CN112489421A - Burst congestion judging method and system based on multi-source traffic big data fusion - Google Patents
Burst congestion judging method and system based on multi-source traffic big data fusion Download PDFInfo
- Publication number
- CN112489421A CN112489421A CN202011295441.0A CN202011295441A CN112489421A CN 112489421 A CN112489421 A CN 112489421A CN 202011295441 A CN202011295441 A CN 202011295441A CN 112489421 A CN112489421 A CN 112489421A
- Authority
- CN
- China
- Prior art keywords
- data
- congestion
- model
- data set
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000004927 fusion Effects 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 53
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 239000000049 pigment Substances 0.000 claims abstract description 9
- 238000010276 construction Methods 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 25
- 238000013528 artificial neural network Methods 0.000 claims description 19
- 238000011478 gradient descent method Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 12
- 230000000694 effects Effects 0.000 abstract description 6
- 238000011156 evaluation Methods 0.000 description 6
- 238000007667 floating Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000035605 chemotaxis Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a burst congestion judging method and a system based on multi-source traffic big data fusion, wherein an acquired real-time road condition map is vectorized, a corresponding topological relation is constructed based on interrupted discrete points, then the acquired multi-source traffic big data is fused, a corresponding data set is constructed and stored, then a neural network model is utilized to train a training data set, a square error sum loss function is utilized to carry out convergence, a corresponding judging model is obtained, and then any plurality of multi-source traffic big data in the data set are input into the verified judging model for training after being zeroed, so that a congestion judging model is obtained; and finally, acquiring a plurality of burst data based on the topological relation, inputting the burst data into the congestion judging model, establishing a corresponding topological grade relation according to the prediction result, and carrying out road condition pigment assignment, so that the processing effect on the burst congestion can be improved.
Description
Technical Field
The invention relates to the technical field of traffic management, in particular to a sudden congestion judging method and system based on multi-source traffic big data fusion.
Background
Compared with the common congestion in the urban road network, the time and place randomness of the sudden congestion is high, and the information of the sudden congestion event is obtained by means of video manual inspection, citizen reporting and the like in urban traffic management. Through food monitoring equipment such as electronic police, bayonet monitoring that distribute in the urban road network, can discern some sudden congestion events through the mode of artificial observation, but because personnel are equipped with inadequately, video manual inspection is all the road network and often needs several hours once, causes the omission of event. In addition, residents report events through crowdsourcing ways such as 122 alarming and public numbers, further verification and examination are needed, and timeliness of congestion events is difficult to guarantee. The internet enterprises perform congestion identification through floating car sampling data, the average road sampling rate is less than 20%, and randomness and contingency have great influence on the accuracy of identification results. In summary, the accuracy and timeliness of the conventional urban sudden traffic congestion detection method cannot be guaranteed, and the prediction analysis of the road network caused by congestion is one-sided, and is easily influenced by human factors, so that the processing effect on sudden congestion is poor.
Disclosure of Invention
The invention aims to provide a method and a system for judging sudden congestion based on multi-source traffic big data fusion, which improve the effect of processing sudden congestion.
In order to achieve the above object, in a first aspect, the present invention provides a method for judging sudden congestion based on multi-source traffic big data fusion, including the following steps:
carrying out road network vectorization on the real-time road condition graph, and constructing a topological relation among discrete points based on traffic flow direction;
fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion, and constructing a data set;
training a training data set by using a neural network model, and converging by using an error square sum loss function to obtain a corresponding discrimination model;
after being zeroed, any plurality of multi-source traffic big data in the data set are input into the verified discrimination model for training to obtain a congestion discrimination model;
and acquiring a plurality of burst data based on the topological relation, and outputting a corresponding prediction result by using the congestion judging model.
Wherein after obtaining the prediction result, the method further comprises:
and establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information.
The method comprises the following steps of carrying out road network vectorization on a real-time road condition map, and constructing a topological relation among discrete points based on traffic flow, wherein the road network vectorization comprises the following steps:
and importing the acquired real-time road condition map into a geographic information system to create a pyramid, creating an auxiliary file to vectorize the real-time road condition map, and marking intersection points.
The method comprises the following steps of training a training data set by using a neural network model, converging by using an error square sum loss function, and obtaining a corresponding discriminant model, wherein the method comprises the following steps:
performing a back-put extraction sampling on the data set in an elastic distributed data set form to obtain a training data set; and inputting the training data set into a neural network model by using an ensemble learning algorithm for reverse iterative training.
Wherein, utilize neural network model to train the training data set to utilize error square sum loss function to converge, obtain corresponding discriminant model, still include:
and carrying out convergence judgment on the training values of the training data set by utilizing a square sum of error loss function and a first threshold value, and then inputting the data except the training data set into the neural network as a test data set for verification to obtain corresponding output values.
Acquiring a plurality of burst data based on the topological relation, and outputting a corresponding prediction result by using the congestion discrimination model, wherein the method comprises the following steps:
and performing back propagation iteration according to an error function between a set output value and the corresponding output value, and simultaneously adjusting the neural network parameters by using a gradient descent method and a batch update method until all the output values meet a threshold range to obtain a corresponding discriminant model.
In a second aspect, the invention provides a sudden congestion judging system based on multi-source traffic big data fusion, and the sudden congestion judging method based on multi-source traffic big data fusion as described in the first aspect is applied to a sudden congestion judging system based on multi-source traffic big data fusion,
the burst congestion distinguishing system based on the multi-source traffic big data fusion comprises a data acquisition module, a road network construction module, a data fusion module, a distinguishing model construction module, a congestion distinguishing model construction module and a data linkage prediction module, wherein the data acquisition module is connected with the road network construction module, the data fusion module and the data linkage prediction module, the data fusion module, the distinguishing model construction module, the congestion distinguishing model construction module and the data linkage prediction module are sequentially connected, and the data linkage prediction module is also connected with the road network construction module;
the data acquisition module is used for acquiring a real-time road condition map and multi-source traffic big data acquired based on a big data platform;
the road network construction module is used for carrying out road network vectorization on the real-time road condition map and constructing a topological relation among discrete points based on traffic flow direction;
the data fusion module is used for fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion and constructing a data set;
the discriminant model construction module is used for inputting the constructed training data set into a neural network model by using an ensemble learning algorithm for reverse iterative training, inputting the test data set into the converged neural network for directional propagation iteration, and adjusting the neural network parameters by using a gradient descent method and a batch update method to obtain a corresponding discriminant model;
the congestion distinguishing model building module is used for performing zeroing on any multiple multi-source traffic big data in the data set and inputting the zeroed data into the verified distinguishing model for training to obtain a congestion distinguishing model;
and the data linkage prediction module is used for acquiring a plurality of burst data based on the topological relation, outputting a corresponding prediction result by using the congestion discrimination model, establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information.
The invention relates to a burst congestion distinguishing method and a system based on multi-source traffic big data fusion, which comprises the steps of firstly leading an acquired real-time road condition map into a geographic information system for vectorization, constructing a corresponding topological relation based on interrupted discrete points, then fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion, constructing a corresponding data set for storage, then inputting a training data set into a neural network model by using an integrated learning algorithm for reverse iterative training, then inputting a test data set into a converged neural network, performing reverse propagation iteration based on an error function between threshold values, simultaneously adjusting parameters of the neural network by using a gradient descent method and a batch updating method to obtain a corresponding distinguishing model, then carrying out zero resetting on any plurality of multi-source traffic big data in the data set, and inputting the verified distinguishing model for training, obtaining a congestion judging model; and finally, acquiring a plurality of burst data based on the topological relation, inputting the burst data into the congestion judging model, establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information, so that the processing effect on the burst congestion can be improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic step diagram of a method for judging sudden congestion based on multi-source traffic big data fusion provided by the invention.
Fig. 2 is a schematic structural diagram of a sudden congestion judging system based on multi-source traffic big data fusion provided by the invention.
The system comprises a data acquisition module, a 2-road network construction module, a 3-data fusion module, a 4-discrimination model construction module, a 5-congestion discrimination model construction module and a 6-data linkage prediction module.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
In the description of the present invention, it is to be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships illustrated in the drawings, and are used merely for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention. Further, in the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
Referring to fig. 1, the present invention provides a method for determining sudden congestion based on multi-source traffic big data fusion, including the following steps:
s101, carrying out road network vectorization on the real-time road condition graph, and constructing a topological relation between discrete points based on traffic flow.
Specifically, a real-time road condition map is imported into a geographic information system to create a pyramid, a shapefile file is created, the real-time road condition is vectorized, a part with road condition information is vectorized, vector lines of the intersected part of a road section are intersected in the vectorization process, the vector lines are required to be completely within the range of the road condition information, and the accuracy of the subsequent pigment identification is ensured. And then, uniformly breaking the vectorized road network into points at equal intervals by using a point turning tool, wherein the breaking intervals are 40-70 m, each section only contains one road condition information, the data volume is overlarge due to too small intervals, and two sections of different road conditions are connected together due to overlarge intervals, so that congestion judgment errors are caused.
And then according to the geographic information system, marking discrete points of the intersection points to indicate that information such as intersections exist, and according to the path information and the traffic flow direction on the geographic information system, constructing a corresponding traffic topological relation network consisting of a plurality of rays, so as to facilitate the subsequent congestion analysis according to the topological relation.
S102, fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion, and constructing a data set.
Specifically, the multi-source traffic data comprises traffic parameter characteristic data, time interval characteristic data and environment characteristic data; the traffic parameter characteristic data are traffic data related to traffic jam, the traffic parameter characteristic data comprise traffic data detected by an electronic police, a microwave device, a multi-target radar, an induction coil and a floating car, the time interval characteristic data are time interval data related to the traffic jam, and the time interval characteristic data comprise day type data and time interval type data; the environment type characteristic data is environment data related to traffic jam, and the environment type characteristic data comprises road section length, lane number and road grade of the road section.
Because the types of the collected data are different, the evaluation indexes have different properties, and generally have different dimensions and orders of magnitude. When the levels of the indexes are greatly different, if the original index values are directly used for analysis, the function of the indexes with higher numerical values in the comprehensive analysis is highlighted, and the function of the indexes with lower numerical levels is relatively weakened. Therefore, in order to ensure the reliability of the result, the raw index data needs to be standardized. The data standardization processing mainly comprises two aspects of data chemotaxis processing and dimensionless processing. The data homochemotaxis processing mainly solves the problem of data with different properties, directly sums indexes with different properties and cannot correctly reflect the comprehensive results of different acting forces, and firstly considers changing the data properties of inverse indexes to ensure that all the indexes are homochemotactic for the acting forces of the evaluation scheme and then sum to obtain correct results. The data dimensionless process mainly addresses the comparability of data. Through the standardization processing, the original data are all converted into non-dimensionalized index mapping evaluation values, namely, all index values are in the same quantity level, and comprehensive evaluation analysis can be carried out.
And constructing corresponding membership functions according to different application environments and an expert system, and fuzzifying the measured values of the evaluation levels by each index. Assuming that m indexes and n evaluation levels are provided, different membership functions r can be obtainedij。
Each measured data XiRespectively substituting into membership functions rijThe fuzzy relation matrix obtained in the method is as follows:
the weight of each evidence is calculated:
calculating the fuzzy matrix after weight modification:
RF＝R×W
and fusing the fuzzy data by using an iterative fusion algorithm to obtain a final fusion result. For example, the corresponding traffic parameter class feature data, the time interval class feature data and the environment class feature data can be arranged and combined according to the calculated weight and the fuzzy matrix according to the same time value, and then the corresponding data set is constructed to store the combined data, so that the subsequent data extraction is facilitated.
S103, training the training data set by using the neural network model, and converging by using the error square sum loss function to obtain a corresponding discriminant model.
Specifically, a task is created and executed based on Spark big data processing technology, a data set is subjected to playback extraction sampling in an elastic distributed data set (RDD) form to obtain a training data set, wherein, the number of the elastic distributed data sets is a plurality, a plurality of corresponding training data sets can be obtained, the training data sets are input into the neural network model by utilizing the integrated learning algorithm for reverse iterative training, a plurality of corresponding predicted values are obtained, calculating function loss values corresponding to the plurality of predicted values by using the error sum of squares loss function, comparing the function loss values with a first threshold value, determining whether the function loss values are larger than the first threshold value, and if the function loss values are smaller than the first threshold value, continuing to train and iterate the neural network model, and if the training iteration is larger than the first threshold value, judging that the neural network model is converged. The purpose of convergence is to avoid large errors in subsequent congestion determination.
And then inputting data except the training data set into the neural network as a test data set for verification to obtain corresponding output values, performing back propagation iteration according to an error function between a set output value and the corresponding output value, and simultaneously adjusting parameters of the neural network by using a gradient descent method and a batch update method until all the output values meet a threshold range to obtain a corresponding discrimination model. The BP neural network is also called as a reverse error neural network, is actually one of multilayer perceptrons and comprises a hidden layer, an input layer and an output layer, wherein the input layer of the BPNN model transmits data signals to the hidden layer, and the hidden layer transmits processed data signals to the output layer. If the output result is opposite to the expected output value, the error is reversely propagated to the hidden layer, the hidden layer corrects the weight value and the threshold value by adopting a chain rule, the process is repeated until the output result is basically consistent with the expected output value, the network training is stopped, and the related parameters of the network are stored. In order to achieve the purpose of training, the weight parameters of all hidden layer neurons are adjusted by a gradient descent method, the actual output of the neural network is close to the expected output as far as possible, and the accuracy of the output data is guaranteed as far as possible.
And S104, performing zeroing on any multiple multi-source traffic big data in the data set, inputting the verified discrimination model, and training to obtain a congestion discrimination model.
Specifically, in order to increase the use scenes and range of the discrimination model and avoid the influence on the normal use discrimination of the model when data are missed or missing and the like, training on the missing data is added on the basis of the discrimination model.
Setting traffic data detected by one or more detection devices in electronic police, microwave equipment, multi-target radar, induction coils and floating cars in a multi-source traffic data set as 0, inputting corresponding other time period class characteristic data and environment class characteristic data into the discrimination model for training to obtain a corresponding output result, if the output result and the training result of normal complete data are not in a set error range, continuing to train by adopting the next 0-returning traffic parameter class characteristic data until the result of the discrimination model and normal data is in the error range, then returning the environment class characteristic data to 0, inputting other two classes of normal abnormal data into the discrimination model, and comparing and analyzing the output result and the normal data result in the same way until the result of the discrimination model trained by the abnormal data and the discrimination model trained by the normal data is in the normal range, and taking the corresponding discrimination model as a final congestion discrimination model. The method can avoid influencing the judgment result and avoid misjudgment under the condition that partial data of the acquired multi-source traffic data is lost.
And S105, acquiring a plurality of burst data based on the topological relation, and outputting corresponding prediction results by using the congestion judging model.
Specifically, after acquiring current congestion data, acquiring current burst data on all topology lines at corresponding intersection points according to the established topological relation, namely traffic data on all paths in traffic connection with a current suddenly congested road section, inputting the burst data into the congestion judging model to obtain corresponding prediction results, connecting all prediction results on the corresponding topology lines according to the connecting relation on the topological relation, establishing a corresponding topological grade relation, and performing early warning on the traffic condition at the next stage according to the real-time change condition of the acquired data, wherein the three colors of green, yellow and red are usually used for representing the three path conditions of smooth, congested and congested, and displayed in real time, so that the congestion spreading condition can be known conveniently, and corresponding solving measures can be found and taken timely, under the accurate judgment condition, the processing effect on the congestion is improved.
Referring to fig. 2, the present invention provides a sudden congestion identification system based on multi-source traffic big data fusion, and the method for judging sudden congestion based on multi-source traffic big data fusion is applied to a sudden congestion identification system based on multi-source traffic big data fusion,
the burst congestion distinguishing system based on the multi-source traffic big data fusion comprises a data acquisition module 1, a road network construction module 2, a data fusion module 3, a distinguishing model construction module 4, a congestion distinguishing model construction module 5 and a data linkage prediction module 6, wherein the data acquisition module 1 is connected with the road network construction module 2, the data fusion module 3 and the data linkage prediction module 6, the data fusion module 3, the distinguishing model construction module 4, the congestion distinguishing model construction module 5 and the data linkage prediction module 6 are sequentially connected, and the data linkage prediction module 6 is also connected with the road network construction module 2;
the data acquisition module 1 is used for acquiring a real-time road condition map and multi-source traffic big data acquired based on a big data platform;
the road network construction module 2 is configured to perform road network vectorization on the real-time road condition map, and construct a topological relation between discrete points based on a traffic flow direction;
the data fusion module 3 is used for fusing the acquired multi-source traffic big data by using an iterative algorithm utilizing fuzzy fusion and constructing a data set;
the discriminant model construction module 4 is configured to input the constructed training data set into a neural network model by using an ensemble learning algorithm to perform reverse iterative training, input the test data set into a converged neural network to perform directional propagation iteration, and adjust the neural network parameters by using a gradient descent method and a batch update method to obtain a corresponding discriminant model;
the congestion distinguishing model building module 5 is configured to input the verified distinguishing model after zeroing any multiple multi-source traffic big data in the data set, and train the input result to obtain a congestion distinguishing model;
and the data linkage prediction module 6 is configured to obtain a plurality of burst data based on the topological relation, output a corresponding prediction result by using the congestion discrimination model, establish a corresponding topological grade relation according to the prediction result, and trigger corresponding road condition pigment assignment early warning information.
In this embodiment, for a specific limitation of the sudden congestion determination system based on the multi-source traffic big data fusion, reference may be made to the above limitation of the sudden congestion determination method based on the multi-source traffic big data fusion, and details are not described herein again. All modules in the system for judging the sudden congestion based on the multi-source traffic big data fusion can be completely or partially realized through software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
The invention relates to a burst congestion distinguishing method and a system based on multi-source traffic big data fusion, which comprises the steps of firstly leading an acquired real-time road condition map into a geographic information system for vectorization, constructing a corresponding topological relation based on interrupted discrete points, then fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion, constructing a corresponding data set for storage, then inputting a training data set into a neural network model by using an integrated learning algorithm for reverse iterative training, then inputting a test data set into a converged neural network, performing reverse propagation iteration based on an error function between threshold values, simultaneously adjusting parameters of the neural network by using a gradient descent method and a batch updating method to obtain a corresponding distinguishing model, then carrying out zero resetting on any plurality of multi-source traffic big data in the data set, and inputting the verified distinguishing model for training, obtaining a congestion judging model; and finally, acquiring a plurality of burst data based on the topological relation, inputting the burst data into the congestion judging model, establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information, so that the processing effect on the burst congestion can be improved.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. A burst congestion judging method based on multi-source traffic big data fusion is characterized by comprising the following steps:
carrying out road network vectorization on the real-time road condition graph, and constructing a topological relation among discrete points based on traffic flow direction;
fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion, and constructing a data set;
training a training data set by using a neural network model, and converging by using an error square sum loss function to obtain a corresponding discrimination model;
after being zeroed, any plurality of multi-source traffic big data in the data set are input into the verified discrimination model for training to obtain a congestion discrimination model;
and acquiring a plurality of burst data based on the topological relation, and outputting a corresponding prediction result by using the congestion judging model.
2. The method for judging the sudden congestion based on the fusion of the multi-source traffic big data as claimed in claim 1, wherein after obtaining the prediction result, the method further comprises:
and establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information.
3. The method for judging the sudden congestion based on the multi-source traffic big data fusion as claimed in claim 1, wherein the road network vectorization is performed on the real-time road condition graph, and the topological relation among the discrete points is constructed based on the traffic flow direction, and the method comprises the following steps:
and importing the acquired real-time road condition map into a geographic information system to create a pyramid, creating an auxiliary file to vectorize the real-time road condition map, and marking intersection points.
4. The method for judging the sudden congestion based on the multi-source traffic big data fusion as claimed in claim 3, wherein the training data set is trained by using a neural network model, and the error square sum loss function is used for convergence to obtain a corresponding judgment model, comprising:
performing a back-put extraction sampling on the data set in an elastic distributed data set form to obtain a training data set; and inputting the training data set into a neural network model by using an ensemble learning algorithm for reverse iterative training.
5. The method of claim 4, wherein a neural network model is used to train a training data set, and a square error sum loss function is used to converge to obtain a corresponding discriminant model, and further comprising:
and carrying out convergence judgment on the training values of the training data set by utilizing a square sum of error loss function and a first threshold value, and then inputting the data except the training data set into the neural network as a test data set for verification to obtain corresponding output values.
6. The method for judging the sudden congestion based on the fusion of the multi-source traffic big data as claimed in claim 1, wherein the step of obtaining a plurality of sudden data based on the topological relation and outputting the corresponding prediction result by using the congestion judgment model comprises:
and performing back propagation iteration according to an error function between a set output value and the corresponding output value, and simultaneously adjusting the neural network parameters by using a gradient descent method and a batch update method until all the output values meet a threshold range to obtain a corresponding discriminant model.
7. A burst congestion judging system based on multi-source traffic big data fusion is applied to the burst congestion judging system based on the multi-source traffic big data fusion according to any one of claims 1 to 6,
the burst congestion distinguishing system based on the multi-source traffic big data fusion comprises a data acquisition module, a road network construction module, a data fusion module, a distinguishing model construction module, a congestion distinguishing model construction module and a data linkage prediction module, wherein the data acquisition module is connected with the road network construction module, the data fusion module and the data linkage prediction module, the data fusion module, the distinguishing model construction module, the congestion distinguishing model construction module and the data linkage prediction module are sequentially connected, and the data linkage prediction module is also connected with the road network construction module;
the data acquisition module is used for acquiring a real-time road condition map and multi-source traffic big data acquired based on a big data platform;
the road network construction module is used for carrying out road network vectorization on the real-time road condition map and constructing a topological relation among discrete points based on traffic flow direction;
the data fusion module is used for fusing the acquired multi-source traffic big data by using an iterative algorithm of fuzzy fusion and constructing a data set;
the discriminant model construction module is used for inputting the constructed training data set into a neural network model by using an ensemble learning algorithm for reverse iterative training, inputting the test data set into the converged neural network for directional propagation iteration, and adjusting the neural network parameters by using a gradient descent method and a batch update method to obtain a corresponding discriminant model;
the congestion distinguishing model building module is used for performing zeroing on any multiple multi-source traffic big data in the data set and inputting the zeroed data into the verified distinguishing model for training to obtain a congestion distinguishing model;
and the data linkage prediction module is used for acquiring a plurality of burst data based on the topological relation, outputting a corresponding prediction result by using the congestion discrimination model, establishing a corresponding topological grade relation according to the prediction result, and triggering corresponding road condition pigment assignment early warning information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011295441.0A CN112489421A (en) | 2020-11-18 | 2020-11-18 | Burst congestion judging method and system based on multi-source traffic big data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011295441.0A CN112489421A (en) | 2020-11-18 | 2020-11-18 | Burst congestion judging method and system based on multi-source traffic big data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112489421A true CN112489421A (en) | 2021-03-12 |
Family
ID=74931625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011295441.0A Pending CN112489421A (en) | 2020-11-18 | 2020-11-18 | Burst congestion judging method and system based on multi-source traffic big data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112489421A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470347A (en) * | 2021-05-20 | 2021-10-01 | 上海天壤智能科技有限公司 | Congestion identification method and system combining bayonet vehicle passing record and floating vehicle GPS data |
| CN116403400A (en) * | 2023-03-15 | 2023-07-07 | 之江实验室 | Method and system for detecting abnormal value of traffic flow data of radar video all-in-one machine |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140114885A1 (en) * | 2012-10-18 | 2014-04-24 | Enjoyor Company Limited | Urban traffic state detection based on support vector machine and multilayer perceptron |
| CN106448167A (en) * | 2016-11-17 | 2017-02-22 | 天津市市政工程设计研究院 | Real-time-road-condition-based congestion point determination method |
| CN106971536A (en) * | 2017-04-17 | 2017-07-21 | 武汉云砥信息科技有限公司 | It is a kind of to merge Floating Car and the highway congestion status detection method of trackside video |
| KR20190109121A (en) * | 2018-03-16 | 2019-09-25 | 울산과학기술원 | A Unified Deep Learning Model for Time Series Data Prediction |
| CN111127888A (en) * | 2019-12-23 | 2020-05-08 | 广东工业大学 | An urban traffic flow prediction method based on multi-source data fusion |
| CN111540199A (en) * | 2020-04-21 | 2020-08-14 | 浙江省交通规划设计研究院有限公司 | High-speed traffic flow prediction method based on multi-mode fusion and graph attention machine mechanism |
| CN111680745A (en) * | 2020-06-08 | 2020-09-18 | 青岛大学 | A method and system for sudden congestion discrimination based on multi-source traffic big data fusion |
| CN111950627A (en) * | 2020-08-11 | 2020-11-17 | 重庆大学 | A multi-source information fusion method and its application |
-
2020
- 2020-11-18 CN CN202011295441.0A patent/CN112489421A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140114885A1 (en) * | 2012-10-18 | 2014-04-24 | Enjoyor Company Limited | Urban traffic state detection based on support vector machine and multilayer perceptron |
| CN106448167A (en) * | 2016-11-17 | 2017-02-22 | 天津市市政工程设计研究院 | Real-time-road-condition-based congestion point determination method |
| CN106971536A (en) * | 2017-04-17 | 2017-07-21 | 武汉云砥信息科技有限公司 | It is a kind of to merge Floating Car and the highway congestion status detection method of trackside video |
| KR20190109121A (en) * | 2018-03-16 | 2019-09-25 | 울산과학기술원 | A Unified Deep Learning Model for Time Series Data Prediction |
| CN111127888A (en) * | 2019-12-23 | 2020-05-08 | 广东工业大学 | An urban traffic flow prediction method based on multi-source data fusion |
| CN111540199A (en) * | 2020-04-21 | 2020-08-14 | 浙江省交通规划设计研究院有限公司 | High-speed traffic flow prediction method based on multi-mode fusion and graph attention machine mechanism |
| CN111680745A (en) * | 2020-06-08 | 2020-09-18 | 青岛大学 | A method and system for sudden congestion discrimination based on multi-source traffic big data fusion |
| CN111950627A (en) * | 2020-08-11 | 2020-11-17 | 重庆大学 | A multi-source information fusion method and its application |
Non-Patent Citations (3)
| Title |
|---|
| J SONG ET AL.: "Mapping spatio-temporal patterns and detecting the factors of traffic congestion with multi-source data fusion and mining techniques", 《COMPUTERS, ENVIRONMENT AND URBAN SYSTEMS》 * |
| 白静: "基于动态交通流信息的城市交通突发拥堵控制与诱导研究", 《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》 * |
| 闫学东等: "基于浮动车大数据与网格模型的城市交通识别和评价研究", 《北京交通大学学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470347A (en) * | 2021-05-20 | 2021-10-01 | 上海天壤智能科技有限公司 | Congestion identification method and system combining bayonet vehicle passing record and floating vehicle GPS data |
| CN116403400A (en) * | 2023-03-15 | 2023-07-07 | 之江实验室 | Method and system for detecting abnormal value of traffic flow data of radar video all-in-one machine |
| CN116403400B (en) * | 2023-03-15 | 2025-08-29 | 之江实验室 | A radar video integrated machine traffic flow data outlier detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280553B (en) | Mountain torrent disaster risk zoning and prediction method based on GIS-neural network integration | |
| Kim et al. | Calibration of microsimulation models using nonparametric statistical techniques | |
| CN111462488A (en) | An Intersection Safety Risk Assessment Method Based on Deep Convolutional Neural Network and Intersection Behavior Feature Model | |
| CN103714185B (en) | Subject event updating method base and urban multi-source time-space information parallel updating method | |
| CN110210681B (en) | A method for predicting PM2.5 value of monitoring sites based on distance | |
| CN109783967B (en) | Landslide prediction method, device and storage medium based on random forest model | |
| CN106021751A (en) | Land utilization change simulation method for coastal zone based on CA and SAR | |
| CN106250667A (en) | The monitoring method of a kind of landslide transition between states of paddling and device | |
| CN112613225B (en) | Intersection traffic state prediction method based on neural network cell transmission model | |
| CN114066288B (en) | Intelligent data center-based emergency detection method and system for operation road | |
| CN112489421A (en) | Burst congestion judging method and system based on multi-source traffic big data fusion | |
| CN103606274A (en) | Urban road traffic state assessment method | |
| CN109360421A (en) | A kind of traffic information predicting method and device, electric terminal based on machine learning | |
| CN103942251A (en) | Method and system for inputting high altitude meteorological data into database based on multiple quality control methods | |
| CN109460914A (en) | Method is determined based on the bridge health grade of semi-supervised error correction study | |
| CN109544926B (en) | Traffic flow restoration method based on intersection correlation | |
| CN118037065A (en) | Dynamic evaluation method for inland water navigation risk under data driving | |
| CN117009772A (en) | Road traffic safety prediction method and prediction system based on risk intelligent perception model | |
| CN118095834A (en) | A traffic accident risk assessment method based on interpretable random forest | |
| CN116433218A (en) | Self-organizing mapping clustering-based mine mechanical equipment online health assessment method | |
| CN118297414A (en) | Mine water hazard forecasting electronic equipment and computer program product | |
| CN114417729A (en) | An early warning method for mining environment safety based on BP neural network | |
| CN118627889A (en) | Preventive maintenance method based on road damage prediction | |
| CN114996318B (en) | Automatic judgment method and system for processing mode of abnormal value of detection data | |
| CN109214937A (en) | The anti-fraud determination method of settlement of insurance claim intelligence and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210312 |
|
| RJ01 | Rejection of invention patent application after publication |