CN116597656A - Method, equipment and medium for predicting road traffic flow based on big data analysis - Google Patents
Method, equipment and medium for predicting road traffic flow based on big data analysis Download PDFInfo
- Publication number
- CN116597656A CN116597656A CN202310872227.4A CN202310872227A CN116597656A CN 116597656 A CN116597656 A CN 116597656A CN 202310872227 A CN202310872227 A CN 202310872227A CN 116597656 A CN116597656 A CN 116597656A
- Authority
- CN
- China
- Prior art keywords
- data
- road section
- traffic flow
- traffic
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000007405 data analysis Methods 0.000 title claims abstract description 8
- 238000011144 upstream manufacturing Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 241000288105 Grus Species 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000004891 communication Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000003442 weekly effect Effects 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims 1
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The application belongs to the field of road traffic prediction, and particularly relates to a method, equipment and medium for predicting road traffic flow based on big data analysis, which are used for solving the problem of low prediction precision when nonlinear traffic flow data are processed by the existing traffic flow prediction method. The method comprises the following steps: firstly, selecting data by using a cycle average spatial correlation coefficient, selecting partial road sections with the largest correlation, then, carrying out feature extraction on flow data of the selected road sections by using an automatic encoder, finally, inputting upstream and downstream flow after feature extraction and current road section t period flow into a network formed by two layers of GRUs, taking a feature vector output by the GRU network as a person inputting of an SVR model, training the SVR model, predicting by using the SVR model, and carrying out denormalization on model output to obtain a prediction result.
Description
Technical Field
The application belongs to the field of traffic prediction, and particularly relates to a road traffic flow prediction method based on big data analysis.
Background
In recent years, with the vigorous development of big data analysis technology, the data of various industries are explosively increased. The increase of traffic data such as vehicle trajectories, vehicle flows, road sensors and the like is geometric times that of the prior art, and how to process various traffic data has become one of the most interesting tasks in constructing intelligent traffic systems. Traffic flow prediction is a widely studied problem, and accurate and timely traffic flow prediction can not only relieve traffic jam and other problems, but also save various resources. Traffic flow data contains both temporal and spatial correlations, so how to efficiently mine the spatio-temporal relationships between data is challenging. With the development of deep learning, more and more researchers have introduced deep learning into traffic prediction problems. Prior scholars have begun to optimize based on kernel functions, and the optimal kernel function will continue to be the focus of research on SVM-based traffic flow prediction models. The SVM maps low-dimensional data onto a high-dimensional feature space when processing nonlinear data, the complexity of an algorithm is necessarily increased due to the increase of dimensions, the cost of time overhead is also required to be solved when processing complex traffic flow data, and the prediction accuracy of traffic flow is required to be further improved due to the nonlinearity of traffic flow.
Disclosure of Invention
In order to solve the defects in the prior art, the application provides a road traffic flow prediction method based on big data analysis.
The technical scheme adopted by the application is that the method for predicting the road traffic flow based on big data analysis comprises the following steps:
the traffic flow of a certain time point after the current road section t+1 is predicted by using traffic flow data of the current road section t+1 and the upstream road section and the downstream road section in the period from t-1 to t+1, firstly, data selection is carried out on the traffic flow of the current road section t+1 by using a cycle average spatial correlation coefficient, a part of road sections with the largest correlation is selected, then, the traffic flow data X ' ua (t) of the selected road sections is subjected to feature extraction by using an automatic encoder, finally, the upstream and downstream traffic X ' ' ua (t) after the feature extraction and the traffic flow of the current road section t are input into a network formed by two layers of GRUs together, then, the feature vector output by the GRU network is used as an input of an SVR model, the SVR model is trained, the SVR model is predicted, and model output is denormalized to obtain a prediction result, and the specific steps comprise:
step1, assume that the current road segment is x 0 N data-bearing two-level upstream and downstream road segments are combined, and the space road segment set of the network is defined as F { x } 0 ，x 1 ,., xn. Wherein x is 0 Representing the current road section, x 1 To x n For its secondary upstream and downstream road segments. Suppose each road section x i A continuous time series with a length q is denoted as F i Representing the road section x i For the sequence of the sky traffic of (1)Then the period input flow F is constructed by taking the day as the period unit d Is F d =[F 1 ，F 2 ， F 3 ， ...F q ] T Setting training times K;
step2, inputting M-dimensional upstream and downstream road section flow data to an automatic encoder, and then obtaining output, wherein the training frequency is increased by 1;
step3, calculating a reconstruction error;
step4, updating the weight of the automatic encoder, returning to Step2 when the training times do not reach the set value and the reconstruction error is greater than the threshold value d, and otherwise entering Step5.
Step5, dimension reduction is carried out on the M-dimension upstream and downstream data to M-dimension based on the trained automatic encoder.
Step6, normalizing the current road section flow data and inputting the normalized current road section flow data and the dimension-reduced m-dimension characteristic data into the GRU network;
step7, training a GRU network model based on the processed data, and learning the input data;
step8, inputting the feature vector output by the GRU network into an SVR model, and then training the SVR model;
step9, after training the model, inputting the traffic flow data of the current road section at the current moment into the model, and finally outputting the traffic flow predicted value of the current road section at the next moment by the model.
Preferably, the normalization process includes: traffic flow data preprocessing, carrying out normalization processing on the current road section flow data, wherein the calculation formula is as follows: f (f) i norm =(f i -f min )/(f max -f min ) In f i The ith data representing a feature takes on value and f max And f min Representing the maximum and minimum values, f, respectively, of the feature i norm The normalized result of the feature item i data.
Preferably, the step of calculating the weekly average spatial correlation coefficient is as follows: for F d Calculating correlation coefficient for each line，/>Representing the correlation coefficient between road segment 0 and road segment i on day d, from which a matrix of peri-phasic relations can be derived>According to this, the weekly average correlation coefficient is calculated +.>And (3) screening out M pieces of data with highest similarity from n pieces of upstream and downstream of the cycle average correlation coefficient, and inputting the M pieces of road section data serving as upstream and downstream road section data.
Preferably, in the network structure of the network layer of the two-layer GRU, the number of input layer units is 12, the number of hidden layer units is 135, the activation function selects sigmoid, and the initialization function selects he_unitorm.
Preferably, the kernel function of the SVR model is radial basis, gamma is 0.3, epsilon is 0.001.
Accordingly, the present application provides an electronic device, comprising: the device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; the memory has stored therein a computer program which, when executed by the processor, causes the processor to perform the steps of any of the feature selection methods described above.
Accordingly, the present application proposes a computer readable storage medium storing a computer program executable by a processor, which when run on the processor causes the processor to perform the steps of any one of the above-mentioned feature selection methods.
The traffic flow method provided by the application has the beneficial effects that the advantages that GRU has a long-time memory function and the model is relatively simple are utilized, so that the high-precision prediction of the traffic flow of the expressway is realized, the decision basis is provided for the decision management of the road management department, the guidance of the traffic flow of the expressway is realized, and the road service quality is further improved; secondly, in order to fully consider the time-space correlation of traffic flow, a cycle average space correlation coefficient is introduced to process data, and an automatic encoder is adopted to keep the characteristics of the data as far as possible; finally, after being processed by an automatic encoder and GRU, the traffic flow is sent to svr, and the traffic flow is predicted better by means of a mixed model of three models, so that the prediction precision is improved, and the defects of svr, such as low operation speed and high time cost, are overcome to a certain extent.
Detailed Description
The present application is described in detail below, and further, if detailed description of known techniques is not necessary for the illustrated features of the present application, it will be omitted. The embodiments described below are exemplary only and are not to be construed as limiting the application.
It will be understood by those skilled in the art that all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs unless defined otherwise. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The following describes the technical scheme of the present application and how the technical scheme of the present application solves the above technical problems in detail with specific embodiments.
And predicting the flow of a certain time point after the current road section t+1 by using the current road section and upstream and downstream road section data in the time period from t-1 to t+1, wherein z (t+1) is the prediction output. For the upstream and downstream data Xua (t) of the period t, firstly, data selection is carried out on the data using a cycle average spatial correlation coefficient, a part of road sections with the largest correlation is selected, then, the flow data X ' ua (t) of the selected road sections are subjected to feature extraction by using an automatic encoder, finally, the upstream and downstream flow X ' ' ua (t) after the feature extraction and the period flow of the current road section t are input into a network formed by two layers of GRUs together, then, the feature vector output by the GRU network is used as the input of an SVR model, the SVR model is trained, the SVR model is used for prediction, and the model output is denormalized to obtain a prediction result.
Step1, assume that the current road segment is x 0 N data-bearing two-level upstream and downstream road segments are combined, and the space road segment set of the network is defined as F { x } 0 ，x 1 ,., xn. Wherein x is 0 Representing the current road section, x 1 To x n For its secondary upstream and downstream road segments. Suppose each road section x i A continuous time series with a length q is denoted as F i Representing the road section x i For the sequence of the sky traffic of (1)Then the period input flow F is constructed by taking the day as the period unit d Is F d =[F 1 ，F 2 ， F 3 ， ...F q ] T Setting training times K;
step2, inputting M-dimensional upstream and downstream road section flow data to an automatic encoder, and then obtaining output, wherein the training frequency is increased by 1;
step3, calculating a reconstruction error;
step4, updating the weight of the automatic encoder, returning to Step2 when the training times do not reach the set value and the reconstruction error is greater than the threshold value d, and otherwise entering Step5.
Step5, dimension reduction is carried out on the M-dimension upstream and downstream data to M-dimension based on the trained automatic encoder.
Step6, normalizing the current road section flow data and inputting the normalized current road section flow data and the dimension-reduced m-dimension characteristic data into the GRU network;
step7, training a GRU network model based on the processed data, and learning the input data;
step8, inputting the feature vector output by the GRU network into an SVR model, and then training the SVR model;
step9, after training the model, inputting the traffic flow data of the current road section at the current moment into the model, and finally outputting the traffic flow predicted value of the current road section at the next moment by the model.
Preprocessing traffic flow data; carrying out maximum and minimum normalization processing on traffic flow data at all moments of an observation point, wherein a calculation formula is as follows: the calculation formula is as follows: f (f) i norm =(f i -f min )/(f max -f min ) In f i The ith data representing a feature takes on value and f max And f min Representing the maximum and minimum values, f, respectively, of the feature i norm The normalized result of the feature item i data.
Preferably, in the network structure of the network layer of the two-layer GRU, the number of input layer units is 12, the number of hidden layer units is 135, the activation function selects sigmoid, and the initialization function selects he_unitorm.
Preferably, the kernel function of the SVR model is radial basis, gamma is 0.3, epsilon is 0.001.
The pearson correlation coefficient of the traffic sequence of two road segments may be used to measure the spatial correlation between the two road segments. Because the traffic flow change of a certain area is closely related to the travel rules of surrounding residents, a certain periodic rule is often presented, and the time rule of the traffic flow is considered in the space correlation calculation process, a weekly average pearson correlation coefficient measurement mode is provided. The method uses data of one continuous week, carries out correlation calculation on data of each day, and finally carries out day correlation coefficient averaging:
where Rd is the day correlation coefficient, k is the calculated number of days, and the week average correlation coefficient k has a value of 70; the specific calculation is as follows:
for F d Calculating correlation coefficient for each line，/>The correlation coefficient of the road segment 0 and the road segment i on the d-th day is represented. According to this, a matrix of the week phase relationship can be obtained>. Calculating the cycle average correlation coefficient +.>And (3) screening out M pieces of data with highest similarity from n pieces of upstream and downstream of the cycle average correlation coefficient, and inputting the M pieces of road section data serving as upstream and downstream road section data.
In addition, in order to construct the input data of the GRU network, the traffic flow data can be processed in a sliding time window mode. Assuming that the lag parameter used for prediction is L, that is, the prediction is performed using a time series of length L, the input of the encoder is a two-dimensional tensor of (L, M), and the dimension of the encoder is M, and the output of the automatic encoder is a two-dimensional tensor of (L, M). And then the m-dimensional feature vector and the normalized current road section flow form an m+1-dimensional vector to be input into a GRU network, and then the output of the GRU is input into an SVM.
Further included is using the MAE as a loss function for svm model training, the function being defined as follows:wherein f i Traffic flow prediction value representing data set at time t +.>Representing the true value of the traffic flow of the data set at the t-th moment, wherein N is the total number of predicted values; continuously updating model parameters through a back propagation algorithm according to the loss function; training a model; after the model is trained, the traffic flow data at the previous moment is input into the model, and finally the model outputs the traffic flow predicted values of all the observation points at the next moment.
Further, the calculation of the reconstruction error is: inputting the data X into an automatic encoder, encoding the data X by an encoder part to obtain Z, re-decoding the Z into Y by a decoder part, enabling the Y to be close to the input X as much as possible, and reconstructing an error calculation formula to be。
The following describes the implementation of the application and the verification of the effect of the method according to the application by means of an example of a simulation data set.
To verify the effectiveness of the methods of the present application, the present application and other algorithms are modeled and analyzed on road segment traffic flow datasets and their prediction accuracy is compared.
Table 1 results of comparative experiments
| Model | RMAE |
| The application is that | 18 |
| SVR | 28 |
| LSTM | 27 |
| GNU | 21 |
As can be seen from table 1, compared with the other methods, the traffic flow prediction method of the present application has higher prediction accuracy, is more favorable for precisely predicting traffic flow, and results in higher RMAE, thereby exhibiting the effectiveness of the present application.
While the application has been described with reference to the presently preferred embodiments, it will be understood by those skilled in the art that the foregoing preferred embodiments are merely illustrative of the present application and are not intended to limit the scope of the application, and any modifications, equivalent substitutions, variations, improvements, etc. that fall within the spirit and scope of the principles of the application are intended to be included within the scope of the appended claims.
Claims (7)
1. A method for road traffic flow prediction based on big data analysis, the method comprising:
the traffic flow of a certain time point after the current road section t+1 is predicted by using traffic flow data of the current road section t+1 and the upstream road section and the downstream road section in the period from t-1 to t+1, firstly, data selection is carried out on the traffic flow of the current road section t+1 by using a cycle average spatial correlation coefficient, a part of road sections with the largest correlation is selected, then, the traffic flow data X ' ua (t) of the selected road sections is subjected to feature extraction by using an automatic encoder, finally, the upstream and downstream traffic X ' ' ua (t) after the feature extraction and the traffic flow of the current road section t are input into a network formed by two layers of GRUs together, then, the feature vector output by the GRU network is used as an input of an SVR model, the SVR model is trained, the SVR model is predicted, and model output is denormalized to obtain a prediction result, and the specific steps comprise:
step1, assume that the current road section is x 0 A total of n two with dataThe set of spatial segments of the network is defined as F { x }, for the upstream and downstream segments of the network 0 ，x 1 ，...，x n X, where x 0 Representing the current road section, x 1 To x n For its secondary upstream and downstream road segments, each road segment x is assumed i A continuous time series with a length q is denoted as F i Representing the road section x i For the sequence of the sky traffic of (1)，/>Representing road section x i At t 1 Flow sequence of moments,/->Represents x i At t 2 Flow sequence of moments,/->Represents x i At t q The flow sequence at the moment constructs a periodical input flow F by taking a day as a periodical unit d Is F d =[F 1 ，F 2 ， F 3 ， ...F n ] T ，F 1 Representing road section x 1 The sequence of the sky traffic of F 2 Representing road section x 2 The sequence of the sky traffic of F 3 Representing road section x 3 The sequence of the sky traffic of F n Representing road section x n Is the sequence of the sky traffic [] T Representing the transposition of the matrix, and setting training times K;
step2, inputting M-dimensional upstream and downstream road section flow data to an automatic encoder, and then obtaining output, wherein the training frequency is increased by 1;
step3, calculating a reconstruction error;
step4, updating the weight of the automatic encoder, returning to the step2 again when the training times do not reach the set value and the reconstruction error is larger than the threshold d, otherwise, entering the step 5;
step5, reducing the dimension of the M-dimension upstream and downstream data to the M dimension based on the trained automatic encoder;
step6, normalizing the current road section flow data and inputting the normalized current road section flow data and the dimension-reduced m-dimension characteristic data into the GRU network;
step7, training a GRU network model based on the processed data, and learning the input data;
step8, inputting the feature vector output by the GRU network into an SVR model, and then training the SVR model;
and 9, after training the model, inputting the traffic flow data of the current road section at the current moment into the model, and finally outputting the traffic flow predicted value of the current road section at the next moment by the model.
2. The method of claim 1, wherein the normalizing process comprises: traffic flow data preprocessing, carrying out normalization processing on the current road section flow data, wherein the calculation formula is as follows: f (f) i norm =(f i -f min )/(f max -f min ) In f i The ith data representing a feature takes on value and f max And f min Representing the maximum and minimum values, f, respectively, of the feature i norm The normalized result of the feature item i data.
3. The method of claim 1, wherein the step of calculating the weekly average spatial correlation coefficient is as follows: for F d Calculating correlation coefficient for each line，/>Representing the correlation coefficient between road segment 0 and road segment i on day d, from which a matrix of peri-phasic relations can be derived>According to this, the weekly average correlation coefficient is calculated +.>And (3) screening out M pieces of data with highest similarity from n pieces of upstream and downstream of the cycle average correlation coefficient, and inputting the M pieces of road section data serving as upstream and downstream road section data.
4. The method of claim 1, wherein in the network structure of the network layer of the two-layer GRU, the number of input layer units is 12, the number of hidden layer units is 135, the activation function selects sigmoid, and the initialization function selects he_unitorm.
5. The method of claim 1, wherein the kernel function of the SVR model is radial basis, gamma is 0.3, epsilon is 0.001.
6. An electronic device, comprising: the device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; the memory has stored therein a computer program which, when executed by the processor, causes the processor to perform the method of any of claims 1-5.
7. A computer readable storage medium, characterized in that it stores a computer program executable by a processor, which when run on the processor causes the processor to perform the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310872227.4A CN116597656A (en) | 2023-07-17 | 2023-07-17 | Method, equipment and medium for predicting road traffic flow based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310872227.4A CN116597656A (en) | 2023-07-17 | 2023-07-17 | Method, equipment and medium for predicting road traffic flow based on big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116597656A true CN116597656A (en) | 2023-08-15 |
Family
ID=87604778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310872227.4A Pending CN116597656A (en) | 2023-07-17 | 2023-07-17 | Method, equipment and medium for predicting road traffic flow based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116597656A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510741A (en) * | 2018-05-24 | 2018-09-07 | 浙江工业大学 | A kind of traffic flow forecasting method based on Conv1D-LSTM neural network structures |
| CN111080029A (en) * | 2019-12-26 | 2020-04-28 | 山东大学 | Urban traffic road speed prediction method and system based on multi-path segment space-time correlation |
| CN111210633A (en) * | 2020-02-09 | 2020-05-29 | 北京工业大学 | Short-term traffic flow prediction method based on deep learning |
| CN111860979A (en) * | 2020-07-01 | 2020-10-30 | 广西大学 | Short-term load prediction method based on TCN and IPSO-LSSVM combined model |
| CN113674524A (en) * | 2021-07-28 | 2021-11-19 | 长安大学 | LSTM-GASVR-based multi-scale short-time traffic flow prediction modeling and prediction method and system |
-
2023
- 2023-07-17 CN CN202310872227.4A patent/CN116597656A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510741A (en) * | 2018-05-24 | 2018-09-07 | 浙江工业大学 | A kind of traffic flow forecasting method based on Conv1D-LSTM neural network structures |
| CN111080029A (en) * | 2019-12-26 | 2020-04-28 | 山东大学 | Urban traffic road speed prediction method and system based on multi-path segment space-time correlation |
| CN111210633A (en) * | 2020-02-09 | 2020-05-29 | 北京工业大学 | Short-term traffic flow prediction method based on deep learning |
| CN111860979A (en) * | 2020-07-01 | 2020-10-30 | 广西大学 | Short-term load prediction method based on TCN and IPSO-LSSVM combined model |
| CN113674524A (en) * | 2021-07-28 | 2021-11-19 | 长安大学 | LSTM-GASVR-based multi-scale short-time traffic flow prediction modeling and prediction method and system |
Non-Patent Citations (2)
| Title |
|---|
| 潘伟靖: "基于GRU-SVR的短时交通流量预测研究", 《计算机技术与发展》, vol. 29, no. 10, pages 11 - 14 * |
| 王浩: "基于大数据的城市交通流短时预测系统研究与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, no. 8, pages 034 - 286 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Long-term traffic prediction based on lstm encoder-decoder architecture | |
| CN114220271B (en) | Traffic flow prediction method, equipment and storage medium based on dynamic space-time diagram convolution circulation network | |
| Shen et al. | Wind speed prediction of unmanned sailboat based on CNN and LSTM hybrid neural network | |
| Wang et al. | Research on healthy anomaly detection model based on deep learning from multiple time-series physiological signals | |
| Han et al. | Network traffic prediction using variational mode decomposition and multi-reservoirs echo state network | |
| CN111598325A (en) | Traffic speed prediction method based on hierarchical clustering and hierarchical attention mechanism | |
| Yang et al. | Prediction model of energy market by long short term memory with random system and complexity evaluation | |
| CN112966871A (en) | Traffic jam prediction method and system based on convolution long-short term memory neural network | |
| CN112257847A (en) | Method for predicting geomagnetic Kp index based on CNN and LSTM | |
| Wang et al. | The trend-fuzzy-granulation-based adaptive fuzzy cognitive map for long-term time series forecasting | |
| CN112766603A (en) | Traffic flow prediction method, system, computer device and storage medium | |
| Bai et al. | A deep neural network based on classification of traffic volume for short-term forecasting | |
| CN114596726B (en) | Parking berth prediction method based on interpretable space-time attention mechanism | |
| Liu et al. | Multi-stage residual life prediction of aero-engine based on real-time clustering and combined prediction model | |
| Abdalla et al. | Deep learning weather forecasting techniques: literature survey | |
| Fu et al. | MCA-DTCN: A novel dual-task temporal convolutional network with multi-channel attention for first prediction time detection and remaining useful life prediction | |
| Galib et al. | Deepextrema: A deep learning approach for forecasting block maxima in time series data | |
| Alam | Recurrent neural networks in electricity load forecasting | |
| Wang et al. | Data imputation for detected traffic volume of freeway using regression of multilayer perceptron | |
| CN116597656A (en) | Method, equipment and medium for predicting road traffic flow based on big data analysis | |
| Zhang et al. | A hierarchical multivariate denoising diffusion model | |
| CN115348182A (en) | Long-term spectrum prediction method based on depth stack self-encoder | |
| CN115423091A (en) | Conditional antagonistic neural network training method, scene generation method and system | |
| CN114529063A (en) | Financial field data prediction method, device and medium based on machine learning | |
| Siraj et al. | Data mining and neural networks: the impact of data representation |
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 |