CN109544929B - Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data - Google Patents
Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data Download PDFInfo
- Publication number
- CN109544929B CN109544929B CN201811534976.1A CN201811534976A CN109544929B CN 109544929 B CN109544929 B CN 109544929B CN 201811534976 A CN201811534976 A CN 201811534976A CN 109544929 B CN109544929 B CN 109544929B
- Authority
- CN
- China
- Prior art keywords
- data
- traffic
- intersection
- algorithm
- signal
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
- G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/081—Plural intersections under common control
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096833—Systems involving transmission of navigation instructions to the vehicle where different aspects are considered when computing the route
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle low-carbon control and induction method based on big data, which comprises the following steps: s100, collecting traffic data through various ways including different detection devices, mobile devices and networks; s200, according to the traffic data acquired in the S100, an algorithm for accelerating data processing speed based on a cloud computing technology is provided, and the data are processed; s300, performing online data analysis according to the data processed in the S200 to obtain traffic conditions of intersections and road sections in different periods; s400, controlling an online traffic signal in real time according to a main road signal coordination method to realize main road signal coordination optimization; s500, planning an optimal travel path through a traffic guidance algorithm, and finally guiding a driver to select the optimal path for travel. The method has the advantages that the optimal path is selected for traveling by guiding a driver through the coordination optimization of the main road signals and the optimal path planning, so that the number of times of stopping the vehicle at the intersection and the congestion time are reduced, and the problem of increased exhaust emission caused by the vehicle at the intersection is solved.
Description
Technical Field
The invention relates to the field of traffic coordination control, in particular to a vehicle low-carbon control and induction method, system, equipment and storage medium based on big data.
Background
As the economic conditions of people become better and better, the travel demand becomes greater and greater, resulting in a significant increase in the automobile ownership. The traffic is more crowded due to the increase of the owned quantity of motor vehicles, the jam phenomenon is more and more obvious, and the proportion of automobile energy consumption in the total energy is increased. In recent years, the consumption of fossil energy in transportation and transportation of the whole society is increased by 9.74% year and 10.8% year in China due to the fact that automobiles consume diesel oil and gasoline extracted from petroleum. In contrast, the amount of petroleum consumed by transportation in China exceeds 1.06% of the whole society, and under the situation of energy globalization, China is a main country in terms of both consumption and discharge, and especially, traffic becomes a main use direction and a discharge source.
In urban roads, an inevitable complex intersection is the junction location of traffic. Correctly and reasonably handling traffic conflicts between traffic flows, thereby improving traffic efficiency, and solving serious conflicts to some extent is the key to solving the problem. Whether the intersection traffic signal control is reasonable or not directly determines the running condition of the vehicle and even the running efficiency of the whole road network. In the existing project at present, traffic operation indexes including vehicle delay, queuing length and the like are mainly considered when signal timing is carried out on an intersection in signal control research, but the problem that the emission of automobile exhaust is increased due to the intersection is often ignored, so that the prior art needs to be improved and enhanced.
Disclosure of Invention
The invention provides a vehicle low-carbon control and induction method, system, equipment and storage medium based on big data, aiming at solving the problem of increasing the emission of automobile exhaust caused by neglecting intersections in the existing traffic coordination control technology.
In order to achieve the above purpose, the technical means adopted is as follows:
a vehicle low-carbon control and induction method based on big data comprises the following steps:
s100, collecting traffic data through various ways including different detection devices, mobile devices and networks;
s200, according to the traffic data acquired in the S100, an algorithm for accelerating data processing speed based on a cloud computing technology is provided, and the data are processed;
s300, performing online data analysis according to the data processed in the S200 to obtain traffic conditions of intersections and road sections in different periods;
s400, controlling an online traffic signal in real time according to a main road signal coordination method to realize main road signal coordination optimization;
s500, planning an optimal travel path through a traffic guidance algorithm, and finally guiding a driver to select the optimal path for travel.
According to the scheme, the traffic conditions of all road sections of the road network are calculated and predicted through comprehensive processing and analysis of traffic data, the online traffic signals are controlled in real time, the coordination and optimization of the trunk road signals are realized, and an optimal travel path is planned for a driver, so that the parking times and delay time of the vehicle at the intersection are reduced, and the energy consumption and exhaust emission of the vehicle are reduced.
Preferably, the collected data in step S100 is acquired by one or more data collection devices including geomagnetic devices, coils, video devices, RFID devices, internet devices, and mobile devices; for different acquisition routes, only the corresponding longitude and latitude need to be set. The collected traffic data comprises traffic states, traffic indexes, traffic flow, average speed, queuing length, parking times, delay time, saturation, travel time, recording time and the like, and provides more complete and accurate information for analysis and prediction of the traffic states.
Preferably, step S200 specifically includes:
s210, calculating the characteristics of the traffic data acquired in S100 by using a Canopy clustering algorithm, putting a traffic data set with the same characteristics into a subset, namely Canopy, wherein each Canopy is regarded as a cluster, is marked as V, and puts the clusters into a set S;
s211, performing dichotomous clustering in each Canopy by using a K-means clustering algorithm: sequentially extracting each cluster from the set S in the S210, wherein the extracted clusters are matched with a limit point criterion, then performing binary clustering by using a K-means clustering algorithm, putting two clusters with the minimum error square sum in the clusters back into the set S, and performing circularly until K clusters are obtained;
and S212, accelerating data processing through the data processing algorithm of S210 and S211.
Preferably, the online data analysis in step S300 includes:
s310, medium-term traffic flow prediction is carried out based on the hourly combination model:
the first step is as follows: screening the data according to the Latt criterion;
the second step is that: carrying out correlation analysis on the screened data;
the third step: according to the flow distribution, time intervals are divided, and 24 hours a day are divided into 24 time intervals; researching the relation between the flow and the speed of each time interval, namely a Q-V model, and predicting the flow in the time interval by using an ARIMA model;
the fourth step: and combining the prediction result with the result of the Q-V model according to a certain weight, and comparing the final calculation result with the actual flow according to three indexes of average absolute error, average absolute percentage error and mean square error.
S311, estimating traffic state information based on an expected optimization extended Kalman filtering algorithm;
wherein the expectation optimization algorithm is an iterative solution algorithm for estimating model parameters alpha and unknown variables Y
The first step is as follows: solving Y, alpha | X; first, find outThe accurate solution of Y | X, alpha and the limitation of the accurate solution are pointed out, the accurate reasoning is realized by recursive Bayes estimation, the posterior probability density is enlarged, and the calculation is carried out in a recursive mode; then, an extensible Kalman filtering algorithm is used as a substitute approximate solution; assuming that the observed value is x and the hidden state is y, the state transformation distribution is as p (y)(t)y(t-1)) And the observed distribution is as in the formula p (x)(t)|y(t)). The hidden state y is followed by the reasoning:
using p (y)(t-1)|x(1：t-1)) The recursive solution formula is:
p(y(t)|x(1：t-1))＝∫p(y(t)|y(t-1)p(y(t-1)|x(1：t-1))dy(t-1)；
for the same reason, p (y)(t)|x(1：t)) Can be represented by the formula p (y)(t)|x(1：t))＝Cp(x(t）|y(t))p(y(t)|x(1：t-1)) And C ═ p ([ j ] p (x)(t)|y(t))p(y(t)|x(1：t-1)dy(t))-1And (4) obtaining.
The second step is that: solving alpha | Y, X; first, the optimal model parameters α are determined, which can be implemented as a simple error minimization, assuming that all traffic conditions are known; obtaining an observation function h () and a pseudo observation from a basic diagram and a link queue model, wherein the calculation formula is X-H (y); then, the observation X and the pseudo-observation X are minimized+The difference between them, resulting in an iteration of a new model parameter α defined in the desired optimization algorithm, with the formula
Wherein alpha is+A new iteration of a is represented which is,
representing the squared euclidean norm, where alpha comprises the traffic event adaptation. Any event detection, as a quantification of capacity reduction, can be added to the equation as an additional constraint on α.
Preferably, the main channel signal coordination method in step S400 includes an unsaturated signal coordination method and a supersaturated signal coordination method;
the unsaturated signal coordination method comprises the following steps:
step1, determining a signal phase allowable setting mode according to the lane canalization condition and the actual traffic condition of each intersection main road direction;
step2, allowing a setting mode aiming at different signal phases of each intersection, and distributing all surplus split ratios to coordinated phases according to the principle that the total split ratio of the signal phases of the main roads of the intersection is not changed on the basis of meeting the traffic demand of the non-coordinated direction;
step3, according to the principle that the running time of the road section is not changed, the running speed in a certain coordination direction is changed, so that the equivalent distance in the direction is equal to the distance between opposite intersections;
step4, according to the value range of the signal period of each intersection on the main road, taking the intersection as the value optimization space of the public signal period;
step5, deducing corresponding ideal intersection distance by using a time distance graph according to different signal phase combinations between the reference intersection and other intersections;
and 7, for a certain driving direction, respectively calculating the green ratio above and below the green light center time line according to the offset green ratio of each intersection, selecting the minimum green ratio above and below, and adding to obtain the width of the green wave band in the driving direction.
Preferably, the oversaturation signal coordinating method comprises the steps of:
step1, defining a queuing length influence coefficient according to different situations:
there are three cases:
(1) the road environment of the road section in the up and down directions has no special requirement on the queuing length;
(2) the overflow phenomenon is easy to occur on the downstream road section in a certain direction of the road section, and the queue length in the direction is properly increased, so that the downstream overflow control is facilitated;
(3) an entrance and an exit of an important place are arranged in a certain direction of the road section, and when the vehicles in the line in the direction exceed the entrance and the exit, the vehicles in the line need to stop and wait until the vehicles in the line dissipate to the entrance and the exit;
therefore, the uplink and downlink weighted total length delta L and the uplink and downlink total delay D of the road section are defined, and the following steps are included:
in the formula:
respectively representing the maximum queue length of uplink and downlink;
αu、αdrespectively representing the influence weight of the uplink and downlink queuing length, and the value rule is as follows: taking 1 under the condition (1), taking a value larger than 1 under the condition (2) and taking a value smaller than 1 under the condition (3); du、DdRespectively representing uplink and downlink delays;
step2, establishing a phase difference optimization flow by taking min delta L and minD as targets in sequence, and solving the phase difference in a layered mode:
step1, inputting the signal parameters and traffic flow parameters of the upstream and downstream intersections, and determining the influence weight alpha of the upstream and downstream queuing lengths according to the environmental characteristics of the road sectionsu、αd；
Step2 let the phase difference
Judging whether the initial period of the section reaches the attribution of the traffic flow, and sequentially establishing a queuing length model and a delay model;
step3 is traversed by using enumeration method with Step size of 1s
Meanwhile, calculating and storing delta L, D under the corresponding phase difference;
step4 takes min delta L as a first-layer optimization target, and searches a corresponding phase difference value range to obtain an effective solution space;
step5 takes minD as the optimization target of the second layer, and obtains the optimal phase difference from the optimized effective solution space of the previous layer.
Preferably, the traffic induction algorithm in step S500 includes:
s510.MAP phase: the Map function takes the section travel time of the corresponding sub-network, intersection delay data and a calculation time range as input, and traverses the structure of the graph according to the hierarchy according to the breadth-first principle; the Map function calculates the time for reaching the next intersection according to the travel time of each time period and intersection delay data, and finally generates a key/value form intermediate value; the key of the intermediate value is the time of arriving at the intersection, and the values are the intersection ID, the precursor intersection ID and the time of arriving at the precursor intersection respectively;
s511.reduce stage: the HaLoop combines all values with the same key value together and transmits the values to the Reduce function, and the Reduce function processes the value set of the intersection with the same arrival time to generate a new bucket;
s512, iteration of HaLoop MapReduce, namely, taking the output of the Reduce stage as the input of the next Map stage, continuously starting the operation of a Map-Reduce task by the operation server, and completing the calculation of all intersections included in the path through multiple iterations; in the iterative process, a Master node machine of HaLoop is responsible for loop control in Job until the iterative computation is finished.
A big data-based vehicle low carbon control and induction system comprises:
the data acquisition module is used for acquiring traffic data through various ways including different detection equipment, mobile equipment and networks;
the data processing module is used for providing an algorithm for accelerating the data processing speed based on the cloud computing technology according to the traffic data acquired by the data acquisition module and processing the data;
the traffic running state module is used for carrying out online data analysis according to the data processed by the data processing module and obtaining the traffic conditions of intersections and road sections in different periods through analysis;
and the coordination optimization module is used for controlling the online traffic signals in real time according to the main road signal coordination method and realizing the main road signal coordination optimization.
And the route induction module is used for planning an optimal travel route through a traffic induction algorithm and finally inducing a driver to select the optimal route for travel.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the above-described method of the invention when the processor executes the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of the invention as described above.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention collects traffic data through different detection devices, mobile devices, networks and other ways; providing an algorithm for accelerating data processing speed based on a cloud computing technology according to acquired traffic data and processing the data; carrying out online data analysis on the traffic data, and analyzing traffic conditions of intersections and road sections in different periods; performing main road signal coordination optimization, realizing one-road green light same-row traffic flow when the traffic flow is not saturated in the main road group traffic flow of the road network, and reducing parking times and congestion time when the traffic flow is in a near-saturated or saturated state; and optimal path planning is carried out, and a driver is induced to select an optimal path to go out, so that the number of times of stopping the vehicle at the intersection and the congestion time are reduced, the problem of increased exhaust emission caused by the vehicle at the intersection is solved, the energy consumption of the vehicle is reduced, and the urban living environment and the humanistic environment are improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Fig. 2 is a schematic diagram of data acquisition according to an embodiment of step S100 of the present invention.
Fig. 3 is a graph a of the relationship between the speed and the flow rate according to one embodiment of step S310 of the present invention.
Fig. 4 is a graph b of the relationship between the speed and the flow rate in step S310 according to an embodiment of the present invention.
FIG. 5 is a graph a comparing the average absolute error with the actual flow rate in step S310 according to one embodiment of the present invention.
FIG. 6 is a graph b illustrating the comparison between the average absolute error and the actual flow rate in step S310 according to an embodiment of the present invention.
FIG. 7 is a comparison graph c of the average absolute error and the actual flow rate in step S310 according to one embodiment of the present invention.
Fig. 8 is a schematic diagram of the traffic status information estimation method in step S311 according to the present invention.
Fig. 9 is a flowchart illustrating an unsaturated signal coordination method in step S400 according to the present invention.
Fig. 10 is a flowchart illustrating a method for coordinating oversaturation signals in step S400 according to the present invention.
Fig. 11 is a diagram of a traffic route guidance algorithm process a in step S500 according to the present invention.
Fig. 12 is a diagram of a traffic route guidance algorithm process b in step S500 according to the present invention.
Fig. 13 is a diagram of a traffic route guidance algorithm process c in step S500 according to the present invention.
FIG. 14 is a schematic diagram of route induction according to an embodiment of the present invention.
FIG. 15 is a block diagram of the system of the present invention.
FIG. 16 is a block diagram of a computer setup of the present invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
The invention provides a vehicle low-carbon control and induction method based on big data, which comprises the following steps:
s100, collecting traffic data through various ways including different detection devices, mobile devices and networks;
s200, according to the traffic data acquired in the S100, an algorithm for accelerating data processing speed based on a cloud computing technology is provided, and the data are processed;
s300, performing online data analysis according to the data processed in the S200 to obtain traffic conditions of intersections and road sections in different periods;
s400, controlling an online traffic signal in real time according to a main road signal coordination method to realize main road signal coordination optimization;
s500, planning an optimal travel path through a traffic guidance algorithm, and finally guiding a driver to select the optimal path for travel.
Wherein, the collected data in step S100 is acquired by one or more data collection devices including geomagnetic sensor, coil sensor, video sensor, RFID sensor, internet sensor, and mobile device; for different acquisition routes, only the corresponding longitude and latitude need to be set.
As shown in fig. 2, in this example, 334 days of traffic flow data are selected from the road junction north of the Xinzhongshan city, zhongshan city, Guangdong province, in 2017, from 1 month to 11 months and 30 days, and the sample collection interval is 1 h. The speed data is from a Gaode map, the collection time period is also 1 month and 1 day to 11 months and 30 days in 2017, the sample collection interval is also 1h, and 8016 groups of data are counted.
Wherein, step S200 specifically includes:
s210, calculating the characteristics of the traffic data acquired in S100 by using a Canopy clustering algorithm, putting a traffic data set with the same characteristics into a subset, namely Canopy, wherein each Canopy is regarded as a cluster, is marked as V, and puts the clusters into a set S;
s211, performing dichotomous clustering in each Canopy by using a K-means clustering algorithm: sequentially extracting each cluster from the set S in the S210, wherein the extracted clusters are matched with a limit point criterion, then performing binary clustering by using a K-means clustering algorithm, putting two clusters with the minimum error square sum in the clusters back into the set S, and performing circularly until K clusters are obtained; the K-means clustering algorithm is a classic algorithm of a data clustering algorithm, and the core idea is as follows: firstly, randomly extracting k data center points from an original data set, and regarding the points as initial clustering centers; second, the distance from the surrounding points to the center point is calculated around the cluster center. Finally, the best cluster is selected based on the distance parameter, data is assigned to the cluster, data points are then assigned to the cluster, and the process is repeated. In the embodiment, a K-means algorithm is used for analyzing S210 data, and the following definitions 1,2 and 3 are used for analyzing clusters; wherein:
definition 1(Canopy definition): given data set U ═ U i1,2,3,. n, wherein,
conform to
Then set xiThat is, they are classified as Canopy set, T1Denotes the radius of the Canopy set, cjRepresenting the Canopy center point.
Definitions 2 (SSE)i): cluster CiRepresents the sum of squares of the distances from all points within the cluster to the center of the cluster, and has the formula
ciIs a cluster CiThe center of mass of the lens.
Define 3 (limit point criterion): given cluster
So that
X is thenp,xqIs the limit point of cluster C, in xp,xqTwo points are a cluster C initial clustering center, namely a limit point principle, in the formula, dist (x)p,xq) Indicating the limit distance.
And S212, accelerating data processing through the data processing algorithm of S210 and S211.
Wherein, the online data analysis in step S300 includes:
s310, medium-term traffic flow prediction is carried out based on the hourly combination model:
the first step is as follows: screening the data according to the Latt criterion;
the second step is that: carrying out correlation analysis on the screened data; from the whole, the positive correlation with the demand of the latest moment (within 2 hours) of the history is relatively high and exceeds 66 percent;
the third step: according to the flow distribution, time intervals are divided, and 24 hours a day are divided into 24 time intervals; researching the relation between the flow and the speed of each time interval, namely a Q-V model, and predicting the flow in the time interval by using an ARIMA model;
the fourth step: and combining the prediction result with the result of the Q-V model according to a certain weight, and comparing the final calculation result with the actual flow according to three indexes of average absolute error, average absolute percentage error and mean square error.
Among them, the reiter criterion (i.e., the 3 σ criterion) is a method for discriminating an abnormal value in the case of normal distribution. The specific contents are as follows: suppose that in a column of equal precision measurement results, the residual error corresponding to the ith measurement value
Is satisfied with
The error is a gross error, corresponding to the measured value xiIf the value is abnormal, it should be rejected. Wherein the standard deviation estimate:
according to the collected data as shown in fig. 2, correlation analysis is performed, and the correlation analysis refers to analyzing two or more variable elements with correlation, so as to measure the degree of closeness of correlation of the two variable elements. The correlation analysis can be performed only if a certain relation or probability exists between the elements subjected to the correlation. As shown in table 1 below, d (w, m) represents the total amount of travel demand at the mth time on the w-th day. As shown in table 1, which is the data correlation analysis result from 1/11/30/2017, it can be seen that:
TABLE 1
On the same day, the flow rate at the current moment is inversely proportional to the historical time, as shown in the last column of table 1, and as a whole, the positive correlation with the flow rate at the latest moment (within 1 hour) in the history is relatively high and exceeds 0.8; whereas the correlation with earlier data decreased significantly, especially to 0.112 at hour 5.
| Correlation coefficient | d(w,m) | d(w+1,m) | d(w+2,m) | d(w+3,m) | d(w+4,m) | d(w+5,m) | d(w+6,m) |
| d(w,m) | 1 | 0.906 | 0.875 | 0.869 | 0.866 | 0.865 | 0.788 |
| d(w+1,m) | 0.906 | 1 | 0.906 | 0.875 | 0.869 | 0.867 | 0.780 |
| d(w+2,m) | 0.875 | 0.906 | 1 | 0.906 | 0.875 | 0.869 | 0.780 |
| d(w+3,m) | 0.869 | 0.875 | 0.906 | 1 | 0.906 | 0.875 | 0.784 |
| d(w+4,m) | 0.866 | 0.869 | 0.875 | 0.906 | 1 | 0.906 | 0.788 |
| d(w+5,m) | 0.865 | 0.867 | 0.869 | 0.875 | 0.906 | 1 | 0.811 |
| d(w+6,m) | 0.788 | 0.780 | 0.780 | 0.784 | 0.788 | 0.811 | 1 |
TABLE 2
As can be seen from table 2, in the same time of the week, the traffic on the historical days at the same time is higher in the traffic correlation on the working day as a whole, and is the lowest in the correlation with the weekday. However, saturday is special, and the correlation with working days is higher than sunday, and is mainly related to working of more enterprise units near the intersection.
Wherein, the annual speed and flow data are counted, and the flow-density-speed relation diagram is shown in fig. 3. The fitted curve is: Q-2.8324V2+148.88V-704.44R2＝0.766；
The 24 hours of the whole day is divided into 24 periods, and the relation between the average flow and the speed of each period is counted. The "0 point" is defined to mean 0:00:00 to 0:59:59, and the "1 point" is defined to mean 1:00:00 to 1:59:59, as shown in FIG. 4.
According to the analysis of the related variables, the input variables of the taxi travel demand prediction model are not limited to the demands at the moments before the calendar, but are more closely related to the historical moments, and the input parameters used in the embodiment are as follows: 1. the current date w; 2. a time period; 3. the last four historical time flows.
Take 6 o' clock as an example to describe the rule. It can be seen that at 6 o' clock, the relationship between speed and flow is as follows:
Q＝0.1878V3-24.18V2+1036V-14529；
the prediction result index is shown in fig. 5 to 7. It can be found that at 6 points, as the weight coefficient α increases, the prediction accuracy of the wavelet neural network combined model and the BP neural network combined model increases, while the prediction accuracy of the ARIMA combined model and the gray combined model decreases, and when α is 1, that is, when only the Q-V model is adopted, the prediction accuracy is between that of the BP neural network combined model and the gray combined model. When alpha is the same, the prediction accuracy of the combined prediction model is as low as: ARIMA (1,0,1) combination model, gray combination model, BP neural network combination model, wavelet neural network combination model.
According to the above, the model prediction precision of each time segment is sorted under the condition of the same weight alpha, wherein A represents an ARMIA combined prediction model, B represents a BP neural network combined prediction model, G represents a gray combined prediction model, and W represents a wavelet neural network combined prediction model. Wherein the prediction precision is reduced from left to right, the model in brackets represents the same or similar prediction precision as the previous model, and the model can be replaced. The results are shown in Table 3.
| Time of day | Selecting model orderings | Time of day | Selecting model orderings |
| Point 0 | 1A(G)>2W> |
12 points | 1A>2W>3G> |
| 1 point | 1A>2G>3W>4B | 13 o' clock | 1A>2W(G)> |
| 2 point | 1A(G)>2W> |
14 points | 1A(G/B)> |
| 3 point | 1B>2A(W)> |
15 points | 1A>2G>3B> |
| 4 points | 1A(G/B)> |
16 points | 1A>2G>3B> |
| 5 point | 1W>2A(G)> |
17 point | 1A>2G(B/W) |
| 6 points | 1A>2G>3B> |
18 points | 1A(W)>2G> |
| 7 point | 1A>2W>3B> |
19 points | 1A>2W>3G> |
| 8 points | 1A>2B>3G(W) | 20 points | 1A>2W>3G>4B |
| 9 o' clock | 1G>2B(A)> |
21 point | 1A(W)>2G>3B |
| 10 o' clock | 1A>2G>3B> |
22 points | 1A>2W>3G> |
| 11 point | 1A(B)>2G>3W | 23 o' clock | 1A(G/W)>2B |
TABLE 3
As can be seen from table 3, in 24 time periods, the ARMIA combined prediction model obtained 21 first choices (13 unique first choices plus 8 parallel first choices), the gray combined prediction model obtained 6 first choices (1 unique first choice plus 5 parallel first choices), the BP neural network combined prediction model obtained 4 first choices (1 unique first choice plus 3 parallel first choices), and the wavelet neural network combined prediction model obtained 4 first choices (1 unique first choice plus 3 parallel first choices). And in the aspect of the worst prediction precision, the BP neural network combined prediction model is worst in 13 time intervals (comprising 12 worst and 1 parallel worst), the wavelet neural network combined prediction model is worst in 10 time intervals (comprising 8 worst and 2 parallel worst), the gray combined prediction model is worst in 3 time intervals (comprising 2 worst and 1 parallel worst), and the ARMIA combined prediction model is worst in 0 time interval. Therefore, the ARMIA combined prediction model has high precision and can be used for a prediction time period of 87.5 percent, and the prediction precision of the two neural network combined models is low.
S311, estimating traffic state information based on an expected optimization extended Kalman filtering algorithm;
wherein the expectation optimization algorithm is an iterative solution algorithm for estimating model parameters alpha and unknown variables
The first step is as follows: y, α | X are obtained. Estimating the unobserved traffic conditions, i.e. solving for Y | X, α, followed by requiring an accurate solution for Y | X, α, and indicating the limitations of the accurate solution, assuming that all model parameters are known, the accurate inference is performed by recursive bayesian estimationThe posterior probability density is expanded and computed in a recursive manner. Assuming that the observed value is x and the hidden state is y, the state transformation distribution is as p (y)(t)y(t-1)) And the observed distribution is as in the formula p (x)(t)|y(t)). The hidden state y is followed by the reasoning:
using p (y)(t-1)|x(1：t-1)) The recursive solution formula is:
p(y(t)|x(1：t-1))＝∫p(y(t)|y(t-1)p(y(t-1)|x(1：t-1)))dy(t-1)；
for the same reason, p (y)(t)|x(1：t)) Can be represented by the formula p (y)(t)|x(1：t))＝Cp(x(t)|y(t))p(y(t)|x(1：t-1)) And C ═ p ([ j ] p (x)(t)|y(t))p(y(t)|x(1：t-1)dy(t))-1And (4) obtaining.
The second step is that: solving alpha | Y, X; first, the optimal model parameters α are determined, which can be implemented as a simple error minimization, assuming that all traffic conditions are known; obtaining an observation function h () and a pseudo observation from a basic diagram and a link queue model, wherein the calculation formula is X-H (y); then, the observation X and the pseudo-observation X are minimized+The difference between them, resulting in an iteration of a new model parameter α defined in the desired optimization algorithm, with the formula
Wherein alpha is+A new iteration of a is represented which is,
representing the squared euclidean norm, where alpha comprises the traffic event adaptation. Any event detection, as a quantification of capacity reduction, can be added to the equation as an additional constraint on α.
As shown in fig. 8, initially, all links are set to null, the simulation duration T is 1.05 (hours), and the time interval Δ T is 1.75 × 10-4(hours). The boundary conditions are all constants: the origin is required to be constant Do(t) 7020vph and the end point supply also being a constant Sd(t) 2340 vph. Further, the basic diagram of the link is the free flow velocity Ff70mph, the congestion index of each road is J125 v/km, and the critical traffic density is dc25 v/km. Length of each link Li＝[1，1，1，1]km, number of lanes per link ni＝[3，1，2，1]And km. Link 1 becomes link 2 and link 3 at the intersection. In addition, there are parking lots along the street on link 3, which can increase or decrease the traffic inrush into link 3 by a random number z-N (0, 500) vph. In the initial condition, all links are empty. In the link queue model, there are 4 traffic flow state variables in the whole road network in a specific time step i.
The trunk signal coordination method in step S400 includes an unsaturated signal coordination method and a supersaturated signal coordination method;
as shown in fig. 9, the unsaturated signal coordination method includes:
step1, determining a signal phase allowable setting mode according to the lane canalization condition and the actual traffic condition of each intersection main road direction;
step2, allowing a setting mode aiming at different signal phases of each intersection, and distributing all surplus split ratios to coordinated phases according to the principle that the total split ratio of the signal phases of the main roads of the intersection is not changed on the basis of meeting the traffic demand of the non-coordinated direction;
step3, according to the principle that the running time of the road section is not changed, the running speed in a certain coordination direction is changed, so that the equivalent distance in the direction is equal to the distance between opposite intersections;
step4, according to the value range of the signal period of each intersection on the main road, taking the intersection as the value optimization space of the public signal period;
step5, deducing corresponding ideal intersection distance by using a time distance graph according to different signal phase combinations between the reference intersection and other intersections;
and 7, for a certain driving direction, respectively calculating the green ratio above and below the green light center time line according to the offset green ratio of each intersection, selecting the minimum green ratio above and below, and adding to obtain the width of the green wave band in the driving direction.
As shown in fig. 10, the supersaturation signal coordination method includes the steps of:
step1, defining a queuing length influence coefficient according to different situations:
there are three cases:
(1) the road environment of the road section in the up and down directions has no special requirement on the queuing length;
(2) the overflow phenomenon is easy to occur on the downstream road section in a certain direction of the road section, and the queue length in the direction is properly increased, so that the downstream overflow control is facilitated;
(3) an entrance and an exit of an important place are arranged in a certain direction of the road section, and when the vehicles in the line in the direction exceed the entrance and the exit, the vehicles in the line need to stop and wait until the vehicles in the line dissipate to the entrance and the exit;
therefore, the uplink and downlink weighted total length delta L and the uplink and downlink total delay D of the road section are defined, and the following steps are included:
in the formula:
respectively representing the maximum queue length of uplink and downlink;
αu、αdrespectively representing the influence weight of the uplink and downlink queuing length, and the value rule is as follows: taking 1 under the condition (1), taking a value larger than 1 under the condition (2) and taking a value smaller than 1 under the condition (3); du、DdRespectively representing uplink and downlink delays;
step2, establishing a phase difference optimization flow by taking min delta L and minD as targets in sequence, and solving the phase difference in a layered mode:
step1, inputting the signal parameters and traffic flow parameters of the upstream and downstream intersections, and determining the influence weight alpha of the upstream and downstream queuing lengths according to the environmental characteristics of the road sectionsu、αd；
Step2 let the phase difference
Judging whether the initial period of the section reaches the attribution of the traffic flow, and sequentially establishing a queuing length model and a delay model;
step3 is traversed by using enumeration method with Step size of 1s
Meanwhile, calculating and storing delta L, D under the corresponding phase difference;
step4 takes min delta L as a first-layer optimization target, and searches a corresponding phase difference value range to obtain an effective solution space;
step5 takes minD as the optimization target of the second layer, and obtains the optimal phase difference from the optimized effective solution space of the previous layer.
Wherein, the traffic guidance algorithm in step S500 includes:
s510.MAP phase is as shown in FIG. 11: the Map function takes the section travel time of the corresponding sub-network, intersection delay data and a calculation time range as input, and traverses the structure of the graph according to the hierarchy according to the breadth-first principle; the Map function calculates the time for reaching the next intersection according to the travel time of each time period and intersection delay data, and finally generates a key/value form intermediate value; the key of the intermediate value is the time of arriving at the intersection, and the values are the intersection ID, the predecessor intersection ID and the time of arriving at the predecessor intersection respectively.
S511.reduce phase is shown in FIG. 12: the HaLoop combines all values with the same key value together and transmits the values to the Reduce function, the Reduce function processes the value set of the intersection with the same arrival time to generate a new bucket, and the bucket list B is used for storing intersection nodes to be accessed at different time nodes.
S512, iteration of HaLoop MapReduce is shown in a figure 13, wherein each time output of a Reduce stage is used as input of a next Map stage, a job server (JobTracker) continuously starts to run Map-Reduce tasks, and calculation of all intersections included in a path is completed through multiple iterations; in the iterative process, a Master node machine of HaLoop is responsible for loop control in Job until the iterative computation is finished.
As shown in fig. 14, in the present route guidance embodiment, conditions such as a start point, an end point, and a route are set, and an optimal travel route is planned by the method in step S500, so as to implement a route guidance scheme from guangzhou city to zhongshan city.
The present invention also provides a system applied to the method of the present invention, as shown in fig. 15, including:
the data acquisition module 1 is used for acquiring traffic data through various ways including different detection devices, mobile devices and networks; the collected traffic data comprises traffic states, traffic indexes, traffic flow, average speed, queuing length, parking times, delay time, saturation, travel time, recording time and the like, and more complete and accurate information is provided for analysis and prediction of the traffic states;
and the data processing module 2 is used for providing an algorithm for accelerating the data processing speed based on the cloud computing technology and processing the data.
And the traffic running state module 3 is used for carrying out online data analysis according to the data processed by the data processing module 2, and analyzing to obtain the traffic conditions of intersections and road sections in different periods. According to different time periods of selection and analysis, traffic state analysis, index analysis, flow analysis, speed analysis, queuing length analysis, parking times analysis, delay time analysis, saturation analysis, travel time analysis and the like are respectively carried out on intersection, road section and region data in time periods of time, day, week, month and year to generate analysis statistical graphs, and traffic conditions in different periods are visually displayed in a graph mode.
And the coordination optimization module 4 is used for controlling the online traffic signals in real time through a main road signal coordination method to realize the main road signal coordination optimization. The intersection traffic signal controller is controlled by the issuing interface through the scheme, so that the traffic flow passes through a green light when the traffic flow of the road network trunk group is not saturated, and the parking times and the congestion time are reduced when the traffic flow is in a near-saturated or saturated state.
And the path induction module 5 is used for planning an optimal travel path through a traffic induction algorithm and finally inducing a driver to select the optimal path for travel.
The invention also provides computer equipment, which can be mobile terminal, desktop computer, notebook, palm computer, server and other computer equipment. As shown in fig. 16, a processor 10, a memory 20 and a display 30 in the computer device. FIG. 16 shows only some of the components of the computer device, but it is to be understood that not all of the shown components are required to be implemented, and that more or fewer components may be implemented instead.
The memory 20 may in some embodiments be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The memory 20 may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device. Further, the memory 20 may also include both an internal storage unit and an external storage device of the computer device. The memory 20 is used for storing application software installed on the computer device and various types of data, such as program codes of the installed computer device. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 20 stores a big data-based vehicle low carbon control and induction program 40, and the big data-based vehicle low carbon control and induction program 40 can be executed by the processor 10, so as to implement the video list switching method based on the educational system according to the embodiments of the present application.
The processor 10 may be, in some embodiments, a Central Processing Unit (CPU), a microprocessor or other data Processing chip, and is configured to run program codes stored in the memory 20 or process data, such as executing the video list switching method based on the educational system.
The display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch panel, or the like in some embodiments. The display 30 is used for displaying information at the computer device and for displaying a visual user interface. The components 10-30 of the computer device communicate with each other via a system bus.
When the processor 10 executes the vehicle low carbon control and induction program based on the big data in the memory 20, the following steps are realized:
s100, collecting traffic data through various ways including different detection devices, mobile devices and networks;
s200, according to the traffic data acquired in the S100, an algorithm for accelerating data processing speed based on a cloud computing technology is provided, and the data are processed;
s300, performing online data analysis according to the data processed in the S200 to obtain traffic conditions of intersections and road sections in different periods;
s400, controlling an online traffic signal in real time according to a main road signal coordination method to realize main road signal coordination optimization;
s500, planning an optimal travel path through a traffic guidance algorithm, and finally guiding a driver to select the optimal path for travel.
The present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of a big data based vehicle low carbon control and inducement method.
The terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (9)
1. A vehicle low-carbon control and induction method based on big data is characterized by comprising the following steps:
s100, collecting traffic data through various ways including different detection devices, mobile devices and networks;
s200, according to the traffic data acquired in the S100, an algorithm for accelerating data processing speed based on a cloud computing technology is provided, and the data are processed;
s300, performing online data analysis according to the data processed in the S200 to obtain traffic conditions of intersections and road sections in different periods;
s400, controlling an online traffic signal in real time according to a main road signal coordination method to realize main road signal coordination optimization;
s500, planning an optimal travel path through a traffic guidance algorithm, and finally guiding a driver to select the optimal path for travel;
the online data analysis described in step S300 includes:
s310, medium-term traffic flow prediction is carried out based on the hourly combination model:
the first step is as follows: screening the data according to the Latt criterion;
the second step is that: carrying out correlation analysis on the screened data;
the third step: according to the flow distribution, time intervals are divided, and 24 hours a day are divided into 24 time intervals; researching the relation between the flow and the speed of each time interval, namely a Q-V model, and predicting the flow in the time interval by using an ARIMA model;
the fourth step: combining the prediction result with the result of the Q-V model according to a certain weight, and comparing the final calculation result with the actual flow according to three indexes of average absolute error, average absolute percentage error and mean square error;
s311, estimating traffic state information based on an extended Kalman filtering algorithm of expectation optimization:
wherein the desired optimization algorithm is an iterative solution algorithm argmax estimating model parameters alpha and unknown variables YY,αP(Y,α|X)：
The first step is as follows: solving Y, alpha | X; firstly, solving an accurate solution of Y | X, alpha, indicating the limitation of the accurate solution, realizing accurate reasoning by recursive Bayes estimation, expanding posterior probability density, and calculating in a recursive mode; then, an extensible Kalman filtering algorithm is used as a substitute approximate solution;
the second step is that: solving alpha | Y, X; first, the optimal model parameters α are determined, which can be implemented as a simple error minimization, assuming that all traffic conditions are known; obtaining an observation function h () and a pseudo observation from a basic diagram and a link queue model, wherein the calculation formula is X-H (y); then, the observation X and the pseudo-observation X are minimized+The difference between them, resulting in an iteration of a new model parameter α defined in the desired optimization algorithm, with the formula
Wherein alpha is+A new iteration of a is represented which is,
representing the squared euclidean norm, where alpha comprises the traffic event adaptation.
2. The big data based vehicle low carbon control and guidance method according to claim 1, wherein the collected data in step S100 is acquired by one or more data collection devices including geomagnetic, coil, video, RFID, internet and mobile devices; for different acquisition routes, only the corresponding longitude and latitude need to be set.
3. The big-data-based vehicle low-carbon control and induction method according to claim 1, wherein the step S200 specifically comprises:
s210, calculating the characteristics of the traffic data acquired in S100 by using a Canopy clustering algorithm, putting a traffic data set with the same characteristics into a subset, namely Canopy, wherein each Canopy is regarded as a cluster, is marked as V, and puts the clusters into a set S;
s211, performing dichotomous clustering in each Canopy by using a K-means clustering algorithm: sequentially extracting each cluster from the set S in the S210, wherein the extracted clusters are matched with a limit point criterion, then performing binary clustering by using a K-means clustering algorithm, putting two clusters with the minimum error square sum in the clusters back into the set S, and performing circularly until K clusters are obtained;
and S212, accelerating data processing through the data processing algorithm of S210 and S211.
4. The big data-based low carbon control and induction method for the vehicle as claimed in claim 1, wherein the arterial road signal coordination method in step S400 comprises an unsaturated signal coordination method and a supersaturated signal coordination method;
the unsaturated signal coordination method comprises the following steps:
step1, determining a signal phase allowable setting mode according to the lane canalization condition and the actual traffic condition of each intersection main road direction;
step2, allowing a setting mode aiming at different signal phases of each intersection, and distributing all surplus split ratios to coordinated phases according to the principle that the total split ratio of the signal phases of the main roads of the intersection is not changed on the basis of meeting the traffic demand of the non-coordinated direction;
step3, according to the principle that the running time of the road section is not changed, the running speed in a certain coordination direction is changed, so that the equivalent distance in the direction is equal to the distance between opposite intersections;
step4, according to the value range of the signal period of each intersection on the main road, taking the intersection as the value optimization space of the public signal period;
step5, deducing corresponding ideal intersection distance by using a time distance graph according to different signal phase combinations between the reference intersection and other intersections;
step 6, determining the optimal public signal period of the trunk road and the optimal signal phase setting mode of each intersection, and enabling the ideal intersection position to be most matched with the actual intersection position; determining the absolute phase difference of each intersection according to the signal phase setting mode of each intersection, the position of the nearest ideal intersection and the green signal ratio of the coordinated direction release phase;
and 7, for a certain driving direction, respectively calculating the green ratio above and below the green light center time line according to the offset green ratio of each intersection, selecting the minimum green ratio above and below, and adding to obtain the width of the green wave band in the driving direction.
5. The big data based vehicle low carbon control and induction method according to claim 4, wherein the supersaturation signal coordination method step comprises:
step1, defining a queuing length influence coefficient according to different situations:
there are three cases:
(1) the road environment of the road section in the up and down directions has no special requirement on the queuing length;
(2) the overflow phenomenon is easy to occur on the downstream road section in a certain direction of the road section, and the queue length in the direction is properly increased, so that the downstream overflow control is facilitated;
(3) an entrance and an exit of an important place are arranged in a certain direction of the road section, and when the vehicles in the line in the direction exceed the entrance and the exit, the vehicles in the line need to stop and wait until the vehicles in the line dissipate to the entrance and the exit;
therefore, the uplink and downlink weighted total length delta L and the uplink and downlink total delay D of the road section are defined, and the following steps are included:
in the formula:
respectively representing the maximum queue length of uplink and downlink; alpha is alphau、αdRespectively representing the influence weight of the uplink and downlink queuing length, and the value rule is as follows: taking 1 under the condition (1), taking a value larger than 1 under the condition (2) and taking a value smaller than 1 under the condition (3); du、DdRespectively representing uplink and downlink delays;
step2, establishing a phase difference optimization flow by taking min delta L and minD as targets in sequence, and solving the phase difference in a layered mode:
step1, inputting the signal parameters and traffic flow parameters of the upstream and downstream intersections, and determining the influence weight alpha of the upstream and downstream queuing lengths according to the environmental characteristics of the road sectionsu、αd；
Step2 let the phase difference
Judging whether the initial period of the section reaches the attribution of the traffic flow, and sequentially establishing a queuing length model and a delay model;
wherein the content of the first and second substances,
is downstream intersection IbPhase 1 relative upstream intersection IaThe red light phase difference of the phase 1, b is an intersection number, and I represents an intersection;
step3 is traversed by using enumeration method with Step size of 1s
Meanwhile, calculating and storing delta L, D under the corresponding phase difference;
step4 takes min delta L as a first-layer optimization target, and searches a corresponding phase difference value range to obtain an effective solution space;
step5 takes minD as the optimization target of the second layer, and obtains the optimal phase difference from the optimized effective solution space of the previous layer.
6. The big data-based vehicle low-carbon control and induction method according to claim 1, wherein the traffic induction algorithm in step S500 comprises:
s510.MAP phase: the Map function takes the section travel time of the corresponding sub-network, intersection delay data and a calculation time range as input, and traverses the structure of the graph according to the hierarchy according to the breadth-first principle; the Map function calculates the time for reaching the next intersection according to the travel time of each time period and intersection delay data, and finally generates a key/value form intermediate value; the key of the intermediate value is the time of arriving at the intersection, and the values are the intersection ID, the precursor intersection ID and the time of arriving at the precursor intersection respectively;
s511.reduce stage: the HaLoop combines all values with the same key value together and transmits the values to the Reduce function, and the Reduce function processes the value set of the intersection with the same arrival time to generate a new bucket;
s512, iteration of HaLoop MapReduce, namely, taking the output of the Reduce stage as the input of the next Map stage, continuously starting the operation of a Map-Reduce task by the operation server, and completing the calculation of all intersections included in the path through multiple iterations; in the iterative process, a Master node machine of HaLoop is responsible for loop control in Job until the iterative computation is finished.
7. The utility model provides a vehicle low carbon control and induction system based on big data which characterized in that includes:
the data acquisition module is used for acquiring traffic data through various ways including different detection equipment, mobile equipment and networks;
the data processing module is used for providing an algorithm for accelerating the data processing speed based on the cloud computing technology according to the traffic data acquired by the data acquisition module and processing the data;
the traffic running state module is used for carrying out online data analysis according to the data processed by the data processing module and obtaining the traffic conditions of intersections and road sections in different periods through analysis;
the online data analysis comprises:
s310, medium-term traffic flow prediction is carried out based on the hourly combination model:
the first step is as follows: screening the data according to the Latt criterion;
the second step is that: carrying out correlation analysis on the screened data;
the third step: according to the flow distribution, time intervals are divided, and 24 hours a day are divided into 24 time intervals; researching the relation between the flow and the speed of each time interval, namely a Q-V model, and predicting the flow in the time interval by using an ARIMA model;
the fourth step: combining the prediction result with the result of the Q-V model according to a certain weight, and comparing the final calculation result with the actual flow according to three indexes of average absolute error, average absolute percentage error and mean square error;
s311, estimating traffic state information based on an extended Kalman filtering algorithm of expectation optimization:
wherein the desired optimization algorithm is an iterative solution algorithm argmax estimating model parameters alpha and unknown variables YY,αP(Y,α|X)：
The first step is as follows: solving Y, alpha | X; firstly, solving an accurate solution of Y | X, alpha, indicating the limitation of the accurate solution, realizing accurate reasoning by recursive Bayes estimation, expanding posterior probability density, and calculating in a recursive mode; then, an extensible Kalman filtering algorithm is used as a substitute approximate solution;
the second step is that: solving alpha | Y, X; first, the optimal model parameters α are determined, which can be implemented as a simple error minimization, assuming that all traffic conditions are known; obtaining an observation function h () and a pseudo observation from a basic diagram and a link queue model, wherein the calculation formula is X-H (y); then, the observation X and the pseudo-observation X are minimized+The difference between them, resulting in an iteration of a new model parameter α defined in the desired optimization algorithm, with the formula
Wherein alpha is+A new iteration of a is represented which is,
represents the squared euclidean norm, where α includes the traffic event adaptation;
the coordination optimization module is used for controlling the on-line traffic signals in real time according to a main road signal coordination method to realize main road signal coordination optimization;
and the route induction module is used for planning an optimal travel route through a traffic induction algorithm and finally inducing a driver to select the optimal route for travel.
8. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811534976.1A CN109544929B (en) | 2018-12-14 | 2018-12-14 | Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811534976.1A CN109544929B (en) | 2018-12-14 | 2018-12-14 | Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109544929A CN109544929A (en) | 2019-03-29 |
| CN109544929B true CN109544929B (en) | 2020-12-15 |
Family
ID=65856396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811534976.1A Active CN109544929B (en) | 2018-12-14 | 2018-12-14 | Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109544929B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110634287B (en) * | 2019-08-26 | 2021-08-17 | 上海电科智能系统股份有限公司 | Urban traffic state refined discrimination method based on edge calculation |
| CN110517492B (en) * | 2019-08-27 | 2020-12-01 | 中国科学院自动化研究所 | Traffic path recommendation method, system and device based on parallel ensemble learning |
| CN111179586A (en) * | 2019-10-24 | 2020-05-19 | 广州市高科通信技术股份有限公司 | Traffic guidance method, equipment and storage medium based on big data analysis |
| CN110969845B (en) * | 2019-11-19 | 2020-11-24 | 武汉理工大学 | Intelligent vehicle speed control method and system based on vehicle-road cooperation |
| CN113411758B (en) * | 2020-03-16 | 2022-08-09 | 成都鼎桥通信技术有限公司 | Method and device for determining group call key event of private network cluster terminal |
| CN111915887A (en) * | 2020-07-10 | 2020-11-10 | 广州运星科技有限公司 | Integration and processing system and method based on multi-source heterogeneous traffic data |
| CN111968383A (en) * | 2020-08-24 | 2020-11-20 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | Method for realizing speed guidance of intersection by using electronic identification information of automobile |
| CN112233445B (en) * | 2020-09-28 | 2022-03-01 | 上海思寒环保科技有限公司 | Intelligent roadblock avoiding method and system |
| CN113299088B (en) * | 2021-05-06 | 2022-03-11 | 东南大学 | Regional multi-directional green wave design and driving speed guiding method based on Internet of vehicles |
| CN114333335A (en) * | 2022-03-15 | 2022-04-12 | 成都交大大数据科技有限公司 | Lane-level traffic state estimation method, device and system based on track data |
| CN116721545B (en) * | 2023-06-28 | 2024-05-28 | 东南大学 | Network-connected time-varying path control method based on traffic carbon emission reduction |
| CN117455087B (en) * | 2023-10-25 | 2024-04-12 | 南京迅集科技有限公司 | Logistics energy-saving control method and system based on Internet of things |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509470A (en) * | 2011-10-14 | 2012-06-20 | 北京掌城科技有限公司 | System and method for realizing energy conservation and emission reduction of vehicle based on dynamic path planning |
| CN104157151A (en) * | 2014-06-13 | 2014-11-19 | 东南大学 | Method and system for cooperation of city traffic guidance and signal control |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101472366A (en) * | 2007-12-26 | 2009-07-01 | 奥城同立科技开发（北京）有限公司 | Traffic light control system suitable for intersection control |
| CN101383091B (en) * | 2008-10-14 | 2011-11-09 | 北京交通大学 | Network control system for traffic induction outdoor screen |
| CN101571997A (en) * | 2009-05-31 | 2009-11-04 | 上海宝康电子控制工程有限公司 | Method and device for fusion processing of multi-source traffic information |
| CN103559797B (en) * | 2013-10-21 | 2016-04-13 | 华南理工大学 | A kind of Universal phase position type artery two-way green wave coordination control |
| CN105206053A (en) * | 2015-09-21 | 2015-12-30 | 河海大学常州校区 | Road comprehensive information processing system and method based on technology of Internet of vehicles |
| CN105702029B (en) * | 2016-02-22 | 2017-11-28 | 北京航空航天大学 | A kind of Expressway Traffic trend prediction method for considering space-time relationship at times |
| CN105865476B (en) * | 2016-06-21 | 2018-08-28 | 北京航空航天大学 | A kind of automobile most low-carbon method for searching path under dynamic traffic environment |
| CN107742420A (en) * | 2017-09-22 | 2018-02-27 | 北京交通大学 | It is a kind of to be used for the method that road traffic flow is predicted under emergent traffic incident |
| CN108269401B (en) * | 2018-01-30 | 2021-02-23 | 银江股份有限公司 | Data-driven viaduct traffic jam prediction method |
-
2018
- 2018-12-14 CN CN201811534976.1A patent/CN109544929B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509470A (en) * | 2011-10-14 | 2012-06-20 | 北京掌城科技有限公司 | System and method for realizing energy conservation and emission reduction of vehicle based on dynamic path planning |
| CN104157151A (en) * | 2014-06-13 | 2014-11-19 | 东南大学 | Method and system for cooperation of city traffic guidance and signal control |
Non-Patent Citations (1)
| Title |
|---|
| Ex-post environmental and traffic assessment of a speed reduction strategy in Madrid"s inner ring-road;FiammaPerez-Prada;《Journal of Transport Geography》;20170112;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109544929A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109544929B (en) | Vehicle low-carbon control and induction method, system, equipment and storage medium based on big data | |
| CN107103383B (en) | Dynamic taxi sharing scheduling method based on taxi-taking hotspot | |
| Yao et al. | Short‐term traffic speed prediction for an urban corridor | |
| CN108286980B (en) | Method for predicting destination and recommending driving route | |
| Kaddoura et al. | Towards welfare optimal operation of innovative mobility concepts: External cost pricing in a world of shared autonomous vehicles | |
| CN109949604B (en) | Large parking lot scheduling navigation method and system | |
| Miao et al. | Taxi dispatch with real-time sensing data in metropolitan areas: A receding horizon control approach | |
| Ma et al. | T-share: A large-scale dynamic taxi ridesharing service | |
| Xu et al. | Bus arrival time prediction with real-time and historic data | |
| CN107919014B (en) | Taxi running route optimization method for multiple passenger mileage | |
| CN104183119B (en) | Based on the anti-arithmetic for real-time traffic flow distribution forecasting method pushed away of section OD | |
| US20130144831A1 (en) | Predicting Taxi Utilization Information | |
| CN107832882A (en) | A kind of taxi based on markov decision process seeks objective policy recommendation method | |
| CN111402616B (en) | Intelligent parking control method, system, storage medium and terminal | |
| Jiang et al. | A multi-period analysis of taxi drivers' behaviors based on GPS trajectories | |
| US12018952B2 (en) | System and method for vehicle routing using big-data | |
| Haliem et al. | A distributed model-free ride-sharing algorithm with pricing using deep reinforcement learning | |
| CN110490365B (en) | Method for predicting network car booking order quantity based on multi-source data fusion | |
| CN114839984A (en) | Regular bus path planning method, device, equipment and storage medium | |
| Haliem et al. | AdaPool: A diurnal-adaptive fleet management framework using model-free deep reinforcement learning and change point detection | |
| Gao et al. | An efficient dynamic ridesharing algorithm | |
| Banerjee et al. | An intelligent hybrid scheme for optimizing parking space: A Tabu metaphor and rough set based approach | |
| CN110986992A (en) | Navigation method and device for unmanned vending vehicle, electronic equipment and storage medium | |
| Kluge et al. | On the computation of the energy-optimal route dependent on the traffic load in Ingolstadt | |
| Xia et al. | A distributed EMDN-GRU model on Spark for passenger waiting time forecasting |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |