CN107103142B - Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network - Google Patents
Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network Download PDFInfo
- Publication number
- CN107103142B CN107103142B CN201710300803.2A CN201710300803A CN107103142B CN 107103142 B CN107103142 B CN 107103142B CN 201710300803 A CN201710300803 A CN 201710300803A CN 107103142 B CN107103142 B CN 107103142B
- Authority
- CN
- China
- Prior art keywords
- network
- prediction
- traffic
- road
- node
- 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
- 238000005516 engineering process Methods 0.000 title claims abstract description 15
- 238000004088 simulation Methods 0.000 claims abstract description 28
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 15
- 230000004941 influx Effects 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 9
- 238000012546 transfer Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000005096 rolling process Methods 0.000 claims description 6
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 108091092195 Intron Proteins 0.000 claims description 3
- 206010027476 Metastases Diseases 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000005265 energy consumption Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000009401 metastasis Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 230000037361 pathway Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000009472 formulation Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 description 11
- 239000002131 composite material Substances 0.000 description 11
- 230000008520 organization Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Geometry (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention discloses a kind of towards the comprehensive traffic network operation situation of highway and railway network deduction emulation technology, it is characterized in that, include: 1) regional complex transportation network emergency simulation framework, 2) needs estimate and prediction model based on trend prediction in conjunction with neural network, 3) the multi-mode comprehensive transport capability simulation model based on intelligent body；This is deduced emulation technology towards the comprehensive traffic network operation situation of highway and the railway network and utilizes existing network of highways, railway network traffic data, emphasis constructs the general frame of regional complex transportation network emergency emulation platform, with time-space network modeling technique, the multi-mode comprehensive traffic simulation model based on intelligent body is constructed.
Description
Technical field
It is specifically a kind of to be transported towards the comprehensive traffic network of highway and the railway network the invention belongs to intelligent transportation field
Row deducing manoeuver emulation technology.
Background technique
In regional response process of emergency system, how to formulate comprehensive traffic road network emergency Traffic Organization ensures friendship of meeting an urgent need
Lead to safe and efficient, orderly function, for improving region emergency managerial ability, reduces casualties and property caused by disaster
Loss, has important practical significance.For transportation network, especially composite communications transport network, it is difficult to pass through mathematics
It its operating status changing trend under event condition of modes quick obtaining such as calculates and different management and control measures or emergency is rung
Answer the visual effect of measure.Therefore, it establishes regional complex transportation network operation situation and deduces emulation, realize corresponding network of highways
Anxious management and control measures on composite communications transport network it is quick load with deduce emulate, be evaluation with optimization road network emergency traffic
The important means of organization scheme.
Existing traffic simulation platform is intended for single travel pattern more, and less research is more in composite communications transport network
How emulation platform is constructed under the conditions of existing for kind travel pattern is common.And existing analogue system, mostly single-threaded system,
Simulation velocity is partially slow, does not adapt to dock with the real-time of operation system.
Summary of the invention
Existing network of highways, railway network traffic data are utilized the object of the present invention is to provide a kind of, emphasis constructs region
The general frame of composite communications transport network emergency emulation platform is constructed with time-space network modeling technique based on intelligent body
Multi-mode comprehensive traffic simulation model towards the comprehensive traffic network operation situation of highway and the railway network deduce emulation technology.
In order to solve the above-mentioned technical problem, the technical scheme is that
It is a kind of to deduce emulation technology towards the comprehensive traffic network operation situation of highway and the railway network, comprising:
1) regional complex transportation network emergency simulation framework,
Including comprehensive travel needs estimate and prediction module, road-net node/section dynamic function Attribute tuning module, multimode
Formula dynamic traffic flow, traffic flow simulation, dynamic shortest path calculate, run assessment, output and interface display module, emulation system
System key data input include in real time with the OD demand data of history, road net traffic state operation data, incident data,
Network foundation data and road control and information distribution scheme；
2) needs estimate and prediction model based on trend prediction in conjunction with neural network,
Comprehensive travel needs estimate and prediction module are mainly by real-time and history OD demand data, road grid traffic shape
State operation data etc. carries out pattern analysis and OD trend is estimated, and influence of the emergency event to transport need is combined to carry out dynamic tune
It is whole, generate road network Dynamic OD requirement matrix；
3) the multi-mode comprehensive transport capability simulation model based on intelligent body,
Multi-mode dynamic traffic mainly solves the Traffic Assignment Problem between a variety of travel patterns with flow module, by institute
There is the calculating of the generation of path set, the selecting module for mode of going on a journey, by vehicles distribution corresponding with people, realizes intelligent body
Travel schedule includes the path of timestamp sequence and sequence node to each intelligent body stroke one.
Further, road-net node/section dynamic function attribute module is the basic data by network, in conjunction with dynamic
Incident data and road control data and information distribution scheme, generate emulation road-net node/section dynamic function attribute.
Further, multi-mode dynamic traffic passes through the choosing to all generations for having path set, mode of going on a journey with flow module
Vehicles distribution corresponding with people is realized the travel schedule of intelligent body, to each intelligent body stroke one by the calculating for selecting module
Path comprising timestamp sequence and sequence node.
Further, traffic flow simulation and dynamic shortest path module constitute the middle nucleus module for seeing traffic simulation, hand over
Through-flow analog module is using section transmission, node-node transmission model, and under the premise of meeting relevant constraint, realization exists to vehicle
The simulation shifted between section internal operation and section, section mode are related to the magnitude of traffic flow-density relations, and theory is current
Ability is mainly by road segment classification and lane quantity；Node metastasis model is related to turning left and Capacity Constraints of keeping straight on, dynamic point
Hourage of the traffic flow simulated with platform on section will feed back into the dynamic shortest path model of next iteration, in turn
Specially designated path is selected to provide decision-making foundation for intelligent body, meanwhile, new pathway can be fed back to next iteration.
Further, operation evaluation module assesses, path, region operation result from section, channel, OD, wraps
In terms of including efficiency, speed, reliability, energy consumption.
Further, the vehicle flowrate or trip predicted using polynomial trend and combine kalman filter method to Ingress node
The volume of the flow of passengers is estimated and is predicted respectively, the method is as follows:
J=road-net node serial number, j=1,2 ..., N；
τ=departure time interval serial number, τ=1,2 ...；
The stage of k=rolling forecast, k=1,2,3 ...；
The departure time intervening sequence number that l=each rolling stage includes；
D(j,τ)The demand influx of=node j in time interval τ；
Prior estimate of=node the j based on historical data in the demand influx of time interval τ；
μ(j,τ)=node j is in time interval τ D(j,τ)With prior estimateDemand disruption；
μ′(j,τ),μ″(j,τ),μ″′(j,τ)=demand disruption μ(j,τ)The single order of variable, second order, third order difference amount；
=D(j,τ)Estimate mean value；
=μ(p) (j,τ)Estimation mean value；
ZkThe state variable vector of=stage k；
YkThe measurement vector of=stage k；
Hk=measurement vector YkWith state vector ZkMatching relationship matrix；
wkThe processing noise of=stage k；
RkThe measurement error of=stage k；
=the Z based on the observation prediction by the end of stage k-1kThat is E (Zk|Y1,Y2,…,Yk-1)；
=the Z based on the observation estimation by the end of the k stagekThat is E (Zk|Y1,Y2,…,Yk)；
Pk,k-1The Z of=stage k-1 predictionkState covariance battle array, i.e.,
Pk,k=stage k, the Z of estimationkState covariance battle array, i.e.,
Polynomial trend prediction is introduced, the prediction of the difference vector of τ+ζ is carried out, constructs the function such as using m order polynomial
Under:
μ(j,τ+ζ)=b0+b1ζ+b2ζ2+…bpζp+…+bmζm；
According to Taylor expansion anticipation function μ(j,τ+ζ)It is extended to μ(j,τ)Multistage difference combination:
The coefficient of the formula of original polynomial distribution can be expressed as:
Three rank multinomial models are taken to carry out μ(j,τ+ζ)Prediction be shown below:
We are defined on forecast period k, τ=kl, and such as each stage forecast Period Length is that l includes 3 time intervals, that
In the k=3 stage, the sequence number of τ is 9；
State-transition matrix is defined as:
The predictor formula in k+1 stage is as follows:
Zk+1=AkZk；
Introduce Kalman filtering algorithm influences to eliminate charge station's influx measured value error bring, since this part is straight
It is identical vector value that measured value, which is connect, with estimated value, therefore H=I；
It is carried out first with the posterior estimate of previous stage, the prediction of current generation and the transmitting of covariance:
After introducing stage k measured value later, the calculating of gain is carried out:
The transmitting of the posterior estimate and covariance matrix of formation stages k:
Pk,k=(I-KkHk)Pk,k-1
Call the μ of above-mentioned formula forecast period k+1(j,τ+ζ), the node-flow for deduction can be generated based on the predicted value
Enter magnitude D(j,τ), and form multistage difference matrix needed for subsequent prediction.
Further, by BP neural network model can respectively in freeway net vehicle OD distribution carry out prediction and
The OD distribution for going out administrative staff in the railway network is predicted, illustrates algorithm by taking the OD forecast of distribution of vehicle in freeway net as an example
Process, algorithm flow are as follows:
Step 1: data decimation and standardization, for n charge tiny node in road network, before choosing predicted time section
The OD data of m days same periods are simultaneously standardized according to formula 4.1；
Step 2: building BP neural network enables i=1, for i-th of charge tiny node, chooses first day OD and is distributed ratio
Example be used as input vector, choose second day data be used as desired output vector, and so on composition m-1 group training sample；
Step 3: training neural network makes network reach the training objective of setting by repeatedly training；
The prediction of step 4:OD distribution proportion, the number for trained neural network, on the day before input prediction objective time interval
According to obtaining prediction distribution ratio, if going to step 5, otherwise go to step 2, adjust network parameter, re -training network；
Step 5: output prediction OD distribution proportion matrix, the output matrix if i=n, otherwise i+1 goes to step 2.
Further, it predicts to obtain road grid traffic respectively with BP neural network algorithm and polynomial trend prediction algorithm
After distribution proportion and each entrance magnitude of traffic flow, two kinds of prediction results are merged, under the conditions of obtaining no event, are predicted on network
The passenger of period or the OD matrix of vehicle, by vehicle or passenger by the flow q of node i to node jijWebsite i entrance can be passed through
Flow OiWith with the distribution proportion r that goes to j websiteijIt determines, under the conditions of no event, road network predicts OD matrix by following formula
It acquires:
By above-mentioned formula respectively to the OD matrix OD of vehiclevWith the OD matrix OD of passengerpIt is calculated, to obtain pre-
Survey the OD matrix without vehicle and passenger under the conditions of event on the period.
Further, Travel Time Budget TTB (p), starting point o are givenp, the departure time τ of intelligent body pp, considering road
Under the constraint condition of appearance of a street amount and Passenger Flow service ability, maximize the space-time accessibility in system scope, based on intelligence
The objective function of the optimization problem of the composite communications transport network dynamic distribution model of body, is shown below:
That is, the optimization aim of model is to minimize the unreachable generalized cost of user.If intelligent body p can be pre- in the given time
It is arrived at the destination in calculation, then inaccessible generalized cost is 0, and otherwise, C (p) is used the overall travel time in path by intelligent body.
The technology of the present invention effect major embodiment is in the following areas: in regional response process of emergency system, how to formulate synthesis
Traffic network emergency Traffic Organization ensures emergency traffic safety, efficient, orderly function, for improving region contingency management energy
Power reduces casualties and property loss caused by disaster, has important practical significance, special for transportation network
It is not composite communications transport network, is difficult its operating status under event condition by modes quick obtainings such as mathematics calculating and becomes
Change situation and therefore the visual effect to different management and control measures or emergency response measure establishes regional complex transportation network
Network emulation platform realizes that the quick load on composite communications transport network to network of highways emergency management and control measures emulates, is to comment
The important means of valence and optimization road network emergency Traffic Organization.
Detailed description of the invention
Fig. 1 is regional complex transportation network emergency simulation framework figure of the invention；
Fig. 2 is comprehensive travel needs estimate and prediction architecture diagram of the invention；
Fig. 3 is the trip distribution modeling algorithm flow chart of the invention based on BP neural network；
Fig. 4 is the algorithm flow chart under the finite path collection of the invention based on column-generation；
Fig. 5 is input and output design drawing of the invention.
Specific embodiment
It is a kind of to deduce emulation technology towards the comprehensive traffic network operation situation of highway and the railway network, including
1) regional complex transportation network emergency simulation framework
Regional complex transportation network emergency emulation platform mainly includes comprehensive travel needs estimate and prediction module, road
Net node/section dynamic function Attribute tuning module, multi-mode dynamic traffic flow, traffic flow simulation, dynamic shortest path meter
It calculates, run several modules such as assessment, and is pre- with the transportation network security risk based on G I S in such a way that interface calls
Alert platform and Decision System of Emergency are connected to each other, architecture design such as Fig. 1 institute of regional complex transportation network emergency emulation platform
Show.
The key data input of analogue system includes running number with the OD demand data of history, road net traffic state in real time
According to incident data, network foundation data and road control data and information distribution scheme.
Comprehensive travel needs estimate and prediction module are mainly by real-time and history OD demand data, road grid traffic shape
State operation data etc. carries out pattern analysis and OD trend is estimated, and influence of the emergency event to transport need is combined to carry out dynamic tune
It is whole, generate road network Dynamic OD requirement matrix.
Road-net node/section dynamic function attribute module is mainly the basic data by network, in conjunction with dynamic burst
Event data and road control data and information distribution scheme, generate emulation road-net node/section dynamic function attribute.
Multi-mode dynamic traffic mainly solves the Traffic Assignment Problem between a variety of travel patterns with flow module, by institute
There is the calculating of the generation of path set, the selecting module for mode of going on a journey, by vehicles distribution corresponding with people, realizes intelligent body
Travel schedule includes the path of timestamp sequence and sequence node to each intelligent body stroke one.
Traffic flow simulation and dynamic shortest path module constitute the middle nucleus module for seeing traffic simulation, traffic flow simulation mould
Block is using section transmission, node-node transmission model, and under the premise of meeting relevant constraint, vehicle is transported in realization inside section
The simulation shifted between capable and section.Section mode is related to the magnitude of traffic flow-density relations, basic capacity mainly by
Road segment classification and lane quantity；Node metastasis model is related to turning left and Capacity Constraints of keeping straight on.Dynamically distribute platform institute mould
Hourage of the quasi- traffic flow on section will feed back into the dynamic shortest path model of next iteration, and then select for intelligent body
It selects specially designated path and decision-making foundation is provided.Meanwhile new pathway can be fed back to next iteration.
Operation evaluation module can comment multiple dimensions such as, path, regions operation result from section, channel, OD
Estimate, including many aspects such as efficiency, speed, reliability, energy consumption.
The input and output of regional complex transportation network emergency emulation platform enter shown in Fig. 5.
2) needs estimate and prediction model based on trend prediction in conjunction with neural network；
Needs estimate and prediction architecture diagram such as Fig. 2.
1. entrance needs estimate
The vehicle flowrate or passenger traffic volume point predicted using polynomial trend and combine kalman filter method to Ingress node
Is not estimated and predicted.
J=road-net node serial number, j=1,2 ..., N；
τ=departure time interval serial number, τ=1,2 ...；
The stage of k=rolling forecast, k=1,2,3 ...；
The departure time intervening sequence number that l=each rolling stage includes；
D(j,τ)The demand influx of=node j in time interval τ；
Prior estimate of=node the j based on historical data in the demand influx of time interval τ；
μ(j,τ)=node j is in time interval τ D(j,τ)With prior estimateDemand disruption；
μ′(j,τ),μ″(j,τ),μ″′(j,τ)=demand disruption μ(j,τ)The single order of variable, second order, third order difference amount；
=D(j,τ)Estimate mean value；
=μ(p) (j,τ)Estimation mean value；
ZkThe state variable vector of=stage k；
YkThe measurement vector of=stage k；
Hk=measurement vector YkWith state vector ZkMatching relationship matrix；
The processing noise of wk=stage k；
RkThe measurement error of=stage k；
=the Z based on the observation prediction by the end of stage k-1kThat is E (Zk|Y1,Y2,…,Yk-1)；
=the Z based on the observation estimation by the end of the k stagekThat is E (Zk|Y1,Y2,…,Yk)；
Pk,k-1The Z of=stage k-1 predictionkState covariance battle array, i.e.,
Pk,k=stage k, the Z of estimationkState covariance battle array, i.e.,
Polynomial trend prediction is introduced, the prediction of the difference vector of τ+ζ is carried out, constructs the function such as using m order polynomial
Under:
μ(j,τ+ζ)=b0+b1ζ+b2ζ2+…bpζp+…+bmζm；
According to Taylor expansion anticipation function μ(j,τ+ζ)It is extended to μ(j,τ)Multistage difference combination:
The coefficient of the formula of original polynomial distribution can be expressed as:
Three rank multinomial models are taken to carry out μ(j,τ+ζ)Prediction be shown below:
We are defined on forecast period k, τ=kl, and such as each stage forecast Period Length is that l includes 3 time intervals, that
In the k=3 stage, the sequence number of τ is 9；
State-transition matrix is defined as:
The predictor formula in k+1 stage is as follows:
Zk+1=AkZk；
Introduce Kalman filtering algorithm influences to eliminate charge station's influx measured value error bring, since this part is straight
It is identical vector value that measured value, which is connect, with estimated value, therefore H=I；
It is carried out first with the posterior estimate of previous stage, the prediction of current generation and the transmitting of covariance:
After introducing stage k measured value later, the calculating of gain is carried out:
The transmitting of the posterior estimate and covariance matrix of formation stages k:
Pk,k=(I-KkHk)Pk,k-1
Call the μ of above-mentioned formula forecast period k+1(j,τ+ζ), the node-flow for deduction can be generated based on the predicted value
Enter magnitude D(j,τ), and form multistage difference matrix needed for subsequent prediction.
2. network OD distribution proportion is predicted
By BP neural network model can respectively in freeway net vehicle OD distribution carry out prediction and to the railway network
In go out the OD distribution of administrative staff and predict that pre- flow gauge is similar, as shown in figure 3, therefore, only with vehicle in freeway net
Illustrate that algorithm flow, algorithm flow are as follows for OD forecast of distribution.
Step 1: data decimation and standardization, for n charge tiny node in road network, before choosing predicted time section
The OD data of m days same periods are simultaneously standardized according to formula 4.1；
Step 2: building BP neural network enables i=1, for i-th of charge tiny node, chooses first day OD and is distributed ratio
Example be used as input vector, choose second day data be used as desired output vector, and so on composition m-1 group training sample；
Step 3: training neural network makes network reach the training objective of setting by repeatedly training；
The prediction of step 4:OD distribution proportion, the number for trained neural network, on the day before input prediction objective time interval
According to obtaining prediction distribution ratio rij(j=1,2,3 ..., n), ifAnd rij>=0, step 5 is gone to, step is otherwise gone to
Rapid 2, adjust network parameter, re -training network；
Step 5: output prediction OD distribution proportion matrix, the output matrix if i=n, otherwise i+1 goes to step 2.
3. comprehensive travel demand OD Matrix prediction
It predicts to obtain road grid traffic distribution proportion respectively with BP neural network algorithm and polynomial trend prediction algorithm
After each entrance magnitude of traffic flow, need to merge two kinds of prediction results, under the conditions of obtaining no event, prediction period on network
Passenger or vehicle OD matrix.By vehicle or passenger by the flow q of node i to node jijThe stream of website i entrance can be passed through
Measure OiWith with the distribution proportion r that goes to j websiteijIt determines, then under the conditions of no event, road network predicts that OD matrix can be by as follows
Formula acquires:
OD matrix OD that can respectively to vehicle by above-mentioned formulavWith the OD matrix OD of passengerpIt is calculated, to obtain
The OD matrix of vehicle and passenger under the conditions of predicted time Duan Shangwu event.
3) the multi-mode comprehensive transport capability simulation model based on intelligent body
In composite communications transport network, traveler is arrived at the destination by different trip modes, such as railway, self-driving and group
Conjunction mode (such as highway, railway transfer).The network of this multi-mode can be modeled as the multilayer that can carry out Used in Dynamic Traffic Assignment
Secondary network, wherein constraint condition is the service of different level or the constraint of road passage capability, the crucial ginseng in modeling process
Number, variable symbol and meaning are as shown in the table.
Given Travel Time Budget TTB (p), starting point op, the departure time τ of intelligent body pp, considering some road capacities
Under the constraint condition of Passenger Flow service ability, maximize the space-time accessibility in system scope, i.e., inaccessible degree
Reach minimum.The objective function of the optimization problem of composite communications transport network dynamic distribution model based on intelligent body, it is as follows
Shown in formula.
That is, the optimization aim of model is to minimize the unreachable generalized cost of user.If intelligent body p can be pre- in the given time
It is arrived at the destination in calculation, then inaccessible generalized cost is 0, and otherwise, C (p) is used the overall travel time in path by intelligent body.
That is, objective function is intended to make the quantity of destination unreachable in all travelers minimum.
There are multiple constraint conditions, i.e. space-time traffic flow Constraints of Equilibrium, traveler trip constraint, traffic flow and space in model
Capacity-constrained and the constraint of railway transport of passengers service ability etc..
1. space-time traffic flow equilibrium constraint
For all p ∈ P
Each only one starting point of intelligent body p and a terminal are indicated, on spatio-temporal state vertex in addition to the start and the end points only
It is constrained in accordance with flow equilibrium, i.e., for intelligent body p, each spatio-temporal state vertex only once reaches and once leaves.
The constraint condition 2. intelligent body is gone on a journey
All p ∈ P
Intelligent body p, which is arrived at the destination, perhaps to be gone on a journey by spatio-temporal activity arc or is gone on a journey by virtual spacetime arc of motion.
3. traffic flow and path space capacity constraints
Traffic flow constraint condition, i.e., the constraint condition of overall influx r are described by seeing traffic flow model in spatial queue
Are as follows:
To arbitrary (i, j) ∈ Ep, t ∈ T wherein, Ai,j(t) it indicates on section (i, j)
Accumulative amount of reach, Di,j(t) indicate that the accumulative amount of setting out, calculation formula are respectively as follows:
For section (i, j),Indicate the free flow travel time,Indicate road section length,Indicate that section is stifled
Density under plug-like state.
4) Passenger Flow service ability constrains
To all (i, j, t, s) ∈ AT
Passenger Flow service constraints: train can only travel on specific space-time arc according to the operation figure of formulation, in this feelings
Under condition, the intelligent body quantity of service is no more than vehicle bearing capacity.
5) rail-road transfer (arc) stand service ability constrain
To all (i, j, t, s) ∈ AT
The constraint of rail-road transfer stop service ability: different grades of junction transfer ability is limited, by hinge parking facility,
The facilities services ability such as facility, safety devices of waiting determines that this part also includes between from road forking node to junction
Section related capacity consistency.
The derivation algorithm of model is the algorithm based on column-generation, and the method based on column-generation is as shown in Figure 4.
Outer circulation is generated and in internal circulant solution certainly needed for the problem under finite path collection based on col-generating arithmetic
The smallest user cost path of time correlation.In each outer circulation iteration m, the cost minimization routing algorithm of time correlation
Applied to the path for being every O-D to the marginal user cost's cost minimization for finding time correlation with each time interval left,
New route (if any) is added in the subset Pm in present feasible path in m iteration, and one based on two-dimensional projection
Gradient descent direction method be used to solve the problems, such as to define on Pm under finite path collection, if do not find new path or
Reach preset convergence standard, algorithm terminates and exports the path flow of the time correlation obtained in current iteration.
The resulting gradient descent direction method based on two-dimensional projection of iteration is formed based on following in col-generating arithmetic frame
Ring, in each interior loop iteration n, with the path flow rn of DNLE (d) model evaluation update, corresponding system user totle drilling cost
TE (rn) and user cost, if in the difference in two continuous iteration (i.e. TE (rn)-TE (rn-1)) between target value
Value is less than preset threshold or reaches preset convergence (such as n=Nmax), interior loop termination, and algorithm returns to outer circulation.
The technology of the present invention effect major embodiment is in the following areas: in regional response process of emergency system, how to formulate synthesis
Traffic network emergency Traffic Organization ensures emergency traffic safety, efficient, orderly function, for improving region contingency management energy
Power reduces casualties and property loss caused by disaster, has important practical significance, special for transportation network
It is not composite communications transport network, is difficult its operating status under event condition by modes quick obtainings such as mathematics calculating and becomes
Change situation and therefore the visual effect to different management and control measures or emergency response measure establishes regional complex transportation network
Network emulation platform realizes that the quick load on composite communications transport network to network of highways emergency management and control measures emulates, is to comment
The important means of valence and optimization road network emergency Traffic Organization.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
The change or replacement expected without creative work, should be covered by the protection scope of the present invention.
Claims (5)
1. a kind of deduce emulation technology towards the comprehensive traffic network operation situation of highway and the railway network characterized by comprising
1) regional complex transportation network emergency simulation framework,
It is dynamic including comprehensive travel needs estimate and prediction module, road-net node/section dynamic function Attribute tuning module, multi-mode
State traffic flow distribution, traffic flow simulation, dynamic shortest path calculate, run assessment, output and interface display module, analogue system
Key data input include in real time with the OD demand data of history, road net traffic state operation data, incident data, network
Basic data and road control and information distribution scheme；
2) needs estimate and prediction model based on trend prediction in conjunction with neural network,
Comprehensive travel needs estimate and prediction module are mainly by real-time and history OD demand data, road net traffic state fortune
Row data carry out pattern analysis and OD trend is estimated, and influence of the emergency event to transport need is combined to carry out dynamic adjustment, raw
At road network Dynamic OD requirement matrix；
3) the multi-mode comprehensive transport capability simulation model based on intelligent body,
Multi-mode dynamic traffic mainly solves the Traffic Assignment Problem between a variety of travel patterns with flow module, by having to all
Vehicles distribution corresponding with people is realized the stroke of intelligent body by the generation of path set, the calculating of the selecting module of trip mode
Scheme, each intelligent body stroke are made of a path comprising timestamp sequence and sequence node；Using polynomial trend
It predicts and combines kalman filter method that the vehicle flowrate or passenger traffic volume of Ingress node are estimated and predicted respectively, method is such as
Under:
J=road-net node serial number, j=1,2 ..., N；
τ=departure time interval serial number, τ=1,2 ...；
The stage of k=rolling forecast, k=1,2,3 ...；
The departure time intervening sequence number that l=each rolling stage includes；
D(j,τ)The demand influx of=node j in time interval τ；
μ(j,τ)=node j is in time interval τ D(j,τ)With prior estimateDemand disruption；
μ′(j,τ),μ″(j,τ),μ″′(j,τ)=demand disruption μ(j,τ)The single order of variable, second order, third order difference amount；
Estimation mean value；
Estimation mean value；
ZkThe state variable vector of=stage k；
ZK jThe state variable vector of=road-net node j in stage k
YkThe measurement vector of=stage k；
Hk=measurement vector YkWith state vector ZkMatching relationship matrix；
wkThe processing noise of=stage k；
RkThe measurement error of=stage k；
I.e.
That is E (Zk|Y1,Y2,…,Yk)；
Pk,k-1The Z of=stage k-1 predictionkState covariance battle array, i.e.,
Pk,k=stage k, the Z of estimationkState covariance battle array, i.e.,
Polynomial trend prediction is introduced, the prediction of the difference vector of τ+ζ is carried out, constructs μ using m order polynomial(j, τ+ζ)Function is such as
Under:
μ(j,τ+ζ)=b0+b1ζ+b2ζ2+…bpζp+…+bmζm；
According to Taylor expansion anticipation function μ(j,τ+ζ)It is extended to μ(j,τ)Multistage difference combination:
The coefficient of the formula of original polynomial distribution can be expressed as:
Three rank multinomial models are taken to carry out μ(j,τ+ζ)Prediction be shown below:
We are defined on forecast period k, τ=kl, include 3 time intervals if each stage forecast Period Length is l, then
K=3 stage, the sequence number of τ are 9；
State-transition matrix is defined as:
The predictor formula in k+1 stage is as follows:
Zk+1=AkZk；
Introduce Kalman filtering algorithm influences to eliminate charge station's influx measured value error bring, since this part is directly surveyed
Magnitude is identical vector value with estimated value, therefore H=I；
It is carried out first with the posterior estimate of previous stage, the prediction of current generation and the transmitting of covariance:
After introducing stage k measured value later, the calculating of gain is carried out:
The transmitting of the posterior estimate and covariance matrix of formation stages k:
Pk,k=(I-KkHk)Pk,k-1
Call the μ of above-mentioned formula forecast period k+1(j,τ+ζ), it is based on predicted value μ(j, τ+ζ)The node inflow for deduction can be generated
Magnitude D(j,τ), and form multistage difference matrix needed for subsequent prediction；
By BP neural network model can respectively in freeway net vehicle OD distribution carry out prediction and in the railway network go out
The OD distribution of administrative staff predicts that the OD forecast of distribution of vehicle illustrates that algorithm flow, algorithm flow are as follows in freeway net:
Step 1: data decimation and standardization are chosen before predicted time section m days for n charge tiny node in road network
The OD data of same period are standardized；
Step 2: building BP neural network enables i=1, for i-th of charge tiny node, chooses first day OD distribution proportion and makees
For input vector, choose second day data and be used as desired output vector, and so on composition m-1 group training sample；
Step 3: training neural network makes network reach the training objective of setting by repeatedly training；
The prediction of step 4:OD distribution proportion, for trained neural network, data on the day before input prediction objective time interval are obtained
To prediction distribution ratio, rijJ=1.2.3 ... n, if ∑ rij=1, and rij≥0；Step 5 is gone to, step 2 is otherwise gone to, is adjusted
Whole network parameter, re -training network；
Step 5: output prediction OD distribution proportion matrix, the output matrix if i=n, otherwise i+1 goes to step 2；Using BP
Neural network algorithm and polynomial trend prediction algorithm are predicted to obtain road grid traffic distribution proportion and each entrance magnitude of traffic flow respectively
Afterwards, two kinds of prediction results are merged, under the conditions of obtaining no event, the OD square of the passenger of prediction period or vehicle on network
Battle array, by vehicle or passenger by the flow q of node i to node jijThe flow O of website i entrance can be passed throughiWith with go to j website
Distribution proportion rijIt determines, under the conditions of no event, road network prediction OD matrix is acquired by following formula:
By above-mentioned formula respectively to the OD matrix OD of vehiclevWith the OD matrix OD of passengerpIt is calculated, to obtain in prediction
Between vehicle and passenger under the conditions of Duan Shangwu event OD matrix；
Given Travel Time Budget TTB (p), starting point Op, the departure time τ of intelligent body pp, considering road capacity and railway passenger
Under the constraint condition for flowing service ability, maximize the space-time accessibility in system scope, the comprehensive traffic fortune based on intelligent body
The objective function of the optimization problem of defeated network dynamic distribution model, is shown below:
P is indicated
All intelligent body set；
That is, the optimization aim of model is to minimize the unreachable generalized cost of user, if intelligent body p can be in given time budget
It arrives at the destination, then inaccessible generalized cost is 0, and otherwise, C (p) is used the overall travel time in path by intelligent body；
There are multiple constraint conditions, i.e. space-time traffic flow Constraints of Equilibrium, traveler trip constraint, traffic flow and spatial content in model
Constraint and the constraint of railway transport of passengers service ability；
1. space-time traffic flow equilibrium constraint
It indicates each only one starting point of intelligent body p and a terminal, is abided by spatio-temporal state vertex in addition to the start and the end points only
Flow equilibrium constraint, i.e., for intelligent body p, each spatio-temporal state vertex only once reaches and once leaves；
The constraint condition 2. intelligent body is gone on a journey
All p ∈ P
Intelligent body p, which is arrived at the destination, perhaps to be gone on a journey by spatio-temporal activity arc or is gone on a journey by virtual spacetime arc of motion；
3. traffic flow and path space capacity constraints
Traffic flow constraint condition, i.e., the constraint condition of overall influx r are described by seeing traffic flow model in spatial queue are as follows:
Wherein, Ai,j(t) it indicates to add up amount of reach, D on section (i, j)i,j(t) the accumulative amount of setting out, calculation formula difference are indicated
Are as follows:
For section (i, j), sI, jIndicate free flow travel time, LI, jIndicate road section length,Indicate section blocked state
Under density；
4) Passenger Flow service ability constrains
To all (i, j, t, s) ∈ AT
Passenger Flow service constraints: train can only travel on specific space-time arc according to the operation figure of formulation, in the case,
Its intelligent body quantity serviced is no more than vehicle bearing capacity；
5) rail-road transfer or the constraint of arc station service ability
Above-mentioned symbol definition is as follows:
The constraint of rail-road transfer stop service ability: different grades of junction transfer ability is limited, by hinge parking facility, waits
Facility, safety devices facilities services ability determine that this part also includes the section between from road forking node to junction
Related capacity consistency.
2. emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network according to claim 1,
Be characterized in that: road-net node/section dynamic function attribute module is the basic data by network, in conjunction with dynamic emergency event
Data and road control data and information distribution scheme, generate emulation road-net node/section dynamic function attribute.
3. emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network according to claim 2,
Be characterized in that: multi-mode dynamic traffic passes through the selecting module to all generations for having path set, mode of going on a journey with flow module
It calculates, by vehicles distribution corresponding with people, realizes the travel schedule of intelligent body, include the time to each intelligent body stroke one
Stab the path of sequence and sequence node.
4. emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network according to claim 3,
Be characterized in that: traffic flow simulation and dynamic shortest path module constitute the middle nucleus module for seeing traffic simulation, traffic flow simulation
Module is using section transmission, node-node transmission model, under the premise of meeting relevant constraint, realizes to vehicle inside section
The simulation shifted between operation and section, section mode are related to the magnitude of traffic flow-density relations, and basic capacity is main
By road segment classification and lane quantity；Node metastasis model is related to turning left and Capacity Constraints of keeping straight on, and dynamically distributes platform institute
Hourage of the traffic flow of simulation on section will feed back into the dynamic shortest path model of next iteration, and then be intelligent body
Specially designated path is selected to provide decision-making foundation, meanwhile, new pathway can be fed back to next iteration.
5. emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network according to claim 4,
Be characterized in that: operation evaluation module assesses, path, region operation result from section, channel, OD, including efficiency, speed
In terms of degree, reliability, energy consumption.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710300803.2A CN107103142B (en) | 2017-07-11 | 2017-07-11 | Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710300803.2A CN107103142B (en) | 2017-07-11 | 2017-07-11 | Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107103142A CN107103142A (en) | 2017-08-29 |
| CN107103142B true CN107103142B (en) | 2019-12-03 |
Family
ID=59657979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710300803.2A Active CN107103142B (en) | 2017-07-11 | 2017-07-11 | Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107103142B (en) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133291B (en) * | 2017-12-22 | 2021-05-11 | 北京交通大学 | Correlation analysis method between social events and passenger traffic demands |
| CN108446759A (en) * | 2018-02-11 | 2018-08-24 | 陕西师范大学 | Tourist flows amount prediction technique based on seasonal shot and long term memory network model |
| CN108491979B (en) * | 2018-04-03 | 2021-11-30 | 中南大学 | Equal-intensity distribution-based time-varying demand prediction method for passenger traveling on high-speed railway |
| CN108520353B (en) * | 2018-04-09 | 2020-02-21 | 交通运输部公路科学研究所 | Method for identifying safety risk points of integrated transportation network |
| CN110555223A (en) * | 2018-06-04 | 2019-12-10 | 北京交通大学 | High-speed train system reliability assessment method based on seepage theory |
| CN109118412B (en) * | 2018-08-15 | 2021-08-03 | 北京交通大学 | Urban rail transit network passenger flow online control system |
| CN109785625B (en) * | 2019-02-18 | 2020-09-08 | 北京航空航天大学 | Method for identifying and evaluating red and green areas of urban road network operation state |
| CN110162931B (en) * | 2019-06-21 | 2021-06-22 | 北京交通大学 | Large-scale road network rapid simulation system for urban rail transit |
| CN110490365B (en) * | 2019-07-12 | 2022-04-05 | 四川大学 | Method for predicting network car booking order quantity based on multi-source data fusion |
| CN110599760B (en) * | 2019-10-17 | 2021-08-10 | 东南大学 | Travel behavior simulation method under multi-mode traffic network |
| CN113129582B (en) * | 2019-12-31 | 2023-05-09 | 阿里巴巴集团控股有限公司 | Traffic state prediction method and device |
| CN112101610B (en) * | 2020-07-29 | 2023-10-17 | 北京交通大学 | Multimode network design method for maximizing reachable flows based on reference dependency theory |
| CN111951553B (en) * | 2020-08-17 | 2022-11-11 | 上海电科智能系统股份有限公司 | Prediction method based on traffic big data platform and mesoscopic simulation model |
| CN112201033B (en) * | 2020-09-08 | 2022-06-07 | 同济大学 | Road traffic operation active control strategy selection method |
| CN112598177B (en) * | 2020-12-22 | 2023-10-24 | 北京市地铁运营有限公司 | Urban rail transit emergency online passenger flow prediction and simulation system |
| CN112633740B (en) * | 2020-12-30 | 2024-04-26 | 北京城市轨道交通咨询有限公司 | Urban rail transit network safety evaluation method |
| CN113065215A (en) * | 2021-03-22 | 2021-07-02 | 广州弘毅软件科技有限公司 | Method and device for predicting operation situation of expressway |
| CN113191528B (en) * | 2021-04-02 | 2022-02-22 | 东南大学 | Traffic state deduction method integrating macroscopic traffic simulation and short-term traffic prediction |
| CN113393044A (en) * | 2021-06-22 | 2021-09-14 | 中远海运科技(北京)有限公司 | Logistics resource emergency optimization system based on big data integration |
| CN116341882B (en) * | 2023-05-29 | 2023-08-01 | 中铁第四勘察设计院集团有限公司 | High-speed rail transfer passenger traffic analysis method and system based on generalized utility model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593535A (en) * | 2013-11-22 | 2014-02-19 | 南京洛普股份有限公司 | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration |
| CN103793760A (en) * | 2014-01-24 | 2014-05-14 | 同济大学 | Multi-mode optimization method for comprehensive transportation junction interior passenger flow transfer distribution ratio |
| CN104217250A (en) * | 2014-08-07 | 2014-12-17 | 北京市交通信息中心 | Rail transit new line opening passenger flow prediction method based on historical data |
| CN105787858A (en) * | 2016-03-29 | 2016-07-20 | 交通运输部公路科学研究所 | Situation deduction method for expressway network |
-
2017
- 2017-07-11 CN CN201710300803.2A patent/CN107103142B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593535A (en) * | 2013-11-22 | 2014-02-19 | 南京洛普股份有限公司 | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration |
| CN103793760A (en) * | 2014-01-24 | 2014-05-14 | 同济大学 | Multi-mode optimization method for comprehensive transportation junction interior passenger flow transfer distribution ratio |
| CN104217250A (en) * | 2014-08-07 | 2014-12-17 | 北京市交通信息中心 | Rail transit new line opening passenger flow prediction method based on historical data |
| CN105787858A (en) * | 2016-03-29 | 2016-07-20 | 交通运输部公路科学研究所 | Situation deduction method for expressway network |
Non-Patent Citations (2)
| Title |
|---|
| "组合出行模式下城市交通流分配模型与算法";孟梦;《中国博士学位论文全文数据库 工程科技Ⅱ辑》;20140515;论文全文 * |
| "考虑路径约束的城市多模式交通流分配模型与算法研究";赵微;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20160715;论文正文第3-6章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107103142A (en) | 2017-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107103142B (en) | Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network | |
| WO2021248607A1 (en) | Deep reinforcement learning-based taxi dispatching method and system | |
| Wang et al. | Passenger-demands-oriented train scheduling for an urban rail transit network | |
| Li et al. | Solving simultaneous route guidance and traffic signal optimization problem using space-phase-time hypernetwork | |
| CN103632212B (en) | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow | |
| Christofa et al. | Person-based traffic responsive signal control optimization | |
| CN102629418B (en) | Fuzzy kalman filtering-based traffic flow parameter prediction method | |
| Chen et al. | Uncertainty in urban mobility: Predicting waiting times for shared bicycles and parking lots | |
| CN102346964A (en) | Real-time jam prediction and intelligent management system for road traffic network area | |
| CN112380398B (en) | Airport passenger multi-traffic mode transfer travel chain construction device | |
| CN106485359A (en) | A kind of urban track traffic section passenger flow estimation method based on train schedule | |
| Carron et al. | Scalable model predictive control for autonomous mobility-on-demand systems | |
| Osorio et al. | A surrogate model for traffic optimization of congested networks: an analytic queueing network approach | |
| CN108364127A (en) | A kind of road network passenger flow Collaborative Control optimization system | |
| CN105787858A (en) | Situation deduction method for expressway network | |
| CN105160201A (en) | Genetic algorithm back propagation (GABP) neural network based controller workload prediction method and system | |
| CN115527369B (en) | Large passenger flow early warning and evacuation method under large-area delay condition of airport hub | |
| Chandra et al. | Cross-correlation analysis and multivariate prediction of spatial time series of freeway traffic speeds | |
| CN113723659A (en) | Urban rail transit full-scene passenger flow prediction method and system | |
| CN110246332A (en) | The real-time passenger flow method for monitoring of rail traffic and system based on multisource data fusion | |
| Chen et al. | A Deep Learning Model with Conv‐LSTM Networks for Subway Passenger Congestion Delay Prediction | |
| Kiani et al. | A novel intelligent traffic recovery model for emergency vehicles based on context-aware reinforcement learning | |
| CN113408189B (en) | Urban multipoint circulating emergency evacuation and simulation deduction method based on variable cells | |
| Xie et al. | An excess-demand dynamic traffic assignment approach for inferring origin-destination trip matrices | |
| CN113641749A (en) | Population mobility network estimation method and system based on bipartite graph |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhang Fan Inventor after: Zhao Li Inventor after: Cui Wei Inventor after: Zhang Jinjin Inventor after: Wang Hua Inventor after: Wang Tibin Inventor after: Niu Shuyun Inventor after: Zhang Jisheng Inventor after: Zhou Xuesong Inventor after: Li Honghai Inventor after: Sun Xiaoliang Inventor after: Liu Jianping Inventor after: Zhang Li Inventor after: Zhu Lili Inventor before: Zhang Fan Inventor before: Zhao Li Inventor before: Cui Wei Inventor before: Zhang Jinjin Inventor before: Wang Hua Inventor before: Wang Tibin Inventor before: Zhang Jisheng Inventor before: Niu Shuyun Inventor before: Zhou Xuesong Inventor before: Li Honghai Inventor before: Sun Xiaoliang Inventor before: Liu Jianping Inventor before: Zhang Li Inventor before: Zhu Lili |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |