CN104732807A - Busy terminal area flow regulating method - Google Patents

Abstract

The invention provides a busy terminal area flow regulating method. The busy terminal area flow regulating method includes the steps that various flight take-off and landing ranking schemes are received, and the multiple kinds of the flight take-off and landing ranking schemes are arrayed in a scale-free network form; according to the array form of the multiple kinds of the flight take-off and landing ranking schemes in a scale-free network, the various flight take-off and landing ranking schemes are differentiated as flight take-off and landing center ranking schemes and flight take-off and landing non-center ranking schemes, and the flight take-off and landing center ranking schemes and the flight take-off and landing non-center ranking schemes are updated in different ways respectively; comparison is performed on the various flight take-off and landing ranking schemes updated before and after according to a flight take-off and landing ranking objective function, and a latest flight take-off and landing overall history optimal ranking scheme is determined; the take-off and landing of flights are controlled according to the latest flight take-off and landing overall history optimal ranking scheme.

Description

Busy termination environment flow control method

Technical field

The present invention relates to information control technology field, particularly relating in a kind of aviation field for regulating the busy termination environment flow control method of flight landing sequencing schemes.

Background technology

Along with the fast development of society globalization process, air transportation, as a kind of transportation trade, based on its convenience and high efficiency, plays a part more and more important in the traffic transport industry of modern society.Because air transportation process is easily subject to the interference of various extraneous factor, thus cause airliner delay or cancellation, cause being discontented with of client.In order to reduce the impact of extraneous factor on the flight landing time as far as possible, promote the satisfaction of client, need adjust the landing sequence of flight and control, and make the flight landing sequencing schemes after adjusting can on the basis ensureing aviation safety, reduce adjustment flight landing as far as possible and to sort the cost paid.

Based on above-mentioned purpose, the various intelligent optimization method for optimizing flight landing sequencing schemes is widely used.Intelligent optimization method is the optimization method come by simulating or disclose some spontaneous phenomenon or process development, and it does not rely on gradient information, has the overall situation, parallel, Optimal performance efficiently, the advantage of robustness and highly versatile.Particle group optimizing method (Particle Swarm Optimization, PSO) is the intelligent optimization method of a kind of mockingbird types of populations behavior, be by individual in population between cooperation and information sharing find optimum solution.The advantage of particle group optimizing method is simple easily realization and is not subject to too much parameter perturbation.When adopting the pattern of particle group optimizing to be optimized flight landing sequence control problem, various flight landing sequencing schemes is also constantly optimized according to the mode of learning of particle each in population, thus obtains flight landing Optimal scheduling scheme.

But in existing particle group optimizing method, the mode of learning of all particles and behavior are all same, such as all particles are all only to best neighbor learning, or all particles are all to all neighbor learnings.Therefore, due to scarcity or the redundancy of information, existing particle group optimizing method exists restrained slowly, or was easily absorbed in the defect of local optimum.When adopting existing particle group optimizing pattern to be optimized flight landing sequence control problem, the optimization efficiency of flight landing sequencing schemes is low, and be difficult to obtain satisfactory flight landing sorting consistence scheme, thus add adjustment flight landing and to sort the cost paid.

Summary of the invention

The embodiment of the present invention provides a kind of busy termination environment flow control method, in order to solve the problem that when busy termination environment flow control method in prior art regulates flight landing sequencing schemes, cost is higher.

The embodiment of the present invention provides a kind of busy termination environment flow control method, and described method comprises:

Receive multiple flight landing sequencing schemes, described multiple flight landing sequencing schemes is arranged with the form of scales-free network;

According to the spread pattern of described multiple flight landing sequencing schemes in described scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and upgrades described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing respectively in a different manner;

According to flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading are compared, determine up-to-date flight landing global history optimal sequencing scheme;

The landing of flight is controlled according to described up-to-date flight landing global history optimal sequencing scheme.

In one embodiment of this invention, also comprise:

Degree threshold value for distinguishing flight landing center sequencing schemes and the non-central sequencing schemes of flight landing is set; Wherein, the degree of flight landing sequencing schemes represents the quantity of flight landing sequencing schemes adjacent with described flight landing sequencing schemes in described scales-free network;

Described according to the spread pattern of described multiple flight landing sequencing schemes in described scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, comprising:

Flight landing sequencing schemes angle value being greater than described degree threshold value is defined as flight landing center sequencing schemes, and flight landing sequencing schemes angle value being less than or equal to described degree threshold value is defined as the non-central sequencing schemes of flight landing;

Described flight landing center sequencing schemes according to

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade flight landing sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the renewal amount of i-th kind of flight landing sequencing schemes, represent the sequencing schemes after the renewal of i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme of n-th neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, K _irepresent the size of the set of neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, represent between random quantity, for being greater than the random number of 0;

The non-central sequencing schemes of described flight landing according to

$\stackrel{\→}{v_i}=\mathrm{\χ}\·(\stackrel{\→}{v_i}+\stackrel{\→}{U}(0,c_1)\·(\stackrel{\→}{p_i}-\stackrel{\→}{x_i})+\stackrel{\→}{U}(0,c_2)\·(\stackrel{\→}{p_{\mathrm{mi}}}-\stackrel{\→}{x_i})),1\≤i\≤I$

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the optimized amount of i-th kind of flight landing sequencing schemes, represent sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, represent the history optimal case of described i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, c ₁represent the self-teaching factor, c ₂represent social learning's factor, represent [0, c ₁] between random quantity, represent [0, c ₂] between random quantity.

In one embodiment of this invention, also comprise:

According to sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, redefine the history optimal sequencing scheme of described i-th kind of flight landing sequencing schemes the history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of flight landing sequencing schemes and the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes

According to described redefine the renewal amount of i-th kind of flight landing center sequencing schemes, and again upgrade described flight landing center sequencing schemes according to described renewal amount;

According to described and redefine the renewal amount of the non-central sequencing schemes of described i kind flight landing, and again upgrade the non-central sequencing schemes of described flight landing according to described optimized amount.

In one embodiment of this invention, the history optimal case redefining described i-th kind of flight landing sequencing schemes described in comprises:

The functional value of sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes is determined by described flight landing sequence objective function, the functional value that the functional value of sequencing schemes after described renewal is corresponding with the history optimal case of described i-th kind of flight landing center sequencing schemes compares, and the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefining n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, compared by functional value corresponding with the history optimal case of described n-th neighbours' sequencing schemes for the functional value of sequencing schemes after the renewal of described n-th neighbours' sequencing schemes, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefined in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of all neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, select the functional value corresponding with the history optimal case in described all neighbours' sequencing schemes of the optimal function value in described functional value to compare, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme in all neighbours' sequencing schemes of described i-th kind of sequencing schemes

In one embodiment of this invention, describedly according to flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading to be compared, determine up-to-date flight landing sequence optimal case, comprising:

According to described flight landing sequence objective function, the various flight landing sequencing schemes before described renewal are compared, select sequencing schemes optimum in described flight landing sequencing schemes as flight landing global history optimal sequencing scheme;

According to described flight landing sequence objective function, the various flight landing sequencing schemes after described renewal are compared, select wherein optimum flight landing sequencing schemes and described flight landing global history optimal sequencing scheme to compare, using in the flight landing sequencing schemes of described optimum and described flight landing global history optimal sequencing scheme preferably sequencing schemes as up-to-date flight landing history optimal sequencing scheme.

In one embodiment of this invention, described flight landing sequence objective function is:

$Z_{\mathrm{total}}=\underset{d=1,...,P}{\mathrm{\Σ}}{Z}_d$

$Z_d=\left\{\begin{array}{cc}{g}_d(T_d-X_d)& \mathrm{if},X_d\≤T_d\\ h_d(X_d-T_d)& \mathrm{if},X_d>T_d\end{array}\right.$

E _d≤X _d≤L _d,d＝1,...,P

X _d-X _f≤S _dfd＝1,...,P,f＝1,...,P,d≠f

Wherein, P represents the total quantity of aircraft, d and f represents the numbering of aircraft, T _drepresent the time of aircraft d expection landing, Z _drepresent that aircraft d shifts to an earlier date or lags behind expection landing time T _dcost cost when landing, Z _totalrepresent total cost cost of the aircraft of all participation landing sorting consistence, X _drepresent the actual landing time of aircraft d, X _frepresent the actual landing time of aircraft f, E _drepresent the earliest time of aircraft d landing, L _drepresent the time the latest of aircraft d landing, g _dthe cost of unit interval when representing that aircraft d shifts to an earlier date landing; h _drepresent the cost of unit interval during aircraft d delayed landing, S _dfrepresent and consider safety factor, required time interval when aircraft d and aircraft f lands.

The busy termination environment flow control method that the embodiment of the present invention provides, adopt the mode of operation of particle group optimizing method, by the spread pattern of multiple flight landing sequencing schemes in scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and upgrade described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing respectively in a different manner, by flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading are compared again, namely flight landing optimal sequencing scheme can be determined.The present invention is directed to different flight landing sequencing schemes and upgrade described flight landing sequencing schemes by different way, satisfactory flight landing Optimal scheduling scheme can be found more fast, thus promote the security of air transportation, the cost paid and the landing of effective reduction adjustment flight is sorted.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, introduce doing one to the accompanying drawing used required in embodiment or description of the prior art simply below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

The process flow diagram of a kind of busy termination environment flow control method that Fig. 1 provides for the embodiment of the present invention one;

The process flow diagram of a kind of busy termination environment flow control method that Fig. 2 provides for the embodiment of the present invention two.

Embodiment

For making the object of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

The present invention busy termination environment flow control method adopts the particle group optimizing method (Selectively-informed Particle Swarm Optimization, SIPSO) introducing heterogeneous learning strategy to be optimized flight landing sequencing schemes.Adopt scales-free network topological structure in the particle group optimizing method of the heterogeneous learning strategy of described introducing, the degree of a particle refers to the population be connected with this particle in described scales-free network.In scales-free network topological structure, the degree distribution of the particle of a Stochastic choice meets power law.Different particle in population is made to have stronger heterogeneity according to the difference of the angle value of each particle in described scales-free network.In the present invention, namely a particle represents a kind of flight landing sequencing schemes, and described particle populations represents multiple flight landing sequencing schemes.Described particle group optimizing method is by the different learning performances of heterogeneity particle, particle is constantly updated to obtain optimal value, described multiple flight landing sequencing schemes follows this optimization characteristics of population, namely can constantly update various flight landing sequencing schemes, thus find optimum flight landing sequencing schemes.

The process flow diagram of a kind of busy termination environment flow control method that Fig. 1 provides for the embodiment of the present invention one.As shown in Figure 1, the busy termination environment flow control method of the present embodiment, can comprise the steps:

Step S101: receive multiple flight landing sequencing schemes, arranges described multiple flight landing sequencing schemes with the form of scales-free network.

Step S102: according to the spread pattern of described multiple flight landing sequencing schemes in described scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and upgrades described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing respectively in a different manner.

In step s 102, specifically comprise:

Degree threshold value for distinguishing flight landing center sequencing schemes and the non-central sequencing schemes of flight landing is set; Wherein, the degree of flight landing sequencing schemes represents the quantity of flight landing sequencing schemes adjacent with described flight landing sequencing schemes in described scales-free network; Described degree threshold value is decided by described flight landing sequence objective function, and in the present embodiment, setting flight landing sequencing schemes quantity is 50, and described degree threshold value is between 3-5.

Flight landing sequencing schemes angle value being greater than described degree threshold value is defined as flight landing center sequencing schemes, and flight landing sequencing schemes angle value being less than or equal to described degree threshold value is defined as the non-central sequencing schemes of flight landing.

Described flight landing center sequencing schemes adopts perfect information (Fully-informed) learning strategy, according to formula

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade flight landing sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the renewal amount of i-th kind of flight landing sequencing schemes, represent the sequencing schemes after the renewal of i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme of n-th neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, K _irepresent the size of the set of neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, represent between random quantity, in order to ensure randomness when flight landing sequencing schemes upgrades according to perfect information learning strategy, wherein for being greater than the random number of 0.

In perfect information learning strategy, the K of described i-th kind of flight landing sequencing schemes _ithe renewal amount of individual neighbours' flight landing sequencing schemes on described i-th kind of flight landing sequencing schemes all has impact, and namely flight landing center sequencing schemes obtains experience from the history optimal case of all neighbours' flight landing sequencing schemes.

The non-central sequencing schemes of described flight landing adopts single information (Single-informed) learning strategy, according to formula

$\stackrel{\→}{v_i}=\mathrm{\χ}\·(\stackrel{\→}{v_i}+\stackrel{\→}{U}(0,c_1)\·(\stackrel{\→}{p_i}-\stackrel{\→}{x_i})+\stackrel{\→}{U}(0,c_2)\·(\stackrel{\→}{p_{\mathrm{mi}}}-\stackrel{\→}{x_i})),1\≤i\≤I$

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the optimized amount of i-th kind of flight landing sequencing schemes, represent sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, represent the history optimal case of described i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, c ₁represent the self-teaching factor, c ₂represent social learning's factor, represent [0, c ₁] between random quantity, represent [0, c ₂] between random quantity, in order to ensure that flight landing sequencing schemes is according to randomness during single information learning policy update, c ₁represent the self-teaching factor, c ₂represent social learning's factor, preferably, the present embodiment arranges c ₁=c ₂=2.05, χ=0.7298.

In single information learning strategy, the Part I of bracket be memory section, represent flight landing sequencing schemes to the maintenance of former sequencing schemes, make it upgrade according to oneself state; Part II be autognosis part, represent various flight landing sequencing schemes to the thinking of be optimized by itself experience, guide various flight landing sequencing schemes to learn to the history optimal case of himself; Part III it is social recognition part, represent the cognition of flight landing sequencing schemes to best neighbours' sequencing schemes, also represent the information sharing between various flight landing sequencing schemes and interaction simultaneously, guide flight landing sequencing schemes to learn to the history optimal case in its all neighbours' flight sequencing schemes.

Step S103: according to flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading are compared, determine up-to-date flight landing global history optimal sequencing scheme;

Particularly, described flight landing sequence objective function is:

$Z_{\mathrm{total}}=\underset{d=1,...,P}{\mathrm{\Σ}}{Z}_d$

$Z_d=\left\{\begin{array}{cc}{g}_d(T_d-X_d)& \mathrm{if},X_d\≤T_d\\ h_d(X_d-T_d)& \mathrm{if},X_d>T_d\end{array}\right.$

E _d≤X _d≤L _d,d＝1,...,P

X _d-X _f≤S _dfd＝1,...,P,f＝1,...,P,d≠f

Wherein, P represents the total quantity of aircraft, d and f represents the numbering of aircraft, T _drepresent the time of aircraft d expection landing, Z _drepresent that aircraft d shifts to an earlier date or lags behind expection landing time T _dcost cost when landing, Z _totalrepresent total cost cost of the aircraft of all participation landing sorting consistence, X _drepresent the actual landing time of aircraft d, X _frepresent the actual landing time of aircraft f, E _drepresent the earliest time of aircraft d landing, L _drepresent the time the latest of aircraft d landing, g _dthe cost of unit interval when representing that aircraft d shifts to an earlier date landing; h _drepresent the cost of unit interval during aircraft d delayed landing, S _dfrepresent and consider safety factor, required time interval when aircraft d and aircraft f lands.

Step S104: the landing controlling flight according to described up-to-date flight landing global history optimal sequencing scheme.

In above-mentioned steps, executive agent can be the computer system with data storage and processing capacity.

The busy termination environment flow control method that the embodiment of the present invention provides, adopt the mode of operation of the particle group optimizing method introducing heterogeneous learning strategy, by the spread pattern of multiple flight landing sequencing schemes in scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and upgrade described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing respectively in a different manner, by flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading are compared again, namely flight landing optimal sequencing scheme can be determined.The present invention is directed to different flight landing sequencing schemes and upgrade described flight landing sequencing schemes by different way, satisfactory flight landing Optimal scheduling scheme can be found more fast, thus promote the security of air transportation, the cost paid and the landing of effective reduction adjustment flight is sorted.

The process flow diagram of a kind of busy termination environment flow control method that Fig. 2 provides for the embodiment of the present invention two.

As shown in Figure 2, further, step S102 also comprises:

According to sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, redefine the history optimal sequencing scheme of described i-th kind of flight landing sequencing schemes the history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of flight landing sequencing schemes and the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes

According to described redefine the renewal amount of i-th kind of flight landing center sequencing schemes, and again upgrade described flight landing center sequencing schemes according to described renewal amount;

According to described and redefine the renewal amount of the non-central sequencing schemes of described i kind flight landing, and again upgrade the non-central sequencing schemes of described flight landing according to described optimized amount.

Particularly, the history optimal case redefining described i-th kind of flight landing sequencing schemes described in comprises:

The functional value of sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes is determined by described flight landing sequence objective function, the functional value that the functional value of sequencing schemes after described renewal is corresponding with the history optimal case of described i-th kind of flight landing center sequencing schemes compares, and the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefining n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, compared by functional value corresponding with the history optimal case of described n-th neighbours' sequencing schemes for the functional value of sequencing schemes after the renewal of described n-th neighbours' sequencing schemes, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefined in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of all neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, select the functional value corresponding with the history optimal case in described all neighbours' sequencing schemes of the optimal function value in described functional value to compare, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme in all neighbours' sequencing schemes of described i-th kind of sequencing schemes

Further, described step S103 also comprises:

Step S1031: according to described flight landing sequence objective function, the various flight landing sequencing schemes before described renewal are compared, select sequencing schemes optimum in described flight landing sequencing schemes as flight landing global history optimal sequencing scheme;

Step S1032: the various flight landing sequencing schemes after described renewal are compared according to described flight landing sequence objective function, select wherein optimum flight landing sequencing schemes and described flight landing global history optimal sequencing scheme to compare, using in the flight landing sequencing schemes of described optimum and described flight landing global history optimal sequencing scheme preferably sequencing schemes as up-to-date flight landing history optimal sequencing scheme.

Particularly, calculated various flight landing sequencing schemes by described flight landing sequence objective function, the flight landing sequencing schemes that functional value is less is more excellent.That is, according to total cost cost Z that the actual landing time of aircraft 1 to aircraft P draws _totalthe minimum flight landing sequencing schemes of value is exactly the excellent scheme that sorts most.

Particularly, in said method, upgrade described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and the step determining up-to-date flight landing global history optimal sequencing scheme has been come by the mode of iteration.Preferably, arranging iterations in the present embodiment is 5000 times, and described iteration ends number of times can be determined according to optimum results, and the present invention does not limit this.Before iteration, also comprise and solution space is set, and the initial scheme of various flight landing sequencing schemes and initial renewal amount are set at random in solution space.Described solution space is by the flight set landing the earliest time E _dthe time L of landing the latest _dbetween time series form.Described solution space, flight landing sequence initial scheme, initial renewal amount and iterations can be stored in described computer system in advance.During first time iteration, substitute into the renewal amount of formulae discovery flight landing sequencing schemes according to described flight landing sequence initial scheme and initial renewal amount and the flight landing sequencing schemes after upgrading during each iteration, the result obtained in an iteration all as iteration basis, thus obtains up-to-date flight landing sequencing schemes and flight landing global history optimal sequencing scheme.When iterations is greater than 5000 times, iteration ends.The flight landing global history optimal sequencing scheme now obtained is the final optimization pass scheme of flight landing sequence.

In the flow control method of the present invention busy termination environment, what described flight landing center sequencing schemes adopted is perfect information learning strategy, know information a large amount of in flight landing sequencing schemes population, in addition, the flight landing center sequencing schemes that the non-central sequencing schemes of flight landing generally tends to angle value in its neighbours larger learns, make flight landing center sequencing schemes that population can be instructed overall toward the motion of better direction more stablely, thus ensure that the speed finding flight landing optimal sequencing scheme; What the non-central sequencing schemes of described flight landing adopted is single information learning strategy, the information that the non-central sequencing schemes of each flight landing receives neighbours is less, the optimization characteristics of self can be kept preferably, thus maintain certain population diversity, there is the problem of local optimum in the optimal case that can prevent flight landing sequencing schemes from searching out.Therefore, the present invention busy termination environment flow control method passes through the flight landing center sequencing schemes learning strategy different from the non-central sequencing schemes of flight landing, improve the Optimal performance of flight landing sequencing schemes, can to find more excellent flight landing sequencing schemes faster velocity-stabilization.

Beneficial effect below by table 1-table 4 couple the present invention busy termination environment flow control method is described in detail.

Table 1 lists eight kinds of conventional trial functions of the Optimal performance for test particle group optimizing method, comprises Rosenbrock function, Sphere function, Schwefel ' sP2.22 function, Quartic function, Rastrigin function, Griewank function (10 peacekeepings 30 are tieed up) and Ackley function.Described eight trial functions are all the problems finding minimum value.Wherein front four functions are all unimodal functions, and optimizing difficulty is relatively low, and Quartic function is the objective function of Noise; Rear four functions are Solving Multimodal Functions, and population is easily absorbed in local optimum, and particle group optimizing method therefore can be checked to solve the ability of more difficult problem.These eight trial functions can the comparatively comprehensive Optimal performance of reaction particle group optimizing method under various scene setting.

Table 1 eight trial functions

As shown in table 1, there is shown the formula of eight trial functions, dimension, optimization range and superior threshold value.Optimization range is the solution space of each objective function, optimal value is the trial function best values that can obtain in solution space in theory, whether successfully superior threshold value judges optimizing standard, iterations when simultaneously reaching superior threshold value also can weigh the speed of particle populations convergence, i.e. speed of searching optimization.

The performance of particle group optimizing method can be weighed by following evaluation index, comprising:

Final optimization pass value: the optimal function value reaching population during the iteration ends number of times of setting;

Superior algebraically: in optimizing process, population optimal function value reaches iterations value during superior threshold value first;

Superior rate: in test of many times, the number of times successfully reaching superior threshold value accounts for the number percent of total experiment number.

Wherein, final optimization pass value is through whole optimizing process, the best optimal value that trial function can obtain, superior algebraically be used for weigh optimizing convergence speed, superior rate can reflect the success ratio of optimizing.

Below by above eight kinds of trial functions respectively to particle group optimizing method (the Selectively-informed Particle Swarm Optimization introducing heterogeneous learning strategy, SIPSO) and the Optimal performance of existing six kinds of particle group optimizing methods test, the advantage of SIPSO can be found out.Particularly, these six kinds of existing particle group optimizing methods are: full UNICOM structured particles group optimizing method (Global-bestParticle Swarm Optimization, GPSO), ring network structure particle group optimizing method (Local-bestParticle Swarm Optimization, LPSO), scales-free network particle group optimizing method (Scale-freeParticle Swarm Optimization, SFPSO), full UNICOM structure perfect information particle group optimizing method (Global-best Fully-informed Particle Swarm Optimization, GFIPSO), ring network structure perfect information particle group optimizing method (Local-best Fully-informed Particle Swarm Optimization, LFIPSO) scales-free network structure perfect information particle group optimizing method fully-informed PSO (Scale-free Fully-informed Particle Swarm Optimization, SFIPSO).

Table 2 lists in above-mentioned eight kinds of trial functions, introduces the comparing result of the particle group optimizing method of heterogeneous learning strategy and the final optimization pass value of existing six kinds of optimization methods.Because all trial functions are all the problems finding minimum value, the performance of the less explanation optimization method of final optimization pass value result is better.

Table 2: the final optimization pass value contrast of seven kinds of optimization methods

Table 3 lists in above-mentioned eight kinds of trial functions, introduces the comparing result of the particle group optimizing method of heterogeneous learning strategy and the superior algebraically of existing six kinds of optimization methods.Superior algebraically is less, illustrates that speed of convergence is faster.

Table 3: the superior algebraically of seven kinds of optimization methods

Table 4 lists in above-mentioned eight kinds of trial functions, introduces the comparing result of the particle group optimizing method of heterogeneous learning strategy and the superior rate of existing six kinds of optimization methods.

Table 4: the superior rate of seven kinds of optimization methods

As can be seen from Table 2, the particle group optimizing method introducing heterogeneous learning strategy can obtain best optimal value on the first five trial function, and the final optimization pass value on rear three functions is also positioned at first three.As can be seen from Table 3, particle group optimizing method of the present invention has speed of convergence faster.As can be seen from Table 4, the superior rate of particle group optimizing method of the present invention on whole trial function is nearly all 100%.

In conjunction with above comparative result, can find out that the particle group optimizing method SIPSO introducing heterogeneous learning strategy can stably find better solution with speed of convergence faster significantly.That is, the busy termination environment flow control method of the particle group optimizing method introducing heterogeneous learning strategy that adopts provided by the invention can obtain more excellent flight landing sequencing schemes compared to existing technology quickly, thus save the cost of flight landing sequence adjustment, ensure the security of air transportation.

Last it is noted that above each embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to foregoing embodiments to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein some or all of technical characteristic; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (6)

1. a busy termination environment flow control method, is characterized in that, comprising:

Receive multiple flight landing sequencing schemes, described multiple flight landing sequencing schemes is arranged with the form of scales-free network;

According to the spread pattern of described multiple flight landing sequencing schemes in described scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, and upgrades described flight landing center sequencing schemes and the non-central sequencing schemes of flight landing respectively in a different manner;

According to flight landing sequence objective function, the various flight landing sequencing schemes before and after upgrading are compared, determine up-to-date flight landing global history optimal sequencing scheme;

The landing of flight is controlled according to described up-to-date flight landing global history optimal sequencing scheme.

2. method according to claim 1, is characterized in that, also comprises:

Degree threshold value for distinguishing flight landing center sequencing schemes and the non-central sequencing schemes of flight landing is set; Wherein, the degree of flight landing sequencing schemes represents the quantity of flight landing sequencing schemes adjacent with described flight landing sequencing schemes in described scales-free network;

Described according to the spread pattern of described multiple flight landing sequencing schemes in described scales-free network, various flight landing sequencing schemes is divided into flight landing center sequencing schemes and the non-central sequencing schemes of flight landing, comprising:

Flight landing sequencing schemes angle value being greater than described degree threshold value is defined as flight landing center sequencing schemes, and flight landing sequencing schemes angle value being less than or equal to described degree threshold value is defined as the non-central sequencing schemes of flight landing;

Described flight landing center sequencing schemes according to

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade flight landing sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the renewal amount of i-th kind of flight landing sequencing schemes, represent the sequencing schemes after the renewal of i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme of n-th neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, K _irepresent the size of the set of neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, represent between random quantity, for being greater than the random number of 0;

The non-central sequencing schemes of described flight landing according to

$\stackrel{\→}{x_i}=\mathrm{\χ}\·(\stackrel{\→}{x_i}+\stackrel{\→}{U}(0,c_1)\·(\stackrel{\→}{p_i}-\stackrel{\→}{x_i})+\stackrel{\→}{U}(0,c_2)\·(\stackrel{\→}{p_{\mathrm{mi}}}-\stackrel{\→}{x_i})),1\≤i\≤I$

Determine renewal amount, according to

$\stackrel{\→}{x_i}=\stackrel{\→}{x_i}+\stackrel{\→}{v_i}$

Upgrade sequencing schemes;

Wherein, I represents the quantity of flight landing sequencing schemes, and i represents the numbering of flight landing sequencing schemes, represent the optimized amount of i-th kind of flight landing sequencing schemes, represent sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, represent the history optimal case of described i-th kind of flight landing sequencing schemes, represent the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes, χ represent control flight landing sequencing schemes convergence in population speed contraction factor, c ₁represent the self-teaching factor, c ₂represent social learning's factor, represent [0, c ₁] between random quantity, represent [0, c ₂] between random quantity.

3. method according to claim 2, is characterized in that, also comprises:

According to sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes, redefine the history optimal sequencing scheme of described i-th kind of flight landing sequencing schemes the history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of flight landing sequencing schemes and the history optimal sequencing scheme in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes

According to described redefine the renewal amount of i-th kind of flight landing center sequencing schemes, and again upgrade described flight landing center sequencing schemes according to described renewal amount;

According to described and redefine the renewal amount of the non-central sequencing schemes of described i kind flight landing, and again upgrade the non-central sequencing schemes of described flight landing according to described optimized amount.

4. method according to claim 3, is characterized in that,

The described history optimal case redefining described i-th kind of flight landing sequencing schemes comprises:

The functional value of sequencing schemes after the renewal of described i-th kind of flight landing sequencing schemes is determined by described flight landing sequence objective function, the functional value that the functional value of sequencing schemes after described renewal is corresponding with the history optimal case of described i-th kind of flight landing center sequencing schemes compares, and the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefining n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, compared by functional value corresponding with the history optimal case of described n-th neighbours' sequencing schemes for the functional value of sequencing schemes after the renewal of described n-th neighbours' sequencing schemes, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme of n-th neighbours' sequencing schemes of described i-th kind of sequencing schemes

The described history optimal sequencing scheme redefined in all neighbours' flight landing sequencing schemes of described i-th kind of flight landing sequencing schemes comprise:

The functional value of sequencing schemes after the renewal of all neighbours' sequencing schemes of described i-th kind of sequencing schemes is determined by described flight landing sequence objective function, select the functional value corresponding with the history optimal case in described all neighbours' sequencing schemes of the optimal function value in described functional value to compare, the sequencing schemes that wherein preferably functional value is corresponding is as the new history optimal sequencing scheme in all neighbours' sequencing schemes of described i-th kind of sequencing schemes

5. the method according to any one of claim 1-4, is characterized in that, describedly compares the various flight landing sequencing schemes before and after upgrading according to flight landing sequence objective function, determines up-to-date flight landing sequence optimal case, comprising:

According to described flight landing sequence objective function, the various flight landing sequencing schemes before described renewal are compared, select sequencing schemes optimum in described flight landing sequencing schemes as flight landing global history optimal sequencing scheme;

According to described flight landing sequence objective function, the various flight landing sequencing schemes after described renewal are compared, select wherein optimum flight landing sequencing schemes and described flight landing global history optimal sequencing scheme to compare, using in the flight landing sequencing schemes of described optimum and described flight landing global history optimal sequencing scheme preferably sequencing schemes as up-to-date flight landing history optimal sequencing scheme.

6. the method according to any one of claim 1-4, is characterized in that, described flight landing sequence objective function is:

$Z_{\mathrm{total}}=\underset{d=1,..,P}{\mathrm{\Σ}}{Z}_d$

$Z_d=\left\{\begin{array}{cc}{g}_d(T_d-X_d)& \mathrm{if}{X}_d\≤T_d\\ h_d(X_d-T_d)& \mathrm{if}{X}_d>T_d\end{array}\right.$

E _d≤X _d≤L _d,d＝1,...,P

X _d-X _f≤S _dfd＝1,...,P,f＝1,...,P,d≠f

Wherein, P represents the total quantity of aircraft, d and f represents the numbering of aircraft, T _drepresent the time of aircraft d expection landing, Z _drepresent that aircraft d shifts to an earlier date or lags behind expection landing time T _dcost cost when landing, Z _totalrepresent total cost cost of the aircraft of all participation landing sorting consistence, X _drepresent the actual landing time of aircraft d, X _frepresent the actual landing time of aircraft f, E _drepresent the earliest time of aircraft d landing, L _drepresent the time the latest of aircraft d landing, g _dthe cost of unit interval when representing that aircraft d shifts to an earlier date landing; h _drepresent the cost of unit interval during aircraft d delayed landing, S _dfrepresent and consider safety factor, required time interval when aircraft d and aircraft f lands.