Expressway travel time reliability analysis method and system based on toll collection data

Abstract

The invention discloses a highway travel time reliability analysis method and system based on toll collection data, and belongs to the technical field of traffic, and the method comprises the steps: firstly obtaining vehicle passing data of a key road section through an edge gateway, obtaining a cache optimization index of an edge node through analysis, and dynamically adjusting the cache depth according to the cache optimization index; the adjusted cache depth is used for extracting historical travel data in the same time period, and analyzing to obtain a confidence level interference trigger value of the vehicle charging data. And judging the confidence level of the data according to the value, and adjusting the width of a confidence interval so as to improve the reliability of the data. And then, the edge gateway aggregates and sends charging data to the cloud in batches, monitors network resource data, and analyzes to obtain a network state characterization value. According to the value, the data packet of the charging data is screened and reissued, so that the integrity and timeliness of data transmission are ensured, the processing and transmission strategy of the charging data is optimized, and the accuracy and efficiency of travel time reliability analysis can be improved subsequently.

Description

Technical Field

[0001] The present invention relates to the field of transportation technologies, and in particular, to a method and system for analyzing the reliability of highway travel time based on toll data. Background Art

[0002] Massive historical toll data is stored in the networked toll collection centers at all levels of highways. The toll data contains a large amount of valuable information. By mining this data, in addition to providing decisions for regional road network OD trip distribution, spatio-temporal changes in traffic demand, road network congestion analysis and reconstruction and expansion, etc., information such as road network operation status and travel time prediction can also be mined to provide more potential value information for the public.

[0003] For example, the patent with the publication number: CN118644978A discloses a method and system for predicting highway travel time under sparse data conditions, belonging to the field of highway management and operation technologies. It obtains highway vehicle trajectory data and calculates the spatio-temporal speed field under real traffic conditions; constructs a physics-informed deep learning model to estimate the spatio-temporal speed field; constructs an STA_LSTM model to predict highway travel time. The present invention constructs a physics-informed deep learning model for the estimation and reconstruction of the spatio-temporal speed field.

[0004] For example, the patent with the publication number: CN118658301A discloses a method for guiding lane change of highway tunnel vehicles based on a travel time prediction model, including: S1. Obtaining the trajectory data of intelligent connected vehicles in the tunnel and preprocessing the trajectory data; S2. Conducting a correlation analysis on travel time features, extracting travel time features with a detection period of Tpre seconds, normalizing the data, and constructing a travel time prediction data set; S3. Building a travel time prediction model of PSO-CNN-LSTM-AM, inputting the travel time prediction data set into the model for training, and then conducting experimental comparison and verification; S4. Using the model trained in step S3 for periodic prediction, and based on the travel time prediction value, proposing a lane change control method for the entrance area of the highway tunnel.

[0005] However, in the process of implementing the technical solutions of the present invention in the embodiments of the present application, it is found that the above technologies have at least the following technical problems: Data transmission often faces problems such as buffer overflow and burst traffic congestion, high latency and packet loss caused by network fluctuations, insufficient sensitivity in detecting abnormal events, and low cloud collaboration efficiency, which may lead to buffer bloat, that is, the data packets queue in the buffer for too long, resulting in high latency and decreased throughput. Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention provides a method and system for analyzing the reliability of highway travel time based on toll data, which solves the problems involved in the above-mentioned background art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for analyzing the reliability of highway travel time based on toll data includes obtaining vehicle passing data of key sections in the edge gateway, analyzing the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node, and adjusting and updating the cache depth of the edge node according to the cache optimization index of the edge node.

[0008] Extract the historical travel data of the key section in the same time period according to the updated cache depth of the edge node and analyze it to obtain the confidence level interference trigger value of the vehicle toll data in the key section.

[0009] Judge the confidence level of the vehicle toll data in the key section according to the confidence level interference trigger value of the vehicle toll data in the key section, and adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle toll data in the key section.

[0010] The edge gateway batches and aggregates the toll data and sends it to the cloud. At the same time, it monitors and obtains the network resource data of the edge node and analyzes it to obtain the network state characterization value of the network resource data of the edge node. Sieve the data packets of the toll data according to the network state characterization value of the network resource data of the edge node, and reissue the data packets of the toll data.

[0011] Further, analyzing the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node, the specific process is as follows: The vehicle passing data of the edge gateway includes the average arrival rate of data in the toll system, the average occupancy rate of the current queue, and the average remaining storage capacity of the node, and obtain the reference arrival rate of data in the toll system, the reference occupancy rate of the current queue, and the reference remaining storage capacity of the node stored in the database.

[0012] Perform ratio analysis on the deviation amounts between the average arrival rate of data in the toll system and the reference arrival rate of data in the toll system, the average occupancy rate of the current queue and the reference occupancy rate of the current queue, and the average remaining storage capacity of the node and the reference remaining storage capacity of the node respectively with their corresponding reference values, and introduce a weight coefficient and then couple them to obtain the cache optimization index of the edge node. The cache optimization index of the edge node is used to evaluate the resource matching efficiency of the edge node for the dynamic data processing requirements.

[0013] Further, the cache depth of the edge node is adjusted and updated according to the cache optimization index of the edge node. The specific process is as follows: Extract the cache optimization index of the edge node during a preset monitoring period and compare it with the set cache optimization index threshold of the edge node. If the cache optimization index of the edge node is less than or equal to the cache optimization index threshold of the edge node, the cache depth of the edge node is adjusted and updated. If the cache optimization index of the edge node is higher than the cache optimization index threshold of the edge node, there is no need to adjust the cache depth of the edge node, and the cache depth of the edge node is directly updated.

[0014] Further, obtain the confidence level interference trigger value of the vehicle toll data in the critical section. The specific process is as follows: The historical travel data of the critical section in the same time period, including the standard deviation of the average travel time, the rainfall in the adjacent period, the traffic flow, and the average vehicle speed.

[0015] Analyze the standard deviation of the average travel time and the set reference average travel time standard deviation, the rainfall in the adjacent period and the adjacent period rainfall boundary value, the traffic flow and the reference traffic flow, and the average vehicle speed and the reference average vehicle speed with their corresponding reference values respectively, and introduce a weight coefficient and a lower limit truncation and then perform coupling to obtain the confidence level interference trigger value of the vehicle toll data in the critical section. The confidence level interference trigger value of the vehicle toll data in the critical section is used to evaluate the abnormal interference influence intensity of the toll data in the dynamic traffic environment.

[0016] Further, judge the confidence level of the vehicle toll data in the critical section according to the confidence level interference trigger value of the vehicle toll data in the critical section. The specific process is as follows: Extract the confidence level interference trigger value of the vehicle toll data in the critical section and match it with the confidence interval shrinkage amount corresponding to each interval of the set confidence level interference trigger value of the vehicle toll data in the critical section to obtain the confidence interval shrinkage amount of the vehicle toll data in the critical section.

[0017] Further, adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle toll data in the critical section. The specific process is as follows: Extract the confidence interval shrinkage amount of the vehicle toll data in the critical section during a preset monitoring time period, and extract the confidence interval of the current vehicle toll data from the database. Shrink the confidence interval of the current vehicle toll data according to the confidence interval shrinkage amount of the vehicle toll data in the critical section and update it.

[0018] Further, obtain the network state characterization value of the network resource data of the edge node. The specific process is as follows: Extract the network resource data of the edge node during a preset monitoring time period, including the bandwidth utilization rate, the round-trip delay, and the cache hit rate.

[0019] The deviation amounts between the bandwidth utilization rate and the reference bandwidth utilization rate, the round-trip delay and the reference round-trip delay, and the cache hit rate and the reference cache hit rate are respectively proportionally analyzed with their corresponding reference values. After combining the confidence interval shrinkage amount of the vehicle toll data and introducing a weight coefficient, they are coupled to obtain the network state characterization value of the network resource data of the edge node. The network state characterization value of the network resource data of the edge node is used to evaluate the comprehensive operation efficiency of the edge computing node.

[0020] Further, the data packets of the toll data are screened according to the network state characterization value of the network resource data of the edge node. The specific process is as follows: extract the network state characterization value of the network resource data of the edge node, and compare it with the set network state characterization threshold of the network resource data of the edge node. If the network state characterization value of the network resource data of the edge node is less than the network state characterization threshold of the network resource data of the edge node, then extract the confidence level interference trigger value of the vehicle toll data. If the confidence level interference trigger value of the vehicle toll data is higher than the preset confidence level interference trigger value of the vehicle toll data, then mark the data packet of the toll data corresponding to the confidence level interference trigger value of the vehicle toll data as a high-priority data packet.

[0021] Further, the data packets of the toll data are reissued. The specific process is as follows: in the preset monitoring time period, if the network state characterization value of the network resource data of the edge node is higher than or equal to the network state characterization threshold of the network resource data of the edge node, then reissue the high-priority data packets.

[0022] The second aspect of the present invention also provides a highway travel time reliability analysis system based on toll data, including a key section data collection and cache optimization module, which is used to obtain the vehicle passing data of the key section in the edge gateway, analyze the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node, and adjust and update the cache depth of the edge node according to the cache optimization index of the edge node.

[0023] A historical travel data trigger analysis module, which is used to extract and analyze the historical travel data of the key section in the same time period according to the updated cache depth of the edge node to obtain the confidence level interference trigger value of the vehicle toll data in the key section.

[0024] A confidence interval dynamic calibration module, which is used to judge the confidence level of the vehicle toll data in the key section according to the confidence level interference trigger value of the vehicle toll data in the key section, and adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle toll data in the key section.

[0025] Cloud aggregation transmission and adaptive retransmission module, which is used to batch-aggregate and send toll data from the edge gateway to the cloud, and at the same time monitor and obtain the network resource data of the edge node, analyze it to obtain the network state characterization value of the network resource data of the edge node, screen the data packets of the toll data according to the network state characterization value of the network resource data of the edge node, and retransmit the data packets of the toll data. One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: The present invention has the following beneficial effects: (1) The method and system for analyzing the reliability of highway travel time based on toll data provided by the present invention first obtain the vehicle passing data of the key section through the edge gateway, analyze to obtain the cache optimization index of the edge node, and dynamically adjust the cache depth accordingly. The adjusted cache depth is used to extract the historical travel data of the same time period, and analyze to obtain the confidence level interference trigger value of the vehicle toll data. According to this value, the confidence level of the data is judged, and the width of the confidence interval is adjusted to improve the reliability of the data. Subsequently, the edge gateway batch-aggregates and sends the toll data to the cloud, and at the same time monitors the network resource data, and analyzes to obtain the network state characterization value. According to this value, the data packets of the toll data are screened and retransmitted to ensure the integrity and timeliness of data transmission, optimize the processing and transmission strategy of the toll data, and help improve the accuracy and efficiency of subsequent travel time reliability analysis.

[0026] (2) By obtaining the cache optimization index of the edge node, the present invention helps the edge gateway to adaptively optimize the cache strategy, which can not only expand the storage and relieve the queuing pressure in time under the condition of data volume surge or heterogeneous network environment, and finally provide high-quality and low-latency data support for the subsequent analysis of the reliability of highway travel time based on toll data, effectively improve the accuracy and stability of travel time prediction, and at the same time reduce the risk of network congestion and data loss.

[0027] (3) By obtaining the confidence level interference trigger value of the vehicle toll data in the key section, the present invention enables the confidence interval of the travel time to improve the detection sensitivity by fusing statistical fluctuations and external disturbance factors in real time, thus significantly improving the accuracy and stability of the analysis of the reliability of highway travel time based on toll data.

[0028] (4) By obtaining the network status characterization value of the network resource data of the edge node, the records with trigger values higher than the threshold are marked as high-priority data packets and written into the edge elastic buffer; when the network status recovers or is excellent, that is, the characterization value is higher than the threshold, the system batches and reissues the previously marked high-priority data packets through a high-reliability channel, so as to ensure that the critical abnormal trip records can be transmitted to the cloud in a timely and complete manner. It can automatically delay the transmission of low-priority data during network fluctuations, reserve sufficient bandwidth for the timely reissue of critical data, and actively resume data synchronization after the network conditions improve, minimizing the packet loss rate and retransmission delay, and significantly improving the capture ability of abnormal events and the overall prediction accuracy in the travel time reliability analysis based on toll data. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the method of the present invention; Figure 2 It is a schematic diagram of the system of the present invention; Figure 3 It is a schematic diagram of the logical flow of the method for analyzing the travel time reliability of expressways based on toll data of the present invention; Figure 4 It is a schematic diagram of the process for screening data packets of toll data of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0031] Please refer to Figure 1 , the embodiments of the present invention provide a technical solution: a method for analyzing the travel time reliability of expressways based on toll data, including obtaining the vehicle passing data of key sections in the edge gateway, analyzing the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node, and adjusting and updating the cache depth of the edge node according to the cache optimization index of the edge node.

[0032] Extract the historical travel data of the key section in the same time period according to the updated cache depth of the edge node and analyze it to obtain the confidence level interference trigger value of the vehicle toll data in the key section.

[0033] Judge the confidence level of the vehicle toll data in the key section according to the confidence level interference trigger value of the vehicle toll data in the key section, and adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle toll data in the key section.

[0034] The edge gateway aggregates and sends toll data to the cloud in batches, while monitoring and obtaining the network resource data of edge nodes and analyzing it to obtain the network status characterization value of the network resource data of edge nodes. The data packets of toll data are screened according to the network status characterization value of the network resource data of edge nodes, and the data packets of toll data are reissued.

[0035] Specifically, analyzing the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node. The specific process is as follows: The vehicle passing data of the edge gateway includes the average arrival rate of data in the toll system, the average occupancy rate of the current queue, and the average remaining storage capacity of the node, and obtains the reference arrival rate of data in the toll system, the reference occupancy rate of the current queue, and the reference remaining storage capacity of the node stored in the database.

[0036] It should be noted that the average arrival rate of data is obtained by counting the number of messages per second reported in real time by the traffic counter built into the gateway monitoring the ingress network interface. The average occupancy rate of the current queue is obtained by periodically reading the ratio of the real-time depth of the gateway message queue to its maximum capacity through the gateway management API. For example, if the system processes 10 requests per second and 15 requests arrive per second on average, then the average occupancy rate of the queue is 1.5; the average remaining storage capacity of the node is obtained by querying the available disk space of the local file system or the edge database through the operating system command.

[0037] Perform ratio analysis on the deviation amounts between the average arrival rate of data in the toll system and the reference arrival rate of data in the toll system, the average occupancy rate of the current queue and the reference occupancy rate of the current queue, and the average remaining storage capacity of the node and the reference remaining storage capacity of the node with their corresponding reference values respectively, and perform coupling after introducing the weight coefficient to obtain the cache optimization index of the edge node. The cache optimization index of the edge node is used to evaluate the resource matching efficiency of the edge node for the dynamic data processing requirements.

[0038] It should be noted that the arrival rate of data in the toll system is the rate at which new data packets enter the queue per unit time; the occupancy rate of the current queue is the ratio of the number of messages to be processed in the queue to the maximum capacity of the queue.

[0039] It should be noted that the cache optimization index of the edge node, the specific analysis conditions are: ; In the formula, represents the cache optimization index of the edge node represents the average arrival rate of data in the toll system, represents the reference arrival rate of data in the toll system, represents the average occupancy rate of the current queue, Indicates the occupancy rate of the current queue reference Indicates the average remaining storage capacity of the node Indicates the reference remaining storage capacity of the node Indicates the weight coefficient corresponding to the data arrival rate of the set charging system Indicates the weight coefficient corresponding to the average occupancy rate of the current queue set Indicates the weight coefficient corresponding to the remaining storage capacity of the set node

[0040] It should be noted that during the data transmission process of the charging system, there is a close correlation among the three parameters: the average data arrival rate, the average occupancy rate of the current queue, and the average remaining storage capacity of the node. The average data arrival rate reflects the amount of data entering the system per unit time, the average occupancy rate of the current queue represents the amount of data waiting to be processed in the system, and the average remaining storage capacity of the node indicates the available storage resources of the system. When the average data arrival rate increases, the data in the queue may accumulate, resulting in an increase in the average occupancy rate of the current queue. And as the queue grows, the system requires more storage space to cache this data, thus consuming the remaining storage capacity of the node. Therefore, these three parameters jointly affect the cache optimization index of the system. By real-time monitoring and analyzing the change trends of these parameters, the system can dynamically adjust the cache strategy, optimize resource allocation, and improve data processing efficiency.

[0041] In a specific embodiment, the value ranges of the weight coefficient corresponding to the data arrival rate of the charging system, the weight coefficient corresponding to the average occupancy rate of the current queue, and the weight coefficient corresponding to the remaining storage capacity of the node are usually set between 0 and 1. First, by establishing a mapping table between the weight coefficient corresponding to the data arrival rate of the charging system and the weight coefficient, the weight coefficient corresponding to the data arrival rate of the charging system measured in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the weight coefficient of the data arrival rate of the charging system; at the same time, for the average occupancy rate of the current queue, by constructing a mapping table between the average occupancy rate of the current queue and the weight coefficient, the average occupancy rate of the current queue detected in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the average occupancy rate of the current queue; for the remaining storage capacity of the node, also through the pre-established mapping table between the remaining storage capacity of the node and the weight coefficient, the remaining storage capacity of the node measured in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the remaining storage capacity of the node.

[0042] Specifically, the cache depth of the edge node is adjusted and updated according to the cache optimization index of the edge node. The specific process is as follows: extract the cache optimization index of the edge node during a preset monitoring period and compare it with the set cache optimization index threshold of the edge node. If the cache optimization index of the edge node is less than or equal to the cache optimization index threshold of the edge node, then adjust and update the cache depth of the edge node. If the cache optimization index of the edge node is higher than the cache optimization index threshold of the edge node, there is no need to adjust the cache depth of the edge node, and directly update the cache depth of the edge node.

[0043] It should be noted that the cache depth of the edge node is initially declared by the system administrator or the operation and maintenance script through the configuration file or the platform management interface during the node deployment phase and persisted in the local storage or the queue attributes of the message broker. When directly updating the cache depth of the edge node, the system does not change the value of the cache depth, but directly updates the cache depth by rewriting the original configuration to ensure the consistency of the runtime configuration with the backend storage or the management interface.

[0044] It should be noted that the steps for adjusting and updating the cache depth of the edge node are as follows: during the monitoring period, when the cache optimization index is lower than the threshold for the first time, increase the cache queue length by 20% to 30% to provide sufficient buffer space to cope with sudden data traffic growth. And re-obtain the cache optimization index of the edge node. If after the initial adjustment, the cache optimization index of the edge node is still less than or equal to the cache optimization index of the edge node, the cache queue length can continue to be increased step by step at a rate of 10% until the maximum capacity limit of the system is reached.

[0045] Specifically, obtain the confidence level interference trigger value of the vehicle toll data in the key section. The specific process is as follows: the historical travel data of the key section in the same time period, including the standard deviation of the average travel time, the rainfall in the adjacent time period, the traffic flow, and the average vehicle speed.

[0046] It should be noted that the standard deviation of the average travel time is obtained by collecting the vehicle passing times of the same section multiple times (using GPS or geomagnetic sensors to record the time difference between the vehicle passing the starting point and the ending point). The rainfall in the adjacent time period is usually automatically recorded by the rain gauge of the weather station for the cumulative precipitation per unit time. For example, if a rain gauge measures 15mm and 20mm respectively in two adjacent hours, the total rainfall in the two time periods is 35mm; the traffic flow is counted in real time by the loop coil or the video recognition system for the number of vehicles passing through per unit time. For example, the loop coil at a certain intersection detects the passing data of 1200 vehicles within 1 hour during the morning rush hour; the average vehicle speed is deduced based on the time taken by the vehicle to pass through a known distance section measured by a fixed detection device. For example, if a section is 1 kilometer long and vehicle A takes 1 minute to pass through, the speed is 60 kilometers per hour.

[0047] The standard deviation of the average travel time is analyzed with the set reference standard deviation of the average travel time, the rainfall in the adjacent period is analyzed with the defined value of the rainfall in the adjacent period, the traffic flow is analyzed with the reference traffic flow, and the average vehicle speed is analyzed with the reference average vehicle speed respectively. Then, the weight coefficient and the lower limit truncation are introduced and coupled to obtain the confidence level interference trigger value of the vehicle toll data in the key section. The confidence level interference trigger value of the vehicle toll data in the key section is used to evaluate the intensity of the abnormal interference of the toll data in the dynamic traffic environment.

[0048] It should be noted that the specific analysis conditions for the confidence level interference trigger value of the vehicle toll data are as follows: ; In the formula, represents the confidence level interference trigger value of the vehicle toll data represents the standard deviation of the average travel time, represents the set reference standard deviation of the average travel time, represents the rainfall in the adjacent period, represents the defined value of the rainfall in the adjacent period, represents the traffic flow, represents the reference traffic flow, represents the average vehicle speed, represents the set reference average vehicle speed, represents the weight coefficient corresponding to the set standard deviation of the average travel time, represents the weight coefficient corresponding to the set rainfall in the adjacent period, represents the weight coefficient corresponding to the set traffic flow, represents the weight coefficient corresponding to the set average vehicle speed.

[0049] It should be noted that for the four core parameters of the standard deviation of the average travel time, the rainfall in the adjacent period, the traffic flow, and the average vehicle speed, their measured values are respectively compared with the preset reference thresholds. When the actual value of any parameter exceeds the corresponding reference value, the excess ratio (i.e., the ratio of the actual value divided by the reference value minus 1) is obtained, and the positive value of the over-limit part is taken and truncated at zero lower limit, which helps to sensitively capture the abnormal fluctuations of the vehicle toll data under extreme weather or high flow conditions.

[0050] It should be noted that in the analysis of the reliability of highway travel time, there is a close correlation among the four parameters of the standard deviation of the average travel time, the rainfall in the adjacent period, the traffic flow, and the average vehicle speed. They jointly affect the stability of road traffic and the accuracy of the analysis of the reliability of highway travel time. First of all, the standard deviation of the average travel time reflects the degree of fluctuation of the vehicle travel time. When the rainfall increases, the road surface becomes slippery and the visibility decreases, resulting in a decrease in the average vehicle speed and possibly causing traffic congestion, which in turn increases the volatility of the traffic flow. Secondly, the change in traffic flow will also affect the stability of the average vehicle speed and travel time. The reduction of the distance between vehicles will lead to the instability of the average vehicle speed, thereby increasing the fluctuation of the travel time. Finally, the change in the average vehicle speed directly affects the stability of the travel time. Under adverse weather conditions such as rainfall, the average vehicle speed usually decreases, which not only prolongs the travel time but also increases the volatility of the travel time. In a specific embodiment, the value ranges of the weight coefficients corresponding to the standard deviation of the average travel time, the weight coefficient corresponding to the rainfall in the adjacent period, the weight coefficient corresponding to the traffic flow, and the weight coefficient corresponding to the average vehicle speed are usually set between 0 and 1. First, through the mapping table between the standard deviation of the average travel time and the weight coefficient, the real-time measured standard deviation of the average travel time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the standard deviation of the average travel time; at the same time, for the rainfall in the adjacent period, by constructing the mapping table between the rainfall in the adjacent period and the weight coefficient, the real-time detected rainfall in the adjacent period is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the rainfall in the adjacent period; for the traffic flow and the average vehicle speed, through the pre-established mapping tables between the traffic flow and the weight coefficient and between the average vehicle speed and the weight coefficient, the traffic flow is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the traffic flow and the weight coefficient corresponding to the average vehicle speed.

[0051] Specifically, the confidence level of the vehicle toll data in the critical section is judged according to the confidence level interference trigger value pair of the vehicle toll data in the critical section. The specific process is as follows: Extract the confidence level interference trigger value of the vehicle toll data in the critical section, and match it with the confidence interval shrinkage amount corresponding to each interval of the set confidence level interference trigger value of the vehicle toll data in the critical section to obtain the confidence interval shrinkage amount of the vehicle toll data in the critical section.

[0052] Specifically, the width of the confidence interval is adjusted according to the confidence level interference trigger value of the vehicle toll data in the critical section. The specific process is as follows: extract the confidence interval shrinkage amount of the vehicle toll data in the critical section during the preset monitoring time period, and extract the confidence interval of the current vehicle toll data from the database. Shrink the confidence interval of the current vehicle toll data according to the confidence interval shrinkage amount of the vehicle toll data in the critical section and update it.

[0053] It should be noted that shrinking and updating the confidence interval of the current vehicle toll data according to the confidence interval shrinkage amount of the vehicle toll data in the critical section specifically involves calling the upper and lower limits of the confidence interval of the current travel time of this section saved in the database, and then shrinking them proportionally according to the confidence interval shrinkage amount of the vehicle toll data in the critical section (for example, tightening the percentage of the shrinkage amount for both the upper and lower limits at the same time), and then rewriting the updated confidence interval into the storage through a unified database write interface.

[0054] Specifically, the network state characterization value of the network resource data of the edge node is obtained. The specific process is as follows: extract the network resource data of the edge node during the preset monitoring time period, including bandwidth utilization rate, round-trip delay, and cache hit rate.

[0055] It should be noted that the bandwidth utilization rate collects the actual transmitted data volume in real time through a traffic monitoring tool and compares it with the preset physical bandwidth upper limit, and finally reflects the network resource utilization efficiency in the form of a percentage; the round-trip delay is achieved by sending test packets and recording the time difference. For example, using the timestamp marking in the TCP protocol to measure the time interval from the packet sending to receiving and returning; the cache hit rate is obtained by counting the proportion of the number of successful cache responses to the total number of requests. Specifically, the system will record each data access request initiated by the user. If the data required for this request already exists in the cache (i.e., a hit), the hit count is incremented by one; if it needs to be retrieved again from the database or other back-end storage (i.e., a miss), the miss count is incremented by one. Finally, divide the hit count by the total number of requests (the sum of the hit count and the miss count) and convert the result to a percentage form to obtain the cache hit rate.

[0056] Perform ratio analysis on the deviation amounts between the bandwidth utilization rate and the reference bandwidth utilization rate, the round-trip delay and the reference round-trip delay, and the cache hit rate and the reference cache hit rate respectively with their corresponding reference values, combine with the confidence interval shrinkage amount of the vehicle toll data, and introduce a weight coefficient for coupling to obtain the network state characterization value of the network resource data of the edge node. The network state characterization value of the network resource data of the edge node is used to evaluate the comprehensive operation efficiency of the edge computing node.

[0057] It should be noted that the network state characterization value of the network resource data of the edge node, the specific analysis conditions are: ; In the formula, represents the network state characterization value of the network resource data of the edge node represents the bandwidth utilization rate, represents the reference bandwidth utilization rate, represents the round-trip delay, represents the set reference round-trip delay, represents the cache hit rate, represents the set reference cache hit rate, represents the confidence interval shrinkage of vehicle toll data, represents the weight coefficient corresponding to the set bandwidth utilization rate, represents the weight coefficient corresponding to the set round-trip delay, represents the weight coefficient corresponding to the set cache hit rate, represents the weight coefficient corresponding to the set confidence interval shrinkage of vehicle toll data.

[0058] It should be noted that during the data transmission process of the toll system, there is a close correlation among the bandwidth utilization rate, round-trip delay, cache hit rate, and the confidence interval shrinkage of vehicle toll data. They jointly affect the stability and efficiency of the toll data transmission system. The bandwidth utilization rate reflects the degree of network resource utilization. When the bandwidth utilization rate approaches or exceeds the network capacity, data transmission may be restricted, resulting in an increase in round-trip delay. The increase in round-trip delay affects the timely transmission of data, which may in turn lead to data backlog in the cache, affecting the cache hit rate. The cache hit rate represents the proportion of data requests in the system that can be directly obtained from the cache. A high cache hit rate can reduce the dependence on the network and lower the round-trip delay. When the confidence interval shrinkage increases, that is, the prediction accuracy improves, the system can more accurately identify key data, and the cache hit rate will increase accordingly. Therefore, these four parameters interact with each other to form a dynamic feedback system. By real-time monitoring and adjustment, the data transmission strategy can be optimized, and the overall performance of the system can be improved.

[0059] In a specific embodiment, the weight coefficient corresponding to the bandwidth utilization, the weight coefficient corresponding to the round-trip delay, the weight coefficient corresponding to the cache hit rate, and the weight coefficient corresponding to the confidence interval contraction of the vehicle charging data are usually set in the range of 0 to 1. First, by establishing a mapping table between the bandwidth utilization and the weight coefficient, the bandwidth utilization measured in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the bandwidth utilization; at the same time, for the round-trip delay, by constructing a mapping table between the round-trip delay and the weight coefficient, the round-trip delay detected in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the round-trip delay; for the cache hit rate, by also using the mapping table between the cache hit rate and the weight coefficient established in advance, the cache hit rate is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the cache hit rate; similarly, for the weight coefficient corresponding to the confidence interval contraction of the vehicle charging data, by also establishing a mapping table between the confidence interval contraction of the vehicle charging data and the weight coefficient, the confidence interval contraction of the vehicle charging data obtained in real time is input into the preset mapping table, so as to quickly obtain the weight coefficient corresponding to the confidence interval contraction of the vehicle charging data.

[0060] Specifically, the data packets of the charging data are screened according to the network status characterization value of the network resource data of the edge node. The specific process is: extract the network status characterization value of the network resource data of the edge node, and compare it with the set network status characterization threshold of the network resource data of the edge node. If the network status characterization value of the network resource data of the edge node is less than the network status characterization threshold of the network resource data of the edge node, then the confidence level interference trigger value of the vehicle charging data is extracted. If the confidence level interference trigger value of the vehicle charging data is higher than the preset confidence level interference trigger value of the vehicle charging data, then the data packet of the charging data corresponding to the confidence level interference trigger value of the vehicle charging data is marked as a high priority data packet.

[0061] Specifically, the data packet of the charging data is reissued, and the specific process is: in the preset monitoring time period, if the network status characterization value of the network resource data of the edge node is higher than or equal to the network status characterization threshold of the network resource data of the edge node, the high priority data packet is reissued.

[0062] It should be noted that by dynamically sensing the network resource status of edge nodes and giving priority to reissuing high-priority data packets when the network is congested or abnormal, it helps to ensure the integrity and continuity of charging data, reduce the loss of core data due to network fluctuations, and avoid deviations in travel time reliability analysis due to data gaps.

[0063] like Figure 2As shown in the figure, the second aspect of the present invention further provides a highway travel time reliability analysis system based on toll data, including a key section data collection and cache optimization module, which is used to obtain the vehicle passing data of the key section in the edge gateway, analyze the vehicle passing data of the edge gateway to obtain the cache optimization index of the edge node, and adjust and update the cache depth of the edge node according to the cache optimization index of the edge node.

[0064] A historical travel data trigger analysis module, which is used to extract the historical travel data of the key section in the same time period according to the updated cache depth of the edge node and perform analysis to obtain the confidence level interference trigger value of the vehicle toll data in the key section.

[0065] A confidence interval dynamic calibration module, which is used to judge the confidence level of the vehicle toll data in the key section according to the confidence level interference trigger value of the vehicle toll data in the key section, and adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle toll data in the key section.

[0066] A cloud aggregation transmission and adaptive retransmission module, which is used to batch aggregate and send toll data from the edge gateway to the cloud, and at the same time monitor and obtain the network resource data of the edge node and perform analysis to obtain the network state characterization value of the network resource data of the edge node. Screen the data packets of the toll data according to the network state characterization value of the network resource data of the edge node, and retransmit the data packets of the toll data.

[0067] It should be noted that the highway travel time reliability analysis system based on toll data further includes a database, which is used to store data such as the cache optimization index threshold of the edge node, the network state characterization threshold of the network resource data of the edge node, and the network state characterization threshold of the network resource data of the edge node.

[0068] It should be noted that Figure 3It is a schematic diagram of the logical process of the freeway travel time reliability analysis method based on toll data according to the present invention. First, real-time analysis capabilities are introduced at the edge layer. The resource matching degree of nodes to dynamic traffic is measured by the cache optimization index, and the cache depth is adaptively adjusted accordingly to avoid overflow and packet loss under sudden fluctuations, while reducing resource waste during idle periods. Then, the system continues to use the updated cache to extract the historical travel fluctuations, weather, and traffic conditions at the same time period for each road section, obtains the confidence level interference trigger value, and dynamically narrows or broadens the confidence interval of the travel time based on this to accurately characterize the data reliability and abnormal interference intensity. And in the data transmission and cloud collaboration stage, the cache adjustment and confidence analysis results are parallel. Finally, when the network condition is poor, the system classifies the data packets in combination with the aforementioned confidence trigger value, and automatically reissues the key records when the link is restored to ensure that abnormal trips are not missed; after the network is restored, the low-priority packets are re-injected.

[0069] Figure 4 It is a schematic diagram of the process of screening data packets of toll data according to the present invention. The global network health and cache efficiency of the edge computing node are combined into one to accurately evaluate the current transmission capacity. On this basis, the system is compared with a preset threshold. If the characterization value is lower than the threshold, indicating that there is network congestion or link degradation, then it enters the next abnormal data prioritization process, otherwise it directly ends, fully avoiding unnecessary processing overhead when the network status is good. It integrates the idea of active load degradation based on the service level and the QoS management and reordering buffer technology of the edge gateway, and realizes a closed-loop strategy from network status perception to priority scheduling, not only jointly adaptively controls network fluctuations and toll data, but also significantly improves the transmission reliability of subsequent key toll data and the overall accuracy of cloud travel time prediction.

[0070] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] The present invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowchart flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or blocks.

[0072] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions specified in the flowchart flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or blocks.

[0073] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and / or block or blocks. Figure 1 in a flow or flows and / or block or blocks Figure 1 or blocks.

[0074] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0075] It is apparent that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A highway travel time reliability analysis method based on toll data, characterized in that: include: Obtain vehicle traffic data on key sections of the edge gateway, analyze the vehicle traffic data of the edge gateway to obtain the cache optimization index of the edge node, and adjust and update the cache depth of the edge node according to the cache optimization index of the edge node; According to the updated cache depth of the edge node, the historical travel data of the key road section in the same period is extracted and analyzed to obtain the confidence level interference trigger value of the vehicle charging data in the key road section; Judging the confidence level of the vehicle charging data in the key section according to the confidence level interference trigger value of the vehicle charging data in the key section, and adjusting the width of the confidence interval according to the confidence level interference trigger value of the vehicle charging data in the key section; The edge gateway aggregates and sends the charging data to the cloud in batches, and at the same time monitors and obtains the network resource data of the edge nodes and analyzes it to obtain the network status representation value of the network resource data of the edge nodes. The charging data packets are screened according to the network status representation value of the network resource data of the edge nodes, and the charging data packets are resent.

2. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The vehicle traffic data of the edge gateway is analyzed to obtain the cache optimization index of the edge node, and the specific process is as follows: The vehicle traffic data of the edge gateway includes the average arrival rate of data of the toll collection system, the average occupancy rate of the current queue, and the average remaining storage capacity of the node, and obtains the reference arrival rate of data of the toll collection system, the reference occupancy rate of the current queue, and the reference remaining storage capacity of the node stored in the database; The average data arrival rate of the charging system and the reference data arrival rate of the charging system, the average occupancy of the current queue and the reference occupancy of the current queue, and the deviation between the average remaining storage capacity of the node and the reference remaining storage capacity of the node are respectively analyzed with their corresponding reference values, and the weight coefficient is introduced and coupled to obtain the cache optimization index of the edge node. The cache optimization index of the edge node is used to evaluate the resource matching efficiency of the edge node for dynamic data processing requirements.

3. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The cache depth of the edge node is adjusted and updated according to the cache optimization index of the edge node. The specific process is as follows: The cache optimization index of the edge node is extracted during the preset monitoring period and compared with the set cache optimization index threshold of the edge node. If the cache optimization index of the edge node is less than or equal to the cache optimization index threshold of the edge node, the cache depth of the edge node is adjusted and updated. If the cache optimization index of the edge node is higher than the cache optimization index threshold of the edge node, there is no need to adjust the cache depth of the edge node, and the cache depth of the edge node is directly updated.

4. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The specific process of obtaining the confidence level interference trigger value of the vehicle charging data in the key road section is as follows: The historical travel data of the key road section in the same period, including the standard deviation of the average travel time, rainfall in the adjacent period, traffic volume and average vehicle speed; The average travel time standard deviation and the set reference average travel time standard deviation, the adjacent period rainfall and the adjacent period rainfall limit value, the traffic volume and the reference traffic volume, and the average vehicle speed and the reference average vehicle speed are analyzed with their corresponding reference values ​​respectively, and the weight coefficient and the lower limit truncation are introduced and coupled to obtain the confidence level interference trigger value of the vehicle toll data in the key section. The confidence level interference trigger value of the vehicle toll data in the key section is used to evaluate the intensity of abnormal interference impact of the toll data in a dynamic traffic environment.

5. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The confidence level of the vehicle charging data in the key section is judged according to the confidence level interference trigger value of the vehicle charging data in the key section. The specific process is: The confidence level interference trigger value of the vehicle charging data in the key section is extracted, and matched with the confidence interval contraction amount corresponding to each interval of the confidence level interference trigger value of the vehicle charging data in the key section, so as to obtain the confidence interval contraction amount of the vehicle charging data in the key section.

6. The highway travel time reliability analysis method based on toll data as claimed in claim 5, characterized in that: The width of the confidence interval is adjusted according to the confidence level interference trigger value of the vehicle charging data in the key road section. The specific process is: The confidence interval contraction amount of the vehicle charging data in the key section is extracted during the preset monitoring time period, and the confidence interval of the current vehicle charging data is extracted from the database, and the confidence interval of the current vehicle charging data is contracted and updated according to the confidence interval contraction amount of the vehicle charging data in the key section.

7. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The specific process of obtaining the network status representation value of the network resource data of the edge node is as follows: Extract network resource data of edge nodes during a preset monitoring period, including bandwidth utilization, round-trip latency, and cache hit rate; The deviations between bandwidth utilization and reference bandwidth utilization, round-trip delay and reference round-trip delay, and cache hit rate and reference cache hit rate are respectively analyzed with their corresponding reference values. The confidence interval contraction of vehicle charging data is combined and the weight coefficient is introduced to couple them to obtain the network status characterization value of the network resource data of the edge node. The network status characterization value of the network resource data of the edge node is used to evaluate the comprehensive operating efficiency of the edge computing node.

8. The highway travel time reliability analysis method based on toll data as claimed in claim 7, characterized in that: The data packets of the charging data are screened according to the network status characterization value of the network resource data of the edge node, and the specific process is as follows: The network status characterization value of the network resource data of the edge node is extracted, and compared with the set network status characterization threshold of the network resource data of the edge node. If the network status characterization value of the network resource data of the edge node is less than the network status characterization threshold of the network resource data of the edge node, the confidence level interference trigger value of the vehicle charging data is extracted. If the confidence level interference trigger value of the vehicle charging data is higher than the preset confidence level interference trigger value of the vehicle charging data, the data packet of the charging data corresponding to the confidence level interference trigger value of the vehicle charging data is marked as a high priority data packet.

9. The highway travel time reliability analysis method based on toll data as claimed in claim 1, characterized in that: The specific process of reissuing the data packet of the charging data is as follows: In a preset monitoring time period, if the network status representation value of the network resource data of the edge node is higher than or equal to the network status representation threshold of the network resource data of the edge node, the high priority data packet is resent.

10. A system using the highway travel time reliability analysis method based on toll data as claimed in any one of claims 1 to 9, characterized in that: include: The key section data collection and cache optimization module is used to obtain the vehicle traffic data of the key sections in the edge gateway, analyze the vehicle traffic data of the edge gateway to obtain the cache optimization index of the edge node, and adjust and update the cache depth of the edge node according to the cache optimization index of the edge node; The historical travel data trigger analysis module is used to extract and analyze the historical travel data of the key sections in the same period according to the updated cache depth of the edge node, and obtain the confidence level interference trigger value of the vehicle charging data in the key sections; A confidence interval dynamic calibration module is used to determine the confidence level of the vehicle charging data in the key section according to the confidence level interference trigger value of the vehicle charging data in the key section, and adjust the width of the confidence interval according to the confidence level interference trigger value of the vehicle charging data in the key section; The cloud-based aggregation transmission and adaptive re-issuance module is used for the edge gateway to aggregate and send charging data to the cloud in batches, and at the same time monitor and obtain the network resource data of the edge node and analyze it to obtain the network status representation value of the network resource data of the edge node. According to the network status representation value of the network resource data of the edge node, the data packets of the charging data are screened and the data packets of the charging data are re-issued.

Images