CN103294823A - Rail transit multi-mode optimal transit transfer inquiring method based on cultural ant colony - Google Patents

Abstract

The invention relates to a rail transit multi-mode optimal transit transfer inquiring method based on a cultural ant colony. The method includes the following steps: (1) a central processing unit receives a query request through a touch screen, obtains website information from a database according to the query request and builds a route selecting model; (2) the central processing unit operates a cultural ant colony system on the basis of the route selecting model, calculates the optimal rail transit transfer scheme under different optimal objects, and outputs the best route; (3) values of the route selecting model are updated, whether optimization is finished is judged, if the answer is positive, a calculation result is fed back to the touch screen, and a step (4) is operate, and if the answer is negative, the step (2) is returned; (4) the touch screen displays the calculation result. Compared with the prior art, the method improves flexibility and efficiency of resident traveling for the rail transit transfer due to the fact that the culture ant colony system rapidly and precisely calculates the optimal rail transit transfer schemes under different optimal objects.

Description

The optimum transfer of track traffic multi-mode querying method based on cultural ant group system

Technical field

The present invention relates to the optimum transfer of a kind of track traffic computing method, especially relate to a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system.

Background technology

City Rail Transit System is to contact one of link the most closely with city dweller's daily life, even determining city dweller's life style to a certain extent, thereby, at present the electronic chart product in numerous cities is all realizing that track traffic network optimal path inquiry is as its most important thing, in the hope of making electronic chart can satisfy user's demand better, but existing inquiry system is not only made mistakes easily, and inefficiency, can not carry out the multi-mode transfer simultaneously:

On the one hand, most of software developers think that track traffic network optimum route analysis is the same with other network analysiss, also should be based on the shortest, but user's optimum is not only shortest path, so it also has very big gap from customer requirements;

On the other hand, most users think that minimum transfer is only key issue.Minimum transfer and shortest path seem unified, but actually this is not so, therefore how to accomplish both unifications, and proposing reality feasible optimization transfer model and algorithm has become the problem that presses for solution.

Summary of the invention

Purpose of the present invention is exactly the optimum transfer of the track traffic multi-mode querying method based on cultural ant group system that provides in order to overcome the defective that above-mentioned prior art exists that a kind of computing velocity is fast, precision is high, has improved dirigibility and the high efficiency of resident trip to orbit traffic transfer.

Purpose of the present invention can be achieved through the following technical solutions:

A kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system, this method may further comprise the steps:

1) central processing unit receives query requests by touch-screen, and obtains site information according to query requests from database, the build path preference pattern;

2) central processing unit is carried out cultural ant group on multiple populations system based on path Choice Model, calculates the optimal trajectory traffic transfer scheme that obtains under the different optimal objectives, the output optimal path;

Whether 3) numerical value to path Choice Model upgrades, and judge to optimize and finish, if then result of calculation is fed back to touch-screen, execution in step 4), if not, return step 2);

4) touch-screen shows result of calculation.

Described query requests comprises initial website and final website.

Described optimal objective comprises that the time is the shortest, transfer is minimum and distance is minimum.

Described cultural ant group system comprises the ant group evolutionary process of group space and the renewal of knowledge process in faith space, and the ant group evolutionary process of described group space may further comprise the steps:

A1) pheromones of initialization group space distributes, and group space is divided into a plurality of subgroups, and each subgroup adopts the ant group system of different behaviors to carry out parallel evolutionary respectively, obtains the locally optimal solution of each subgroup;

A2) between each subgroup according to based on the information interaction policy update local message element separately of study mechanism;

A3) locally optimal solution according to each subgroup upgrades globally optimal solution, and stores it into faith space by accepting function;

A4) carry out the plain renewal of global information according to the output in faith space;

A5) judge whether to satisfy the algorithm end condition, if satisfy, then algorithm stops; Otherwise, change step a2);

The renewal of knowledge process in described faith space may further comprise the steps:

B1) initialization faith space;

B2) receive the current globally optimal solution that group space provides by receiver function;

B3) the 2-OPT operation is implemented in the faith space, optimized the faith space;

B4) export optimum solution, and provide it to step a4 by influence function).

Described each subgroup is adopted the ant group system of different behaviors to carry out parallel evolutionary respectively and is specially:

A101) each subgroup places m ant of varying number on the website of n website randomly;

A102) state transitions is carried out according to behavior separately in each subgroup, selects next node, carries out simultaneously that local message is plain to be upgraded, and described behavior comprises at random, comforms, greedy or mix;

A103) repeating step a102), all form a fullpath until every ant, namely each subgroup travels through all nodes respectively, obtains locally optimal solution separately.

Described information interaction strategy based on study mechanism is:

Information interaction is carried out in each subgroup and neighbour other two subgroups, the locally optimal solution of current locally optimal solution and neighbour other two subgroups is compared, and upgrade the local message element of self with more excellent locally optimal solution.

The described function Accept () that accepts is:

Accept()＝T

The constant of T for setting.

It is described that operation is specially to faith space enforcement 2-OPT:

B301) r is set ₀Be a given constant in [0,1], produce the random number r of [0, a 1] scope, if r＞r ₀Then change step b4);

B302) if there is node c in the current optimal path _i, c _j, j 〉=i+2 wherein, and

d(c _i，c _i+1)+d(c _j，c _j+1)＞d(c _i，c _j)+d(c _i+1，c _j+1)

So with limit (c _i, c _j), (c _I+1, c _J+1) replacement (c _i, c _I+1), (c _j, c _J+1), the path (c in the circuit of exchange back _j..., c _I+1) be reversed; Otherwise change step b4).

Described influence function Influence () is:

$\mathrm{Influence}\left(\right)=\left\{\begin{array}{cc}\mathrm{BaseNum},& \mathrm{CurrentStep}\≤C\\ \mathrm{BaseNum}*\frac{\mathrm{EndStep}-\mathrm{CurrentStep}}{\mathrm{EndStep}-C},& \mathrm{otherwise}\end{array}\right.$

Wherein, EndStep is the maximum evolution algebraically of predefined ant group system, and CurrentStep is the ant group current algebraically that develops, and BaseNum and C are constant.

Compared with prior art, the present invention has the following advantages:

1, the present invention adopts cultural ant group system to carry out optimal path to find the solution, culture ant group system is a kind of new effectively optimizing method of ant group system being included in cultural algorithm frame, this computation model comprises based on the group space of ant group system with based on the faith space of current optimum solution, two spaces have colony and independent parallel evolution separately, have improved speed and precision that algorithm is found the solution;

2, the present invention adopts the ant group system of parallel evolutionary on multiple populations, and is undertaken having improved the precision of algorithm alternately by the information interaction strategy based on study mechanism between each subgroup;

3,2-OPT swap operation is at random adopted in the faith space of the present invention's culture ant group system, to the optimum solution optimization that makes a variation, solution individuality after developing is used for upgrading the group space global information is plain, the evolutionary process of group space is instructed in help, thereby reach the diversity that improves population, prevent precocity, reduce the purpose of calculation cost;

4, the inventive method has better degree of accuracy and robustness, even for extensive problem, and also can be to try to achieve less population number and short working time the less satisfactory solution of relative error.

In a word, beneficial effect of the present invention is: the present invention designs and has realized a kind of novel intelligence computation method, can carry out the efficient optimization design to different resident's multi-modes transfers, has improved dirigibility, the high efficiency of resident trip to orbit traffic transfer.The present invention has adapted to the tomorrow requirement of Traffic Development, sustainable scale and the growing orbit traffic transfer of complexity is carried out the management of science.

Description of drawings

Fig. 1 is schematic flow sheet of the present invention;

Fig. 2 is the system framework figure of the present invention cultural ant group on multiple populations system;

Fig. 3 is the framework synoptic diagram of the present invention's parallel evolutionary on multiple populations;

Fig. 4 is the principle schematic of the present invention's culture ant group system.

Embodiment

The present invention is described in detail below in conjunction with the drawings and specific embodiments.Present embodiment is that prerequisite is implemented with the technical solution of the present invention, provided detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

As shown in Figure 1, a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system, this method may further comprise the steps:

1) central processing unit receives query requests by touch-screen, comprises initial website and final website, and obtains site information from database, the build path preference pattern according to query requests;

2) central processing unit is carried out cultural ant group on multiple populations system based on path Choice Model, calculates to obtain different optimal objectives (comprise that the time is the shortest, transfer is minimum and distance minimum etc.) optimal trajectory traffic transfer scheme down, exports optimal path;

Whether 3) numerical value to path Choice Model upgrades, and judge to optimize and finish, if then result of calculation is fed back to touch-screen, execution in step 4), if not, return step 2);

4) touch-screen shows result of calculation, shows path Choice Model, represents content and comprises orbit traffic transfer information and wiring diagram.

As Fig. 2-shown in Figure 4, described cultural ant group system comprises the ant group evolutionary process of group space and the renewal of knowledge process in faith space, and the ant group evolutionary process of described group space may further comprise the steps:

A1) pheromones of initialization group space distributes, and group space is divided into a plurality of subgroups (as the subgroup 1 among Fig. 2, Fig. 3, subgroup 2, subgroup 3, subgroup 4), each subgroup places m ant of varying number on the website of n website randomly, carry out state transitions according to behavior separately, select next node, carry out simultaneously that local message is plain to be upgraded, described behavior comprises at random, comforms, greedy or mix; All form a fullpath until every ant, namely each subgroup travels through all nodes respectively, obtains locally optimal solution separately.

The concrete steps that each subgroup develops are:

A101) initialization: t=0, Nc=0, τ _Ij(t)=τ ₀, Δ τ _Ij(t)=0, m ant placed on n the website randomly;

A102) put taboo table index s=1, and its starting point website is added separately in the taboo table, judge whether the taboo table is full, if, execution in step a104 then), if not, s=s+1 then, execution in step a103),

A103) each ant is by its transition probability that calculates separately

Select next website, and this website added in the taboo table, carry out the plain renewal of local message simultaneously:

τ _ij＝(1-ρ)τ _ij+ρτ ₀

Wherein: ρ (0＜ρ＜1) is the part volatilization factor of pheromones; τ ₀It is the plain concentration value of initial information on each paths;

A104) every ant all forms a fullpath, namely travels through all nodes, calculates the length L of traveling round that all ants pass by _k, upgrade current optimum solution, obtain locally optimal solution.

A2) between each subgroup according to based on the information interaction policy update local message element separately of study mechanism.

As shown in Figure 3, each subgroup connects in the form of a ring, information interaction is carried out in each subgroup and neighbour other two subgroups, the locally optimal solution of current locally optimal solution and neighbour other two subgroups is compared, and upgrade the local message element of self with more excellent locally optimal solution.

A3) locally optimal solution according to each subgroup upgrades globally optimal solution, and stores it into faith space by accepting function;

The described function Accept () that accepts is:

Accept()＝T

The constant of T for setting can be made as 20;

A4) carry out the plain renewal of global information according to the output in faith space:

${\mathrm{\τ}}_{\mathrm{ij}}=(1-\∂){\mathrm{\τ}}_{\mathrm{ij}}+\∂\mathrm{\Δ}{\mathrm{\τ}}_{\mathrm{ij}}$

Wherein:

Be the overall situation volatilization factor of pheromones,

L _GbRepresent the path (from the length in on-test resulting global optimum path) of current globally optimal solution;

A5) judge whether to satisfy the algorithm end condition, if satisfy, then algorithm stops; Otherwise, empty all taboo tables, change step a103):

The algorithm end condition is that the maximum evolution algebraically or the globally optimal solution that reach setting do not change in setting algebraically continuously.

The ant group system that adopt in each space, subgroup is that the pheromones that a kind of local updating rule with pheromones and overall update rule carry out on the path is upgraded, thereby makes the search volume and the algorithm that enlarge algorithm be restrained the energy organic unity.As shown in Figure 3, A represents each subgroup by local cooperation and exchange optimum solution renewal local message element, and B represents that each subgroup offers globally optimal solution faith spatial update global information element alternately.

The renewal of knowledge process in described faith space may further comprise the steps:

B1) initialization faith space;

B2) receive the current globally optimal solution that group space provides by receiver function;

B3) the 2-OPT operation is implemented in the faith space, optimized the faith space;

It is described that operation is specially to faith space enforcement 2-OPT:

B301) r is set ₀Be a given constant in [0,1], produce the random number r of [0, a 1] scope, if r＞r ₀Then change step b4);

B302) if there is node c in the current optimal path _i, c _j, j 〉=i+2 wherein, and

d(c _i，c _i+1)+d(c _j，c _j+1)＞d(c _i，c _j)+d(c _i+1，c _j+1)

So with limit (c _i, c _j), (c _I+1, c _J+1) replacement (c _i, c _I+1), (c _j, c _J+1), the path (c in the circuit of exchange back _i..., c _I+1) be reversed; Otherwise change step b4).

B4) export optimum solution, and provide it to step a4 by influence function).

Described influence function Influence () is:

$\mathrm{Influence}\left(\right)=\left\{\begin{array}{cc}\mathrm{BaseNum},& \mathrm{CurrentStep}\≤C\\ \mathrm{BaseNum}*\frac{\mathrm{EndStep}-\mathrm{CurrentStep}}{\mathrm{EndStep}-C},& \mathrm{otherwise}\end{array}\right.$

Wherein, EndStep is the maximum evolution algebraically of predefined ant group system, and CurrentStep is the ant group current algebraically that develops, and BaseNum and C are constant, are set by the user.Usually the BaseNum value is 30, the C:EndStep value is 1: 3, like this in starting stage that the ant group develops, the knowledge solution in faith space is less to its influence, can enough guarantee rapid evolution, in the later stage that the ant group develops, the knowledge solution strengthens gradually to its influence, can reach the guiding of accepting knowledge space more, enlarge the search volume simultaneously, possess better ability of searching optimum.

The group space individuality forms individual experience during evolution, by function accept () individual experience is delivered to the faith space, and the faith space compares the individual experience of receiving and optimizes according to certain rule of conduct, form optimum solution.The optimum solution of faith space to finding in the evolutionary process, adopt 2-OPT swap operation at random, to the optimum solution optimization that makes a variation, and take full advantage of 2-OPT algorithm simple and high-efficient characteristics at random, and finish the variation of self, the solution individuality after developing is used for upgrading the group space global information is plain, the evolutionary process of group space is instructed in help, thereby reach the diversity that improves population, prevent precocity, reduce the purpose of calculation cost.The faith space after forming colony's experience by influence function to group space in individual rule of conduct make amendment so that individual space obtains higher efficiency of evolution.

Claims (9)

1. the optimum transfer of track traffic multi-mode querying method based on cultural ant group system is characterized in that this method may further comprise the steps:

1) central processing unit receives query requests by touch-screen, and obtains site information according to query requests from database, the build path preference pattern;

2) central processing unit is carried out cultural ant group on multiple populations system based on path Choice Model, calculates the optimal trajectory traffic transfer scheme that obtains under the different optimal objectives, the output optimal path;

Whether 3) numerical value to path Choice Model upgrades, and judge to optimize and finish, if then result of calculation is fed back to touch-screen, execution in step 4), if not, return step 2);

4) touch-screen shows result of calculation.

2. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 1 is characterized in that described query requests comprises initial website and final website.

3. the optimum transfer plan querying method of a kind of track traffic multi-mode based on cultural ant group system according to claim 1 is characterized in that described optimal objective comprises that the time is the shortest, transfer is minimum and distance is minimum.

4. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 1, it is characterized in that, described cultural ant group system comprises the ant group evolutionary process of group space and the renewal of knowledge process in faith space, and the ant group evolutionary process of described group space may further comprise the steps:

A1) pheromones of initialization group space distributes, and group space is divided into a plurality of subgroups, and each subgroup adopts the ant group system of different behaviors to carry out parallel evolutionary respectively, obtains the locally optimal solution of each subgroup;

A2) between each subgroup according to based on the information interaction policy update local message element separately of study mechanism;

A3) locally optimal solution according to each subgroup upgrades globally optimal solution, and stores it into faith space by accepting function;

A4) carry out the plain renewal of global information according to the output in faith space;

A5) judge whether to satisfy the algorithm end condition, if satisfy, then algorithm stops; Otherwise, change step a2);

The renewal of knowledge process in described faith space may further comprise the steps:

B1) initialization faith space;

B2) receive the current globally optimal solution that group space provides by receiver function;

B3) the 2-OPT operation is implemented in the faith space, optimized the faith space;

B4) export optimum solution, and provide it to step a4 by influence function).

5. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 4 is characterized in that described each subgroup is adopted the ant group system of different behaviors to carry out parallel evolutionary respectively and is specially:

A101) each subgroup places m ant of varying number on the website of n website randomly;

A102) state transitions is carried out according to behavior separately in each subgroup, selects next node, carries out simultaneously that local message is plain to be upgraded, and described behavior comprises at random, comforms, greedy or mix;

A103) repeating step a102), all form a fullpath until every ant, namely each subgroup travels through all nodes respectively, obtains locally optimal solution separately.

6. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 4 is characterized in that described information interaction strategy based on study mechanism is:

Information interaction is carried out in each subgroup and neighbour other two subgroups, the locally optimal solution of current locally optimal solution and neighbour other two subgroups is compared, and upgrade the local message element of self with more excellent locally optimal solution.

7. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 4 is characterized in that the described function Accept () that accepts is:

Accept()＝T

The constant of T for setting.

8. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 4 is characterized in that, describedly the 2-OPT operation is implemented in the faith space is specially:

B301) r is set ₀Be a given constant in [0,1], produce the random number r of [0, a 1] scope, if r＞r ₀Then change step b4);

B302) if there is node c in the current optimal path _i, c _j, j 〉=i+2 wherein, and

d(c _i，c _i+1)+d(c _j，c _j+1)＞d(c _i，c _j)+d(c _i+1，c _j+1)

So with limit (c _i, c _j), (c _I+1, c _J+1) replacement (c _i, c _I+1), (c _j, c _J+1), the path (c in the circuit of exchange back _j..., c _I+1) be reversed; Otherwise change step b4).

9. a kind of optimum transfer of track traffic multi-mode querying method based on cultural ant group system according to claim 4 is characterized in that described influence function Influence () is:

$\mathrm{Influence}\left(\right)=\left\{\begin{array}{cc}\mathrm{BaseNum},& \mathrm{CurrentStep}\≤C\\ \mathrm{BaseNum}*\frac{\mathrm{EndStep}-\mathrm{CurrentStep}}{\mathrm{EndStep}-C},& \mathrm{otherwise}\end{array}\right.$

Wherein, EndSep is the maximum evolution algebraically of predefined ant group system, and CurrentStep is the ant group current algebraically that develops, and BaseNum and C are constant.

Images