CN101188002A - A city traffic dynamic prediction system and method with real time and continuous feature - Google Patents
A city traffic dynamic prediction system and method with real time and continuous feature Download PDFInfo
- Publication number
- CN101188002A CN101188002A CNA2007103039821A CN200710303982A CN101188002A CN 101188002 A CN101188002 A CN 101188002A CN A2007103039821 A CNA2007103039821 A CN A2007103039821A CN 200710303982 A CN200710303982 A CN 200710303982A CN 101188002 A CN101188002 A CN 101188002A
- Authority
- CN
- China
- Prior art keywords
- traffic
- module
- traffic behavior
- real
- real time
- 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
Images
Landscapes
- Traffic Control Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an urban traffic state predicting system with real-time and continuous characteristics. A real-time traffic data pre-treatment module carries out real-time reception of outside traffic data flows and also pre-treatment of online generalization and characteristic extraction of the data flows and inputs the online traffic parameter prediction done by the traffic state parameter predicting module; an active monitoring module of a predicting model performs real-time monitoring of characteristic changes of traffic data flows and provides real-time feedback and warning in case of limit threshold exceeding; a self-adapting adjusting module of the predicting model actively conducts self-adapting adjustment on the predicting model according to the warning information of the monitoring module; an automatic traffic state recognition module automatically recognizes the present traffic state according to the various predicted traffic state parameters. The invention can help traffic management departments to formulate traffic control and management strategies, thereby improving domestic study and application level in the transportation field, shortening the distance with the international level, enlarging the management capacity and effectively relieving traffic congestion.
Description
Technical field
The present invention relates to a kind of traffic behavior prediction and measuring technique, relate in particular to a kind of forecasting techniques of urban road traffic state, especially have in real time and the city traffic prognoses system and the method for continuation property, belong to the intelligent transport technology field.
Background technology
Traffic congestion has become the serious problems of current puzzlement socio-economic development.Traffic control is to alleviate the valid approach of traffic congestion with inducing, and in real time, traffic behavior prediction accurately is to realize having caused the extensive concern in international traffic field in traffic control and the prerequisite and the basis of inducing.Along with the development of intelligent transportation, more and more higher to the requirement of forecasting techniques, the deficiencies in the prior art highlight day by day.Be in particular in: (1) can not handle magnanimity " stream " the formula data of quick arrival in real time; (2) forecast model is not considered the process feature that traffic behavior changes, and accuracy is not high; (3) model can not dynamically be adjusted to adapt to the variation that traffic data distributes.These problems have proposed severe challenge to existing traffic behavior forecasting techniques.
Summary of the invention
In order to overcome the deficiency of prior art structure, the invention provides a kind of city traffic prognoses system and method with real-time and continuation property.
Described this urban road traffic state prognoses system will solve the prediction real-time of city traffic forecasting techniques in the prior art and continuity is not enough and inapplicable technical matters.
A kind of have in real time and the city traffic prognoses system of continuation property, mainly comprises:
The pretreatment module of described real time traffic data is accepted the extraneous transport data stream that imports in real time, and it is carried out online generalization and feature extraction;
The prediction module of described traffic behavior parameter is accepted pretreated traffic parameter information, makes online traffic parameter prediction;
The active monitoring module of described forecast model, the variation of monitoring and controlling traffic data stream feature is in real time fed back in real time and is reported to the police for exceeding the situation that limits threshold value;
The self-adaptation adjusting module of described forecast model is according to the warning message of monitoring module, initiatively forecast model is carried out adaptive adjustment, make forecast model to change with the external world automatically, keep the precision that predicts the outcome;
The automatic identification module of described traffic behavior according to all kinds of traffic behavior parameters (as parameters such as flow, occupation rates) of traffic behavior parameter prediction module output, according to the traffic behavior of setting, is discerned current traffic behavior automatically.
Data flow relationship description between above-mentioned each module is as follows:
Pretreatment module is accepted the input of real time traffic data stream, and the pre-service result is imported prediction module, active monitoring module and self-adaptation adjusting module; Prediction module produces predicting the outcome of traffic parameter in real time according to the input transport data stream, and is input to the state recognition module; Initiatively whether monitoring module has concept drift to take place according to input real time traffic data flow monitoring.As monitor the concept drift generation, then notify the self-adaptation adjusting module; The initiatively self-adaptation adjustment notice of monitoring module is received in the self-adaptation adjustment, is accepting on the current pretreated transport data stream basis forecast model to be carried out adaptive adjustment; The automatic identification module of traffic behavior is on the basis of Prediction Parameters, and the traffic behavior that identification is obtained outputs to the final user.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of have in real time and the city traffic Forecasting Methodology of continuation property, contains following steps;
Online generalization and the feature extraction of step 1, transport data stream are carried out dimensionality reduction to transport data stream, obtain and the maximally related data subset of prediction task, but have but kept the feature of data former;
The foundation of step 2, traffic behavior forecast model, the process feature and the non-process feature of fusion transport data stream are efficiently set up decision making package knowledge network system;
The online adaptive mechanism of step 3, traffic behavior forecast model, the initiatively generation of monitor concept drift, and dynamically the traffic behavior model is carried out adaptive adjustment;
The automatic identification of step 4, traffic behavior determines current traffic behavior according to all kinds of traffic parameters are fast automatic.
A kind of have in real time and the city traffic Forecasting Methodology of continuation property,
(1) adopt up-to-date data processing technique-data flow technique after, under the situation of linear access (having avoided traditional repeatedly iteration access module) data stream, traffic flow is handled and is analyzed, thereby satisfy the demand of using real-time at memory headroom;
(2) adopt the process neural network technology, can accept data stream continuously, the input of process type, after efficiently finishing the traffic behavior prediction, and export (current method only can support to export a time point state), thereby satisfy the demand of using following plurality of continuous traffic behavior in the time in a continuous time period mode;
(3) adopt
The concept drift treatment technology, the traffic behavior forecast model is carried out the adjustment of dynamic self-adapting, thereby makes the precision of prediction of model and variation that performance can adapt to external environment and corresponding the variation.
The traffic behavior prediction is the basis of launching traffic guidance, and traffic guidance is the key that transport solution blocks.Present existing achievement has non real-time, discontinuous characteristic, can not finely support actual application.Therefore, the characteristic that beneficial effect of the present invention had can help administrative authority's science to formulate the traffic control and management strategy, provides support for realizing traffic intelligentization to a greater extent.For promoting research and the application level of China at intelligent transportation field, the gap of shortening and world level has important significance for theories; For the management level that improve China's road traffic, effectively alleviate traffic congestion, have important practical significance.The art of this patent has the very big market space.
Description of drawings
Fig. 1 is the functional schematic of traffic behavior prediction of the present invention in system for traffic guiding;
Fig. 2 is a traffic behavior forecast function Module Division synoptic diagram of the present invention;
Fig. 3 discerns synoptic diagram automatically for the traffic behavior of traffic behavior prediction of the present invention;
Fig. 4 is the flow chart of steps of traffic behavior prediction of the present invention.
The present invention is further described below in conjunction with drawings and Examples.
Embodiment
As shown in Figure 1, the traffic behavior prognoses system provides real-time estimate according to real-time traffic flow data to the traffic behavior in the following period, and will predict the outcome offers traffic control and inducible system, helps supvr's specified control to induce strategy.Simultaneously, will predict the outcome offers traveler, helps them to select suitable trip route.Measure that supvr and traveler are taked and selection exert an influence to the traffic behavior of next period again, become the further foundation of prediction of prognoses system.Thereby the traffic behavior prognoses system has been set up interactional relation between supvr, traveler and the traffic behavior, realized the information of real-time traffic states is effectively utilized and manual intervention, and be the part of the core of traffic control and inducible system.
As shown in Figure 2, be core of the present invention, mainly comprise the pretreatment module of real time traffic data, the prediction module of traffic behavior parameter, the active monitoring module of forecast model, the self-adaptation adjusting module and the final automatic identification module of traffic behavior of forecast model.
Wherein, the major function of the pretreatment module of transport data stream is: because transport data stream has characteristics such as multi-source, magnanimity, low granularity and higher-dimension, the needs of efficient modeling be can't satisfy, must online generalization (Online Generalization) and feature extraction be carried out transport data stream.Online generalization mainly finished under the prerequisite that keeps the traffic data process feature, by dynamic foundation and the extensive hierarchical tree of maintenance data stream transport data stream mapped directly to higher conceptual level from the primary fine ganglionic layer; Feature extraction is to finish the adjacent position relevant with the target location, data sequence length and traffic parameter Feature Extraction.Based on online generalization and the feature extracting method of data flow technique, can make the transport data stream behind the dimensionality reduction is the reduction of legacy data, keeps original process feature again, thereby for efficiently, forecast modeling lays the foundation accurately.
Wherein, the major function of the prediction module of traffic behavior parameter (flow, occupation rate and speed etc.) is: transport data stream is a kind of complex data that comprises space time information, intricate ground of various features weave in, both comprised regularity factor, comprised various uncertain enchancement factors (as accident etc.) again with process feature.Set up the traffic behavior forecast model and must take all factors into consideration various factors, construct a traffic behavior forecast model with decision making package ability.The present invention has set up the traffic behavior forecast model of procedure-oriented neural network.This model comprises dissimilar neurons, and the process neuron of existing processing procedure feature has the common neuron of handling non-process feature (as uncertain incident etc.) again.Model had both comprised the function of handling regular knowledge, comprised the function of handling uncertain incident again.Adopt the decision-making mechanism that multiple neuron merges to form a kind of effective decision making package system, improved the robustness of prediction accuracy and prediction.
Wherein, the major function of the active monitoring module of forecast model is: in the transport data stream environment, when DATA DISTRIBUTION does not have (or very little) to change, do not need to carry out the continuous updating of model.Need the initiatively bigger variation of monitor data distribution, the renewal of trigger process neural network model when update condition is satisfied.The present invention proposes to propose the DATA DISTRIBUTION variation monitoring method based on the kernel function density Estimation on the basis of data stream sampling theory.This monitoring function is independent of network structure, with the change of the real-time monitor data distribution of the cost of minimum.In a single day the model modification condition satisfies, and then starts the adaptive updates function.
Wherein, the major function of the self-adaptation adjusting module of forecast model is: when the active update condition satisfied, model need carry out Real-time and Dynamic to be upgraded.Adjusted model can reflect the feature of current data stream, has guaranteed prediction accuracy.The present invention proposes the adaptive updates technology based on the traffic behavior forecast model of process neural network, enable to catch the variation of feature in the transport data stream environment, the structure of self-optimizing model self and weights reduce the extensive error of network; According to the characteristics of process neural network, the online incremental learning method of model is proposed on the basis of data flow technique, improve quick, the lasting learning ability of model, thereby set up the adaptation mechanism of large scale process neural network model.
Wherein, the major function of the automatic identification module that traffic behavior is final is: the traffic parameter data (as parameters such as flow, occupation rates) that the traffic behavior prediction obtains all are real-valued, these information are very useful to the domain expert, but for domestic consumer, then details and be unfavorable for understanding too.As shown in Figure 3, the present invention proposes and utilize Hidden Markov Model (HMM) to carry out the automatic identifying method of traffic behavior, high model can carry out modeling, study and online traffic status identification automatically according to the traffic behavior rank of default.
Claims (7)
1. one kind has in real time and the city traffic prognoses system of continuation property, comprise the pretreatment module of real time traffic data, the prediction module of traffic behavior parameter, the active monitoring module of forecast model, the self-adaptation adjusting module of forecast model and the automatic identification module of traffic behavior, it is characterized in that:
The pretreatment module of described real time traffic data is accepted the extraneous transport data stream that imports in real time, and it is carried out online generalization and feature extraction;
The prediction module of described traffic behavior parameter is accepted pretreated traffic parameter information, makes online traffic parameter prediction;
The active monitoring module of described forecast model, the variation of monitoring and controlling traffic data stream feature is in real time fed back in real time and is reported to the police for exceeding the situation that limits threshold value;
The self-adaptation adjusting module of described forecast model is according to the warning message of monitoring module, initiatively forecast model is carried out adaptive adjustment, make forecast model to change with the external world automatically, keep the precision that predicts the outcome;
The automatic identification module of described traffic behavior according to all kinds of traffic behavior parameter flows, the occupation rate parameter of traffic behavior parameter prediction module output, according to the traffic behavior of setting, is discerned current traffic behavior automatically;
Pretreatment module is accepted the input of real time traffic data stream, and the pre-service result is imported prediction module, active monitoring module and self-adaptation adjusting module; Prediction module produces predicting the outcome of traffic parameter in real time according to the input transport data stream, and is input to the state recognition module; Initiatively whether monitoring module has concept drift to take place according to input real time traffic data flow monitoring; As monitor the concept drift generation, then notify the self-adaptation adjusting module; The initiatively self-adaptation adjustment notice of monitoring module is received in the self-adaptation adjustment, is accepting on the current pretreated transport data stream basis forecast model to be carried out adaptive adjustment; The automatic identification module of traffic behavior is on the basis of Prediction Parameters, and the traffic behavior that identification is obtained outputs to the final user.
2. a kind of city traffic prognoses system with real-time and continuation property according to claim 1 is characterized in that: the pretreatment module of described transport data stream, adopt and finish based on online generalization and the Feature Extraction Technology of data flow technique.
3. a kind of city traffic prognoses system with real-time and continuation property according to claim 1, it is characterized in that: described traffic behavior forecast model adopts the process neural network technology, catches the process feature that traffic behavior changes.
4. a kind of city traffic prognoses system with real-time and continuation property according to claim 1, it is characterized in that: the active monitoring module of described forecast model is based on the DATA DISTRIBUTION variation monitoring method of kernel function density Estimation.
5. a kind of city traffic prognoses system with real-time and continuation property according to claim 1, it is characterized in that: the self-adaptation adjusting module of described forecast model is that the online incremental learning method that proposes on the basis of data flow technique is finished.
6. a kind of city traffic prognoses system with real-time and continuation property according to claim 1 is characterized in that: the automatic identification module of described traffic behavior adopts the Hidden Markov Model (HMM) technology to carry out the on-line automatic identification of traffic behavior.
7. one kind has in real time and the city traffic Forecasting Methodology of continuation property, it is characterized in that containing following steps;
Online generalization and the feature extraction of step 1, transport data stream are carried out dimensionality reduction to transport data stream, obtain and the maximally related data subset of prediction task, have kept the feature of data former;
The foundation of step 2, traffic behavior forecast model, the process feature and the non-process feature of fusion transport data stream are efficiently set up decision making package knowledge network system;
The online adaptive mechanism of step 3, traffic behavior forecast model, the initiatively generation of monitor concept drift, and dynamically the traffic behavior model is carried out adaptive adjustment;
The automatic identification of step 4, traffic behavior determines current traffic behavior according to all kinds of traffic parameters are fast automatic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007103039821A CN101188002A (en) | 2007-12-24 | 2007-12-24 | A city traffic dynamic prediction system and method with real time and continuous feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007103039821A CN101188002A (en) | 2007-12-24 | 2007-12-24 | A city traffic dynamic prediction system and method with real time and continuous feature |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101188002A true CN101188002A (en) | 2008-05-28 |
Family
ID=39480377
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007103039821A Pending CN101188002A (en) | 2007-12-24 | 2007-12-24 | A city traffic dynamic prediction system and method with real time and continuous feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101188002A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101977351A (en) * | 2010-10-14 | 2011-02-16 | 北京邮电大学 | Method and system for adjusting vehicle travelling state information transmitting rate |
| WO2011060730A1 (en) * | 2009-11-19 | 2011-05-26 | 北京世纪高通科技有限公司 | Traffic flow predicting method and device thereof |
| CN102169631A (en) * | 2011-04-21 | 2011-08-31 | 福州大学 | Manifold-learning-based traffic jam event cooperative detecting method |
| CN102708684A (en) * | 2012-06-21 | 2012-10-03 | 陕西师范大学 | Short-term traffic flow Volterra-DFP self-adaption prediction method |
| CN102882745A (en) * | 2012-09-29 | 2013-01-16 | 摩卡软件(天津)有限公司 | Method and device for monitoring business server |
| CN102930732A (en) * | 2012-11-19 | 2013-02-13 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Payne model |
| CN102930730A (en) * | 2012-11-19 | 2013-02-13 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Phillips model |
| CN102938208A (en) * | 2012-11-19 | 2013-02-20 | 西安费斯达自动化工程有限公司 | On-line traffic bottleneck predictive control method based on field programmable gate array (FPGA) and improved Payne-Whitham model |
| CN102945610A (en) * | 2012-11-19 | 2013-02-27 | 西安费斯达自动化工程有限公司 | Method for predicting and controlling traffic bottlenecks on line based on field programmable gate array (FPGA) and improved Zhang improved model |
| CN102945608A (en) * | 2012-11-19 | 2013-02-27 | 西安费斯达自动化工程有限公司 | On-line predictive control method of traffic bottlenecks based on field programmable gate array (FPGA) and improved Whitham model |
| CN103413443A (en) * | 2013-07-03 | 2013-11-27 | 太原理工大学 | Short-term traffic flow forecasting method based on hidden Markov model |
| CN105160866A (en) * | 2015-08-07 | 2015-12-16 | 浙江高速信息工程技术有限公司 | Traffic flow prediction method based on deep learning nerve network structure |
| CN105575113A (en) * | 2015-12-14 | 2016-05-11 | 清华大学 | Sensing method of traffic running states |
| CN108629976A (en) * | 2018-05-17 | 2018-10-09 | 同济大学 | Urban traffic blocking predetermined depth learning method based on GPS |
| CN109313718A (en) * | 2016-05-20 | 2019-02-05 | 渊慧科技有限公司 | The Internet |
| CN109410575A (en) * | 2018-10-29 | 2019-03-01 | 北京航空航天大学 | A kind of road network trend prediction method based on capsule network and the long Memory Neural Networks in short-term of nested type |
| CN109858631A (en) * | 2019-02-02 | 2019-06-07 | 清华大学 | The automaton learning system and method for stream data analysis for concept migration |
| CN110135647A (en) * | 2019-05-21 | 2019-08-16 | 中国太平洋保险（集团）股份有限公司 | A kind of control method and control device for realizing trend prediction based on feature modeling |
| CN110866638A (en) * | 2019-11-07 | 2020-03-06 | 怀化学院 | Traffic volume prediction model construction method and device, computer equipment and storage medium |
| CN111815275A (en) * | 2020-07-06 | 2020-10-23 | 广东省交通集团有限公司 | Road transportation integrated system |
-
2007
- 2007-12-24 CN CNA2007103039821A patent/CN101188002A/en active Pending
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011060730A1 (en) * | 2009-11-19 | 2011-05-26 | 北京世纪高通科技有限公司 | Traffic flow predicting method and device thereof |
| CN101977351B (en) * | 2010-10-14 | 2013-06-12 | 北京邮电大学 | Method and system for adjusting vehicle travelling state information transmitting rate |
| CN101977351A (en) * | 2010-10-14 | 2011-02-16 | 北京邮电大学 | Method and system for adjusting vehicle travelling state information transmitting rate |
| CN102169631A (en) * | 2011-04-21 | 2011-08-31 | 福州大学 | Manifold-learning-based traffic jam event cooperative detecting method |
| CN102708684A (en) * | 2012-06-21 | 2012-10-03 | 陕西师范大学 | Short-term traffic flow Volterra-DFP self-adaption prediction method |
| CN102882745A (en) * | 2012-09-29 | 2013-01-16 | 摩卡软件(天津)有限公司 | Method and device for monitoring business server |
| CN102945608B (en) * | 2012-11-19 | 2014-09-03 | 西安费斯达自动化工程有限公司 | On-line predictive control method of traffic bottlenecks based on field programmable gate array (FPGA) and improved Whitham model |
| CN102938208A (en) * | 2012-11-19 | 2013-02-20 | 西安费斯达自动化工程有限公司 | On-line traffic bottleneck predictive control method based on field programmable gate array (FPGA) and improved Payne-Whitham model |
| CN102945610A (en) * | 2012-11-19 | 2013-02-27 | 西安费斯达自动化工程有限公司 | Method for predicting and controlling traffic bottlenecks on line based on field programmable gate array (FPGA) and improved Zhang improved model |
| CN102945608A (en) * | 2012-11-19 | 2013-02-27 | 西安费斯达自动化工程有限公司 | On-line predictive control method of traffic bottlenecks based on field programmable gate array (FPGA) and improved Whitham model |
| CN102930730A (en) * | 2012-11-19 | 2013-02-13 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Phillips model |
| CN102930732A (en) * | 2012-11-19 | 2013-02-13 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Payne model |
| CN102930730B (en) * | 2012-11-19 | 2015-02-04 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Phillips model |
| CN102930732B (en) * | 2012-11-19 | 2015-02-04 | 西安费斯达自动化工程有限公司 | Online traffic bottleneck prediction control method based on FPGA and improved Payne model |
| CN103413443A (en) * | 2013-07-03 | 2013-11-27 | 太原理工大学 | Short-term traffic flow forecasting method based on hidden Markov model |
| CN103413443B (en) * | 2013-07-03 | 2015-05-20 | 太原理工大学 | Short-term traffic flow forecasting method based on hidden Markov model |
| CN105160866A (en) * | 2015-08-07 | 2015-12-16 | 浙江高速信息工程技术有限公司 | Traffic flow prediction method based on deep learning nerve network structure |
| CN105575113A (en) * | 2015-12-14 | 2016-05-11 | 清华大学 | Sensing method of traffic running states |
| CN109313718A (en) * | 2016-05-20 | 2019-02-05 | 渊慧科技有限公司 | The Internet |
| CN109313718B (en) * | 2016-05-20 | 2022-08-16 | 渊慧科技有限公司 | Systems, methods, and media for predicting a future state of an interactive object |
| CN108629976A (en) * | 2018-05-17 | 2018-10-09 | 同济大学 | Urban traffic blocking predetermined depth learning method based on GPS |
| CN109410575A (en) * | 2018-10-29 | 2019-03-01 | 北京航空航天大学 | A kind of road network trend prediction method based on capsule network and the long Memory Neural Networks in short-term of nested type |
| CN109858631A (en) * | 2019-02-02 | 2019-06-07 | 清华大学 | The automaton learning system and method for stream data analysis for concept migration |
| CN109858631B (en) * | 2019-02-02 | 2021-04-27 | 清华大学 | Automatic machine learning system and method for streaming data analysis for concept migration |
| CN110135647A (en) * | 2019-05-21 | 2019-08-16 | 中国太平洋保险（集团）股份有限公司 | A kind of control method and control device for realizing trend prediction based on feature modeling |
| CN110866638A (en) * | 2019-11-07 | 2020-03-06 | 怀化学院 | Traffic volume prediction model construction method and device, computer equipment and storage medium |
| CN111815275A (en) * | 2020-07-06 | 2020-10-23 | 广东省交通集团有限公司 | Road transportation integrated system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101188002A (en) | A city traffic dynamic prediction system and method with real time and continuous feature | |
| CN108846517B (en) | Integration method for predicating quantile probabilistic short-term power load | |
| Jing et al. | Short-term prediction of urban rail transit passenger flow in external passenger transport hub based on LSTM-LGB-DRS | |
| CN106448151A (en) | Short-time traffic flow prediction method | |
| CN104715292A (en) | City short-term water consumption prediction method based on least square support vector machine model | |
| CN111985619B (en) | Urban single intersection control method based on short-time traffic flow prediction | |
| Bao et al. | Prediction of train arrival delay using hybrid ELM-PSO approach | |
| CN107045785A (en) | A kind of method of the short-term traffic flow forecast based on grey ELM neutral nets | |
| Kaibei et al. | Passenger flow forecast of railway station based on improved LSTM | |
| Chen et al. | Dynamic traffic light optimization and Control System using model-predictive control method | |
| Chen et al. | Short-term traffic flow prediction with recurrent mixture density network | |
| Lin et al. | Insights into Travel Pattern Analysis and Demand Prediction: A Data-Driven Approach in Bike-Sharing Systems | |
| Lin et al. | A model‐based demand‐balancing control for dynamically divided multiple urban subnetworks | |
| Liu et al. | Research on the short-term traffic flow prediction method based on BP neural networks | |
| Yi et al. | Forecasting of vessel traffic flow using BPNN based on genetic algorithm optimization | |
| Tian et al. | Active Control Method of Traffic Signal Based on Parallel Control Theory | |
| Chen et al. | Prediction model of commercial economic index based on BPNN optimization algorithm | |
| Huang et al. | Save or waste: Real data based energy-efficient driving | |
| Song et al. | Vehicle relocation scheduling method for car sharing service system based on Markov chain and genetic algorithm | |
| Zhang et al. | Data-driven model for traffic signal control | |
| Ye et al. | Urban Traffic Dynamic OD Prediction Based on Multi-Source Data | |
| CN108830509A (en) | One kind is cruised taxi transport power scale dynamic adjusting method | |
| Kang et al. | Analysis of the Predictive Power of Highway Traffic Flow with Weather Conditions | |
| Wang et al. | Prediction method of traffic congestion volume in traffic jam road section based on genetic algorithm. | |
| CN108198447A (en) | Driver information alarm set and system based on vehicle flowrate prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080528 |