CN105513350A - Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics - Google Patents
Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics Download PDFInfo
- Publication number
- CN105513350A CN105513350A CN201510859229.5A CN201510859229A CN105513350A CN 105513350 A CN105513350 A CN 105513350A CN 201510859229 A CN201510859229 A CN 201510859229A CN 105513350 A CN105513350 A CN 105513350A
- Authority
- CN
- China
- Prior art keywords
- time
- traffic flow
- traffic
- prediction
- short
- 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
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
- 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 time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics. According to the time-phased multi-parameter short-term traffic flow prediction method, on a basis of acquiring real-time and historical data of three traffic parameters including speed, traffic flow and time occupancy at a target monitoring point as well as upstream and downstream monitoring points, a TS-WNN prediction model is established by combining the time-space characteristics (Time-Space, TS) of the traffic flow with a wavelet neural network (Wavelet Neural Network, WNN) prediction algorithm, and the tree traffic parameters are subjected to short-term traffic flow prediction at workdays and non-workdays separately by adopting the time-phased multi-parameter prediction method. The time-phased multi-parameter short-term traffic flow prediction method fully considers the time-space characteristics of the traffic flow, performs time-phased multi-parameter prediction, increases prediction accuracy and prediction universality, can better satisfy prediction requirements of highway traffic, helps traffic managers in performing effective traffic control, and plans better travel routes for travelers.
Description
Technical field
The present invention relates to intelligent transport system field, be specifically related to a kind of Short-time Traffic Flow Forecasting Methods of multiparameter at times based on space-time characterisation.
Background technology
In the ingredient of intelligent transportation system, the performance analysis of road traffic state and prediction are important basic theories, and one of its core carries out short-time traffic flow forecast in real time, exactly.
Forecasting traffic flow, according to the length of prediction step, can be divided into long-term, mid-term and short-term three kinds of type of prediction.Short-term prediction refers generally to prediction step Δ t≤15min.In the middle of urban road network, the general traffic flow conditions needed in the prediction following short time, requirement of real-time is higher, and short-time traffic flow forecast can meet this requirement better.
Short-term traffic flow has very strong time variation and randomness, and therefore set up high, the real-time and stable short-term traffic flow forecasting model that predicts the outcome of precision of prediction is one of Research Challenges of intelligent transport system field always.
Summary of the invention
The object of the present invention is to provide the Short-time Traffic Flow Forecasting Methods of multiparameter at times based on space-time characterisation, take into full account the space-time characterisation of traffic flow data, both the correlativity of surrounding time section traffic flow data had been considered, also the correlativity of upstream and downstream traffic flow data is considered, and build TS-WNN forecast model, the Forecasting Methodology of multiparameter is at times adopted to carry out short-time traffic flow forecast, to improve forecasting traffic flow precision and universality.
The Short-time Traffic Flow Forecasting Methods of multiparameter at times based on space-time characterisation: obtaining the speed of target monitoring point and upstream and downstream monitoring point, on the basis of the real-time and historical data of the magnitude of traffic flow and time occupancy three kinds of traffic parameters, by utilizing the space-time characterisation (Time-Space of traffic flow, TS) with wavelet neural network (WaveletNeuralNetwork, WNN) prediction algorithm combines, build TS-WNN forecast model, and utilize the Forecasting Methodology of multiparameter at times, three kinds of traffic parameters are carried out short-time traffic flow forecast respectively with nonworkdays on weekdays.
Further, above-mentioned Forecasting Methodology obtains in real time and the time interval of historical traffic flow data is 5min, and comprises speed (speed), the magnitude of traffic flow (flow) and time occupancy (occupancy) three kinds of traffic flow datas; Described historical data at least need comprise the data of month, to ensure enough data training forecast models.
Further, above-mentioned Forecasting Methodology carries out space-time characterisation analysis to traffic flow data, determine best Time and place dimension, as predicted time interval of delta t=5min, time dimension is set to 2, namely previous moment x (t-Δ t) and current time x (t), be set to 4 by space dimensionality, namely selects current point traffic flow data, two upstream point traffic datas and a point downstream traffic data.
Further, the concrete steps of described structure TS-WNN forecast model comprise:
1) TS-WNN forecast model is based on BP neural network topology structure, and select Morlet wavelet basis function to substitute the transport function of hidden layer node, its expression formula is:
x is parametric variable,
2) initialization input vector: the Time and place dimension optimum configurations input vector according to the best:
X=[x(p-2,t
0),x(p-1,t
0),x(p,t
0),x(p+1,t
0),x(p-2,t
0-Δt),
x(p-1,t
0-Δt),x(p,t
0-Δt),x(p+1,t
0-Δt)],
3) wavelet neural network WNN builds: arrange input layer, hidden layer node, output layer node;
4) wavelet neural network training: select the traffic flow data of month as training data, frequency of training is set, learning rate lr1 and lr2 of WNN is set;
5) wavelet neural network prediction: according to the TS-WNN forecast model trained, short-time traffic flow forecast is carried out to road.
Further, choose multiparameter prediction method at times, refer to because traffic flow presents the different regularities of distribution from nonworkdays on weekdays at times, there is significantly early evening peak in workaday traffic distribution curve, the traffic distribution curve of nonworkdays is then relatively more steady, not obvious early evening peak; Multiparameter refers to that Prediction Parameters comprises speed, the magnitude of traffic flow and time occupancy three kinds of parameters, and three kinds of traffic parameters are carried out short-time traffic flow forecast with nonworkdays by this Forecasting Methodology on weekdays respectively.
The present invention is compared with existing traffic forecasting technique, and tool has the following advantages:
(1) consider the space-time characterisation of traffic flow data, construct TS-WNN forecasting traffic flow model, decrease predicated error, improve precision of prediction.
(2) multi-parameter prediction is at times carried out, traffic flow presents the different regularities of distribution with nonworkdays on weekdays, the forecast model that the present invention builds can be predicted respectively to the working day of three traffic parameters and nonworkdays traffic, improves precision of prediction and universality.
Accompanying drawing explanation
Fig. 1 a is prediction model based on wavelet neural network schematic diagram;
Fig. 1 b is the short-time traffic flow forecast of the multiparameter at times process flow diagram based on space-time characterisation.
Fig. 2 is the forecasting traffic flow schematic diagram based on space-time characterisation;
Fig. 3 is multiparameter prediction method schematic diagram at times;
Fig. 4 is medium velocity of the present invention predicted value and observed reading fitting result chart on weekdays;
Fig. 5 is magnitude of traffic flow predicted value and observed reading fitting result chart on weekdays in the present invention;
Fig. 6 is time occupancy predicted value and observed reading fitting result chart on weekdays in the present invention;
Fig. 7 is that medium velocity of the present invention is at nonworkdays predicted value and observed reading fitting result chart;
Fig. 8 be in the present invention the magnitude of traffic flow at nonworkdays predicted value and observed reading fitting result chart;
Fig. 9 be in the present invention time occupancy at nonworkdays predicted value and observed reading fitting result chart.
Embodiment
Below in conjunction with accompanying drawing, specific embodiment of the invention process is elaborated; but the scope of protection of present invention is not limited to the scope of lower example statement; be pointed out that; if have process or the parameter of not special detailed description below; such as, in neural network, related function is the parameter in conventional expression formula, without the need to illustrating implication.
The Short-time Traffic Flow Forecasting Methods of multiparameter at times based on space-time characterisation of the present invention, take into full account the correlativity of surrounding time section traffic flow data and the correlativity of upstream and downstream traffic flow data, and utilize wavelet neural network algorithm, build TS-WNN forecasting traffic flow model, and carry out multi-parameter prediction at times.
As Fig. 1 b, this example adopts following technical scheme to realize:
(1) obtain real-time traffic flow data, the time interval of data is 5min, and comprises speed (speed), the magnitude of traffic flow (flow) and time occupancy (occupancy) three kinds of traffic flow datas.In addition, historical data at least need comprise the data of month, to ensure enough data training forecast models;
(2) space-time characterisation analysis is carried out to traffic flow data, determine best Time and place dimension;
(3) TS-WNN forecasting traffic flow model is built:
TS-WNN forecast model is based on BP neural network topology structure, and the transport function wavelet basis function selecting wavelet basis function to substitute hidden layer node selects its expression formula of Morlet wavelet basis function to be formula:
By carrying out flexible and translation transformation to formula (1), can obtain wavelet basis function, its expression formula is shown in formula (2).Wherein, a is
contraction-expansion factor, b is
shift factor.
According to the knowwhy of wavelet analysis and neural network, can show that its output quantity y is:
Wherein, N is hidden layer neuron number, and M is input layer number.TS-WNN forecast model is based on BP neural network topology structure, selects wavelet basis function to substitute the transport function of hidden layer node, its forecast model as shown in Figure 1a, in figure, x
1, x
2..., x
mbe input, y is that prediction exports, ω
ijrepresent the connection weights between input layer and hidden layer, w
jrepresent the connection weights between output layer and hidden layer.
(4) carry out multiparameter traffic forecast at times, namely speed, the magnitude of traffic flow and time occupancy three traffic parameters are predicted respectively with the traffic of nonworkdays on weekdays.
The roughly step of TS-WNN forecasting traffic flow model is as follows:
1. space-time characterisation analysis:
As shown in Figure 2, urban road is interconnected, and the traffic in upstream and downstream section can interact, and the traffic flow in section, upstream increases suddenly, downstream road section can be caused to occur traffic congestion, and the traffic of downstream road section equally also can affect the traffic in section, upstream conversely.Arranging of the present embodiment Time and place dimension is as follows:
(1) time dimension: current time and previous moment
(2) Spatial Dimension: monitored upstream point selection 2 points, monitored down point selection 1 point (P is monitoring point, place, and P-2, P-1 and P+1 represent monitoring point, 2, upstream and 1 monitored down point respectively)
Therefore, input vector:
X=(x(p-2,t
0),x(p-1,t
0),x(p,t
0),x(p+1,t
0),x(p-2,t
0-Δt),
x(p-1,t
0-Δt),x(p,t
0-Δt),x(p+1,t
0-Δt))(4)
Output vector (x is parametric variable, for prediction of speed):
Y=x(p,t
0+Δt)(5)
2. wavelet neural network builds:
The wavelet neural network structure that this example adopts is 8-6-1, and input layer has 8 nodes (such as formula (4) Suo Shi), and node in hidden layer is 6, and output layer only has 1 node (such as formula (5) Suo Shi).
3. wavelet neural network training:
Train wavelet neural network with training data, frequency of training selects 500 times, and learning rate lr1 and lr2 of WNN gets 0.01 and 0.001 respectively.
4. wavelet neural network prediction:
With the prediction of the wavelet neural network based on the space-time characterisation short-term traffic flow trained, and analyze predicting the outcome.
5. pair to predict the outcome and analyze.
As example further, the present embodiment select certain highway in March, 2014 traffic flow data as experimental data, for working day, in March, 2014 has at 21 days on working day, using the data of first 20 days as training data, the data of last day (March 31) are as predicted data.For nonworkdays, in March, 2014 has nonworkdays (weekend) 10 days.Because the partial data on March 9 lacks, therefore the data on March 9 are rejected.The experimental data of surplus 9 days altogether, using the data of first 8 days as training data, the data of last day (March 30) are as predicted data.In addition, prediction step is set to 5min.
The present embodiment adopts mean absolute error MAE, root-mean-square error RMSE, impartial coefficient EC assesses estimated performance.The expression formula of three estimated performance evaluation indexes is shown in formula (4) ~ (6).
In formula, N represents sample number; x
realrepresent the actual observed value of traffic flow parameter; x
prerepresent the predicted value of traffic flow parameter.
MAE is mainly used in the absolute average of the error represented between predicted value and actual value, and its value is less, illustrates that prediction effect is better.RMSE is mainly used in the distribution situation representing error, and its value is less, and specification error distribution is more concentrated, and therefore prediction effect is better.EC is mainly used in representing the fitting degree between predicted value and actual value, and its value is more close to 1, and prediction effect is better.Generally can think, as EC>0.9, the prediction effect of system is better.
As can be seen from above-described embodiment, the multiparameter prediction method at times that the present invention proposes considers the correlativity of surrounding time section traffic flow data and the correlativity of upstream and downstream traffic flow data simultaneously, and three traffic parameters are predicted in different time sections (working day and nonworkdays), as shown in Figure 3.In order to compare with classic method estimated performance, the present embodiment selects identical prediction modeling and experimental data, gives the traffic flow three parameter prediction performance considering space-time characterisation, and does not consider the traffic flow three parameter prediction performance of space-time characterisation.Estimated performance evaluation index result of calculation is in above-mentioned two situations in table 1.
Table 1 three traffic parameters on weekdays with the estimated performance of nonworkdays
As can be seen from Table 1, the Forecasting Methodology that the present invention proposes is better than forecast model or the method for not considering space-time characterisation, the root-mean-square error (RMSE) that speed, the magnitude of traffic flow and time occupancy three traffic parameters are predicted on weekdays reduces 18%, 9%, 8% respectively, reduces 7%, 3%, 11% respectively in the root-mean-square error (RMSE) of nonworkdays prediction.Fig. 3 ~ Fig. 5 is predicting the outcome on March 31st, 2014, sets forth the speed of method proposed by the invention, the magnitude of traffic flow and time occupancy predicted value on weekdays and the fitting effect of actual value.Fig. 6 ~ Fig. 8 is predicting the outcome on March 30th, 2014, sets forth the speed of method proposed by the invention, the magnitude of traffic flow and the time occupancy predicted value at nonworkdays and the fitting effect of actual value.
Claims (5)
1. based on the Short-time Traffic Flow Forecasting Methods of multiparameter at times of space-time characterisation, it is characterized in that: obtaining the speed of target monitoring point and upstream and downstream monitoring point, on the basis of the real-time and historical data of the magnitude of traffic flow and time occupancy three kinds of traffic parameters, by utilizing the space-time characterisation (Time-Space of traffic flow, TS) with wavelet neural network (WaveletNeuralNetwork, WNN) prediction algorithm combines, build TS-WNN forecast model, and utilize the Forecasting Methodology of multiparameter at times, three kinds of traffic parameters are carried out short-time traffic flow forecast respectively with nonworkdays on weekdays.
2. as claimed in claim 1 based on the Short-time Traffic Flow Forecasting Methods of multiparameter at times of space-time characterisation, it is characterized in that: obtaining the time interval that is real-time and historical traffic flow data is 5min, and comprises speed (speed), the magnitude of traffic flow (flow) and time occupancy (occupancy) three kinds of traffic flow datas; Described historical data at least need comprise the data of month, to ensure enough data training forecast models.
3. as claimed in claim 1 based on the Short-time Traffic Flow Forecasting Methods of multiparameter at times of space-time characterisation, it is characterized in that: space-time characterisation analysis is carried out to traffic flow data, determine best Time and place dimension, as predicted time interval of delta t=5min, time dimension is set to 2, i.e. previous moment x (t-+t) and current time x (t), space dimensionality is set to 4, namely selects current point traffic flow data, two upstream point traffic datas and a point downstream traffic data.
4. as claimed in claim 1 based on the Short-time Traffic Flow Forecasting Methods of multiparameter at times of space-time characterisation, it is characterized in that: the concrete steps building TS-WNN forecast model comprise:
1) TS-WNN forecast model is based on BP neural network topology structure, and select Morlet wavelet basis function to substitute the transport function of hidden layer node, its expression formula is:
x is parametric variable,
2) initialization input vector: the Time and place dimension optimum configurations input vector according to the best:
X=[x(p-2,t
0),x(p-1,t
0),x(p,t
0),x(p+1,t
0),x(p-2,t
0-Δt),
x(p-1,t
0-Δt),x(p,t
0-Δt),x(p+1,t
0-Δt)],
3) wavelet neural network WNN builds: arrange input layer, hidden layer node, output layer node;
4) wavelet neural network training: select the traffic flow data of month as training data, frequency of training is set, learning rate lr1 and lr2 of WNN is set;
5) wavelet neural network prediction: according to the TS-WNN forecast model trained, short-time traffic flow forecast is carried out to road.
5. as claimed in claim 1 based on the Short-time Traffic Flow Forecasting Methods of multiparameter at times of space-time characterisation, it is characterized in that: choose multiparameter prediction method at times, refer to because traffic flow presents the different regularities of distribution from nonworkdays on weekdays at times, there is significantly early evening peak in workaday traffic distribution curve, the traffic distribution curve of nonworkdays is then relatively more steady, not obvious early evening peak; Multiparameter refers to that Prediction Parameters comprises speed, the magnitude of traffic flow and time occupancy three kinds of parameters, and three kinds of traffic parameters are carried out short-time traffic flow forecast with nonworkdays by this Forecasting Methodology on weekdays respectively.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510859229.5A CN105513350A (en) | 2015-11-30 | 2015-11-30 | Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510859229.5A CN105513350A (en) | 2015-11-30 | 2015-11-30 | Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105513350A true CN105513350A (en) | 2016-04-20 |
Family
ID=55721291
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510859229.5A Pending CN105513350A (en) | 2015-11-30 | 2015-11-30 | Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105513350A (en) |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105788270A (en) * | 2016-05-13 | 2016-07-20 | 广州运星科技有限公司 | Internet of things-based traffic data prediction method and processing server |
CN106355879A (en) * | 2016-09-30 | 2017-01-25 | 西安翔迅科技有限责任公司 | Time-space correlation-based urban traffic flow prediction method |
CN106971547A (en) * | 2017-05-18 | 2017-07-21 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods for considering temporal correlation |
CN106981198A (en) * | 2017-05-24 | 2017-07-25 | 北京航空航天大学 | Deep learning network model and its method for building up for predicting travel time |
CN107085942A (en) * | 2017-06-26 | 2017-08-22 | 广东工业大学 | A kind of traffic flow forecasting method based on wolf pack algorithm, apparatus and system |
CN107085941A (en) * | 2017-06-26 | 2017-08-22 | 广东工业大学 | A kind of traffic flow forecasting method, apparatus and system |
CN107230351A (en) * | 2017-07-18 | 2017-10-03 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods based on deep learning |
CN107967803A (en) * | 2017-11-17 | 2018-04-27 | 东南大学 | Traffic congestion Forecasting Methodology based on multi-source data and variable-weight combined forecasting model |
CN108229724A (en) * | 2017-12-06 | 2018-06-29 | 华南理工大学 | A kind of transport data stream Forecasting Methodology in short-term based on Spatial-temporal Information Fusion |
WO2018122804A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Road traffic anomaly detection method using non-isometric time/space division |
CN109064748A (en) * | 2018-09-18 | 2018-12-21 | 浙江工业大学 | Traffic average speed prediction method based on temporal clustering analysis and variable convolution neural network |
CN109493599A (en) * | 2018-11-16 | 2019-03-19 | 南京航空航天大学 | A kind of Short-time Traffic Flow Forecasting Methods based on production confrontation network |
CN109872535A (en) * | 2019-03-27 | 2019-06-11 | 深圳市中电数通智慧安全科技股份有限公司 | A kind of current prediction technique of wisdom traffic, device and server |
CN110220527A (en) * | 2019-05-31 | 2019-09-10 | 中国四维测绘技术有限公司 | A kind of paths planning method and device based on public activity prediction |
CN110853347A (en) * | 2019-10-14 | 2020-02-28 | 深圳市综合交通运行指挥中心 | Short-time traffic road condition prediction method and device and terminal equipment |
CN110992685A (en) * | 2019-11-20 | 2020-04-10 | 安徽百诚慧通科技有限公司 | Traffic safety early warning method based on sudden change of highway traffic flow |
CN111127888A (en) * | 2019-12-23 | 2020-05-08 | 广东工业大学 | Urban traffic flow prediction method based on multi-source data fusion |
CN112365705A (en) * | 2020-08-27 | 2021-02-12 | 招商局重庆交通科研设计院有限公司 | Method for determining road traffic volume |
CN112651361A (en) * | 2020-12-31 | 2021-04-13 | 维特瑞交通科技有限公司 | Monitoring method based on dynamic traffic flow |
CN112884190A (en) * | 2019-11-29 | 2021-06-01 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
CN113159374A (en) * | 2021-03-05 | 2021-07-23 | 北京化工大学 | Data-driven urban traffic flow rate mode identification and real-time prediction early warning method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001084479A (en) * | 1999-09-14 | 2001-03-30 | Matsushita Electric Ind Co Ltd | Method and device for forecasting traffic flow data |
JP2007122359A (en) * | 2005-10-27 | 2007-05-17 | Matsushita Electric Ind Co Ltd | Traffic information processor, and traffic information processing method |
CN102629418A (en) * | 2012-04-09 | 2012-08-08 | 浙江工业大学 | Fuzzy kalman filtering-based traffic flow parameter prediction method |
CN104464291A (en) * | 2014-12-08 | 2015-03-25 | 杭州智诚惠通科技有限公司 | Traffic flow predicting method and system |
CN105046953A (en) * | 2015-06-18 | 2015-11-11 | 南京信息工程大学 | Short-time traffic-flow combination prediction method |
-
2015
- 2015-11-30 CN CN201510859229.5A patent/CN105513350A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001084479A (en) * | 1999-09-14 | 2001-03-30 | Matsushita Electric Ind Co Ltd | Method and device for forecasting traffic flow data |
JP2007122359A (en) * | 2005-10-27 | 2007-05-17 | Matsushita Electric Ind Co Ltd | Traffic information processor, and traffic information processing method |
CN102629418A (en) * | 2012-04-09 | 2012-08-08 | 浙江工业大学 | Fuzzy kalman filtering-based traffic flow parameter prediction method |
CN104464291A (en) * | 2014-12-08 | 2015-03-25 | 杭州智诚惠通科技有限公司 | Traffic flow predicting method and system |
CN105046953A (en) * | 2015-06-18 | 2015-11-11 | 南京信息工程大学 | Short-time traffic-flow combination prediction method |
Non-Patent Citations (5)
Title |
---|
余国强: "基于小波神经网络的短时交通流预测算法的研究", 《中国优秀硕士学位论文全文数据库•工程科技Ⅱ辑》 * |
张菁菁: "基于浮动车数据的城市快速路短时交通状态预测的研究", 《中国优秀硕士学位论文全文数据库·工程科技Ⅱ辑》 * |
朱顺应 等: "《交通流参数及交通事件动态预测方法》", 31 May 2008 * |
谭政: "城市道路交通流预测及应用", 《中国优秀硕士学位论文全文数据库·工程科技Ⅱ辑》 * |
金玉婷 等: "基于小波神经网络的短时交通流预测", 《交通科技与经济》 * |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105788270A (en) * | 2016-05-13 | 2016-07-20 | 广州运星科技有限公司 | Internet of things-based traffic data prediction method and processing server |
CN106355879A (en) * | 2016-09-30 | 2017-01-25 | 西安翔迅科技有限责任公司 | Time-space correlation-based urban traffic flow prediction method |
WO2018122804A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Road traffic anomaly detection method using non-isometric time/space division |
CN110168520A (en) * | 2016-12-30 | 2019-08-23 | 同济大学 | A kind of intelligence road traffic method for detecting abnormality |
CN109997179A (en) * | 2016-12-30 | 2019-07-09 | 同济大学 | A kind of road traffic method for detecting abnormality that non-equidistant space-time divides |
CN106971547B (en) * | 2017-05-18 | 2019-06-04 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods considering temporal correlation |
CN106971547A (en) * | 2017-05-18 | 2017-07-21 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods for considering temporal correlation |
CN106981198A (en) * | 2017-05-24 | 2017-07-25 | 北京航空航天大学 | Deep learning network model and its method for building up for predicting travel time |
CN106981198B (en) * | 2017-05-24 | 2020-11-03 | 北京航空航天大学 | Deep learning network model for travel time prediction and establishing method thereof |
CN107085942A (en) * | 2017-06-26 | 2017-08-22 | 广东工业大学 | A kind of traffic flow forecasting method based on wolf pack algorithm, apparatus and system |
CN107085941A (en) * | 2017-06-26 | 2017-08-22 | 广东工业大学 | A kind of traffic flow forecasting method, apparatus and system |
CN107230351A (en) * | 2017-07-18 | 2017-10-03 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods based on deep learning |
CN107967803A (en) * | 2017-11-17 | 2018-04-27 | 东南大学 | Traffic congestion Forecasting Methodology based on multi-source data and variable-weight combined forecasting model |
CN108229724A (en) * | 2017-12-06 | 2018-06-29 | 华南理工大学 | A kind of transport data stream Forecasting Methodology in short-term based on Spatial-temporal Information Fusion |
CN108229724B (en) * | 2017-12-06 | 2020-12-22 | 华南理工大学 | Short-term traffic data flow prediction method based on temporal-spatial information fusion |
CN109064748A (en) * | 2018-09-18 | 2018-12-21 | 浙江工业大学 | Traffic average speed prediction method based on temporal clustering analysis and variable convolution neural network |
CN109493599A (en) * | 2018-11-16 | 2019-03-19 | 南京航空航天大学 | A kind of Short-time Traffic Flow Forecasting Methods based on production confrontation network |
CN109872535A (en) * | 2019-03-27 | 2019-06-11 | 深圳市中电数通智慧安全科技股份有限公司 | A kind of current prediction technique of wisdom traffic, device and server |
CN110220527A (en) * | 2019-05-31 | 2019-09-10 | 中国四维测绘技术有限公司 | A kind of paths planning method and device based on public activity prediction |
CN110853347A (en) * | 2019-10-14 | 2020-02-28 | 深圳市综合交通运行指挥中心 | Short-time traffic road condition prediction method and device and terminal equipment |
CN110992685A (en) * | 2019-11-20 | 2020-04-10 | 安徽百诚慧通科技有限公司 | Traffic safety early warning method based on sudden change of highway traffic flow |
CN112884190A (en) * | 2019-11-29 | 2021-06-01 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
CN112884190B (en) * | 2019-11-29 | 2023-11-03 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
CN111127888A (en) * | 2019-12-23 | 2020-05-08 | 广东工业大学 | Urban traffic flow prediction method based on multi-source data fusion |
CN112365705A (en) * | 2020-08-27 | 2021-02-12 | 招商局重庆交通科研设计院有限公司 | Method for determining road traffic volume |
CN112365705B (en) * | 2020-08-27 | 2022-05-27 | 招商局重庆交通科研设计院有限公司 | Method for determining road traffic volume |
CN112651361A (en) * | 2020-12-31 | 2021-04-13 | 维特瑞交通科技有限公司 | Monitoring method based on dynamic traffic flow |
CN113159374A (en) * | 2021-03-05 | 2021-07-23 | 北京化工大学 | Data-driven urban traffic flow rate mode identification and real-time prediction early warning method |
CN113159374B (en) * | 2021-03-05 | 2022-04-22 | 北京化工大学 | Data-driven urban traffic flow rate mode identification and real-time prediction early warning method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105513350A (en) | Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics | |
Xiao et al. | An improved seasonal rolling grey forecasting model using a cycle truncation accumulated generating operation for traffic flow | |
Liu et al. | A hybrid statistical method to predict wind speed and wind power | |
CN103903430B (en) | Dynamic fusion type travel time predicting method with multi-source and isomorphic data adopted | |
Dharia et al. | Neural network model for rapid forecasting of freeway link travel time | |
CN102629418B (en) | Fuzzy kalman filtering-based traffic flow parameter prediction method | |
CN104408913B (en) | A kind of traffic flow three parameter real-time predicting method considering temporal correlation | |
Lei et al. | A review on the forecasting of wind speed and generated power | |
CN104933862A (en) | Urban traffic jam intelligent combination prediction method based on track of floating vehicle | |
CN103177570B (en) | Method for predicting traffic jam indexes for rush hours in morning and evening | |
CN102521989B (en) | Dynamic-data-driven highway-exit flow-quantity predicting method | |
Murat | Comparison of fuzzy logic and artificial neural networks approaches in vehicle delay modeling | |
CN112071062B (en) | Driving time estimation method based on graph convolution network and graph attention network | |
CN106017496A (en) | Real-time navigation method based on road condition | |
Yu et al. | Bus travel-time prediction based on bus speed | |
CN107170234A (en) | A kind of method predicted based on SVM algorithm traffic in short-term | |
Chandra et al. | Cross-correlation analysis and multivariate prediction of spatial time series of freeway traffic speeds | |
Ivanjko et al. | Ramp metering control based on the Q-learning algorithm | |
CN113053120B (en) | Traffic signal lamp scheduling method and system based on iterative learning model predictive control | |
Li et al. | A k-nearest neighbor locally weighted regression method for short-term traffic flow forecasting | |
CN103745602B (en) | A kind of traffic flow forecasting method average based on sliding window | |
CN106370198A (en) | Route selection method taking outgoing delayed reaction into account | |
CN103500362B (en) | A kind of urban road speed predicting method based on analysis of spectrum | |
CN108345955A (en) | A kind of powerline ice-covering short term prediction method, device and storage medium | |
CN110211375A (en) | Based on the traffic flow forecasting method for improving space time correlation KNN algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160420 |
|
WD01 | Invention patent application deemed withdrawn after publication |