CN101807222B - Station-based urban public traffic network optimized configuration method - Google Patents
Station-based urban public traffic network optimized configuration method Download PDFInfo
- Publication number
- CN101807222B CN101807222B CN2010101121819A CN201010112181A CN101807222B CN 101807222 B CN101807222 B CN 101807222B CN 2010101121819 A CN2010101121819 A CN 2010101121819A CN 201010112181 A CN201010112181 A CN 201010112181A CN 101807222 B CN101807222 B CN 101807222B
- Authority
- CN
- China
- Prior art keywords
- bus
- bus station
- network
- public
- value
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a station-based urban public traffic network optimized configuration method, which comprises the following steps of: optimized layout of bus stops under the constraint condition; calculation of trip generation of the bus stops; generation of candidate routes based on bus stop generation and route constraint condition; and genetic algorithm-based bus route optimization and index evaluation thereof. The station-based urban public traffic network optimized configuration method can search new routes based on the conventional routes, particularly realizes conventional public traffic network optimized layout in the presence of rapid transit. The station-based urban public traffic network optimized configuration method can directly evaluate the generation of the bus stops based on urban land utilization and population distribution, and also can directly utilize the generation and origin and destination data acquired by field transit survey. The system can be used for spatial distribution optimization of conventional metropolitan public traffic networks, and the efficiency for public traffic network regulation and layout is improved.
Description
1 technical field
The invention belongs to digital data processing field, particularly based on the urban public traffic network optimized configuration method of website.
2 background technologies
2.1 classical public bus network spatial configuration method
Classical public bus network planing method is present in the urban transportation systems organization model (Urban TransportationModeling System).This system serves as that the basis constitutes traffic analysis unit (TAZ) with the statistic unit of socio-economic activity, and the classical Four-stage Method of utilization is found the solution link flow.Four-stage is for trip generates, trip distributes, pattern is cut apart, assignment of traffic.It is generation and the traffic attraction that obtains TAZ that trip generates; Methods such as trip distribution utilization Gravity Models are calculated the travel amount (Origin-Destination between the TAZ; Be the OD amount); It is the ratio (comprising public transport) that various modes of transportation are used in prediction that pattern is cut apart, and assignment of traffic is that the OD amount with merotype is assigned on the road network.The planing method of stage is also among constantly improving, but the not too big change of its overall thinking.Recently during the last ten years, scholar and the engineers of being engaged in Urban Traffic Planning and management have proposed the non-polymerization analysis method based on activity, the direction of having represented traffic programme to analyze.But this is that the method for analytic target has very high requirement to social economy's individual data items, Computer Storage performance, computer computation ability with the microcosmic activity, extensive implement still difficult.
In public transport planning, mainly adopt the two-stage flow process of line configuring-site configuration, promptly lay public bus network respectively according to the amount of the OD on the road network, and layout line way station point.Circuit is laid one by one, and judgement is optimized perfect through the overall situation again.
Relevant document: [1] Meyer, M.D.and E.J.Miller (1984) .Urban Transportation Planning:ADecision-Oriented Approach.New York:McGraw-Hill Book Company; [2] Wang Wei, Yang Xin's seedling, Chen Xuewu (2002) urban public tranlport system planing method and administrative skill, Science Press.
2.2 data decomposition
If the object space unit is less than the source space unit, and do not exist space boundary to intersect between them, then slotting problem becomes face territory resolution problem in face the territory in.Utilize the area ratio method can find the solution this problem, but when object space unit too little the little graticule mesh of microvisual model (as be used for), and the soil that is used to retrain is when utilizing the unit big, solution procedure is comparatively complicated.The effective ways that address this problem are to adopt the Random assignment algorithm, and that practical is Monte Carlo simulation approach (Monte CarloSimulation).The Monte Carlo simulation method, decomposes the socioeconomic statistics data the little graticule mesh system from statistic unit as weight with the soil utilization.The weighted value of soil utilization is confirmed through on-site inspection and statistical study.As can utilize Monte Carlo simulation approach family and employment data to be decomposed on 30 meters the little graticule mesh, be used to simulate soil utilization and traffic activity, to propose the urban development policy.Similarly can add relevant environmental element in the microcosmic Simulation model, be used to estimate the influence of development policies urban environment and social activities.
Some metropolitan soils utilize more complicated, in the algorithm of Monte Carlo, can adopt a kind of weight of double constraints to confirm method.The soil of considering the city zones of different utilizes characteristic there are differences; The weight of similar soil utilization is distinguishing in different zones; Just there are different on the density like the residential estate of downtown area and the residential estate in peripheral district newly developed; It is more outstanding that land used that the more important thing is the downtown area mixes phenomenon, need make comprehensive judgement.Two bounding algorithms comprise soil utilization constraint and position constraint, in generalized information system, have realized computation process.Utilize this method, can metropolitan socio-economic activity statistics total amount be decomposed, obtain the population of microcosmic and the space distribution of employment.This micro-data distributes and is more suitable in the analysis of public transport activity than general large space unit.
Relevant document: [1] Wegener; M. (2001). " New spatial planning models. " International Journalof Applied Earth Observation and Geoinformation.3 (3): 224-237. [2] Huang; Z.; H.F.L.Ottens; Et al. (2007). " A doubly weighted approach to urban data disaggregation in GIS-Acase study of Wuhan, China. " Transactions in GIS 1l (2): 197-211.
2.3 bus passenger number system
Automatically passenger's counting (Automatic passenger counter APC) is the effective ways of collecting the passenger getting on/off when and where automatically, in conjunction with vehicle locate automatically, the wireless messages transmission techniques, can transmit real-time passenger flow information; Through data management system and GIS-Geographic Information System, can obtain runing required various, data information widely through data statistics and spatial analysis.The comprehensive counting that main equipment types has the board-like automatic passenger's counting of pressure, the automatic passenger's counting of infrared type and several different methods to merge.
From the eighties in last century, the public transportation department of developed country begins to utilize this system.With respect to manual research, this system can the continual data information that obtains a large amount of high accuracy in real time, combines with statistical analysis software to produce the required various forms of public transit system operation management automatically.It is the necessary component of intelligent public transportation system, need coordinate jointly to use with other system, and cost and operating environment are had relatively high expectations.
Through the get on the bus passenger flow and the passenger flow of getting off of each bus station of APC device acquisition, their are the generating capacity and the traffic attraction of corresponding bus station respectively.These data are the Customer information of actual acquisition, can replace the bus station generating capacity and the traffic attraction that obtain based on the accessibility model.
2.4 random walk algorithm
Random walk algorithm is based under the simulation resident trip environment, calculates the number of times that road candidate point and part of path are walked under given trip step-length restriction.Because the space structure characteristic of network, the number of times that different candidate bus stations or circuit are walked under going on a journey with equiprobability has difference.Through Traveling simulator repeatedly, can obtain the statistical nature of number of times that each node passes through (that is topological potential value).This topology potential value index is to differentiate the important evidence of node status in road network.
The calculating of algorithm is based on following hypothesis: the resident from the circuit that starting point is breasted the tape and walked, does not exist the loop line or the phenomenon of retracing one's steps in trip once; The length of random walk preestablishes, and surpasses this length and then reselects starting point.
Since considered the path of process, the node topology potential value that random walk algorithm obtained is than the architectural feature that more can reflect network merely based on the connective index that algorithm obtained of topology.
Relevant document: Li; Z., S.Zhao et al. (2009) .Position potential for analysing optimal locationsin urban road networks.11th International Conference on Computers in Urban Planning andUrban Management (CUPUM2009) .Honkong.
2.5 the bus station is optimized
Position collection covering problem (LSCP) model resulted from for the 1970's, with the target that is minimised as of the facilities Allocation cost that satisfies certain service level, was used for emergent communal facility optimized Site Selection at first, and this method is used to confirm the position and the quantity of bus station subsequently.Along with the spatial data quality improves constantly, the GIS technology is used to make up the model solution environment, particularly combines accessibility tolerance and location model, can analyze the bus station effectively and cover redundancy issue, to produce more efficiently analysis conclusion.As in the public transit system in Brisbane ,Australia city is analyzed, the LSCP model is combined with GIS, contrasted the positioning efficiency of existing bus station and desirable bus station.In addition, the Reachability question of public bus network also can use traditional p-central value model to find the solution, and its core is to find the solution the minimum service scope of each bus station, removes redundant bus station.In addition,, can obtain the significance index of road network node, with this important evidence that is provided with as the bus station according to the correlation technique of graph theory.
The Flowmap of Holland Utrecht university development can be optimized layout to the locus of communal facility, and covering collection model and extended model that it adopts can be used for optimizing distribution of bus station.
Relevant document: [1] Murray; A.T. (2003). " A Coverage Model for Improving Public TransitSystem Accessibility and Expanding Access. " Annals of Operations Research 123 (1): 143-156. [2] Breukelman; J.; G.Brink; Et al. (2009) .Manual Flowmap 7.3, Faculty of GeographicalSciences, Utrecht University.http: //flowmap.geog.uu.nl/. [3] Geertman; S.; T.de Jong & C.Wessels (2003), Flowmap:A Support Tool for Strategic Network Analysis.In:Geertman, S.&J.Stillwell (Eds): Planning Support Systems in Practise.Berlin:Springer Verlag. ' 2.6 is based on the bus station trip prediction of accessibility
The accessibility model is the basis with the range attenuation theory.Its core concept does, along with the increase of distance, for the use of the space specific service facility trend that tapers off.For bus trip, this distance is meant the distance of being walked to the bus station by the passenger.Under the hypothesis of reasonable consumption, the increase of walking time means the rising of cost, will cause the passenger to select the plan of travel that substitutes, as changes and row line or mode.This phenomenon of more function representation is arranged at present, is representative with negative exponential function and Rochester function wherein.
Based on the prediction of the bus trip of accessibility can be regarded as to spatial data decompose, analyze, the process of polymerization again.Mainly comprise data decomposition, allocation of space, accessibility calculation, public transport demand calculating several stages.Data decomposition is the process that statistic unit transforms---the consensus data from bigger administrative unit, is assigned in the less space cell.The effective attraction scope of bus station for peripheral passenger represented in the coverage of bus station, and in service range, along with distance increases, passenger's trip proportion reduces.Use the GIS analysis tool,, try to achieve the distance of each space cell, and be stored in the raster data figure layer to nearest bus station according to the area dividing that Thiessen polygon is served the bus station.Calculate through function model, the content update of grid cell is the ratio of bus trip.The bus trip ratio of each space cell is multiplied each other with corresponding population total amount, can try to achieve the bus trip amount on the space cell.Public transport demand on all grid cells in the same coverage gathers, and is the bus trip amount on the bus station.
Relevant document: [1] Kimpel; T.J., K.J.Dueker, et al. (2007). " Using GIS to Measurethe Effect of Overlapping Service Areas on Passenger Boardings at Bus Stops. " URISA Journal 19 (1): 5-11. [2] Huang; Z.; X.Liu, et al. (2008) .Measuring transitaccessibility based on disaggregate data in GIS-the case of Wuhan, China.TheBuilt Environment and Its Dynamics-Geoinformatics 2008 and Joint Conferenceon GIS and Built Environment.L.Liu; X.Li; K.Liu, X.Zhang and X.Wang.Guangzhou, China.SPIE.7144:2V1-8.
2.7 Combinatorial Optimization algorithm
Typical case's representative of combinatorial optimization problem heuritic approach is a genetic algorithm.Genetic algorithm is expressed as individuality (chromosome) with the feasible solution of optimization problem, population of several individual formations, and algorithm begins from initial population; Principle according to the survival of the fittest and the survival of the fittest; Produce the population of becoming better and better by the generation evolution,, select individual according to fitness size individual in the Problem Areas in each generation; And the screening mechanism of simulation biological heredity makes up and intersects and variation, produces the population of the new disaggregation of representative.
The typical application of genetic algorithm in public transport optimization relates to the problem aspect the public transport Optimization Dispatching, but some application are also arranged in public traffic network optimization.According to the characteristic of genetic algorithm, generally be to construct a candidate line collection earlier, from this sets of lines, pass through the result that genetic operator obtains Combinatorial Optimization again.To concrete application demand and evaluation emphasis, combination evaluation has different adaptation functional forms.Owing to have nothing in common with each other in optimization aim and data basis, utilize genetic algorithm to implement public traffic network optimization and having their own characteristics each aspect the details such as candidate line collection, adaptation function, genetic operator.Simultaneously, each is for the evaluation procedure that possibly add other between the circulation again, like network allocation.
List of references: [1] Chakroborty; P. (2003). " Genetic Algorithms for Optimal Urban TransitNetwork Design. " Computer Aided Civil and Infrastructure Engineering 18 (3): 184-200. [2] Bielli; M.; M.Caramia, et al. (2002). " Genetic algorithms in bus network optimization. " Transportation Research Part C:Emerging Technologies 10 (1): 19-34. [3] Poli, R.; W.B.Langdon; Et al. (2008) .A Field Guide to Genetic Programming, Lulu Enterprises, UK Ltd.
2.8 the problem that exists
[1] classical urban transportation model system is with a kind of pattern of public transport as trip, and total system is comparatively huge, and the needed data volume of the process of setting up is big, and computation process is complicated.
[2] existing line configuring method is applicable to that based on the method for " laying one by one " circuit of main passenger flow direction is laid, and the process that additional thereafter other circuit constitutes whole public traffic network still lacks effective method.
[3] configuration is implemented in existing public transport planning earlier, is that the location, bus station is carried out on the basis again with the circuit, because the constraint of bus station spacing, the choice of bus station is difficulty comparatively.Because the position, bus station relates to the microcosmic demand, the potential bus station that some demands are big is not owing to there is the circuit existence to be left in the basket.
3 summary of the invention
3.1 purpose
To the problems referred to above; The present invention proposes the public traffic network optimized configuration method of a cover based on website; This method can be optimized the bus station layout, obtains the public transport network of global optimization fast, particularly can improve the efficient of the newly-increased public bus network of configuration on existing public bus network basis.
3.2 technical scheme
Be optimized for the basis with the bus station, utilize locus and generating capacity thereof and the traffic attraction of bus station constraint condition as line configuring, and utilization genetic algorithm optimization public transport network.Concrete steps are following:
Step 2; Obtain the trip generating capacity and the traffic attraction of bus station: for existing bus station; Automatic passenger's counting assembly (APC) that utilization is loaded into bus obtains each bus station passengers quantity that gets on and off; The bus station data are through gathering and by after the time classification, extract the passengers quantity of getting on the bus peak period as the generating capacity of bus station, the traffic attraction of passengers quantity as the bus station of getting off; For newly-increased bus station, utilization is predicted the trip generating capacity and the traffic attraction of bus station based on the bus station Travel Demand Forecasting method of space accessibility model;
Step 3, find the solution candidate line: the first terminal of the public transport in the step 1 is matched in twos, and it is right to constitute the first terminal point of public transport; In basic road network, use general K-shortest path algorithm, in conjunction with public bus network constraint condition, seek the first terminal point of each public transport between K bar candidate public bus network, the value of K is between 3 and 20; Owing to there is constraint condition, the first terminal point of each public transport between actual candidate's public bus network bar that can obtain count N≤K;
Step 4; In candidate's public bus network that step 3 generates, utilize to keep the individual genetic algorithm screening public transport candidate line of high adaptive value, constitute and optimize public transport network: at the N bar of the first terminal of every pair of public transport (in the candidate line of N≤K); Select one arbitrarily; Constitute a cover initial line road network, this cover circuit is the body one by one of genetic algorithm, that is: indiv
i=R1, R2 ... Rn}, 40 individuals constitute the population of genetic algorithm; Utilize genetic algorithm to be directed against this population and carry out preferably, promptly competition selection, heredity, the mutation operation with genetic algorithm obtains a new generation's individuality, according to its adaptive value of adaptation function calculation; Optimum individual adaptive value in a new generation population should be greater than or equal to previous generation, if do not satisfy, then directly the optimum individual of previous generation is replaced the poorest individuality in this generation; Each is on behalf of a circulation, cycle calculations 600-800 generation, chooses the maximum individuality of adaptive value as final optimization sets of lines.
Random walk algorithm can adopt improved random walk algorithm in the above-mentioned steps 1; The improvement of random walk algorithm is, when when a node is judged the roadside, road of next step walking at random, is the walking probability that weight obtains each roadside, road with the width in roadside, road, and the roadside, road of broad will obtain bigger walking probability.The maximum that the maximum of space covering collection model covers with node topology potential value turns to target, and the node conduct of using known covering collection model preferably to satisfy condition increases the bus station newly.
In the above-mentioned steps 2, do, distribute, use in 500 meters scopes, calculate each bus station generating capacity of acquisition by the space accessibility model of range attenuation based on the micro-space of population based on the bus station Travel Demand Forecasting method of space accessibility model; Under the prerequisite that the generating capacity total amount remains unchanged, the micro-space that utilizes based on non-inhabitation soil distributes, and utilization space accessibility model calculates the traffic attraction that obtains each bus station.
In the above-mentioned steps 3; The constraint condition of candidate line comprises line length, non-linear coefficient, circuit benefit value and overall size, and wherein the line length value is the 6-24 km, and the maximal value of non-linear coefficient is got 1.5-2.5; The circuit benefit value is the generating capacity P of all bus stations on candidate's public bus network and the equilibrium value between the traffic attraction A; Be that 1-|P-A|/(P+A), this value minimum gets 0.2, overall size be all bus stations on candidate's public bus network generating capacity P and traffic attraction A's and divided by this Route Length L; I.e. (P+A)/L, this value minimum is got 1000 people/km.
In the above-mentioned steps 4; The adaptation function of the adaptation function genetic algorithm of genetic algorithm is the weighted sum of bus station coverage rate and the individual average benefit value of all circuits that comprises; Be E=a*Cov+b*W, wherein coverage rate Cov in bus station is the ratio that concentrated website number that is covered of optimization and master station count; Website coverage rate W is the average of all circuit benefit value, W=∑ Wi; The circuit benefit value is the generating capacity P of all websites on the circuit and the equilibrium value between the traffic attraction A, i.e. Wi=1-|Pi-Ai|/(Pi+Ai).Pi, Ai are respectively the summations that circuit i goes up all website P and A.
3.3 characteristics and technique effect
The higher public transport network global optimization collocation method based on website of [1] one cover efficient can once obtain whole public transport network allocation plan; The advantage of generalized information system management space and non-space data be can make full use of, population and land use data integrated; The optimization of circuit because of the present circumstance can be under the condition that the present situation circuit exists, search and the optimum newly-increased circuit of configuration.
[2] in the optimizing process of bus station, improved random walk algorithm and space are covered the collection model combine, improved the efficient of bus station configuration; In bus station generating capacity and the traffic attraction computation process, each website is outwards expansion simultaneously, and the population in each grid that each bus station attracted is removed from each grid total population, eliminating the influence of overlapping service, thereby obtains bus trip data more accurately; Not only utilized circuit in the differentiation of candidate's public bus network validity the characteristic of circuit itself (being line length and non-linear coefficient); Also utilized the characteristic (being circuit benefit value and overall size) of website that circuit passes through; These constraint conditions guarantee that every candidate's public bus network all is feasible; And not all first terminal point between candidate's public bus network is all arranged, those do not have candidate's public bus network website to will not appearing in the optimized choice process; Utilize the genetic algorithm to carry out in the public transport network optimizing process, kept the highest individuality of adaptive value that each generation produces, thus guarantee optimum scheme exist with last generation in.
The present invention can optimize the bus station layout, obtains the public transport network of global optimization fast, particularly can improve the efficient of the newly-increased public bus network of configuration on existing public bus network basis.
4 description of drawings
Fig. 1: the calculation process of bus station generating capacity and traffic attraction
Fig. 2: public transport network Optimization result example
Fig. 3: the improvement effect that is used for the public transport network Genetic Algorithms for Optimization; Wherein (a) adapts to the situation of change (800 generation) of functional value for optimum individual before improving; (b) for improving the situation of change (800 generation) that the back optimum individual adapts to functional value.
5 embodiments
5.1 the bus station optimizes distribution under the constraint condition
[1] with present situation bus station and intersection as the road segment segment between network node, the network node as the network limit, the formation base road network; Even for guaranteeing the node space distribution, need the roadside, road above 800 meters is cut apart;
[2] random walk algorithm of application enhancements obtains the topological potential value of each network node.Improvement is: consider road width, when selecting the roadside, road of next walking at random at the node place, the selected probability of broad road is higher; Because the resident trip distance is the change value, the length threshold of each trip is a stochastic distribution, but total amount is normal distribution;
[3] in basic road network, be the basis with the present situation bus station, the utilization space covers the searching of collection model and satisfies maximum other node that covers in space as new bus station.The maximum that the maximum of space covering collection model covers with node topology potential value turns to target, and according to node topology potential value, the covering collection pattern search of application Flowmap obtains newly-increased bus station (http://flowmap.geog.uu.nl/);
[4] distribute with reference to present situation bus station and the first terminal of public transport, confirm the first terminal of new public transport, concrete grammar has description in (Wang Wei, Yang Xin's seedling, Chen Xuewu, 2002 urban public tranlport system planing method and administrative skills, Science Press);
5.2 the obtaining and calculating of website bus trip growing amount
[5] for existing website; Automatic passenger's counting assembly (APC) that utilization is loaded into bus obtains each website passengers quantity that gets on and off; The bus station data are through gathering and by after the time classification; Extract the passengers quantity of getting on the bus peak period as the generating capacity of website, the traffic attraction of passengers quantity of getting off as website; The product (like the prosperous ball in the Hangzhou http://www.hzxqkj.cn/ of Science and Technology Ltd.) that automatic passenger's counting assembly (APC) can be bought for market
[6] for newly-increased website; Utilization is based on the bus station Travel Demand Forecasting method of space accessibility model; The bus trip generating capacity and the traffic attraction (Fig. 1 is the detailed calculated flow process) of prediction website: decompose according to soil utilization and statistic unit, obtain to utilize weight distribution based on the population distribution and the soil of grid;
[7] calculate each website total generating capacity, calculate total generating capacity of all websites: the micro-space based on population distributes; Use the space accessibility model (Kimpel that presses range attenuation in 500 meters scopes; T.J.; K.J.Dueker; Et al. (2007). " Using GIS to Measure the Effect of Overlapping Service Areas on Passenger Boardings atBus Stops. " URISA Journal 19 (1): 5-11.), calculate the generating capacity that obtains each bus station; Generate apart from template, accessibility probability template according to service range, begin from the center successively outwards to calculate the generating capacity of each grid from each bus station, and from each grid total population number, deduct this amount.
[8] the total generating capacity with all websites distributes as traffic attraction: the micro-space that utilizes based on the soil of non-inhabitation distributes, and calculates the attraction weight of each bus station, and from the grid weight map, deducts this attraction weight; Utilization space accessibility model calculates the traffic attraction that obtains each website.
5.3 the candidate line based on bus station generating capacity and circuit constraint condition generates
[9] in basic road network, use general K-shortest path algorithm, in conjunction with public bus network constraint condition, seek the first terminal point of each public transport between K bar candidate public bus network, the value of K is between 3 and 20; Owing to there is constraint condition, the first terminal point of each public transport between actual candidate's public bus network bar that can obtain count N≤K;
[10] constraint condition of candidate line comprises line length, non-linear coefficient, circuit benefit value, overall size; Wherein the line length value is the 6-24 km; The maximal value of non-linear coefficient is got 1.5-2.5, and the circuit benefit value is the generating capacity P of all bus stations on candidate's public bus network and the equilibrium value between the traffic attraction A, i.e. 1-|P-A|/(P+A); This value minimum gets 0.2; Overall size be all bus stations on candidate's public bus network generating capacity P and traffic attraction A's and divided by this Route Length L, i.e. (P+A)/L, this value minimum is got 1000 people/km.
[11] the important characteristic of another one is, if the first terminal point of public transport between existences need the existing public bus network that keeps, then this circuit is as permanent haulage line adding candidate public bus network collection, this public transport head terminal point between no longer search for other public bus networks.
[candidate line instance]
Certain big city is 1287 through the website number of optimizing, and first terminal is 28, and it is right then can to constitute 378 first terminal points.
Following condition is set: generate 8 candidate lines, every line length 6-24km, line nonlinear factor between every pair of first terminal and be 2, circuit generating capacity and traffic attraction equilibrium value are 0.2, circuit growing amount overall balance value is 1000.
Since some first terminal point between article one just do not satisfy these conditions; After these first terminals are filtered; Have 289 first terminal points to generating candidate line, wherein have 2 first terminal points between for present situation permanent haulage line (not other candidate line of regeneration).287 remaining first terminal point centerings, owing to being restricted by conditions, not all website between candidate line bar number can both reach 8.The total candidate line of the overall situation is 2013, when following table is listed K=8,
The pairing candidate line total number of N=1...K:
N | The candidate line |
1 | 289 |
2 | 279 |
3 | 269 |
4 | 257 |
5 | 247 |
6 | 235 |
7 | 227 |
8 | 210 |
Add up to | 2013 |
5.4 the public bus network based on genetic algorithm is preferred
[12] be the basis with the right N bar shortest path of all first terminals, optimize wherein a part as final line configuring.Preferred process adopts genetic algorithm, to the N bar candidate line of the first terminal of every pair of public transport, selects one arbitrarily, constitutes a cover initial line road network, and this cover circuit is the body one by one of genetic algorithm, and 40 individuals constitute the initial population of genetic algorithm.
Below be individual formation example:
[13] utilize competition selection, heredity, the mutation operation of genetic algorithm to obtain population of new generation; According to adapt to each individual fitness of function calculation (select, heredity, mutation operation algorithm principle is referring to Poli; R., W.B.Langdon, et al. (2008) .A Field Guide to Genetic Programming; Lulu Enterprises, UK Ltd.);
[14] the adaptation functional form of genetic algorithm is the weighted sum of website coverage rate and the average benefit value of all circuits, that is: E=a*Cov+b*W
Wherein: website coverage rate Cov is the ratio that concentrated website number that is covered of optimization and master station count; W is the average of all circuit benefit value, W=∑ Wi; The circuit benefit value is the generating capacity P of all websites on the circuit and the equilibrium value between the traffic attraction A, i.e. Wi=1-|Pi-Ai|/(Pi+Ai).Pi, Ai are respectively the summations that circuit i goes up all website P and A.
[15] the genetic algorithm process adopts the optimum individual retention strategy, and iteration all remains the optimum individual that a preceding iteration obtains each time, and population is continued to optimize, and the method can be accelerated convergence of algorithm speed.Optimum individual adaptive value in a new generation population should be greater than or equal to previous generation, if do not satisfy, then directly the optimum individual of previous generation is replaced the poorest individuality in this generation.Each is on behalf of a circulation, cycle calculations 600-800 generation, chooses the maximum individuality of adaptive value as final optimization sets of lines.
[genetic algorithm instance]
In 2013 candidate lines of last step generation, utilization heredity is calculated and is selected wherein 289, constitutes the optimization sets of lines.
The parameter that genetic algorithm is used is: population scale: 40; Crossover probability 0.8; Variation probability 0.3; Cycle index: 800 generations
Adapt to function:
E=a*Cov+b*W
W=∑Wi=∑(1-|Pi-Ai|/(Pi+Ai))
Wherein Cov is the website coverage rate, and W is the average benefit value of all circuits, the default a=0.8 that gets, b=0.2
Through the circulation in 800 generations, last optimum individual E=0.925555
Below be in the optimum individual in the 800th generation, effectively website between the N bar (N≤K, K=8) the selected quantity statistics of candidate line:
N | The quantity that the N bar is selected |
1 | 44 |
2 | 43 |
3 | 34 |
4 | 39 |
5 | 35 |
6 | 29 |
7 | 36 |
8 | 29 |
Add up to | 289 |
Fig. 2 is the diagrammatic representation of Optimization result.Black color dots is represented the first terminal of public transport, and dark line representes to optimize the distribution of public transport network, and light dotted line is represented not by the basic road of public bus network covering.
Fig. 3 is for to carry out improved effect to genetic algorithm, and last figure is the situation of change that optimum individual adapts to functional value before improving, and after the improvement of figure below for the reservation optimum individual, optimum individual adapts to the situation of change of functional value.Can find out that after improvement, cycle calculations is after 200 generations, the adaptation functional value of optimum individual no longer changes.
Claims (5)
1. urban public traffic network optimized configuration method based on website is characterized by:
Step 1 is optimized the bus station, confirms the first terminal of public transport: as network node, the road segment segment between the network node is as the network limit with present situation bus station and intersection, the formation base road network; Use the topological potential value that random walk algorithm obtains each network node; In basic road network, be the basis with the present situation bus station, the utilization space covers the searching of collection model and satisfies maximum other node that covers in space as new bus station; According to node topology potential value and locus, distribute in conjunction with the first terminal of present situation public transport, confirm the first terminal of new public transport;
Step 2; Obtain the trip generating capacity and the traffic attraction of bus station: for existing bus station; Automatic passenger's counting assembly that utilization is loaded into bus obtains each bus station passengers quantity that gets on and off; The bus station data are through gathering and by after the time classification, extract the passengers quantity of getting on the bus peak period as the generating capacity of bus station, the traffic attraction of passengers quantity as the bus station of getting off; For newly-increased bus station, utilization is predicted the trip generating capacity and the traffic attraction of bus station based on the bus station Travel Demand Forecasting method of space accessibility model;
Step 3, find the solution candidate line: the first terminal of the public transport in the step 1 is matched in twos, and it is right to constitute the first terminal point of public transport; In basic road network, use general K-shortest path algorithm, in conjunction with public bus network constraint condition, seek the first terminal point of each public transport between K bar candidate public bus network, the value of K is between 3 and 20; Owing to there is constraint condition, the first terminal point of each public transport between actual candidate's public bus network bar that can obtain count N≤K; Said public bus network constraint condition comprises line length, non-linear coefficient, circuit benefit value and overall size; Wherein the line length value is the 6-24 km; The maximal value of non-linear coefficient is got 1.5-2.5, and the circuit benefit value is the generating capacity P of all bus stations on candidate's public bus network and the equilibrium value between the traffic attraction A, i.e. 1-|P-A|/(P+A); This value minimum gets 0.2; Overall size be all bus stations on candidate's public bus network generating capacity P and traffic attraction A's and divided by this Route Length L, i.e. (P+A)/L, this value minimum is got 1000 people/km;
Step 4; In candidate's public bus network that step 3 generates, utilize to keep the individual genetic algorithm screening public transport candidate line of high adaptive value, constitute and optimize public transport network: in the N bar candidate line of the first terminal of every pair of public transport; Select one arbitrarily; Constitute a cover initial line road network, this cover circuit is the body one by one of genetic algorithm, and 40 individuals constitute the population of genetic algorithm; Utilize genetic algorithm to be directed against this population and carry out preferably, promptly competition selection, heredity, the mutation operation with genetic algorithm obtains a new generation's individuality, according to its adaptive value of adaptation function calculation; Optimum individual adaptive value in a new generation population should be greater than or equal to previous generation, if do not satisfy, then directly the optimum individual of previous generation is replaced the poorest individuality in this generation; Each is on behalf of a circulation, cycle calculations 600-800 generation, chooses the maximum individuality of adaptive value as final optimization sets of lines.
2. method according to claim 1; It is characterized in that: in the random walk algorithm that step 1 adopts; When a node is judged the roadside, road of next step walking at random; Width with the roadside, road is the walking probability that weight obtains each roadside, road, and the roadside, road of broad will obtain bigger walking probability.
3. method according to claim 1 and 2; It is characterized in that: in the step 1; The maximum that the maximum of space covering collection model covers with node topology potential value turns to target, and the node conduct of using known covering collection model preferably to satisfy condition increases the bus station newly.
4. method according to claim 1 and 2; It is characterized in that: in the step 2; Bus station Travel Demand Forecasting method based on space accessibility model does; Micro-space based on population distributes, and uses the space accessibility model of pressing range attenuation in 500 meters scopes, calculates to obtain each bus station generating capacity; Under the prerequisite that the generating capacity total amount remains unchanged, the micro-space that utilizes based on non-inhabitation soil distributes, and utilization space accessibility model calculates the traffic attraction that obtains each bus station.
5. method according to claim 1 and 2; It is characterized in that: in the step 4; The adaptation function of genetic algorithm is the weighted sum of bus station coverage rate and the individual average benefit value of all circuits that comprises; Wherein the bus station coverage rate is to optimize to concentrate the ratio that the bus station that is covered counts and total bus station counts, and the average benefit value of all circuits is the average of all public bus network benefit value in this individuality.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010101121819A CN101807222B (en) | 2010-02-09 | 2010-02-09 | Station-based urban public traffic network optimized configuration method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010101121819A CN101807222B (en) | 2010-02-09 | 2010-02-09 | Station-based urban public traffic network optimized configuration method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101807222A CN101807222A (en) | 2010-08-18 |
CN101807222B true CN101807222B (en) | 2012-07-04 |
Family
ID=42609012
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010101121819A Expired - Fee Related CN101807222B (en) | 2010-02-09 | 2010-02-09 | Station-based urban public traffic network optimized configuration method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101807222B (en) |
Families Citing this family (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101944288B (en) * | 2010-08-25 | 2012-02-22 | 东南大学 | Method for setting stop stations of urban bus line |
CN102737500B (en) * | 2012-06-05 | 2015-01-28 | 东南大学 | Method for acquiring arrival interval reliability of urban bus |
CN103471603B (en) | 2012-06-06 | 2015-10-21 | 腾讯科技(深圳)有限公司 | The matching process of a kind of public bus network and road network and device |
CN102831767B (en) * | 2012-08-13 | 2014-07-09 | 东南大学 | Multi-mode station stopping method at optimized comprehensive cost for urban public transport |
CN102881155A (en) * | 2012-09-29 | 2013-01-16 | 山东浪潮齐鲁软件产业股份有限公司 | Taxi intelligent terminal based stand hotspot zone analysis method |
CN103942948B (en) * | 2014-04-10 | 2016-03-16 | 中南大学 | Based on the generation method of the urban public bus lines network of sectionally smooth join |
CN105206037B (en) * | 2014-05-30 | 2017-10-20 | 中国电信股份有限公司 | Public bus network analysis method and system |
CN104036130A (en) * | 2014-06-06 | 2014-09-10 | 中国交通通信信息中心 | High-resolution remote-sensing image based public travel analysis method and device |
CN105468595A (en) * | 2014-08-12 | 2016-04-06 | 中国科学院深圳先进技术研究院 | Bus route planning method and system |
CN104134105B (en) * | 2014-08-18 | 2017-06-13 | 东南大学 | A kind of public transport network lays optimization method |
CN104318758B (en) * | 2014-11-05 | 2016-06-29 | 东南大学 | Based on multi-level multimodal Public transport network planning method |
CN104699906B (en) * | 2015-03-19 | 2017-08-04 | 北京大学 | A kind of spatial accessibility Measurement Method of the vehicle-mounted track data of combination and landform |
CN104933479B (en) * | 2015-06-10 | 2018-01-30 | 江苏省城市规划设计研究院 | A kind of city bus hinge based on accessibility is layouted planing method |
CN104966135B (en) * | 2015-06-16 | 2018-04-13 | 西南交通大学 | Public bus network network optimization method based on accessibility and up to intensity |
CN105069522B (en) * | 2015-07-24 | 2018-08-07 | 重庆大学 | Scenic spot Touring-line evaluation and improved method |
CN105118014A (en) * | 2015-09-18 | 2015-12-02 | 常州普适信息科技有限公司 | Method for evaluating grids of bus network |
CN107806882B (en) * | 2016-09-08 | 2020-05-08 | 高德信息技术有限公司 | Route planning method and device |
CN109690643B (en) * | 2016-09-13 | 2021-08-20 | 本田技研工业株式会社 | Probe data management device |
CN106779163B (en) * | 2016-11-18 | 2020-06-19 | 华南理工大学 | Customized public transport network optimization method based on intelligent search |
CN107194086B (en) * | 2017-05-26 | 2020-12-22 | 中国科学院遥感与数字地球研究所 | Indoor space configuration method and device |
CN107305590B (en) * | 2017-06-14 | 2020-11-10 | 北京市交通信息中心 | Urban traffic trip characteristic determination method based on mobile phone signaling data |
CN108961804B (en) * | 2018-06-20 | 2020-07-31 | 北京市交通运行监测调度中心 | Bus route adjustment alternative set determination method based on multi-index classification intersection |
CN111223288B (en) * | 2018-11-26 | 2023-02-03 | 厦门雅迅网络股份有限公司 | Line blockage avoiding method based on Internet of vehicles data and storage medium |
CN109753694B (en) * | 2018-12-13 | 2023-10-03 | 东南大学 | Method for designing medium and small city public transportation network based on whole-process travel sensing time |
CN109978267B (en) * | 2019-03-28 | 2023-01-03 | 东南大学 | Urban microcirculation bus route planning method based on urban rail transit data |
CN110889622A (en) * | 2019-11-22 | 2020-03-17 | 深圳市综合交通设计研究院有限公司 | Bus network layout determination method, device and equipment based on bus service level |
CN110675646B (en) * | 2019-12-04 | 2020-03-31 | 武汉元光科技有限公司 | Method and device for acquiring position of bus station |
CN112289066B (en) * | 2019-12-30 | 2021-09-21 | 南京行者易智能交通科技有限公司 | Bus driving plan approach station and arrival time setting method thereof |
CN111047228B (en) * | 2020-03-13 | 2020-08-04 | 武汉元光科技有限公司 | Bus station separating method and system |
CN111581759B (en) * | 2020-05-27 | 2022-08-02 | 北京大学 | Assessment method for bus network optimization based on Rich curvature |
CN112149919B (en) * | 2020-10-15 | 2024-01-16 | 武汉市公用电子工程有限责任公司 | Bus operation line evaluating method, device, equipment and storage medium |
CN112329989A (en) * | 2020-10-19 | 2021-02-05 | 北京中恒云科技有限公司 | Bus route planning method and device based on cloud computing and storage medium |
CN114822013B (en) * | 2021-01-19 | 2023-03-24 | 南京行者易智能交通科技有限公司 | Bus stop optimization method and device based on stop coverage rate and stop passenger flow big data |
CN112466122B (en) * | 2021-01-28 | 2021-06-15 | 深圳市城市交通规划设计研究中心股份有限公司 | Method and device for generating alternative line set and planning line of public traffic line network |
CN113218411B (en) * | 2021-03-31 | 2022-12-06 | 华录智达科技股份有限公司 | Square partition matrix transfer method suitable for multiple buses |
CN114399233B (en) * | 2022-03-25 | 2022-06-07 | 南京大学 | Traffic planning method and device based on deep learning and OD completion |
CN114999668B (en) * | 2022-04-13 | 2024-06-07 | 浙江大学 | Public transportation system epidemic situation transmission blocking method based on integer programming |
CN114819633B (en) * | 2022-04-25 | 2023-11-24 | 重庆市规划设计研究院 | Traffic facility layout evaluation method applied to homeland space planning |
CN115331470B (en) * | 2022-10-11 | 2023-01-17 | 山东恒宇电子有限公司 | Single-ticket bus passenger getting-off station calculation method based on IC and travel chain |
CN115809729B (en) * | 2022-11-28 | 2023-08-22 | 北京交通大学 | Urban rail transit junction transfer facility optimal configuration method considering newly-added line |
-
2010
- 2010-02-09 CN CN2010101121819A patent/CN101807222B/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
CN101807222A (en) | 2010-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101807222B (en) | Station-based urban public traffic network optimized configuration method | |
Zhao et al. | The impacts of rail transit on future urban land use development: A case study in Wuhan, China | |
Arribas et al. | Urban solid waste collection system using mathematical modelling and tools of geographic information systems | |
Samanta et al. | Modeling a rail transit alignment considering different objectives | |
Yamu et al. | Spatial accessibility to amenities, natural areas and urban green spaces: using a multiscale, multifractal simulation model for managing urban sprawl | |
CN107872808B (en) | WLAN station address prediction analysis method and device | |
CN103049829B (en) | Integrated fusion method of urban and rural passenger line network and hub station | |
CN104809112A (en) | Method for comprehensively evaluating urban public transportation development level based on multiple data | |
Mougiakou et al. | Urban green space network evaluation and planning: Optimizing accessibility based on connectivity and raster gis analysis | |
Chiarazzo et al. | The effects of environmental quality on residential choice location | |
CN112184282A (en) | Cinema site selection model establishing method, cinema site selection method and cinema site selection platform | |
CN104282142A (en) | Bus station arrangement method based on taxi GPS data | |
Jiang et al. | Healthy urban streams: The ecological continuity study of the Suzhou creek corridor in Shanghai | |
Wu et al. | Urban Traffic Planning and Traffic Flow Prediction based on ulchis gravity model and Dijkstra algorithm | |
CN117196197A (en) | Public transportation site layout optimization method | |
CN111008730B (en) | Crowd concentration prediction model construction method and device based on urban space structure | |
Karabakan et al. | Measuring the green infrastructure resilience in Turkey | |
CN113554221B (en) | Method for simulating and predicting town development boundary under view angle of' flow space | |
P Shastri et al. | Integrated land-use zoning, using topographical data: Optimizing vacant space for urbanization at Akole Taluka, Maharashtra, India | |
CN101115244A (en) | Propagation model distributing method with unit of cell | |
Mirzahossein et al. | Land-Use Modification Based on Transit-Oriented Development Adjacent to Historical Context (Case Study: Qazvin City) | |
Aulia et al. | Impact of transportation on urban compactness index in South Tangerang City, Indonesia | |
Li et al. | High-Resolution Flood Numerical Model and Dijkstra Algorithm Based Risk Avoidance Routes Planning | |
Shamsuddin et al. | Predicting and simulating future land use pattern: A case study of Seremban District | |
Surya et al. | Population mobility and sustainable development in the Mamminasata Metropolitan South Sulawesi, Indonesia |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120704 Termination date: 20130209 |