CN-121393153-B - Urban traffic linkage early warning method and system based on intelligent monitoring

CN121393153BCN 121393153 BCN121393153 BCN 121393153BCN-121393153-B

Abstract

The invention belongs to the technical field of urban traffic early warning, and provides an urban traffic linkage early warning method and system based on intelligent monitoring, comprising the steps of acquiring multi-source traffic monitoring data, and carrying out multi-mode fusion processing on the multi-source traffic monitoring data to form a traffic state data sequence of a target road network; and constructing a traffic change curve based on the traffic state data sequence, separating change characteristics of the traffic change curve, classifying the change characteristics into a rapid change component and a trend change component, and carrying out quantization processing on the separated change characteristics. The components representing rapid disturbance and components representing gradual change accumulation are separated from the mixed traffic data through principal component analysis and wavelet transformation, and a differential clustering model and a data-driven triggering rule are respectively constructed aiming at two types of modes with different essence, so that sudden accidents and slow traffic jams can be distinguished from the origin of time sequence evolution, and the problems of instantaneous noise false report and trend event missing report are optimized.

Inventors

Hong Xiaozheng
Guo Liangshuang
WANG YIPIN
XIE XIANMING
WU CHENGWEI
CHEN JIAZHI
ZHANG GUOZHONG
LI ZHENZHI
LIN ZHONG
Su Jianshuo
ZHAO BO
ZHANG PENG
CHEN YULIN

Assignees

厦门市市政工程设计院有限公司

Dates

Publication Date: 20260512
Application Date: 20251223

Claims (7)

1. The urban traffic linkage early warning method based on intelligent monitoring is characterized by comprising the following steps: Acquiring multi-source traffic monitoring data, and performing multi-mode fusion processing on the multi-source traffic monitoring data to form a traffic state data sequence of a target road network; constructing a traffic change curve based on the traffic state data sequence, separating change characteristics of the traffic change curve, classifying the change characteristics into a rapid change component and a trend change component, and carrying out quantization treatment on the separated change characteristics; Respectively carrying out feature clustering and event triggering rule extraction on the rapid change components and the trend change components through the combination of the change features and the statistical analysis results of the analysis historical traffic state data, and determining the traffic event triggering type; Wherein the traffic event trigger type includes a rapid burst event and a non-rapid burst event; the characteristic clustering process for the rapid change component and the trend change component comprises the following steps: based on a historical time sample feature library comprising a historical trend event sample set and a historical fast event sample set; For a historical trend event sample set, carrying out cluster analysis by adopting a density-based clustering algorithm, dividing the historical trend event sample set into a plurality of non-rapid burst clusters, wherein each cluster corresponds to one non-rapid burst event; for a historical rapid event sample set, carrying out cluster analysis by adopting a hierarchical clustering algorithm, dividing the sample set into a plurality of rapid burst clusters, wherein each cluster corresponds to one rapid burst event; the acquisition process of the historical time sample feature library comprises the following steps: generating a corresponding historical feature vector for each historical event sample by using the historical traffic state data; For each historical event period, extracting trend change component features to construct a historical trend feature vector, and extracting quick change component features to construct a historical quick feature vector; respectively collecting trend feature vectors and quick feature vectors of all historical events to form a historical trend event sample set and a historical quick event sample set which are used as a historical event sample feature library; the acquisition process of the event triggering rule comprises the following steps: For each non-rapid burst event cluster obtained by clustering a historical trend event sample set, analyzing the statistical distribution of all historical trend feature vectors in the cluster, and for each feature, calculating the mean value and standard deviation in the cluster, wherein an event triggering rule is set as a threshold condition combination; For each rapid burst event cluster obtained by clustering a historical rapid event sample set, analyzing the statistical distribution of each characteristic of the historical rapid characteristic vector in the cluster, and for each characteristic, calculating the mean value and standard deviation in the cluster, wherein an event triggering rule is set as a threshold condition combination; the threshold condition combination, i.e. the sum of the standard deviation of the mean and k times, is positive and negative depending on the correlation direction of the feature and the event; based on the determined traffic event trigger type, a traffic linkage early warning instruction is generated, and an early warning or linkage scheduling strategy is issued to a traffic control system.
2. The urban traffic linkage early warning method based on intelligent monitoring according to claim 1, wherein the acquisition process of the traffic state data sequence is as follows: Preprocessing the collected multi-source traffic monitoring data, and extracting characteristic parameters representing traffic states from the preprocessed multi-source traffic monitoring data; the characteristic parameters comprise vehicle flow, average vehicle speed, vehicle position distribution, vehicle stagnation time, road occupancy and weather visibility; The extracted characteristic parameters are spliced into uniform multidimensional characteristic vectors according to time segments, the dimension reduction processing is carried out on the high-dimensional characteristics through an automatic encoder, the fused multidimensional characteristic vectors are sampled according to fixed time intervals and are arranged according to time sequence, and a traffic state data sequence of a target road network is formed.
3. The urban traffic linkage early warning method based on intelligent monitoring according to claim 1, wherein the process of constructing the traffic curve is as follows: Extracting traffic flow, average speed and road occupancy reflecting traffic state from the traffic state data sequence by adopting a principal component analysis method; and selecting the score sequence of the first main component as a comprehensive traffic state index, taking time as a horizontal axis, taking the comprehensive traffic state index as a vertical axis, and drawing to obtain a traffic change curve of the target road network.
4. The urban traffic linkage early warning method based on intelligent monitoring according to claim 1, wherein the process of separating the change features comprises the following steps: performing signal decomposition on the traffic change curve by adopting wavelet transformation, selecting a wavelet basis function and the number of decomposition layers, and performing discrete wavelet transformation on the traffic change curve to obtain a group of approximation coefficients and detail coefficient sequences; the trend change component is obtained by reconstructing the deepest approximation coefficient, and the quick change component is obtained by reconstructing and superposing the detail coefficients of each layer.
5. The urban traffic linkage early warning method based on intelligent monitoring according to claim 1, wherein the process of carrying out quantization processing on the separated change characteristics is as follows: calculating a first derivative quantization change rate for the trend change component, and extracting accumulated change quantity and trend duration in a sliding window; And detecting local extremum points of the rapid change component, calculating mutation amplitude and change rate between adjacent extremum points, and counting mutation frequency exceeding a preset noise threshold value in unit time.
6. The urban traffic linkage early warning method based on intelligent monitoring according to claim 1, wherein the process of determining the traffic event trigger type is as follows: Acquiring and processing current traffic state data in real time to respectively acquire real-time trend change component characteristics and quick change component characteristics; comparing the real-time rapid component characteristics with the traffic event triggering rules of all the extracted rapid burst event clusters one by one; if the rule condition of a certain rapid burst event cluster is met, judging that a rapid burst event occurs currently, wherein the event type is a cluster label corresponding to the rule; If the real-time rapid component characteristics do not meet any rapid event triggering rules, comparing the real-time trend component characteristics with triggering rules of all extracted non-rapid burst event clusters one by one; If the rule condition of a certain trend event cluster is met, judging that a non-rapid burst event occurs currently, wherein the event type is a cluster label corresponding to the rule.
7. Urban traffic linkage early warning system based on intelligent monitoring, characterized in that it is adapted to perform the method according to any one of the preceding claims 1-6, comprising: The traffic data acquisition module is used for acquiring multi-source traffic monitoring data, and carrying out multi-mode fusion processing on the multi-source traffic monitoring data to form a traffic state data sequence of a target road network; The traffic change analysis module is used for constructing a traffic change curve based on the traffic state data sequence, separating change characteristics of the traffic change curve, classifying the change characteristics into a rapid change component and a trend change component, and carrying out quantization treatment on the separated change characteristics; The traffic event determining module is used for respectively carrying out feature clustering and event triggering rule extraction on the rapid change component and the trend change component through the combination of the change features and the statistical analysis result of the analysis historical traffic state data to determine the traffic event triggering type; Wherein the traffic event trigger type includes a rapid burst event and a non-rapid burst event; and the linkage early warning module is used for generating a traffic linkage early warning instruction based on the determined traffic event trigger type and issuing an early warning or linkage scheduling strategy to the traffic control system.

Description

Urban traffic linkage early warning method and system based on intelligent monitoring Technical Field The invention belongs to the technical field of urban traffic early warning, and particularly relates to an intelligent monitoring-based urban traffic linkage early warning method and system. Background With the continuous expansion of the urban road traffic scale and the continuous increase of the vehicle storage quantity, the urban traffic running state presents the characteristics of frequent flow fluctuation, diversified event types and the like, and an intelligent traffic monitoring system based on multi-source information such as video monitoring, radar detection, perception terminals, floating car data and the like gradually becomes a core infrastructure of traffic management for guaranteeing the road traffic efficiency and improving the response capability of traffic management departments; The existing traffic linkage early warning technology generally depends on single monitoring data or a judging mechanism based on a simple threshold rule, has limited description capability on traffic state change, and under a complex traffic environment, the traffic state change can not only suddenly jump, but also form congestion or delay in a slow accumulation mode, so that the evolution characteristics of different events are difficult to distinguish; Due to the lack of effective distinction between rapid burst traffic events and non-rapid burst trend events, misjudgment and missed judgment often occur in practical application, on one hand, instantaneous noise may trigger false early warning, so that the traffic control system excessively responds to influence the overall operation efficiency, on the other hand, slowly evolving trend congestion is difficult to identify in time, so that a traffic management department misses the optimal intervention time, and in addition, event triggering logic lacks pertinence, so that early warning instruction triggering is unstable, the matching degree between a linkage strategy and actual road conditions is insufficient, and the overall urban traffic control effect is influenced; Therefore, the invention provides an urban traffic linkage early warning method and system based on intelligent monitoring. Disclosure of Invention In order to overcome the deficiencies of the prior art, at least one technical problem presented in the background art is solved. The technical scheme adopted by the invention for solving the technical problems is that the urban traffic linkage early warning method based on intelligent monitoring comprises the following steps: Acquiring multi-source traffic monitoring data, and performing multi-mode fusion processing on the multi-source traffic monitoring data to form a traffic state data sequence of a target road network; constructing a traffic change curve based on the traffic state data sequence, separating change characteristics of the traffic change curve, classifying the change characteristics into a rapid change component and a trend change component, and carrying out quantization treatment on the separated change characteristics; Respectively carrying out feature clustering and event triggering rule extraction on the rapid change components and the trend change components through the combination of the change features and the statistical analysis results of the analysis historical traffic state data, and determining the traffic event triggering type; Wherein the traffic event trigger type includes a rapid burst event and a non-rapid burst event; based on the determined traffic event trigger type, a traffic linkage early warning instruction is generated, and an early warning or linkage scheduling strategy is issued to a traffic control system. Urban traffic linkage early warning system based on wisdom monitoring, this system includes: The traffic data acquisition module is used for acquiring multi-source traffic monitoring data, and carrying out multi-mode fusion processing on the multi-source traffic monitoring data to form a traffic state data sequence of a target road network; The traffic change analysis module is used for constructing a traffic change curve based on the traffic state data sequence, separating change characteristics of the traffic change curve, classifying the change characteristics into a rapid change component and a trend change component, and carrying out quantization treatment on the separated change characteristics; The traffic event determining module is used for respectively carrying out feature clustering and event triggering rule extraction on the rapid change component and the trend change component through the combination of the change features and the statistical analysis result of the analysis historical traffic state data to determine the traffic event triggering type; Wherein the traffic event trigger type includes a rapid burst event and a non-rapid burst event; and the linkage early warning module is used for