A highway congestion prediction method and device based on big data and a medium

Abstract

This invention discloses a method, device, and medium for predicting highway congestion based on big data, relating to the field of road traffic control technology. The method includes: analyzing the traffic status of road segments and identifying the traffic relationships between road segments from multi-source traffic data, outputting road segment relationship status data; identifying bottleneck road segments in the highway from the road segment relationship status data, outputting bottleneck structure data; organizing the bottleneck road segments into paths according to their temporal impact relationships, outputting a congestion propagation path map; performing temporal extrapolation of the influence extension direction and expansion range of the bottleneck road segments along the continuous path relationships in the congestion propagation path map, outputting congestion trend data; and determining the congestion development status and propagation path direction of the highway based on the congestion trend data, outputting highway congestion prediction information. This invention, through the synergy of bottleneck structure data construction and congestion propagation path map generation, achieves clear identification of the transmission path and expansion range of highway congestion impacts.

Description

Technical Field

[0001] This invention relates to the field of road traffic control technology, and in particular to a method, device and medium for predicting highway congestion based on big data. Background Technology

[0002] In recent years, with the development of intelligent transportation IoT application services and the continuous expansion of highway operation scale, congestion monitoring and prediction methods based on multi-source traffic data have gradually become a research hotspot. In the field of road traffic control, traffic flow status is typically analyzed by collecting vehicle traffic data, road segment status data, and environmental data, and time-series data processing methods are combined to achieve early prediction of traffic congestion. Meanwhile, with the application of vehicle-road cooperation and IoT sensing methods, traffic data acquisition methods are constantly being enriched, making it possible to finely characterize highway traffic conditions. This has gradually formed a methodological system centered on traffic data collection, status analysis, and trend prediction, playing an important role in traffic guidance, travel management, and congestion mitigation.

[0003] However, existing methods have shortcomings in modeling congestion propagation relationships and utilizing spatiotemporal evolution. Most existing methods analyze traffic conditions independently based on single road segments or local areas, lacking a systematic expression of inter-segment traffic relationships. Particularly in bottleneck segment identification and characterizing the impact on adjacent road segments, they typically rely on simple correlations or empirical rules, failing to accurately reflect the propagation path and expansion process of congestion on highways. Furthermore, they lack effective organization of the continuous evolution of congestion states over time, resulting in incomplete descriptions of congestion development trends and impact ranges in the prediction results. This limits the ability of traffic management and control to accurately predict and dynamically respond to complex congestion scenarios. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a highway congestion prediction method based on big data to address the shortcomings in modeling congestion propagation relationships and utilizing spatiotemporal evolution.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, this invention provides a highway congestion prediction method based on big data, comprising: collecting multi-source traffic data of highways; analyzing the traffic status of road segments and identifying the traffic relationships between road segments in the multi-source traffic data; outputting road segment relationship status data; identifying bottleneck road segments in the highway from the road segment relationship status data; extracting the traffic status change relationships of bottleneck road segments and bottleneck-related road segments; mapping the time-series influence relationship between bottleneck-related road segments and bottleneck road segments according to the traffic status change relationships; outputting bottleneck structure data; organizing the bottleneck road segments into paths according to the time-series influence relationship based on the time-series influence relationship; outputting a congestion propagation path map; performing time-series extrapolation on the influence extension direction and influence expansion range of the bottleneck road segments along the path continuity relationship in the congestion propagation path map; outputting congestion trend data; determining the congestion development status and propagation path direction of the highway based on the congestion trend data; integrating them to form congestion prediction data; and arranging the congestion prediction data in time sequence and defining its spatial range; outputting highway congestion prediction information.

[0007] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for outputting the road segment relationship status data are as follows: Information on changes in traffic speed, vehicle queues, and road occupancy on various sections of highways is extracted from multi-source traffic data and integrated to form road segment traffic discrimination data. The traffic status data of road segments is generated by comparing the time sequence and classifying the magnitude of change of the traffic data. Based on the traffic status data of road segments, the sequential changes in traffic status and the transmission of traffic status influence between adjacent road segments are marked accordingly to form road segment relationship status data.

[0008] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for extracting the traffic status change relationship of bottleneck sections and bottleneck-related sections are as follows: Based on the road segment relationship status data, the changes in traffic status of each road segment of the expressway are continuously unfolded over time, and the traffic status change sequence is output. Based on the sequence of traffic status changes, the direction, magnitude, and duration of traffic status changes in the traffic segments are compared and determined to identify bottleneck segments. The bottleneck road segment and the road segment relationship status data are matched for time-series impact and filtered for traffic relationship, and the traffic status change relationship and bottleneck associated road segments are output.

[0009] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for outputting bottleneck structure data are as follows: Based on the traffic status change relationship, the traffic status change times between bottleneck-related road segments and bottleneck segments are sorted, and the temporal impact relationship is output. The system classifies the direction of influence and determines the scope of influence based on the temporal impact correlation, and outputs the direction of influence extension and the scope of influence expansion. The system aggregates and correlates temporal impact relationships, bottleneck-related road segments, bottleneck road segments, impact extension direction, and impact expansion range to output bottleneck structure data.

[0010] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for outputting the congestion propagation path map are as follows: The temporal impact correlations in the bottleneck structure data are organized according to the order of transmission between bottleneck road segments to generate a road segment temporal correlation sequence. The corresponding positions of the bottleneck road segments connected one after the other are identified along the temporal correlation sequence of the road segments, which serve as the road segment transmission chain data; In the road segment transmission chain data, bottleneck road segments with the same transmission direction and continuous connection are connected and combined to generate road segment path connection data. Perform route sorting operations on the road segment connectivity data to identify the road segment transmission chains of each bottleneck road segment; The congestion propagation path map is output by sequentially connecting the transmission chains of each road segment according to the continuous path relationship of the highway.

[0011] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for outputting congestion trend data are as follows: By analyzing the continuous path relationships in the congestion propagation path map, the temporal sequence of the impact extension direction of each bottleneck road segment is formed, thus creating a road segment directional extension sequence. Based on the bottleneck road segment extension sequence, the influence expansion range of each bottleneck road segment in the direction of influence extension is determined segment by segment through time series analysis, thus forming a road segment range expansion sequence. The congestion trend data is output by cross-merging the road segment direction extension sequence and the road segment range expansion sequence.

[0012] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for integrating and forming congestion prediction data are as follows: Extract the direction and extent of the impact of each bottleneck road segment from the congestion trend data, divide the direction and extent of the impact into stages, and output the congestion development status. Based on the state of congestion development, the propagation path of each bottleneck section is determined along the direction of influence extension in the congestion trend data; By linking and coupling the development status of congestion with the direction of its propagation, congestion prediction data can be generated.

[0013] As a preferred embodiment of the highway congestion prediction method based on big data described in this invention, the specific steps for outputting highway congestion prediction information are as follows: Based on the propagation path direction, influence extension direction and influence expansion range of each bottleneck road segment in the congestion prediction data, the bottleneck road segments are ranked in order of congestion occurrence to form congestion time sequence ranking data. Boundary division is performed on the congestion time-series ranking data to determine the congestion impact range of each bottleneck section on the highway. The congestion time-series ranking data and congestion impact intervals are structured and arranged to output highway congestion prediction information.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the highway congestion prediction method based on big data as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the highway congestion prediction method based on big data as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By coordinating the construction of bottleneck structure data with the generation of congestion propagation path maps, the clear identification of the transmission path and expansion range of highway congestion impacts is achieved. By mapping the temporal impact correlation between bottleneck-related road segments according to traffic state change relationships, the traffic state change relationships between road segments are transformed into structured relationships with transmission direction and impact range, providing a clear foundation for expressing traffic state correlation information. Based on the temporal impact correlation relationships in the bottleneck structure data, path continuation is organized, allowing the impact transmission between bottleneck road segments to be presented continuously along the highway path, thus providing a stable path basis for the temporal extrapolation of congestion trend data. This enables highway congestion prediction information to serve traffic state perception and traffic guidance. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a highway congestion prediction method based on big data. Figure 2 To output a flowchart of traffic status changes and bottleneck-related road segments; Figure 3 A flowchart for outputting congestion trend data; Figure 4 A flowchart for outputting highway congestion prediction information. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Reference Figures 1-4 This is one embodiment of the present invention, which provides a highway congestion prediction method based on big data, including the following steps: S1. Collect multi-source traffic data of highways, analyze the traffic status of road segments and identify the traffic relationship between road segments, and output the road segment relationship status data.

[0023] Collect multi-source traffic data of highways, extract information on changes in traffic speed, vehicle queues, and road occupancy of each section of the highway from the multi-source traffic data, and integrate them to form road segment traffic discrimination data.

[0024] Specifically, the collection of multi-source traffic data for highways involves using the Intelligent Transportation Internet of Things (IoT) to synchronously aggregate vehicle passage records, roadside sensing records, video monitoring records, toll collection records, and environmental status records across different areas, including the main highway, ramp connection areas, toll station connection areas, and service area entrance / exit connection areas. This results in multi-source traffic data covering different road segments, time locations, and status categories (multi-source traffic data includes vehicle passage records, roadside sensing records, video monitoring records, toll collection records, and environmental status records). After the multi-source traffic data is aggregated, it is time-aligned according to a unified time sequence and then grouped into road segments according to a unified road segment location correspondence, allowing the multi-source traffic data corresponding to the same road segment within the same time range to be centrally arranged.

[0025] After completing time alignment and road segment aggregation, abnormal, duplicate, and missing records in the multi-source traffic data are sequentially organized, retaining data that reflects the traffic changes on each road segment. Based on the organized multi-source traffic data, the speed changes of vehicles on each highway segment within a continuous time range are extracted to obtain traffic speed change information. The formation, growth, and decline of queues on the road segments are extracted to obtain queue change information. The distribution, increase, decrease, and duration of lane occupancy by vehicles on the road segments are extracted to obtain road segment occupancy change information. The traffic speed change information, queue change information, and road segment occupancy change information are then organized according to the road segment and time sequence, ensuring that each road segment has corresponding status discrimination basic content at each time location, forming road segment traffic discrimination data.

[0026] The traffic status data of road segments is generated by comparing the time sequence and classifying the magnitude of change of the traffic data.

[0027] Specifically, according to the arrangement of each traffic segment within a continuous time range, the information on changes in traffic speed, vehicle queues, and road occupancy at adjacent time positions of the same traffic segment is compared to determine the direction of traffic status changes. After completing the comparison, the degree of change in traffic speed, vehicle queues, and road occupancy is divided into amplitudes to form distinguishable change levels for different change processes. After dividing the change amplitudes, the change direction and amplitude are combined and organized to determine the traffic status of each traffic segment at each time position, forming traffic status data for each road segment.

[0028] Furthermore, the magnitude of changes in traffic speed, vehicle queue, and road occupancy information is categorized. This is based on the statistical analysis of the differences in these changes at adjacent time points within a continuous timeframe for the same road segment, forming corresponding change value sequences. These change value sequences are then arranged sequentially according to their numerical values, dividing them into low, medium, and high change intervals from smallest to largest. When the change value falls within the low change interval, it is classified as a weak change level; when it falls within the medium change interval, it is classified as a medium change level; and when it falls within the high change interval, it is classified as a strong change level. After classifying these levels, the direction and intensity of traffic status changes are determined by considering the positive or negative direction of the change values, thus providing a unified basis for classifying the changes in different road segments at different time points.

[0029] The change value sequence is divided into low-change, medium-change, and high-change intervals from smallest to largest. This interval division is based on the arrangement of the change value sequence. Since the change value sequence reflects the distribution of the degree of change of the same traffic segment within a continuous time range, to ensure that the division of different change levels depends on the relative position of the change value in the overall sequence rather than the specific numerical value, thus guaranteeing consistency in interval division across different traffic segments, the change value sequence is divided into multiple segments of equal quantity, with each segment corresponding to a sequence of change levels from low to high. The sorted change value sequence is then divided into equal segments according to the total quantity, with the first part of the change values ​​corresponding to the low-change interval, the middle part corresponding to the medium-change interval, and the last part corresponding to the high-change interval, so that each interval corresponds to a different change level.

[0030] Based on the traffic status data of road segments, the sequential changes in traffic status and the transmission of traffic status influence between adjacent road segments are marked accordingly to form road segment relationship status data.

[0031] Specifically, based on the spatial connectivity of each road segment in the highway network, adjacent road segments with direct connections in the network topology are paired. Adjacent road segments refer to those with direct connections in the highway network topology, including continuous mainline segments, ramp merging segments, tributary exit segments, and service area entrance / exit connecting segments. Traffic status data for adjacent road segments is read within a continuous time range, and the order of traffic status is compared to determine the sequential changes in traffic status. After obtaining the sequential changes in traffic status, the system checks whether there is continuation or expansion of status changes between adjacent road segments (continuation refers to a traffic status change that occurred in a previous road segment at a previous time position continuing to occur in adjacent road segments at a later time position in the same direction; expansion refers to a traffic status change that occurred in a previous road segment spreading from a single road segment to multiple adjacent road segments at a later time position, forming a distribution of changes with increasing range) segment by segment to determine the transmission relationship of traffic status influence.

[0032] When determining the transmission relationship of traffic status influence, a consistency constraint judgment is applied to the formation mechanism of traffic status changes: Based on satisfying the sequential change relationship of traffic status, further verification is made as to whether the traffic status changes between adjacent traffic segments meet the continuous transmission characteristic. When the traffic status change of the later traffic segment lags behind the earlier traffic segment, and the change direction is consistent, the change amplitude has a corresponding response relationship, and there is no synchronous change between the two traffic segments caused by common external factors, it is determined to be a traffic status influence transmission relationship. When adjacent traffic segments experience traffic status changes simultaneously at the same time and location or without significant time lag, it is determined to be a related change caused by a common disturbance and is not considered a traffic status influence transmission relationship. After marking, the sequential change relationship and the traffic status influence transmission relationship of adjacent traffic segments are compiled and collected to form road segment relationship status data. The road segment relationship status data includes adjacent traffic segment identification information, time stamp information, sequential change relationship of traffic status, and traffic status influence transmission relationship, used to describe the state change association between different traffic segments in terms of time dimension and spatial connection relationship.

[0033] S2. Identify bottleneck sections in the highway from the road segment relationship status data, extract the traffic status change relationship of the bottleneck section and the bottleneck associated road segment, perform time-series influence association mapping between the bottleneck associated road segment and the bottleneck section according to the traffic status change relationship, and output the bottleneck structure data.

[0034] Based on the road segment relationship status data, the traffic status changes of each road segment of the expressway are continuously unfolded over time, and the traffic status change sequence is output.

[0035] Specifically, according to the time sequence of each traffic segment in the road segment relationship status data, the traffic status data of the same traffic segment within a continuous time range (a continuous time range refers to a time sequence interval where data is continuously collected at a uniform time interval with no missing time markers and adjacent time positions are connected end to end) are connected hourly, so that the traffic status of each traffic segment at different time positions forms a continuous arrangement structure in chronological order. After completing the continuous arrangement, the traffic status change process of each traffic segment within the continuous time range is recorded segment by segment, so that the change trajectory of traffic status in the time dimension can be fully presented. After completing the segment-by-segment recording, the traffic status change process of each traffic segment within the continuous time range is uniformly organized in chronological order, so that each traffic segment forms a corresponding traffic status change sequence.

[0036] Based on the sequence of traffic status changes, the direction, magnitude, and duration of traffic status changes in the traffic segments are compared and judged to identify bottleneck segments.

[0037] Specifically, the traffic status data of road segments at adjacent time positions in the traffic status change sequence are compared to determine the direction of traffic status change. The magnitude of changes in the traffic status change sequence is statistically analyzed to determine the strength of the change. The duration of continuous changes in the traffic status change sequence is statistically analyzed to determine the duration of the change (the direction of traffic status change refers to the trend of traffic status changing from smooth to congested or from congested to smooth in adjacent time positions; the strength of the change refers to the magnitude of the decrease in traffic speed, increase in queues, or increase in occupancy in adjacent time positions; the duration of the change refers to the length of time that the same direction of change is maintained without interruption in consecutive time positions). After completing the analysis of the direction, magnitude, and duration of traffic status changes, these factors are combined and analyzed. Road segments that simultaneously meet the criteria of continuously deteriorating traffic status, increased magnitude of change, and prolonged duration of change are screened to identify bottleneck sections.

[0038] Furthermore, bottleneck sections are identified through a combined analysis of the direction, magnitude, and duration of traffic status changes. This is based on the statistical results of the direction of traffic status changes for the same traffic section within a continuous time range in the traffic status change sequence. Traffic sections showing a continuous decline in traffic status at multiple consecutive time points are identified as having a continuously deteriorating traffic status. The magnitude of change of the corresponding traffic section at adjacent time points is compared, and when the magnitude of change at a later time point is greater than that at a previous time point, it is identified as having an increased magnitude of change. The duration of continuous decline in traffic status is statistically analyzed, and changes that continue for an uninterrupted duration are identified as having a prolonged duration of change. Traffic sections that simultaneously meet the criteria of continuously deteriorating traffic status, increased magnitude of change, and prolonged duration of change are selected as bottleneck sections.

[0039] When screening bottleneck sections, the stability and causal characteristics of traffic status changes are constrained to distinguish between persistent bottleneck states and short-term fluctuations or occasional disturbances: when the traffic status change of a traffic section occurs only at a single time location or discrete time location, or when the change process is interrupted, reversed, or rapidly recovered within a continuous time range, it is judged as a short-term fluctuation or occasional change and is not considered a bottleneck section; when the traffic status change is caused by an anomaly in a single indicator and does not form a consistent trend with other traffic status indicators, it is judged as a local anomaly or monitoring deviation and is not considered a bottleneck section; when the traffic status change of a traffic section maintains a stable deterioration trend within a continuous time range, and the change forms a continuous influence relationship with adjacent traffic sections in space, and shows a consistent direction of change in traffic speed change information, vehicle queue change information, and road occupancy change information, it is judged as a valid bottleneck section with continuous constraint characteristics.

[0040] When identifying bottleneck sections, the external impact of traffic condition changes is constrained by environmental condition records. Specifically, the causes of traffic condition changes are identified by using information on weather conditions, visibility, precipitation, and road environment changes in the environmental condition records. When traffic condition changes and environmental condition changes have the same trend, environmental factors are used as the influence weight to correct the intensity of traffic condition changes. When traffic condition changes and environmental condition changes are inconsistent, the original traffic condition determination result is maintained to identify the bottleneck section.

[0041] The bottleneck road segment and the road segment relationship status data are matched for time-series impact and filtered for traffic relationship, and the traffic status change relationship and bottleneck associated road segments are output.

[0042] Specifically, based on the bottleneck road segment, adjacent traffic segments connected to the bottleneck road segment are read from the road segment relationship status data. Time and location matching is performed according to the sequential changes in traffic status in the road segment relationship status data, ensuring a corresponding relationship between the bottleneck road segment and adjacent traffic segments. After time and location matching, the traffic status influence transmission relationship between the bottleneck road segment and adjacent traffic segments is filtered, retaining traffic segments with status continuity and change response characteristics to form bottleneck-related road segments. During the filtering of traffic status influence transmission relationships, candidate relationships are also constrained and filtered based on the time lag characteristics of traffic status changes, consistency of change direction, and spatial connectivity continuity, eliminating traffic segment relationships that only satisfy time correlation but lack status transmission characteristics. After filtering, the traffic status change process between the bottleneck road segment and bottleneck-related road segments is correspondingly organized, outputting the traffic status change relationship and bottleneck-related road segments.

[0043] Based on the relationship of traffic status changes, the occurrence time of traffic status changes between bottleneck-related road segments is sorted, and the temporal impact relationship is output.

[0044] Specifically, based on the correspondence of state changes recorded in the traffic status change relationship, the bottleneck road segment and the bottleneck-related road segment are read at the time of the traffic status change and sorted in chronological order so that the state changes between different traffic segments can form a clear time transmission sequence; after completing the time sorting, the start time and response time of the state change between the bottleneck road segment and the bottleneck-related road segment are correspondingly organized so that the chronological relationship of the state changes can be clearly expressed; after the organization is completed, the bottleneck road segment and the bottleneck-related road segment are connected in chronological order to form a temporal influence relationship.

[0045] The system classifies the influence direction and determines the influence scope of the temporal influence correlation, and outputs the influence extension direction and influence expansion range.

[0046] Specifically, based on the temporal sequence and spatial connection between bottleneck segments and their associated segments in the temporal impact correlation, the bottleneck segment where the traffic status change occurs first is taken as the starting position, and the bottleneck associated segment where the traffic status change occurs later is taken as the extension position. According to the spatial arrangement order of each traffic segment of the expressway, the connection path between the bottleneck segment and its associated segment is determined segment by segment, so that the spatial transmission path of the traffic status change forms a clear directional sequence, and the impact extension direction is obtained. After the impact extension direction is determined, the bottleneck associated segments that appear consecutively in the same impact extension direction are statistically analyzed segment by segment, the segment length corresponding to each bottleneck associated segment is extracted, and the traffic status change intensity weight is determined according to the strength of the traffic status change corresponding to each bottleneck associated segment. The segment length of each bottleneck associated segment and the traffic status change intensity weight are weighted and calculated, and then accumulated along the impact extension direction to obtain the quantitative value of the impact expansion range.

[0047] The formula for calculating the quantization value that affects the range of expansion is as follows: ; in, This represents the quantized value that affects the range of expansion. Indicates the order of spatial connections in the direction of influence extension. The sequence number of each bottleneck-related road segment. This indicates the number of bottleneck-related road segments affecting continuous connections in the extended direction. Indicates the first The length of each bottleneck-related road segment. Indicates the first Weighting of the intensity of traffic status changes in each bottleneck-related road segment.

[0048] Based on the quantified value of the influence expansion range, spatial interval mapping is performed on the influence expansion range: according to the spatial connection order of the road segments corresponding to the influence extension direction, the bottleneck-related road segments are accumulated segment by segment. When the accumulated length reaches the quantified value of the influence expansion range, the road segment corresponding to the current accumulation termination position is determined as the spatial boundary endpoint, and the location of the bottleneck road segment is taken as the spatial starting position. A continuous road segment interval from the starting position to the termination position is determined on the spatial path as the spatial interval corresponding to the influence expansion range. When a road segment whose traffic status does not meet the propagation conditions appears during the accumulation process, the accumulation is stopped and the current interval is determined as the effective spatial range of the influence expansion range.

[0049] The system aggregates and correlates temporal impact relationships, bottleneck-related road segments, bottleneck road segments, impact extension direction, and impact expansion range to output bottleneck structure data.

[0050] Specifically, using bottleneck road segments as the index benchmark, the system matches the bottleneck-related road segments connected to the bottleneck road segments in the temporal influence correlation relationship. Based on the chronological order in the temporal influence correlation relationship, the system unifies the connection relationships of traffic status changes between bottleneck road segments and their related segments, ensuring each bottleneck road segment corresponds to a complete temporal transmission structure. According to the direction of influence extension, the spatial transmission paths between bottleneck road segments and their related segments are directionally matched, ensuring a consistent relationship between the temporal transmission structure and the spatial extension direction. Based on the scope of influence expansion, the spatial expansion intervals between bottleneck road segments and their related segments in the corresponding influence extension direction are range-matched, ensuring a unified expression of spatial expansion characteristics and temporal transmission structure. The resulting bottleneck structure data includes bottleneck road segment identification information, a set of bottleneck-related road segments, temporal influence correlation relationships, influence extension direction, and the spatial interval corresponding to the influence expansion range, used to describe the propagation structure of the bottleneck road segment in both the temporal and spatial dimensions.

[0051] S3. Based on the temporal impact correlation in the bottleneck structure data, the bottleneck road segments are organized into paths according to the temporal impact correlation, and a congestion propagation path map is output. The direction and extent of the impact on the bottleneck road segments are extrapolated along the path continuity relationship in the congestion propagation path map, and congestion trend data is output.

[0052] The temporal impact relationships in the bottleneck structure data are organized according to the order of transmission between bottleneck road segments to generate a road segment temporal association sequence.

[0053] Specifically, based on the temporal impact relationships between bottleneck road segments and their associated road segments recorded in the bottleneck structure data, the occurrence time of traffic status changes in each temporal impact relationship is read, and the bottleneck road segments are sorted according to the occurrence time of traffic status changes, forming an ordered arrangement of bottleneck road segments according to the chronological order of traffic status changes. The temporal impact relationships between the sorted bottleneck road segments are matched accordingly, and bottleneck road segments with direct temporal transmission relationships are sequentially connected, forming a continuous link structure between bottleneck road segments. The formed continuous link structure is then uniformly organized, ensuring that the temporal impact relationships between each bottleneck road segment form a complete and continuous transmission order in the time dimension, generating a road segment temporal association sequence. The road segment temporal association sequence is an ordered sequence of road segments arranged according to the chronological order of traffic status changes based on the temporal impact relationships between bottleneck road segments.

[0054] Furthermore, sequentially connecting bottleneck road segments with direct temporal transmission relationships involves determining the sequential changes in traffic status between adjacent bottleneck road segments in the road segment relationship status data. When a subsequent bottleneck road segment changes its traffic status immediately after the preceding bottleneck road segment in time sequence, and there is a traffic status influence transmission relationship between the two bottleneck road segments, a direct temporal transmission relationship is determined between the two bottleneck road segments. Sequentially connecting bottleneck road segments that satisfy both temporal sequence connection and traffic status influence transmission relationship provides a clear basis for determining the temporal transmission relationship.

[0055] The corresponding positions of bottleneck road segments connected one after the other are identified by the temporal correlation sequence of the road segments, which serve as the road segment transmission chain data.

[0056] Specifically, based on the temporal arrangement of bottleneck road segments in the road segment temporal correlation sequence, bottleneck road segments at adjacent time positions are read one by one. Combined with the spatial connection relationship between bottleneck road segments and bottleneck-related road segments in the bottleneck structure data, bottleneck road segments with direct spatial connections are matched to form a correspondence between time-adjacent and spatially connected bottleneck road segments. The spatial location of the matched bottleneck road segments in the highway is calibrated to ensure that the connection relationship between bottleneck road segments is consistent in both the temporal and spatial dimensions. All bottleneck road segments that meet the conditions of temporal adjacency and spatial connection are uniformly organized and output as road segment transmission chain data.

[0057] In the road segment transmission chain data, bottleneck road segments with the same transmission direction and continuous connection are combined to generate road segment path connection data.

[0058] Specifically, based on the connection relationship between bottleneck road segments in the road segment transmission chain data, the transmission direction of each bottleneck road segment in the road segment transmission chain data is determined. Then, based on the corresponding influence extension direction in the bottleneck structure data, bottleneck road segments with consistent transmission directions are filtered to ensure that bottleneck road segments are arranged in the same direction in the spatial path. For bottleneck road segments that have completed the direction consistency filtering, they are spliced ​​together segment by segment according to the sequential connection order in the road segment transmission chain data, forming a continuous connection between adjacent bottleneck road segments. The spliced ​​set of bottleneck road segments is then uniformly organized to form a continuous through structure in the spatial path, generating road segment path connectivity data. Road segment path connectivity data is path structure data formed by combining road segment transmission chains with consistent transmission directions and spatial continuity, used to describe a continuously connected congestion propagation path.

[0059] The road segment connection data is sorted and classified to identify the road segment transmission chain of each bottleneck road segment.

[0060] Specifically, each bottleneck road segment in the road segment path connectivity data is read one by one, and the spatial connection path and transmission direction information corresponding to each bottleneck road segment are extracted. Bottleneck road segments with the same spatial connection path and the same transmission direction are grouped and classified, so that bottleneck road segments with the same path direction are classified into the same set. The bottleneck road segments in each set are arranged segment by segment according to the sequential connection order in the road segment path connectivity data, so that the bottleneck road segments form a continuous path connection relationship within the set. The arranged bottleneck road segment sets are uniformly organized so that each set corresponds to a complete path transmission structure, that is, a road segment transmission chain.

[0061] The congestion propagation path map is output by sequentially connecting the transmission chains of each road segment according to the continuous path relationship of the highway.

[0062] Specifically, the bottleneck segments within each road segment transmission chain are read one by one, and the start and end positions of each road segment transmission chain are extracted. The path continuity relationships between different road segment transmission chains are matched to ensure that the end position corresponds continuously in space to the start position of other road segment transmission chains. Road segment transmission chains that satisfy the path continuity relationship are sequentially spliced ​​to form a continuous connection structure on the highway path. The spliced ​​road segment transmission chains are then uniformly arranged according to the spatial path order to form a continuous expression of the transmission relationship between bottleneck segments on the overall path, outputting a congestion propagation path map. The congestion propagation path map is path structure data constructed based on the temporal influence correlation and spatial connection relationship in the bottleneck structure data. It includes multiple bottleneck segment sequences connected in time order to represent the propagation path of congestion in the highway network.

[0063] Furthermore, when connecting continuous path relationships, for road segments with branching structures, corresponding propagation paths are constructed based on the continuity and impact range of traffic status changes in each branch path, allowing multiple propagation paths to exist in parallel within the branch structure. For road segments with merging structures, traffic status changes from different upstream paths are converged and analyzed. When multiple paths form a consistent trend in the same road segment, the current road segment is included as a convergence node in the propagation path. For road segments with bidirectional separation or multi-lane structures, independent path relationships are constructed according to lane direction or driving direction to avoid interference between traffic flows in different directions.

[0064] By analyzing the continuous relationships along the congestion propagation path map, the chronological order of the impact extension directions of each bottleneck road segment is determined, forming a road segment directional extension sequence.

[0065] Specifically, the bottleneck road segments connected by the continuous path relationship in the congestion propagation path map are unfolded sequentially to determine the position of each bottleneck road segment in the overall propagation path. The influence extension direction corresponding to each bottleneck road segment is matched with the time sequence in the temporal influence correlation, so that the influence extension direction of the bottleneck road segment at different time positions forms a sequential relationship. The corresponding influence extension directions are arranged continuously according to time sequence, so that the influence extension direction forms a clear temporal change pattern in the propagation process. The arranged influence extension directions are then uniformly collected to form a road segment direction extension sequence.

[0066] Based on the extension sequence of bottleneck road sections, the influence expansion range of each bottleneck road section in the direction of influence extension is recursively determined segment by segment using time series analysis methods, thus forming a road section range expansion sequence.

[0067] Specifically, the influence extension direction of each bottleneck road segment in the bottleneck road segment direction extension sequence is mapped one-to-one with the influence expansion range at the corresponding time position, and an influence expansion range sequence is formed according to the time sequence. Time series analysis methods are used to analyze the changing trend and difference statistics of the influence expansion range at adjacent time positions to determine the growth, contraction or maintenance status of the influence expansion range in the time dimension. The influence expansion range of the subsequent time position is updated recursively based on the previous time position, so that the influence expansion range forms a segment-by-segment recursive change relationship along the time sequence. The influence expansion ranges obtained from the recursion of each time position are uniformly arranged to form a road segment range expansion sequence.

[0068] Furthermore, time series analysis is used to analyze the changing trends of the influence expansion range of adjacent time positions. This involves arranging the influence expansion range of consecutive time positions in chronological order and comparing the changes in the influence expansion range of preceding and following time positions. Based on the direction and continuity of change of the influence expansion range in consecutive time positions, the influence expansion range is classified into states. When the influence expansion range of subsequent time positions shows an increasing trend relative to the previous time position, it is determined to be an expanding state. When the influence expansion range of subsequent time positions shows a decreasing trend relative to the previous time position, it is determined to be a contracting state. When the influence expansion range of consecutive time positions remains consistent, it is determined to be a stable state. This provides clear change basis for the step-by-step recursive judgment.

[0069] The congestion trend data is output by cross-merging the road segment direction extension sequence and the road segment range expansion sequence.

[0070] Specifically, the influence extension direction of each bottleneck road segment in the road segment direction extension sequence at continuous time positions is matched one-to-one with the influence extension range at the corresponding time positions in the road segment range expansion sequence, so that the direction information and range information at each time position form a pairing relationship; the matched direction information and range information are combined and organized, with the influence extension direction as the path extension direction and the influence extension range as the degree of expansion on the corresponding path, so that the propagation direction and propagation range at each time position form a unified expression; the combination results of each time position are continuously arranged in chronological order, so that the direction change and range change of the bottleneck road segment in the propagation process form an overall evolutionary relationship, and the congestion trend data is output.

[0071] By organizing the temporal impact relationships in bottleneck structure data into a path-connected structure, the transmission relationship between bottleneck road segments is transformed from a discrete correlation to a continuous path structure. This enables the congestion propagation process to have a traceable path expression capability. By combining the continuous path relationship to perform temporal extrapolation of the direction and extent of impact, the dynamic evolution process of congestion from formation to expansion is structurally characterized. This improves the accuracy and predictive foresight of identifying highway congestion propagation trends, and enables accurate description and application support of congestion development paths and impact ranges.

[0072] S4. Based on congestion trend data, determine the congestion development status and propagation path of highways, integrate them to form congestion prediction data, and arrange the congestion prediction data in chronological order and define its spatial scope to output highway congestion prediction information.

[0073] Extract the direction and extent of the impact of each bottleneck road segment from the congestion trend data, divide the direction and extent of the impact into stages, and output the congestion development status.

[0074] Specifically, the impact direction and range of each bottleneck road segment are extracted sequentially from congestion trend data at continuous time locations. The impact direction and range of the same bottleneck road segment at different time locations are then organized to ensure a continuous change relationship between them over time. The changes in the organized impact direction and range at continuous time locations are compared and judged to establish a correspondence between the stability of the impact direction and the trend of the impact range. Based on the continuity of the impact direction and the changing state of the impact range, each bottleneck road segment is divided into congestion formation, congestion expansion, congestion continuation, and congestion dissipation stages at different time locations, providing a clear stage division of the congestion development process and outputting the congestion development status.

[0075] Furthermore, the changes in the direction and range of influence after processing are compared and judged in consecutive time positions. This involves reading the direction of influence extension and the value of influence range one by one according to adjacent time positions to see if they remain consistent, and whether they increase, remain unchanged, or decrease. When the direction of influence extension remains consistent in multiple consecutive time positions and the influence range continues to increase, it is judged as an expansion trend. When the direction of influence extension remains consistent and the influence range remains basically unchanged, it is judged as a continuous state. When the direction of influence extension changes or the influence range changes from increasing to decreasing, it is judged as a dissipation trend, thus forming the corresponding change judgment results.

[0076] The congestion is divided into stages based on the continuity of the direction of influence and the changes in the range of influence. This involves hourly comparisons of the direction and range of influence for the same bottleneck road segment at continuous time locations. When the direction of influence changes from non-existent to present and the range of influence begins to appear, it is determined to be the congestion formation stage. When the direction of influence remains consistent at continuous time locations and the range of influence continues to expand along the spatial path, it is determined to be the congestion expansion stage. When the direction of influence remains stable and the range of influence remains basically unchanged at continuous time locations, it is determined to be the congestion continuation stage. When the direction of influence is interrupted or reversed and the range of influence gradually shrinks, it is determined to be the congestion dissipation stage.

[0077] Based on the state of congestion development, the propagation path of each bottleneck road segment is determined along the direction of influence extension in the congestion trend data.

[0078] Specifically, the congestion development status is correlated with the direction of influence extension in the congestion trend data, so that the stage attributes of each bottleneck segment at different time locations correspond to the direction of influence extension. For bottleneck segments in the congestion expansion stage and the congestion continuation stage, the influence extension direction is extended segment by segment along the spatial path for confirmation (i.e., starting from the current bottleneck segment, according to the spatial connection order corresponding to the influence extension direction, the downstream passing segments directly connected to the current segment in the current direction are searched one by one, and the traffic status of the current passing segment at the corresponding time location is read; when the traffic status of the passing segment meets the requirements...). When congestion is present or worsening, the current traffic segment is included in the propagation path, and the current traffic segment is used as a new starting point to continue searching for the next adjacent segment in the same direction. When the traffic status of the adjacent segment does not meet the congestion requirements or does not show a worsening trend, the extension stops, thus determining the continuous propagation path along the direction of influence extension, so that the direction of influence extension forms a clear propagation direction in the highway path. The direction of influence extension of each bottleneck segment in continuous time position is uniformly arranged so that the propagation path is consistently expressed in the time and space dimensions, and the direction of the propagation path of each bottleneck segment is determined.

[0079] By linking and coupling the development status of congestion with the direction of its propagation, congestion prediction data can be generated.

[0080] Specifically, the congestion development status and propagation path are mapped one-to-one according to bottleneck road segments and time locations, so that each bottleneck road segment has clear stage attributes and propagation direction information at each time location; the corresponding congestion development status and propagation path are combined and organized so that the congestion development status and propagation path are expressed in a unified result structure; the combined results are arranged continuously in chronological order so that the development stages and path directions of each bottleneck road segment in the congestion propagation process form an overall evolutionary relationship, forming congestion prediction data.

[0081] When processing congestion trend data, it is necessary to define the prediction time range and extrapolate the congestion evolution state in future time intervals based on the congestion trend data at the current time location. Specifically, the current time location is used as the prediction start point, and a prediction time window is formed by extending it forward at fixed time intervals. The direction and extent of the impact of each bottleneck road segment within the prediction time window are then extrapolated to ensure a continuous evolution of the congestion propagation path in the future time interval. During the extrapolation process, the direction and extent of the impact at the current time location are used as the initial state, and subsequent time locations are extrapolated based on the changing trends of adjacent time locations. This ensures that the direction and extent of congestion propagation form a prediction result at the future time location. The extrapolation result is verified by combining the changing information in the environmental status records and the continuity of traffic status changes. Predictions that are consistent with historical trends are retained, while those that show abnormal changes are corrected or truncated, resulting in a congestion prediction result for the future time interval.

[0082] Based on the propagation path, influence extension direction, and influence expansion range of each bottleneck road segment in the congestion prediction data, the bottleneck road segments are ranked in order of their occurrence to form congestion time-series ranking data.

[0083] Specifically, the propagation path direction, influence extension direction, and influence expansion range corresponding to each bottleneck road segment are extracted from the congestion prediction data. The congestion occurrence sequence of each bottleneck road segment is then ordered chronologically to establish a clear temporal relationship. The ordered bottleneck road segments are further organized according to their propagation path direction and influence extension direction to ensure consistency between temporal order and spatial path. Finally, the organized bottleneck road segments are uniformly arranged to create a continuous temporal representation of the congestion occurrence process, forming congestion time-series ranking data. This congestion time-series ranking data, based on the temporal order of congestion occurrence and propagation path relationships of each bottleneck road segment, is used to represent the temporal sequence of different bottleneck road segments during the congestion process.

[0084] By dividing the congestion time-series ranking data into boundaries, the congestion impact range of each bottleneck section on the highway is determined.

[0085] Specifically, the impact range of each bottleneck road segment in the congestion time-series ranking data is expanded segment by segment, so that the expansion range of each bottleneck road segment in the spatial path forms a continuous interval; the starting and ending positions of each bottleneck road segment in the direction of impact extension are sorted out, so that the congestion impact range forms a clear boundary in the spatial dimension; the bottleneck road segments with completed boundary sorting are uniformly divided, so that the congestion impact range corresponding to each bottleneck road segment has a clear spatial range, and the congestion impact range of each bottleneck road segment in the highway is determined.

[0086] The congestion time-series ranking data and congestion impact intervals are structured and arranged to output highway congestion prediction information.

[0087] Specifically, the congestion time-series ranking data and congestion impact intervals are matched according to bottleneck road segments and time sequence, so that each bottleneck road segment has both time sequence information and spatial range information at each time location; the matched congestion time-series ranking data and congestion impact intervals are uniformly arranged to form an integrated expression structure of time and space dimensions; the arranged results are comprehensively sorted to output highway congestion prediction information, which is used for intelligent transportation IoT application services in highway network vehicle information interaction scenarios.

[0088] This embodiment also provides a computer device applicable to the highway congestion prediction method based on big data, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the highway congestion prediction method based on big data as proposed in the above embodiment.

[0089] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0090] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the highway congestion prediction method based on big data as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0091] In summary, the collaborative construction of bottleneck structure data and the generation of congestion propagation path maps enabled the clear identification of the transmission path and expansion range of highway congestion impacts. By mapping the temporal impact relationships between bottleneck-related road segments according to traffic status changes, the traffic status change relationships between road segments are transformed into structured relationships with transmission direction and impact range, providing a clear foundation for expressing traffic status correlation information. Based on the temporal impact correlation relationships in the bottleneck structure data, path continuation is organized so that the impact transmission between bottleneck road segments can be presented continuously along the highway path, thus providing a stable path support for the temporal extrapolation of congestion trend data. This allows highway congestion prediction information to serve traffic status perception and traffic guidance.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A highway congestion prediction method based on big data, characterized in that: include, Collect multi-source traffic data of highways, analyze the traffic status of road segments and identify the traffic relationship between road segments, and output road segment relationship status data; Bottleneck sections in highways are identified from road segment relationship status data. The traffic status change relationship of bottleneck sections and bottleneck-related sections are extracted. The bottleneck-related sections and bottleneck sections are mapped according to the traffic status change relationship, and the bottleneck structure data is output. Based on the temporal impact correlation in the bottleneck structure data, the bottleneck road segments are organized into paths according to the temporal impact correlation, and a congestion propagation path map is output. The direction and range of the impact of the bottleneck road segments on the path continuity relationship in the congestion propagation path map are temporally extrapolated, and congestion trend data is output. Based on congestion trend data, the congestion development status and propagation path of highways are determined, and congestion prediction data is integrated to form congestion prediction data. The congestion prediction data is then arranged in chronological order and spatially defined to output highway congestion prediction information.

2. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for outputting the road segment relationship status data are as follows: Information on changes in traffic speed, vehicle queues, and road occupancy on various sections of highways is extracted from multi-source traffic data and integrated to form road segment traffic discrimination data. The traffic status data of road segments is generated by comparing the time sequence and classifying the magnitude of change of the traffic data. Based on the traffic status data of road segments, the sequential changes in traffic status and the transmission of traffic status influence between adjacent road segments are marked accordingly to form road segment relationship status data.

3. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for extracting the traffic status changes of bottleneck road sections and the associated road sections are as follows: Based on the road segment relationship status data, the changes in traffic status of each road segment of the expressway are continuously unfolded over time, and the traffic status change sequence is output. Based on the sequence of traffic status changes, the direction, magnitude, and duration of traffic status changes in the traffic segments are compared and determined to identify bottleneck segments. The bottleneck road segment and the road segment relationship status data are matched for time-series impact and filtered for traffic relationship, and the traffic status change relationship and bottleneck associated road segments are output.

4. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for outputting the bottleneck structure data are as follows: Based on the traffic status change relationship, the traffic status change times between bottleneck-related road segments and bottleneck segments are sorted, and the temporal impact relationship is output. The system classifies the direction of influence and determines the scope of influence based on the temporal impact correlation, and outputs the direction of influence extension and the scope of influence expansion. The system aggregates and correlates temporal impact relationships, bottleneck-related road segments, bottleneck road segments, impact extension direction, and impact expansion range to output bottleneck structure data.

5. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for outputting the congestion propagation path diagram are as follows: The temporal impact correlations in the bottleneck structure data are organized according to the order of transmission between bottleneck road segments to generate a road segment temporal correlation sequence. The corresponding positions of the bottleneck road segments connected one after the other are identified along the temporal correlation sequence of the road segments, which serve as the road segment transmission chain data; In the road segment transmission chain data, bottleneck road segments with the same transmission direction and continuous connection are connected and combined to generate road segment path connection data. Perform route sorting operations on the road segment connection data to identify the road segment transmission chain of each bottleneck road segment; The congestion propagation path map is output by sequentially connecting the transmission chains of each road segment according to the continuous path relationship of the highway.

6. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for outputting congestion trend data are as follows: By analyzing the continuous path relationships in the congestion propagation path map, the temporal sequence of the impact extension direction of each bottleneck road segment is formed, thus creating a road segment directional extension sequence. Based on the bottleneck road segment extension sequence, the influence expansion range of each bottleneck road segment in the direction of influence extension is determined segment by segment through time series analysis, thus forming a road segment range expansion sequence. The congestion trend data is output by cross-merging the road segment direction extension sequence and the road segment range expansion sequence.

7. The highway congestion prediction method based on big data according to claim 1, characterized in that: The integration of these data to form congestion prediction data involves the following steps: Extract the direction and extent of the impact of each bottleneck road segment from the congestion trend data, divide the direction and extent of the impact into stages, and output the congestion development status. Based on the state of congestion development, the propagation path of each bottleneck section is determined along the direction of influence extension in the congestion trend data; By linking and coupling the development status of congestion with the direction of its propagation, congestion prediction data can be generated.

8. The highway congestion prediction method based on big data according to claim 1, characterized in that: The specific steps for outputting highway congestion prediction information are as follows: Based on the propagation path direction, influence extension direction and influence expansion range of each bottleneck road segment in the congestion prediction data, the bottleneck road segments are ranked in order of congestion occurrence to form congestion time sequence ranking data. Boundary division is performed on the congestion time-series ranking data to determine the congestion impact range of each bottleneck section on the highway. The congestion time-series ranking data and congestion impact intervals are structured and arranged to output highway congestion prediction information.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the highway congestion prediction method based on big data as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the highway congestion prediction method based on big data as described in any one of claims 1 to 8.

Images