CN111861027A - Urban traffic flow prediction method based on deep learning fusion model - Google Patents
Urban traffic flow prediction method based on deep learning fusion model Download PDFInfo
- Publication number
- CN111861027A CN111861027A CN202010741049.8A CN202010741049A CN111861027A CN 111861027 A CN111861027 A CN 111861027A CN 202010741049 A CN202010741049 A CN 202010741049A CN 111861027 A CN111861027 A CN 111861027A
- Authority
- CN
- China
- Prior art keywords
- traffic flow
- deep learning
- time
- urban
- fusion 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
- 230000004927 fusion Effects 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013135 deep learning Methods 0.000 title claims abstract description 23
- 238000010586 diagram Methods 0.000 claims abstract description 17
- 230000000737 periodic effect Effects 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims abstract description 11
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 238000013527 convolutional neural network Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000011161 development Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
Images
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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides an urban traffic flow prediction method based on a deep learning fusion model, which comprises the following steps: step 1, constructing a traffic flow prediction data attribute library; step 2, dividing an urban road network into different blocks according to different purposes, preprocessing historical traffic flow data, respectively counting the traffic flow in different time periods, constructing a traffic flow input matrix, and establishing a binary vector diagram; step 3, constructing a deep learning fusion model combining a 3DLSACN model and a Resnet residual error network as traffic flow characteristic extraction, performing characteristic extraction by using a binary vector diagram as input, and respectively extracting space-time characteristics and periodic characteristics and fusing to obtain a primary fusion result; and step 4, manually extracting external influence factors to form external features, fusing the external features with the primary fusion result again, and finally outputting traffic flow information required to be predicted. The invention can provide the prediction of the traffic flow under the timing space information, is beneficial to scheduling in time and relieves the problem of urban traffic pressure.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a method for predicting urban traffic flow based on a deep learning fusion model.
Background
People's work and life can not leave traffic, modern traffic is the product of social and economic development, and at present, China is in the process of urbanization, whether urban traffic is unobstructed or not is not only related to the economic development of cities, but also is related to the development of urban politics and culture, how to better develop the traffic and transportation industry, and great challenge is brought to urban managers. With the increasing popularization of a plurality of car renting service platforms such as a drip-and-answer trip, a goods-and-pull trip and the like, the data of the car rented by people can be continuously collected. At present, the problems of traffic flow overload and difficulty in getting on and queuing in certain hot areas exist, and the supervision efficiency needs to be improved so as to relieve traffic pressure.
Disclosure of Invention
The invention provides a traffic flow prediction method for effectively predicting and monitoring traffic flow, which aims to solve the problems of traffic flow overload and difficult driving in certain hot areas at present.
The method provided by the invention comprises the following steps: a city traffic flow prediction method based on a deep learning fusion model comprises the following steps:
and 4, manually extracting external influence factors to form external features, fusing the external features with the primary fusion result in the step 3 again, and finally outputting traffic flow information required to be predicted.
Further, the traffic flow prediction data attribute library includes data attribute names required to be used for traffic flow prediction, including time: the accuracy is to year, month, day, hour, minute and second; event: including weekdays, weekends, holidays; geographic location: the longitude and the latitude of the departure point and the arrival point are accurately obtained; the block to which the code belongs: dividing an urban road network into a working area, a business area, a residential area and a living service area according to different purposes; air temperature, weather includes: the weather station collects the information of temperature and weather in the Chinese weather net and provides comprehensive weather information including rainfall, wind speed, visibility, humidity and air pressure; the air quality data is collected in the national environmental monitoring center, and the acquired daily air quality summary information comprises SO2、NO2、CO、O3Pm2.5, for some newly generated influence factors, the influence factors are timely supplemented into the traffic flow prediction data attribute library through periodic updating.
Furthermore, the traffic flow input matrix is composed of the historical traffic flow of each block in different time periods, the urban road network is divided into different blocks according to different purposes, the different blocks are divided into a working area, a business area, a residential area and a living service area, the requirements and the influence degrees of the different purposes on the traffic flow are different, the historical traffic flow information is respectively counted at intervals of 30 minutes according to the different blocks, and a space-time matrix is constructed to establish a binary vector diagram.
Further, the 3DLSACN model and the Resnet residual error network are combined to be used as a mixed deep learning model capable of extracting traffic flow characteristics.
Further, in the proposed hybrid deep learning framework, a 3D LSACN model is used for extracting space-time characteristics of traffic flow, a Resnet residual error network is used for extracting proximity, periodicity and trend characteristics of GPS data of the traffic flow, a time axis is divided into three segments which respectively represent adjacent time periods of the same day, the same time period of the previous day and the same time period of the previous week or the previous month; the three segments are input into three Resnet residual network branches respectively to extract the proximity, periodicity and tendency characteristics of traffic flow in time.
Further, some other influencing factors are extracted from the external data set to form external characteristics, including historical time weather conditions, temperature conditions, air quality conditions, rainfall, wind speed, humidity and air pressure, and are input into a two-layer fully-connected neural network. And (3) fusing the output of the external influence factors with the primary fusion result obtained in the step (3), mapping the output to the (-1, 1) interval by using the tanh activation function as an output, and obtaining the predicted traffic flow information as the final output result of the model.
Further, training two neural networks, including a 3DLSACN model for extracting traffic flow space-time characteristics, a Resnet residual error network for extracting traffic flow periodic characteristics and other external factor correlation characteristic extraction, and finally fusing and outputting a prediction result; when training is started, the weight is initialized randomly, after the last layer of result is obtained through calculation of a neural network, the cross entropy between a predicted value and a true value is calculated to serve as a loss function, the loss function is minimized through an adaptive moment estimation algorithm, and the learning rate is adjusted according to the training process; in the training process, in order to improve the training efficiency, one batch of data is input every time, and meanwhile, in order to prevent overfitting, a certain proportion of weight values are randomly set to be 0 in the training process.
The invention has the advantages that the historical traffic flow is processed by using three-dimensional convolution in the deep learning fusion model to quickly extract the characteristics, and the dynamic change of the traffic flow is considered so as to efficiently and quickly predict the traffic flow information of a given time space section which needs to be predicted, thereby providing the demand of the traffic flow, assisting in timely scheduling treatment and further realizing the effect of relieving traffic pressure.
Drawings
FIG. 1 is a schematic flow chart of a deep learning fusion model-based urban traffic flow prediction method provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of the 3DLSACN model;
FIG. 3 is a frame of a traffic flow of each time zone and traffic flows of adjacent time zones;
FIG. 4 is a schematic diagram of the structure of one of the Resnet residual network branches and one of the residual units;
FIG. 5 is a schematic diagram of a deep learning fusion model.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.
According to an embodiment of the present invention, as shown in fig. 1, a method for predicting urban traffic flow based on a deep learning fusion model is provided, which includes the following steps:
and 4, manually extracting external influence factors to form external features, fusing the external features with the primary fusion result in the step 3 again, and finally outputting traffic flow information required to be predicted.
The traffic flow prediction data attribute library comprises data attribute names required to be used for traffic flow prediction, including time (accurate to year, month, day, hour, minute and second); events (including weekdays, weekends, holidays), geographic location (latitude, longitude accurate to the point of departure and arrival); the block to which the city road network belongs (the city road network can be divided into a working area, a business area, a residential area, a living service area and the like according to different purposes); air temperature, weather (including clear days, cloudy days, rain fall, thunderstorm, light rain, medium rain, heavy rain, rainstorm, snow-filled rain, snow fall, small snow, medium snow, heavy snow, fog, haze, sand storm and the like), air temperature and weather information are collected in a Chinese weather net, and comprehensive weather information including rainfall, wind speed, visibility, humidity, air pressure and the like is provided by a weather station; the air quality data is collected in the national environmental monitoring center, and the acquired daily air quality summary information comprises SO2、NO2、CO、O3Pm2.5 etc. And for some newly generated influence factors, the influence factors are timely supplemented into the traffic flow prediction data attribute library through periodic updating.
Referring to fig. 1, an overall schematic diagram of the method provided by the invention is shown, firstly, an urban road network is divided according to different purposes, then historical traffic data information is preprocessed, historical traffic flow data of each block is respectively counted, a space-time matrix is constructed, a binary vector diagram is established as input, space-time characteristics and periodic characteristics are extracted through a neural network, and the space-time characteristics and the periodic characteristics are fused with manually extracted external influence factors to construct a prediction model, so that a result of predicted traffic flow information is obtained through calculation.
In the embodiment shown in fig. 2, the statistics of the historical traffic flow data are:
and converted into a vector diagram, where s representss blocks, t denotes the t-th time period, Ft sRepresenting the traffic flow of the s-th block in the t-th time period. Each row of the matrix is used as a vector and represents spatial information among blocks in the same time period. Extracting spatial features of the traffic flow from the preprocessed historical traffic flow, and then inputting the spatial features into an LSTM layer to extract time features. Considering that the traffic flow in one time period can affect the traffic flow in the next time period of the adjacent blocks, three-dimensional convolution is used for processing the data, and the traffic flow in each time period and the traffic flow in the front and back adjacent time periods are combined into one frame as shown in fig. 3, so as to consider the dynamic information characteristics of the traffic flow in the time and space aspect. Acquiring convolution information of a local perception domain by using a three-dimensional convolution kernel filter, and then aggregating local features to obtain global features:
Gxyt=F(∑i∑j∑kwijkv(x+i)(y+j)(t+k)+b),
f is an activation function, wijkIs the filter weight, v(x+i)(y+j)(t+k)Is the input value and b is the deviation.
Followed by a pooling layer. The size of the generated signature sequence C is reduced to half of the original dimension by pooling, and the output result becomes a time-series vector Ct ═ C1, C2, C3, …, Ct. Each element in the vector represents a spatial correlation of traffic flow between each block. The resulting time series vector Ct is the input data of LSTM.
In the embodiment shown in fig. 4, the traffic flows of different blocks of the urban road network are converted into a binary vector diagram, and the time axis is divided into three segments according to different time intervals, which respectively represent adjacent time periods of the same day, the same time period of the previous day and the same time period of the previous week or the previous month. Each portion is used to extract the temporal proximity, periodicity and trending characteristics of traffic flow by being input into three Resnet residual network branches, respectively. Fig. 4 shows the structure of one of the Resnet residual network branches and one of the residual units. Different blocks are affected by time proximity, periodicity and trend characteristics to different degrees due to geographical positions, and a fusion method based on a parameter matrix is proposed to fuse the three parts of the periodic characteristics and the spatio-temporal characteristics:
Xcis a temporal proximity feature, XpIs a time-periodic feature, XrIs a time-trending feature, XqIs a space-time feature, Wc、Wp、Wr、WqIs the respective weight of each portion.
In the embodiment shown in fig. 5, some other influencing factors extracted from the external data material constitute external features including historical time weather conditions, temperature conditions, air quality conditions, and rainfall, wind speed, humidity, and air pressure, and are input into a two-layer fully-connected neural network. And (4) fusing the output of the external influence factors with the preliminary fusion result obtained in the step (3), and mapping the output to the (-1, 1) interval by using the tanh activation function as an output.
The external influencing factors are finally fused with the former part, and are directly added and reactivated:
wherein XLaRIs the output of the preceding section, XExtIs the output of the external influencing factor.
Although the embodiments of the present invention have been described above in order to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined by the appended claims, and all matters produced by the invention using the inventive concept are protected.
Claims (7)
1. A deep learning fusion model-based urban traffic flow prediction method is characterized by comprising the following steps:
step 1, constructing a traffic flow prediction data attribute library;
step 2, dividing an urban road network into different blocks according to different purposes, preprocessing historical traffic flow data, respectively counting the traffic flow in different time periods, constructing a traffic flow input matrix, and establishing a binary vector diagram;
step 3, constructing a 3DLSACN, namely combining the LSTM and CNN three-dimensional convolution model with the Resnet residual error network as a deep learning fusion model for traffic flow characteristic extraction, performing characteristic extraction by using the binary vector diagram established in the step 2 as input, and respectively extracting space-time characteristics and periodic characteristics and fusing to obtain a preliminary fusion result;
and 4, manually extracting external influence factors to form external features, fusing the external features with the primary fusion result in the step 3 again, and finally outputting traffic flow information required to be predicted.
2. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
in step 1, the traffic flow prediction data attribute library includes data attribute names required to be used for traffic flow prediction, including time: the accuracy is to year, month, day, hour, minute and second; event: including weekdays, weekends, holidays, geographical locations: the longitude and the latitude of the departure point and the arrival point are accurately obtained; the block to which the code belongs: dividing an urban road network into a working area, a business area, a residential area and a living service area according to different purposes; air temperature, weather: the method comprises the following steps: the weather station collects the information of temperature and weather in the Chinese weather net and provides comprehensive weather information including rainfall, wind speed, visibility, humidity and air pressure; collecting air quality data in national environment monitoring center to obtain daily air qualityThe quantity summary information includes SO2、NO2、CO、O3Pm2.5, for some newly generated influence factors, the influence factors are timely supplemented into the traffic flow prediction data attribute library through periodic updating.
3. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
in the step 2, the traffic flow input matrix is composed of the historical traffic flow of each block in different time periods, the urban road network is divided into different blocks according to different purposes, the different blocks are divided into a working area, a business area, a residential area and a living service area, the requirements and the influence degrees of the different purposes on the traffic flow are different, the historical traffic flow information is respectively counted at intervals of 30 minutes according to the different blocks, and a space-time matrix is constructed to establish a binary vector diagram.
4. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
in the step 3, the training 3DLSACN model and the Resnet residual error network are combined to be used as a deep learning fusion model capable of extracting traffic flow characteristics.
5. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
in the proposed deep learning fusion model framework, a 3DLSACN model is utilized to extract space-time characteristics of traffic flow, a Resnet residual error network is utilized to extract proximity, periodicity and trend characteristics of GPS data of the traffic flow, a time axis is divided into three segments which respectively represent adjacent time periods of the same day, the same time period of the previous day and the same time period of the previous week or the previous month; the three segments are input into three Resnet residual network branches respectively to extract the proximity, periodicity and tendency characteristics of traffic flow in time.
6. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
extracting some other influencing factors from the external data set to form external characteristics, wherein the external characteristics comprise weather conditions and temperature conditions, air quality conditions, rainfall, wind speed, humidity and air pressure of historical time, and inputting the external characteristics into a two-layer fully-connected neural network; and (3) fusing the output of the external influence factors with the primary fusion result obtained in the step (3), mapping the output to the (-1, 1) interval by using the tanh activation function as an output, and obtaining the predicted traffic flow information as the final output result of the model.
7. The urban traffic flow prediction method based on the deep learning fusion model according to claim 1, characterized in that:
in step 3, the method further comprises: training two neural networks, including a 3DLSACN model for extracting traffic flow space-time characteristics, a Resnet residual error network for extracting traffic flow periodic characteristics and other external factor correlation characteristic extraction, and finally fusing and outputting a prediction result; when training is started, the weight is initialized randomly, after the last layer of result is obtained through calculation of a neural network, the cross entropy between a predicted value and a true value is calculated to serve as a loss function, the loss function is minimized through an adaptive moment estimation algorithm, and the learning rate is adjusted according to the training process; in the training process, in order to improve the training efficiency, one batch of data is input every time, and meanwhile, in order to prevent overfitting, a certain proportion of weight values are randomly set to be 0 in the training process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010741049.8A CN111861027A (en) | 2020-07-29 | 2020-07-29 | Urban traffic flow prediction method based on deep learning fusion model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010741049.8A CN111861027A (en) | 2020-07-29 | 2020-07-29 | Urban traffic flow prediction method based on deep learning fusion model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111861027A true CN111861027A (en) | 2020-10-30 |
Family
ID=72948244
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010741049.8A Pending CN111861027A (en) | 2020-07-29 | 2020-07-29 | Urban traffic flow prediction method based on deep learning fusion model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111861027A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112466117A (en) * | 2020-11-24 | 2021-03-09 | 南通大学 | Road network short-term traffic flow prediction method based on deep space-time residual error network |
CN112766600A (en) * | 2021-01-29 | 2021-05-07 | 武汉大学 | Urban area crowd flow prediction method and system |
CN112766597A (en) * | 2021-01-29 | 2021-05-07 | 中国科学院自动化研究所 | Bus passenger flow prediction method and system |
CN113240179A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Method and system for predicting orbital pedestrian flow by fusing spatio-temporal information |
CN114462679A (en) * | 2022-01-04 | 2022-05-10 | 广州杰赛科技股份有限公司 | Network traffic prediction method, device, equipment and medium based on deep learning |
CN114662805A (en) * | 2022-05-26 | 2022-06-24 | 山东融瓴科技集团有限公司 | Traffic flow prediction method based on similar time sequence comparison |
CN115019504A (en) * | 2022-05-17 | 2022-09-06 | 汕头大学 | Short-term traffic flow prediction method based on new deep space time self-adaptive fusion graph network |
CN115017990A (en) * | 2022-06-01 | 2022-09-06 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Traffic flow prediction method, device, equipment and storage medium |
CN115913996A (en) * | 2022-12-07 | 2023-04-04 | 长春理工大学 | Mobile flow prediction system and method based on regional space-time characteristics |
CN116359851A (en) * | 2022-01-06 | 2023-06-30 | 上海交通大学 | Radar active interference detection and identification method and device based on converged network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108205889A (en) * | 2017-12-29 | 2018-06-26 | 长春理工大学 | Freeway traffic flow Forecasting Methodology based on convolutional neural networks |
CN110223517A (en) * | 2019-06-20 | 2019-09-10 | 青岛科技大学 | Short-term traffic flow forecast method based on temporal correlation |
CN110633871A (en) * | 2019-09-25 | 2019-12-31 | 大连理工大学 | Regional traffic demand prediction method based on convolution long-term and short-term memory network |
-
2020
- 2020-07-29 CN CN202010741049.8A patent/CN111861027A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108205889A (en) * | 2017-12-29 | 2018-06-26 | 长春理工大学 | Freeway traffic flow Forecasting Methodology based on convolutional neural networks |
CN110223517A (en) * | 2019-06-20 | 2019-09-10 | 青岛科技大学 | Short-term traffic flow forecast method based on temporal correlation |
CN110633871A (en) * | 2019-09-25 | 2019-12-31 | 大连理工大学 | Regional traffic demand prediction method based on convolution long-term and short-term memory network |
Non-Patent Citations (1)
Title |
---|
JUNBO ZHANG等: "Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction", 《ARXIV:1610.00081V2 [CS.AI]》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112466117A (en) * | 2020-11-24 | 2021-03-09 | 南通大学 | Road network short-term traffic flow prediction method based on deep space-time residual error network |
CN112766600A (en) * | 2021-01-29 | 2021-05-07 | 武汉大学 | Urban area crowd flow prediction method and system |
CN112766597A (en) * | 2021-01-29 | 2021-05-07 | 中国科学院自动化研究所 | Bus passenger flow prediction method and system |
CN112766597B (en) * | 2021-01-29 | 2023-06-27 | 中国科学院自动化研究所 | Bus passenger flow prediction method and system |
CN113240179A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Method and system for predicting orbital pedestrian flow by fusing spatio-temporal information |
CN114462679A (en) * | 2022-01-04 | 2022-05-10 | 广州杰赛科技股份有限公司 | Network traffic prediction method, device, equipment and medium based on deep learning |
CN116359851A (en) * | 2022-01-06 | 2023-06-30 | 上海交通大学 | Radar active interference detection and identification method and device based on converged network |
CN115019504A (en) * | 2022-05-17 | 2022-09-06 | 汕头大学 | Short-term traffic flow prediction method based on new deep space time self-adaptive fusion graph network |
CN114662805A (en) * | 2022-05-26 | 2022-06-24 | 山东融瓴科技集团有限公司 | Traffic flow prediction method based on similar time sequence comparison |
CN115017990A (en) * | 2022-06-01 | 2022-09-06 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Traffic flow prediction method, device, equipment and storage medium |
CN115913996A (en) * | 2022-12-07 | 2023-04-04 | 长春理工大学 | Mobile flow prediction system and method based on regional space-time characteristics |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111861027A (en) | Urban traffic flow prediction method based on deep learning fusion model | |
CN113487066B (en) | Long-time-sequence freight volume prediction method based on multi-attribute enhanced graph convolution-Informer model | |
CN107045673B (en) | Public bicycle flow variation prediction method based on stack model fusion | |
CN112200351A (en) | Urban area passenger flow volume prediction method based on mobile phone signaling data | |
CN110866649A (en) | Method and system for predicting short-term subway passenger flow and electronic equipment | |
CN105719019A (en) | Public bicycle peak time demand prediction method considering user reservation data | |
CN106507315A (en) | A kind of urban traffic accident Forecasting Methodology and system based on network social intercourse media data | |
CN114881381B (en) | Urban ponding water level prediction method and system based on improved convolutional neural network | |
CN115204477A (en) | Bicycle flow prediction method of context awareness graph recursive network | |
CN113192325B (en) | Traffic incident correlation analysis method and system | |
Zhang et al. | Crowd flow forecasting with multi-graph neural networks | |
CN116227362A (en) | Municipal drainage pipe network state prediction method based on graph and deep learning | |
CN113112076B (en) | Passenger flow distribution mode discovery and prediction method | |
CN110490365A (en) | A method of based on the pre- survey grid of multisource data fusion about vehicle order volume | |
CN116629460B (en) | Subway passenger flow prediction method based on ST-RANet model | |
CN116703132B (en) | Management method and device for dynamic scheduling of shared vehicles and computer equipment | |
CN113344290B (en) | Method for correcting sub-season rainfall weather forecast based on U-Net network | |
Jiao et al. | Multi-step traffic flow prediction method based on the Conv1D+ LSTM | |
CN115456238A (en) | Urban trip demand prediction method based on dynamic multi-view coupling graph convolution | |
CN113554221B (en) | Method for simulating and predicting town development boundary under view angle of' flow space | |
CN113191568B (en) | Meteorological-based urban operation management big data analysis and prediction method and system | |
CN110659758A (en) | AI (Artificial intelligence) technology-based short-term high-precision air quality prediction model | |
CN114139773A (en) | Public traffic flow prediction method based on space-time graph convolutional network | |
CN113034195A (en) | Network taxi booking demand prediction method based on convolutional network and non-local network | |
Benchelha et al. | Modeling dynamic urban growth using cellular automata and geospatial technique: case of Casablanca in Morocco |
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: 20201030 |
|
RJ01 | Rejection of invention patent application after publication |