cloud server, using 3D modeling software, a digital twin model that corresponds one-to-one to the physical toll station is constructed based on the actual operating data of the toll station, and the digital twin model reflects the operating status of the toll station in real time.

[0031] In the embodiment of the present invention, high-precision and high-reliability sensors and monitoring equipment are selected, such as vehicle identification sensors, lane occupancy sensors, equipment status monitoring sensors, and high-definition cameras. Vehicle identification sensors are installed at the entrance and exit of the lanes of the toll station to capture information such as the arrival time and license plate number of the vehicle. Lane occupancy sensors are installed on the lanes to monitor the use of the lanes in real time, such as whether there are vehicles passing and whether the lanes are idle. Equipment status monitoring sensors are installed in the toll booths to monitor the working status of toll collection equipment (such as toll machines, printers, display screens, etc.). High-definition cameras are configured to monitor the toll station in all directions, capture real-time images, and provide visual support for subsequent abnormal detection and event processing.

[0032] Step 113, establish a stable and efficient data transmission network to ensure that the data collected by sensors and monitoring equipment can be transmitted to the cloud server in real time and accurately. Use encryption transmission technology to ensure the security of data during transmission and prevent data from being stolen or tampered with. Design a reasonable data transmission protocol to ensure the integrity and order of data and avoid data loss or disorder.

[0033] Step 114: On the cloud server, use 3D modeling software (such as AutoCAD, SketchUp, Revit, etc.) to model the toll station according to its actual layout and size. Associate the real-time data collected by sensors and monitoring equipment (such as vehicle arrival time, lane usage, equipment status, etc.) with the 3D model to achieve real-time synchronization between the model and the physical toll station. Design a model update mechanism to ensure that the digital twin model can be updated in a timely manner as the actual operating status of the toll station changes, maintaining the accuracy and timeliness of the model.

[0034] Take a highway toll station as an example: Install RFID vehicle identification sensors at the entrance and exit of the lane to obtain the arrival time and license plate number of the vehicle by reading the RFID tag on the vehicle. Install geomagnetic lane occupancy sensors on the lane to monitor in real time whether there are vehicles passing through the lane. Install equipment status monitoring sensors in the toll booth to monitor the working status of equipment such as toll machines, printers, and display screens. The data collected by sensors and monitoring equipment are transmitted to the cloud server in real time through the 4G / 5G network. On the cloud server, use AutoCAD software to perform three-dimensional modeling based on the actual layout and size of the toll station, and associate real-time data with the model. Managers can view the digital twin model of the toll station in real time through the cloud platform, understand the operating status of the toll station, such as lane usage, equipment working status, etc., and remotely control and adjust according to actual conditions.

[0035] By collecting and transmitting the actual operating data of the toll station in real time and building a digital twin model on the cloud server, comprehensive and real-time monitoring of the toll station is achieved, and the monitoring efficiency is improved. The digital twin model can reflect the operating status of the toll station in real time, provide accurate and intuitive data support for managers, and help make more scientific and reasonable decisions. The application of encrypted transmission technology and equipment status monitoring sensors improves the security of data transmission and the security of toll station equipment, and reduces safety risks. By real-time monitoring and analysis of the traffic flow and equipment status of the toll station, lane resources can be more reasonably allocated, and the traffic efficiency and service level of the toll station can be improved. The digital twin model can detect equipment anomalies and failures in a timely manner, facilitate timely maintenance and repair, and reduce maintenance costs and time costs.

[0036] In a preferred embodiment of the present invention, the above step 12, setting an edge computing node at the edge of the physical toll station, processing the actual operation data and performing anomaly detection analysis to obtain processing results, and uploading the processing results to the cloud server, may include: Step 123, the edge computing node monitors and receives the raw data streams from the toll station sensors and monitoring equipment, including vehicle arrival time, lane occupancy status, and equipment working status; Step 124, cleaning and formatting the original data stream to obtain cleaned and formatted data, and extracting key features from the cleaned and formatted data, including vehicle passing speed, lane occupancy time, and equipment status change frequency; Step 125, inputting the key features into a preset random forest model, the random forest model evaluates each feature point to determine whether it is an abnormality, the abnormality includes equipment failure, traffic congestion, special events; Step 126, for the detected anomaly, the edge computing node generates an anomaly report, including the anomaly type, occurrence time, and impact scope; Step 127 , packaging the cleaned and formatted data and the anomaly detection report into a data package, and uploading the data package to a cloud server.

[0037] In the embodiment of the present invention, the edge computing node is equipped with a dedicated data receiving module, which continuously monitors the data signals sent by the toll station sensors and monitoring equipment. When new data arrives, the receiving module immediately captures and caches the data to ensure the integrity and continuity of the data. The original data stream includes various types of data such as vehicle arrival time, lane occupancy status, equipment working status, etc., which are transmitted in a specific format (such as JSON, XML, etc.).

[0038] In step 124, the edge computing node preprocesses the received raw data stream, including removing invalid data, filling missing values, correcting erroneous data, etc., to ensure the accuracy and reliability of the data. The cleaned data is formatted into a unified format to facilitate subsequent processing and analysis. Key features are extracted from the cleaned and formatted data, such as vehicle passing speed (obtained by calculating the difference in arrival time of adjacent vehicles), lane occupancy time (the time difference between a vehicle entering a lane and leaving the lane), and device status change frequency (the number of times the device status changes within a period of time).

[0039] In step 125, the edge computing node inputs the extracted key features into a preset random forest model.

[0040] The random forest model is an integrated learning model that includes multiple decision trees. It can independently evaluate each feature point and determine whether it is an anomaly based on the evaluation results. Anomaly types include equipment failure (such as toll machine failure, camera failure, etc.), traffic congestion (lane occupancy time is too long), special events (such as vehicle accidents, malicious toll evasion, etc.). The model outputs the results of anomaly detection, including anomaly type, anomaly degree, etc.

[0041] Step 126: For the detected anomaly, the edge computing node automatically generates an anomaly report. The anomaly report includes detailed information such as the anomaly type, occurrence time, and impact scope (such as the lanes and equipment involved).

[0042] The reports are stored in a structured format (such as JSON, XML, etc.) to facilitate subsequent analysis and processing.

[0043] In step 127, the edge computing node packages the cleaned and formatted data and the anomaly detection report into a data packet, which is uploaded to the cloud server via a secure communication protocol (such as HTTPS, FTP, etc.).

[0044] Take a highway toll station as an example: The edge computing node continuously monitors and receives data streams from lane sensors, equipment status monitoring sensors, etc. The received data is cleaned and formatted to remove invalid and erroneous data, and key features are extracted, such as vehicle speed, lane occupancy time, etc. These features are input into the preset random forest model, which evaluates the features and determines whether there are any anomalies. If an anomaly is detected (such as a lane being occupied for too long, which may indicate traffic congestion), the edge computing node generates an anomaly report containing information such as the anomaly type (traffic congestion), time of occurrence, and scope of impact (the lane). Finally, the edge computing node packages the cleaned data and anomaly detection report into a data packet and uploads it to the cloud server via the HTTPS protocol.

[0045] By processing and analyzing the raw data stream in real time through edge computing nodes, anomalies can be discovered and reported in a timely manner, improving the efficiency and accuracy of anomaly detection. Edge computing nodes perform data processing and anomaly detection locally, reducing the amount of data that needs to be uploaded to the cloud server and alleviating the burden on the cloud server. Edge computing nodes can independently process data and detect anomalies, ensuring the normal operation and anomaly detection capabilities of the system even when the network connection is unstable or the cloud server fails. Through real-time uploaded cleaned data and anomaly detection reports, the cloud server can more accurately understand the operating status and anomalies of the toll station, providing managers with more powerful decision-making support.

[0046] In a preferred embodiment of the present invention, in step 13, the processing results are predicted and analyzed in the cloud server to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment and emergency response measures under special events, which may include: Step 131, the cloud server receives the processing results of the edge computing node, including the cleaned vehicle flow data and anomaly detection report; Step 132, using a time series prediction model to predict the traffic flow data, identifying the changing trend and peak hours of the traffic flow, and obtaining prediction and congestion analysis results; Step 133, based on the prediction and congestion analysis results, the lane opening or closing state, the lane ratio adjustment plan and the emergency response measures under special events are used as optimization variables to initialize an artificial fish school, where each fish represents a traffic flow monitoring plan for a toll station; Step 134, defining an objective function to quantitatively evaluate the comprehensive performance of each toll station's traffic flow monitoring solution under given traffic flow prediction results and congestion analysis conditions, including traffic efficiency, resource utilization, and response capabilities to special events; Step 135, during the iteration process of the artificial fish school, each fish performs foraging, grouping, and tailgating according to the current position and the objective function value. When the preset number of iterations is reached, the final traffic flow monitoring plan for the toll station is obtained, including lane opening or closing suggestions, lane ratio adjustment plans, and emergency response measures for special events.

[0047] In the embodiment of the present invention, the cloud server is configured with a dedicated data receiving interface for receiving the processing results from the edge computing node. The received data includes cleaned traffic flow data (such as vehicle arrival time, lane occupancy, vehicle passing speed, etc.) and abnormal detection reports (such as equipment failure, traffic congestion, special events, etc.). The cloud server stores and pre-processes the received data.

[0048] Step 132: The cloud server uses a time series prediction model (such as ARIMA, LSTM, etc.) to predict the cleaned traffic flow data. Through model analysis, the changing trend of traffic flow (such as growth, decline, stability, etc.) and peak hours (such as morning and evening peaks, holiday peaks, etc.) are identified. Combined with the anomaly detection report, the prediction results are analyzed for congestion to identify possible traffic congestion points and congestion levels.

[0049] Step 133, based on the prediction and congestion analysis results, the cloud server initializes an artificial fish swarm, where each fish represents a traffic flow monitoring plan for a toll station. Each fish in the fish swarm contains optimization variables such as lane opening or closing status, lane ratio adjustment plan, and emergency response measures under special events. The initial fish swarm can be generated randomly.

[0050] Step 134, the cloud server defines an objective function to quantitatively evaluate the comprehensive performance of each toll station traffic flow monitoring solution. The objective function takes into account multiple factors, including traffic efficiency (such as vehicle passing speed, queuing time, etc.), resource utilization (such as lane utilization efficiency, equipment utilization, etc.) and response capabilities to special events (such as emergency response speed, processing effect, etc.). The smaller the value of the objective function, the better the comprehensive performance of the solution.

[0051] Step 135, during the iteration process of the artificial fish school, each fish performs foraging, grouping, and tail-chasing behaviors according to the current position and the objective function value. Each fish in the fish school tries to search for a better solution (i.e., a better traffic flow monitoring solution) near its current position to reduce the value of the objective function. The fish in the fish school will gather together and share information to find the final global solution faster. The fish in the fish school will follow the best performing fish near the current position (i.e., the fish with the smallest objective function value) in the hope of obtaining a better solution. When the preset number of iterations is reached, the iteration process ends and the final traffic flow monitoring solution for the toll station is obtained.

[0052] By predicting traffic flow and congestion, and adjusting the lane opening or closing status, lane ratio, etc. accordingly, traffic congestion can be effectively alleviated and traffic efficiency can be improved. According to the traffic flow prediction results and congestion analysis, the lane ratio and emergency response measures can be reasonably adjusted to make full use of toll station resources and avoid resource waste.

[0053] By formulating emergency response measures for special events and incorporating them into the traffic flow monitoring plan, the response speed and processing capabilities of toll stations to special events can be improved. Using the artificial fish swarm algorithm for iterative optimization, a more scientific and reasonable traffic flow monitoring plan can be obtained, providing strong support for the management and decision-making of toll stations.

[0054] In a preferred embodiment of the present invention, the calculation formula of the objective function is: ; in, Representation scheme The objective function value of , , , , , , , represents the weight coefficient; Indicated in the plan The total number of vehicles passing through the toll booth; Representation scheme The average speed of vehicles passing through the toll booth; Indicated in the plan the number of lanes available at the lower toll booth; Indicates the time period hours; Indicated in the plan The effective length of each lane below; Indicated in the plan The number of lanes actually opened; Indicated in the plan The operating costs required for each lane to be opened; Indicates the total number of lanes; Indicated in the plan the total operating expenses of the lower toll booth; Indicated in the plan Average response time to special events at lower toll booths; Indicated in the plan The duration of the special event; Indicated in the plan The number of lanes affected by the special event; Indicated in the plan The degree of traffic congestion caused by special events; Indicated in the plan The total number of vehicles expected to pass through the toll booth; Indicated in the plan The number of vehicles or the length of the queue before the toll booth; Indicates the maximum queue length.

[0055] In the embodiment of the present invention, the cloud server first receives the processing results from the edge computing node, including the cleaned traffic flow data and the anomaly detection report. The received data is stored and organized according to dimensions such as timestamp, lane number, and vehicle type. Based on historical data and real-time data, various variables (such as , , etc.)

[0056] Set the weight coefficient according to the actual situation and management requirements of the toll station , , , , as well as , , These weight coefficients reflect the importance of different factors in the objective function. The weight coefficients can be set by expert experience, historical data analysis or experimental adjustment to ensure that the objective function can accurately reflect the comprehensive performance of the toll station traffic flow monitoring scheme. For each feasible traffic flow monitoring scheme x, calculations are performed according to the formula of the objective function.

[0057] The specific calculation steps include: Calculation of vehicle traffic efficiency , reflecting the speed and efficiency of vehicles passing through toll booths.

[0058] Computing resource utilization ; It reflects the efficiency of lane opening and operating costs.

[0059] Calculating the Special Incident Response Capability Section , reflecting the response speed and handling effect of the toll station on special events.

[0060] Calculate the queue status part ; Reflect the queue situation in front of the toll booth to avoid excessively long queues.

[0061] Calculate the deviation from the expected total number of vehicles , reflecting the deviation between the actual number of vehicles passing and the expected number of vehicles passing. Substituting the calculation results of each part into the objective function formula, we get the solution The objective function value of The final traffic flow monitoring plan is output, including lane opening or closing suggestions, lane ratio adjustment plan and emergency response measures under special events, etc. The optimization plan is implemented in the actual operation of the toll station, and the effectiveness and adaptability of the plan are ensured through real-time monitoring and adjustment.

[0062] By optimizing the lane opening and closing status, lane ratio, etc., vehicles can pass through the toll station faster and more smoothly, reducing waiting time in queues and improving traffic efficiency. According to traffic flow forecasts and congestion analysis, the number of lane openings can be reasonably adjusted to avoid waste of resources and reduce operating costs. Emergency response measures for special events are formulated and incorporated into the traffic flow monitoring plan to ensure that special events can be responded to quickly and effectively when they occur, reducing the impact on traffic. Through the influence of the queue situation in the objective function, the optimization plan can effectively control the queue length in front of the toll station and avoid traffic congestion and dissatisfaction caused by too long queues.

[0063] In a preferred embodiment of the present invention, the above step 14, sending the traffic flow monitoring scheme to the edge computing node and the toll station field control system, the edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring scheme, may include: Step 141, the edge computing node analyzes the traffic flow monitoring solution and extracts key information, including lane opening or closing instructions, lane direction adjustment requirements, and signal light control parameters; Step 142, using traffic flow monitoring equipment, including cameras and geomagnetic sensors, the edge computing node collects and processes traffic flow data in real time to obtain the real-time status of current traffic flow; Step 143, determining the final lane configuration and signal light control plan according to the traffic flow monitoring plan and the real-time status of the current traffic flow; In step 144, the edge computing node sends the control instructions of the lane configuration and signal light control scheme to the toll station on-site control system, and performs corresponding lane configuration adjustment and signal light control operations, including physically opening or closing the lane gate, adjusting the lane indicator sign, and changing the signal light display status.

[0064] In an embodiment of the present invention, the edge computing node receives an encrypted data packet of the vehicle flow monitoring solution from the cloud server. The data packet is decrypted using a predefined encryption algorithm, and a data integrity check is performed to ensure that the solution has not been tampered with. The data packet is parsed to extract key information, such as lane opening / closing instructions, lane direction adjustment requirements (such as straight, left turn, right turn, etc.), and signal light control parameters (such as green light duration, red light duration, yellow light duration, etc.). The extracted information is stored in a local database, and necessary preprocessing is performed, such as format conversion, data verification, etc.

[0065] Step 142, configure traffic flow monitoring equipment such as cameras and geomagnetic sensors, ensure that the equipment works properly and is connected to the edge computing node. The camera captures the video image of the vehicle on the lane, and the geomagnetic sensor detects the change of the magnetic field when the vehicle passes, and collects the traffic flow data in real time. The collected data is processed, including vehicle identification, speed calculation, traffic flow statistics, etc., to obtain the real-time status of the current traffic flow. The processed traffic flow data is stored in the local database and updated regularly for subsequent analysis and decision-making.

[0066] Step 143, combining the traffic flow monitoring plan and the current real-time status of traffic flow, perform data analysis to identify traffic flow peak hours, congestion points, special events, etc. According to the data analysis results, formulate the final lane configuration and signal light control plan. For example, open more lanes during peak hours, adjust lane directions to adapt to traffic flow changes, and optimize signal light control parameters to reduce waiting time. Perform simulation verification on the formulated plan to ensure its feasibility and effectiveness. If necessary, adjust and optimize the plan.

[0067] Step 144, generate corresponding control instructions according to the final lane configuration and signal light control scheme. The instructions should include specific operation content, execution time, execution order, etc. The control instructions are sent to the toll station field control system through a safe and reliable communication protocol. Ensure accurate transmission and timely arrival of the instructions. After receiving the instructions, the toll station field control system performs corresponding lane configuration adjustments and signal light control operations. Including physically opening or closing lane gates, adjusting lane indicator signs, changing signal light display status, etc. After the field control system executes the operation, it feeds back the execution results and status information to the edge computing node.

[0068] Assume that a highway toll station has three lanes, namely lane 1, lane 2, and lane 3. During peak hours in the morning and evening, the traffic volume is large and congestion often occurs. According to the traffic flow monitoring solution, the edge computing node decides to open all lanes during peak hours and adjusts the signal light control parameters to optimize traffic efficiency.

[0069] The edge computing node receives the traffic flow monitoring plan and analyzes it to find out that all lanes need to be opened during peak hours and the green light duration of the signal lights needs to be adjusted. The traffic flow data is collected in real time through cameras and geomagnetic sensors. It is found that the traffic flow in lanes 1 and 2 is large, and the traffic flow in lane 3 is relatively small. Combined with the traffic flow monitoring plan and real-time data, a lane configuration and signal light control plan is formulated. It is decided to open all lanes during peak hours and extend the green light duration of the signal lights by 10 seconds to reduce vehicle waiting time. A control command is generated and sent to the on-site control system of the toll station. After the system executes the command, lanes 1, 2, and 3 are all opened, and the green light duration of the signal lights is extended by 10 seconds. The execution results and status information are fed back to the edge computing node.

[0070] By adjusting the lane configuration and signal light control scheme in real time, the vehicle travel path and time are optimized, waiting time and congestion are reduced, and traffic efficiency is improved. The lane configuration and signal light control parameters are dynamically adjusted according to the real-time traffic flow conditions to enhance the flexibility and adaptability of the toll station and better cope with changes in traffic flow. By optimizing the lane configuration and signal light control, the conflict and collision risks between vehicles are reduced, and the safety of the toll station is improved. Through reasonable lane configuration and signal light control, unnecessary lane opening and signal light energy consumption are reduced, and the operating cost of the toll station is reduced. Waiting time and congestion are reduced, and the smoothness and user experience of vehicles passing through the toll station are improved.

[0071] In a preferred embodiment of the present invention, the above step 15, according to real-time adjustment of lane configuration and signal light control of the toll station, remotely monitors the real-time operation status and traffic flow of the toll station through the cloud platform, and remotely controls the toll station, including adjusting the number of lane openings and modifying toll rules, may include: In an embodiment of the present invention, a stable and efficient cloud platform is built for remote monitoring and control of toll stations. The cloud platform should have functions such as data processing, storage, analysis and remote control. Edge computing nodes are deployed on site at the toll station to collect and process traffic flow data in real time and communicate with the cloud platform. The on-site control system of the toll station (such as lane gates, signal lights, toll collection equipment, etc.) is integrated with the edge computing node to ensure that remote control instructions from the cloud platform can be received and executed. Traffic flow monitoring equipment such as cameras, geomagnetic sensors, radars, etc. are used to collect traffic flow data of the toll station in real time, including information such as the number of vehicles, speed, and vehicle model. The edge computing node uploads the collected traffic flow data to the cloud platform through a secure and reliable communication protocol (such as MQTT, HTTP, etc.). The cloud platform analyzes the received traffic flow data in real time, identifies peak traffic flow periods, congestion points, abnormal events, etc., and provides a decision-making basis for remote control. The cloud platform displays the real-time operating status and traffic flow of the toll station to the management personnel in the form of charts, reports, etc., so that they can understand the operation status of the toll station at any time.

[0072] According to the traffic data and the operating status of the toll station, the cloud platform provides decision support functions, such as suggestions on the number of lanes to be opened and suggestions on adjusting toll rules. Based on the decision support results, managers generate remote control instructions through the cloud platform, such as adjusting the number of lanes to be opened and modifying toll rules. The cloud platform encrypts the remote control instructions to ensure the security of the instructions. Then, the instructions are sent to the edge computing node through the communication protocol. After receiving the instructions, the edge computing node decrypts and parses them. Then, the instructions are converted into control signals that can be recognized by the on-site control system of the toll station and sent to the corresponding control device.

[0073] After the on-site control system of the toll station executes the command, it will feed back the execution result and status information to the edge computing node. The edge computing node then uploads this information to the cloud platform for management personnel to view and confirm. Management personnel evaluate the execution effect of remote control commands based on the operating status and traffic flow of the toll station displayed on the cloud platform. Based on the evaluation results, management personnel can continuously optimize and adjust strategies such as the number of lane openings and charging rules to improve the operating efficiency and user experience of the toll station. Regularly upgrade and maintain the cloud platform and edge computing nodes to ensure the stability and security of the system. At the same time, continuously optimize and expand the functions of the system according to actual needs and technological development.

[0074] In a preferred embodiment of the present invention, in step 16, the actual operation effect after remote control is fed back to the digital twin model, and the digital twin model compares and analyzes the actual operation effect feedback with the actual operation status, and continuously optimizes the vehicle flow monitoring solution through continuous interaction and iteration, which may include: The digital twin model simulates the actual operating status of the toll station based on the real-time feedback effect data and the actual operating data of the toll station, including the arrival distribution of vehicles, the use of lanes and the control logic of traffic lights, and generates virtual operating status data corresponding to the actual operating status; Step 161, comparing the feedback effect data with the virtual operation status data generated in the digital twin model, including comparing the improvement of lane traffic efficiency, the reduction of vehicle delay time and the change of queue length, identifying the differences and trends between the feedback effect data and the virtual operation status data, and obtaining the results of comparative analysis; Step 162: Based on the results of the comparative analysis, the digital twin model and the physical toll station are continuously interacted to continuously iterate and optimize the vehicle flow monitoring solution.

[0075] In the embodiment of the present invention, actual operation effect feedback data is obtained from the remote control system, including key indicators such as lane efficiency, vehicle delay time, and queue length. Virtual operation status data for the corresponding time period is extracted from the digital twin model. These data should be generated based on the toll station operation status simulated by the model. The actual operation effect data and virtual operation status data are cleaned to remove outliers or invalid data. The data is normalized to ensure that data from different sources are comparable.

[0076] Select K-means clustering algorithm to cluster the improvement data set. Define the improvement / reduction, vehicle delay time reduction rate (to evaluate the effectiveness of remote control solutions), and queue length change rate (to determine the effectiveness of lane configuration and signal light control adjustments). Perform K-means clustering on the above three indicators (improvement, reduction rate, and change rate), and set the number of clusters to 3 (significant improvement / reduction, slight improvement / reduction, no improvement / reduction or decrease).

[0077] Calculate the distance between a sample and the centroid: ; in, Represents sample points With the centroid The distance between represents the sample point; Indicates the total number of features; The index representing the feature; Indicates The weight of each feature; Represents sample points No. eigenvalues; Indicates the number of samples in the data set; , Represents an index variable; Indicates the data set The sample point eigenvalues; Indicates the data set The sample point eigenvalues; Represents the centroid No. eigenvalues. Iterate and update the centroid until it no longer changes. Draw a comparative chart of the improvement in lane efficiency, the reduction rate of vehicle delay time, and the change rate of queue length, such as a bar chart or a line chart. Mark the clustering results in the chart, and use different colors or shapes to represent different clustering categories. According to the clustering results, analyze the difference between the actual operation effect and the model prediction effect. Identify the difference points, such as the lack of obvious improvement in traffic efficiency in certain time periods or lanes, or the reduction rate of vehicle delay time is lower than expected. Analyze the reasons for the difference, such as traffic flow fluctuations, improper signal control, unreasonable lane configuration, etc. Based on the current data and analysis results, predict the future trend of lane efficiency, vehicle delay time, and queue length. Consider the impact of factors such as traffic demand growth, road reconstruction, and policy changes on the trend. Based on the difference analysis, identify optimization points, such as adjusting the signal control logic, optimizing lane configuration, and strengthening traffic management. Propose specific optimization suggestions and improvement measures to improve lane efficiency, reduce vehicle delay time, and queue length.

[0078] Step 162, based on the results of the comparative analysis, identify the problems or deficiencies in the traffic flow monitoring solution. Determine the key parameters that need to be optimized or the strategic direction to be adjusted. Adjust the corresponding parameters or logic in the digital twin model to reflect the new situation or optimization needs in actual operation. Re-run the model to generate new virtual operation status data for the next round of comparison. Based on the results of the model adjustment, formulate a new traffic flow monitoring solution or adjust the existing solution. Consider the feasibility and cost-effectiveness of the actual operation to ensure the feasibility of the solution. Apply the optimized solution to the physical toll station and implement it through the remote control system.

[0079] The actual operating effect data after implementation is collected and fed back into the digital twin model to form a closed-loop iteration.

[0080] Assume that at a highway toll station, the lane configuration and signal light control logic are adjusted through remote control to improve traffic efficiency. After implementation, the actual operation effect data is collected and compared with the virtual operation status data in the digital twin model.

[0081] Actual operation data show that lane efficiency has increased by 10%, vehicle delay time has decreased by 15%, but queue length has only decreased by 5%. Data predicted by the digital twin model show that lane efficiency should be increased by 12%, vehicle delay time should be reduced by 20%, and queue length should be reduced by 10%. Through comparative analysis, it is found that the improvement of lane efficiency and vehicle delay time in actual operation is slightly lower than the model prediction, and the improvement effect of queue length is particularly not obvious. Further analysis shows that this may be due to the deviation between the vehicle arrival distribution in actual operation and the model prediction, and the failure of the signal light control logic to fully adapt to the traffic flow changes in certain specific periods. Based on the above analysis, the vehicle arrival distribution in the digital twin model is corrected, and the signal light control logic is adjusted to adapt to more variable traffic conditions. A new traffic flow monitoring plan is formulated, including further optimizing lane configuration during peak hours and dynamically adjusting the green light duration of the signal light according to real-time traffic flow. Through continuous comparison and analysis of the digital twin model and the actual operation effect, deficiencies in the monitoring plan can be discovered and corrected in a timely manner, improving the effectiveness and pertinence of the plan. The digital twin model provides a scientific basis and prediction for decision-making, reduces the influence of subjective judgment and empiricism, and enhances the scientificity and accuracy of decision-making. By continuously optimizing the traffic flow monitoring solution, the traffic efficiency of toll stations can be effectively improved, vehicle delays and queue time can be reduced, and user experience and service quality can be improved. The optimized monitoring solution can make more rational use of resources, reduce unnecessary energy consumption and labor costs, and reduce the operating costs of toll stations.

[0082] The application of digital twin technology has promoted the intelligent transformation of toll stations and provided strong support for the future construction of smart transportation and smart cities.

[0083] like Figure 2 As shown, an embodiment of the present invention further provides a cloud-edge-end toll station vehicle flow monitoring system based on digital twins, including: Data collection module, used to capture the actual operation data of toll stations in real time, including vehicle arrival time, lane usage, equipment status, lane efficiency, vehicle delay duration and queue length traffic flow information; The edge computing processing module is used to perform preliminary sorting and abnormal identification analysis on the actual operation data obtained by the data collection module, obtain the analysis results, and upload the analysis results to the cloud server; The cloud service management module is used to build a digital twin model corresponding to the physical toll station, receive and process the data uploaded by the edge computing processing module to generate a traffic flow monitoring plan, and send the traffic flow monitoring plan to the edge computing processing module and the toll station on-site control system; Field control execution module, used to adjust the lane layout and signal light control logic of the toll station in real time according to the traffic flow monitoring plan; Remote monitoring and control module, which is used to remotely monitor the real-time operating status and traffic flow of toll stations through the cloud platform, and supports remote control operations, including adjustment of the number of lanes open and modification of toll rules; The effect feedback and optimization iteration module is used to transmit the actual operation data after remote control back to the digital twin model, and compare and analyze it with the actual operation status to continuously iterate and optimize the vehicle flow monitoring solution. It should be noted that this system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0084] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0085] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0086] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cloud-edge toll station traffic flow monitoring method based on digital twins, characterized in that: The method comprises: Collect the actual operation data of the toll station in real time, including vehicle arrival time, lane usage and equipment status information, and build a digital twin model corresponding to the physical toll station on the cloud server; Set up edge computing nodes at the edge of the physical toll station to process and analyze the actual operation data to obtain processing results, and upload the processing results to the cloud server; In the cloud server, the processing results are predicted and analyzed to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment, and emergency response measures under special events; The traffic flow monitoring plan is sent to the edge computing node and the on-site control system of the toll station. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring plan; According to the real-time adjustment of the lane configuration and signal light control of the toll station, the real-time operation status and traffic flow of the toll station are remotely monitored through the cloud platform, and the toll station is remotely controlled, including adjusting the number of lanes open and modifying the toll rules; The actual operating results after remote control are fed back to the digital twin model. The digital twin model compares and analyzes the actual operating status based on the actual operating effect feedback, and continuously optimizes the traffic flow monitoring solution through continuous interaction and iteration.

2. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 1 is characterized in that: Collect the actual operation data of the toll station in real time, including vehicle arrival time, lane usage and equipment status information, and build a digital twin model corresponding to the physical toll station on the cloud server, including: Place sensors and monitoring equipment at key locations of toll booths, including lane entrances, exits and toll booths, to capture real-time data on the actual operation of toll booths, including vehicle arrival times, lane usage and equipment status information; Transmit the actual operation data of the toll station to the cloud server in real time; On the cloud server, 3D modeling software is used to build a digital twin model that corresponds one-to-one to the physical toll station based on the actual operating data of the toll station, and the digital twin model reflects the operating status of the toll station in real time.

3. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 2 is characterized in that: Set up edge computing nodes at the edge of the physical toll station to process the actual operation data and perform anomaly detection and analysis to obtain processing results, and upload the processing results to the cloud server, including: The edge computing node monitors and receives the raw data streams from the toll station sensors and monitoring equipment, including vehicle arrival time, lane occupancy status, and equipment working status; Clean and format the original data stream to obtain cleaned and formatted data, and extract key features from the cleaned and formatted data, including vehicle passing speed, lane occupancy time, and equipment status change frequency; The key features are input into the preset random forest model, which evaluates each feature point to determine whether it is an anomaly. Anomalies include equipment failure, traffic congestion, and special events. For detected anomalies, the edge computing node generates an anomaly report, including the anomaly type, occurrence time and impact range; The cleaned and formatted data and anomaly detection reports are packaged into data packets, and the data packets are uploaded to the cloud server.

4. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 3 is characterized in that: In the cloud server, the processing results are predicted and analyzed to generate a traffic flow monitoring plan for the toll station, including lane opening or closing suggestions, lane ratio adjustment, and emergency response measures for special events, including: The cloud server receives the processing results from the edge computing node, including cleaned traffic flow data and anomaly detection reports; Use time series prediction models to predict traffic flow data, identify traffic flow trends and peak hours, and obtain prediction and congestion analysis results; According to the prediction and congestion analysis results, the lane opening or closing status, lane ratio adjustment plan and emergency response measures under special events are used as optimization variables to initialize an artificial fish swarm, where each fish represents a traffic flow monitoring plan for a toll station. Define an objective function to quantitatively evaluate the comprehensive performance of each toll station's traffic flow monitoring solution under given traffic flow prediction results and congestion analysis conditions, including traffic efficiency, resource utilization, and responsiveness to special events; During the iteration process of the artificial fish school, each fish forages, clusters and rear-ends according to its current position and objective function value. When the preset number of iterations is reached, the final traffic flow monitoring plan for the toll station is obtained, including suggestions for lane opening or closing, lane ratio adjustment plans and emergency response measures under special events.

5. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 4 is characterized in that: The calculation formula of the objective function is: ； in, Representation scheme The objective function value of , , , , , , , represents the weight coefficient; Indicated in the plan The total number of vehicles passing through the toll booth; Representation scheme The average speed of vehicles passing through the toll booth; Indicated in the plan the number of lanes available at the lower toll booth; Indicates the time period hours; Indicated in the plan The effective length of each lane below; Indicated in the plan The number of lanes actually opened; Indicated in the plan The operating costs required for each lane to be opened; Indicates the total number of lanes; Indicated in the plan the total operating expenses of the lower toll booth; Indicated in the plan Average response time to special events at lower toll booths; Indicated in the plan The duration of the special event; Indicated in the plan The number of lanes affected by the special event; Indicated in the plan The degree of traffic congestion caused by special events; Indicated in the plan The total number of vehicles expected to pass through the toll booth; Indicated in the plan The number of vehicles or the length of the queue before the toll booth; Indicates the maximum queue length.

6. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 5 is characterized in that: The traffic flow monitoring solution is sent to the edge computing node and the on-site control system of the toll station. The edge computing node adjusts the lane configuration and signal light control of the toll station in real time according to the traffic flow monitoring solution, including: The edge computing node analyzes the traffic flow monitoring solution and extracts key information, including lane opening or closing instructions, lane direction adjustment requirements, and signal light control parameters; Using traffic flow monitoring equipment, including cameras and geomagnetic sensors, edge computing nodes collect and process traffic flow data in real time to obtain the real-time status of current traffic flow; Determine the final lane configuration and signal light control plan based on the traffic flow monitoring plan and the real-time status of the current traffic flow; The edge computing node sends the control instructions of the lane configuration and signal light control scheme to the toll station on-site control system, and performs corresponding lane configuration adjustments and signal light control operations, including physically opening or closing lane gates, adjusting lane signs, and changing signal light display status.

7. The method for monitoring vehicle flow at a toll station based on cloud-edge digital twin according to claim 6 is characterized in that: The actual operation effect after remote control is fed back to the digital twin model. The digital twin model compares and analyzes the actual operation status based on the actual operation effect feedback. Through continuous interaction and iteration, the vehicle flow monitoring solution is continuously optimized, including: The digital twin model simulates the actual operating status of the toll station based on the real-time feedback effect data and the actual operating data of the toll station, including the arrival distribution of vehicles, the use of lanes and the control logic of traffic lights, and generates virtual operating status data corresponding to the actual operating status; Compare the feedback effect data with the virtual operation status data generated in the digital twin model, including comparing the improvement in lane traffic efficiency, the reduction in vehicle delay time, and the change in queue length, identifying the differences and trends between the feedback effect data and the virtual operation status data, and obtaining the results of the comparative analysis; Based on the results of the comparative analysis, the digital twin model continues to interact with the physical toll station, and the traffic flow monitoring solution is continuously iterated and optimized.

8. A cloud-edge toll station traffic flow monitoring system based on digital twins, the system implements the method as described in any one of claims 1 to 7, characterized in that: include: Data collection module, used to capture the actual operation data of toll stations in real time, including vehicle arrival time, lane usage, equipment status, lane efficiency, vehicle delay duration and queue length traffic flow information; The edge computing processing module is used to perform preliminary sorting and abnormal identification analysis on the actual operation data obtained by the data collection module, obtain the analysis results, and upload the analysis results to the cloud server; The cloud service management module is used to build a digital twin model corresponding to the physical toll station, receive and process the data uploaded by the edge computing processing module to generate a traffic flow monitoring plan, and send the traffic flow monitoring plan to the edge computing processing module and the toll station on-site control system; Field control execution module, used to adjust the lane layout and signal light control logic of the toll station in real time according to the traffic flow monitoring plan; Remote monitoring and control module, which is used to remotely monitor the real-time operating status and traffic flow of toll stations through the cloud platform, and supports remote control operations, including adjustment of the number of lanes open and modification of toll rules; The effect feedback and optimization iteration module is used to transmit the actual operation data after remote control back to the digital twin model, and compare and analyze it with the actual operation status, so as to continuously iterate and optimize the traffic flow monitoring solution.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images