CN103200125B - Electric power data network node congestion bypassing method and system - Google Patents

Abstract

The invention provides the congested bypassing method of a kind of electric power data network and system, its method comprises step: the length obtaining nodal cache queue, this length and the length threshold value preset is compared; According to the Congestion Level SPCC of described length with the comparative result determination node of the threshold value preset; When described Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue; Abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route, the Congestion Level SPCC of node and the priority of energy communication service due to reasonable contemplation, efficiently avoid node congestion, meet the communication requirement of energy communication service.

Description

Electric power data network node congestion bypassing method and system

Technical field

The present invention relates to field of power communication, particularly a kind of electric power data network node congestion bypassing method and system.

Background technology

Along with electric power system is to the development in digitlization direction, the market demands such as the image information of power dispatching automation information, management of power use information, marketing message, financial information, transformer station and Enterprise MIS information increase gradually, also more and more higher to the requirement of the communication technology, the traditional communication carrier mode of electric power system at transmission rate, fail safe, autgmentability, adaptability and QoS(Quality of Service, service quality) etc. aspect be more and more difficult to the demand meeting user.IP network, as the key data bearing mode of electric power data network, is developed greatly.

Occurring congested is the intrinsic attribute of network.In service process, if the demand of a certain resource has exceeded the available service that this resource can provide in network, will network congestion be caused, tremendous influence has been caused to the service quality of business.Node congestion evasion tactics can the congestion phenomenon effectively in circumvent network, the load in equalizing network, make network provide high-quality service to business.

Traditional congestion avoidance mechanisms is generally based on the ant group algorithm of ant group algorithm or improvement, when link congestion being detected or be about to occur congested, send " congested response ant " from destination node and again explore path, when " congested response ant " arrives congested node, show the non-congested path that existence one is new.Congested node will be switched to new route immediately, slow down congestion state.Although it is congested that traditional congested bypassing method has effectively been evaded on link, it well can not evade node congestion.

Summary of the invention

The object of the present invention is to provide a kind of electric power data network node congestion bypassing method and system, effectively can evade node congestion, meet the communication requirement of energy communication service.

Object of the present invention is achieved through the following technical solutions:

The congested bypassing method of a kind of electric power data network, comprises the steps:

Obtain the length of nodal cache queue, this length and the length threshold value preset are compared;

According to the Congestion Level SPCC of the comparative result determination node of described length and length threshold value;

When described Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue;

Abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route.

The congested avoidance system of a kind of electric power data network, comprising:

Comparison module, for obtaining the length of nodal cache queue, compares this length and the length threshold value preset;

Update module, for the Congestion Level SPCC of the comparative result determination node according to described length and length threshold value;

Adjusting module, for when described Congestion Level SPCC reaches default congestion threshold, carries out position adjustment to the energy communication service in described nodal cache queue by the service priority mark preset;

Discard module, for abandoning the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route.

According to the solution of the present invention, it is the length obtaining nodal cache queue, this length and the threshold value preset are compared, and according to the Congestion Level SPCC of comparative result determination node, when this Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue, abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route, it can avoid node congestion effectively, meet the communication requirement of energy communication service.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of the congested bypassing method embodiment of electric power data network of the present invention;

Fig. 2 is the format structure schematic diagram of expansion TLV in Link State Advertisement (LSA);

Fig. 3 is the schematic diagram data in nodal cache queue being carried out to position adjustment by service priority;

Fig. 4 is the refinement schematic flow sheet selecting route in the congested bypassing method of electric power data network of the present invention;

Fig. 5 is the structural representation of an electric power data network of the present invention congested avoidance system embodiment;

Fig. 6 is the structural representation of congested another embodiment of avoidance system of electric power data network of the present invention;

Fig. 7 is the refined structure schematic diagram of the routing selecting module in Fig. 6.

Embodiment

Below in conjunction with embodiment and accompanying drawing, the present invention is further elaborated, but implementation of the present invention is not limited thereto.

Shown in Figure 1, be the schematic flow sheet of the congested bypassing method embodiment of electric power data network of the present invention.As shown in Figure 1, the congested bypassing method of electric power data network in this embodiment comprises the steps:

Step S101: the length obtaining nodal cache queue, carries out ratio by this length and the length threshold value preset, enters step S102;

Length that is real-time or periodically detection node buffer queue, the quantity of the energy communication service namely in detection node buffer queue, and the length of buffer queue and length threshold value are carried out size compare, wherein, the size of length threshold value and quantity (i.e. the number of length threshold value) can set according to actual conditions;

Step S102: according to the Congestion Level SPCC of described length with the comparative result determination node of the threshold value preset, enter step S103;

The power of the whether congested and Congestion Level SPCC of the comparative result determination present node compared according to step S101;

Step S103: when described Congestion Level SPCC reaches default congestion threshold, carries out position adjustment to the energy communication service in described nodal cache queue by the service priority mark preset, enters step S104;

Congestion threshold can be arranged according to actual conditions; Each energy communication service correspond to a service priority mark, from high to low energy communication service is arranged from head of the queue to tail of the queue according to this service priority mark, what namely priority-level was high comes head of the queue, and the energy communication service that priority-level is low is queued at the end of queue;

Step S104: abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route;

The quantity of the low priority traffice abandoned or abandon which low priority traffice, can be abandon number by setting to determine, also can being that the threshold value arranged according to Congestion Level SPCC is determined, being positioned at the energy communication service after a certain Congestion Level SPCC threshold value as abandoned in tail of the queue.

Accordingly, according to the scheme of above-mentioned the present embodiment, it is the length obtaining nodal cache queue, this length and the threshold value preset are compared, and according to the Congestion Level SPCC of comparative result determination node, when this Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue, abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route, the Congestion Level SPCC of node and the priority of energy communication service due to reasonable contemplation, efficiently avoid node congestion, meet the communication requirement of energy communication service.

Wherein, dividing the priority of energy communication service and can have different implementations according to actual needs, wherein a kind of implementation, can be that the priority to energy communication service divides according to singal reporting code time delay and bandwidth; Such as, according to singal reporting code time delay and bandwidth, energy communication service can be divided into six ranks, this six classes priority service and service priority mark thereof are followed successively by from height to low: broadband delay sensitive business, Pr=5; Arrowband delay sensitive business, Pr=4; Broadband time delay time sensitive traffic, Pr=3; Arrowband time delay time sensitive traffic, Pr=2; Broadband non-real-time service, Pr=1; Arrowband non-real-time service, Pr=0, wherein Pr is service priority mark, and the form of service priority mark is also not limited to the digital mode of employing 0 ~ 5 these six.

This implementation, in being preferably applied to that electric power data network is congested and evading, has good application prospect.This is because, evade electric power data network is congested, singal reporting code time delay and bandwidth are two factors needing to consider emphatically, and according to the priority of singal reporting code time delay and bandwidth partition energy communication service, this two factors are taken into full account, high-priority service can be made preferentially to be served, ensure that the QoS of high-priority service.

In addition, wherein in an embodiment, can service priority mark be configured in the classification control field of network control message, so that when described Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue.

Fig. 2 is the format structure schematic diagram of expansion TLV in Link State Advertisement (LSA).Energy communication service priority tag is joined the classification control field of network control message.As shown in Figure 2, in generalized multiprotocol label switching (GMPLS) protocol (GMPLS), add the control field that the service priority with energy communication service identifies, be specially the service priority mark of the communication service that to increase electric power in the TLV field of opaque Link State Advertisement (Opaque LSA).In Fig. 2, Type represents the field type of this part messages; Length represents the length of Value field; Value is the variable length data byte of this part messages.In the present embodiment, add the service priority identification field of energy communication service in the TLV in LSA, be specially: according to RFC2370, Type is taken as 2, represent link TLV information; Length value is 3, and the length representing Value field is 3 bits; Value value is 0 ~ 5, represents 6 priority levels of energy communication service.

Based on above-mentioned several embodiment, introduce a specific embodiment below, but following specific embodiment is not construed as limiting the invention.

Specific embodiment

First, detect congestion state, and upgrade the congestion state mark of this node according to congestion state;

If Q<R _l, then the congestion state of node is without congested, Flag _c=0;

If R _l≤ Q<R _h, then the congestion state of node is that moderate is congested, Flag _c=1;

If R _h≤ Q, then the congestion state of node is heavy congestion, Flag _c=2.

Wherein Q represents the length of nodal cache queue, R _lrepresent the moderate congestion threshold of node, R _hrepresent the heavy congestion threshold value of node, Flag _cfor node congestion status indicator;

In this specific embodiment, identify corresponding congestion state with congestion state, corresponding aforesaid Congestion Level SPCC, and congestion state is divided into three classes, in the specific implementation, be not limited thereto.

Secondly, monitoring congestion state mark, if the congested or heavy congestion that is moderate of congestion state corresponding to congestion state mark, then carries out position adjustment to the data in nodal cache queue by service priority; Fig. 3 illustrates the process of the data in congested node buffer queue according to priority being carried out to position adjustment, supposes that the number-of-packet in now nodal cache queue is 9, is respectively S1, S2, L, S9; By service priority, position adjustment is carried out to the data in congested node buffer queue, as shown in Figure 3;

Wherein, the service priority in Fig. 3 is, according to singal reporting code time delay and bandwidth, energy communication service is divided into six ranks, this six classes priority service and state service priority mark be followed successively by from height to low: broadband delay sensitive business, Pr=5; Arrowband delay sensitive business, Pr=4; Broadband time delay time sensitive traffic, Pr=3; Arrowband time delay time sensitive traffic, Pr=2; Broadband non-real-time service, Pr=1; Arrowband non-real-time service, Pr=0, wherein Pr is service priority mark, and the form of service priority mark is also not limited to the digital mode of employing 0 ~ 5 these six;

Can join in the classification control field of network control message by the service priority of energy communication service mark in advance, concrete implementation as previously mentioned, repeat them here;

Then, abandon the low priority traffice after being positioned at heavy congestion threshold value, and notify that the source node that these low priority traffices are corresponding reselects route;

If situation shown in Fig. 3, then abandon S5, the S3 in the buffer queue after adjustment.

On the other hand, point out to need to notify that the source node that low priority traffice is corresponding reselects route in above-mentioned steps S104, after these source nodes receive notice, enter routing procedure.This selection course is solving of a multi-objective problem.Along with the development of artificial intelligence technology, solving of multi-objective problem has many algorithms.The heuritic approach such as genetic algorithm, quantum genetic algorithm, neural network algorithm, simulated annealing, particle swarm optimization algorithm, ant group algorithm is widely used.

In a preferred embodiment, particle swarm optimization algorithm is selected to solve multi-object routing select permeability.Multi-objective optimization question is merged by Triangle Module and is converted into single-object problem, and utilize particle swarm optimization algorithm to select optimal path.Particle swarm optimization algorithm is a kind of iteration optimization algorithms based on swarm intelligence, is a kind of heuristic search algorithm.This algorithm has following characteristics: in particle swarm optimization algorithm, and the potential solution of each optimization problem is a bird in search volume, is referred to as " particle ".All particles have an adaptive value determined by optimised function, and each particle also has speed to determine the direction that they circle in the air and distance, and then particles are just followed current optimal particle and searched in solution space.Compared with other intelligent algorithms, the advantage of particle swarm optimization algorithm is that simple easily realization has again deep intelligent background simultaneously; Particle swarm optimization algorithm do not require optimised function have can micro-, can lead, the character such as continuous, and have fast convergence rate, algorithm is simple, the features such as easily realize, controling parameters is few, and computational speed is fast.

Thus, implementing for one wherein, can be alternative particle to the path of destination node with source node corresponding to described low priority traffice, with $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ For fitness function, utilize particle swarm optimization algorithm, at described source node place for this low priority traffice reselects route, wherein, x is path, f _dx () represents the propagation delay time function on the x of path, f _bx () represents path x available bandwidth function.Particularly, different implementations can be had.

Shown in Figure 4, reselect the schematic flow sheet of route embodiment for of the present invention for low priority traffice.As shown in Figure 4, comprising the steps: for low priority traffice reselects route in this embodiment

Step S1051: the total iterations of initialization is LoopCount, current iteration number of times t=0, population scale is N, population X (t)=X (0);

Can adopt binary system population coded system, get the path of source node to destination node as alternative particle, the dimension of particle is m, represents the positional information of particle, gets 1 or 0, and on the path whether corresponding node respectively;

The initial position of population is random binary vector, meets the condition that on path, each node is connected; The kth dimension velocity component V of particle i in Binary Particle Swarm Optimization _ikdetermine its location components X _ikget the probability of 1 or 0, V _iklarger, then X _ikget 1 probability larger; In binary particle swarm algorithm, by restriction Vi _k, sig (V can be made _ik) can not too close to 0.0 or 1.0; For avoiding the Premature Convergence of algorithm, by V _ikbe limited in ± 4 between, corresponding to sig (-4)=0.018, sig (4)=0.982;

Step S1052: the fitness value determining each particle of population X (t), obtains the fitness value fitness (X of N number of particle _i(t)), i=1,2, L, N;

Bring the routing information (propagation delay time and available bandwidth) of each particle into fitness function, obtain the fitness value of each particle;

The fitness function that the present invention constructs has following features: one, function can meet the different requirements of energy communication service to singal reporting code; Two, functional form meets feature required by particle cluster algorithm; Three, function convergence is more satisfactory, and particle swarm optimization algorithm can be made to converge to optimal solution in the short period of time;

In the present invention, particle fitness function is

$\mathrm{fitness}\left(X_i\left(t\right)\right)=\frac{f_D\left(X_i\left(t\right)\right)\&CenterDot;f_B\left(X_i\left(t\right)\right)}{1-f_D\left(X_i\left(t\right)\right)-f_B\left(X_i\left(t\right)\right)+2\&CenterDot;f_D\left(X_i\left(t\right)\right)\&CenterDot;f_B\left(X_i\left(t\right)\right)}$

Wherein, X _it () represents the i-th paths.

(1) f _d(X _i(t)) computing formula:

$f_D\left(X_i\left(t\right)\right)=\left\{\begin{array}{cc}\frac{{\mathrm{Delay}}_{\max }-\mathrm{Delay}\left(X_i\left(t\right)\right)}{{\mathrm{Delay}}_{\max }}& \mathrm{Delay}\left(X_i\left(t\right)\right)<{\mathrm{Delay}}_{\max }\\ 0& \mathrm{Delay}\left(X_i\left(t\right)\right)\&GreaterEqual;{\mathrm{Delay}}_{\max }\end{array}\right.$

Wherein,

Delay _maxfor the tolerable time delay maximum of energy communication service;

$\mathrm{Delay}\left(X_i\left(t\right)\right)=\underset{n\&Element;X_i\left(t\right)}{\mathrm{\&Sigma;}}\mathrm{Delay}\left(n\right)+\underset{l\&Element;X_i\left(t\right)}{\mathrm{\&Sigma;}}\mathrm{delay}\left(l\right)$ For path X _it the propagation delay time of (), wherein n represents path X _it the node on (), l represents path X _ilink on (t).

(2) f _b(X _i(t)) computing formula:

$f_B\left(X_i\left(t\right)\right)=\frac{\mathrm{Bandwidth}\left(X_i\left(t\right)\right)-\mathrm{Bandwidt}{h}_{\min }}{\mathrm{Bandwidt}{h}_{\max }-\mathrm{Bandwidt}{h}_{\min }}$

Wherein,

Bandwidth (X _i(t)) be path X _ithe available bandwidth of (t);

Bandwidth _maxfor the available bandwidth maximum in X (t);

Bandwidth _minfor the available bandwidth minimum value in X (t);

Step S1053: select fitness value is maximum in population particle position as the current globally optimal solution P of population _g, select the particle i particle position that fitness value is maximum to current iteration number of times as the history optimal solution P of particle i _i;

Using initialized particle position as the initial P of particle _i, from population, find out the maximum particle position of fitness value as P _g.To particle i, by its current fitness value and P _icorresponding fitness value is made comparisons, if compare P _igood, then reset P _i; By P current for particle i _iwith P _gcorresponding fitness value is made comparisons, if compare P _ggood, then reset P _g;

Step S1054: judge current globally optimal solution P _gwhether meet Optimization route condition, if so, then terminate and export optimal solution; If not, then upgrade population X (t) according to the history optimal solution of current globally optimal solution and each particle, obtain new population X (t+1), return step S1052.

According to the congested bypassing method of electric power data network of the invention described above, the present invention also provides a kind of electric power data network congested avoidance system.

Shown in Figure 5, be the structural representation of the congested avoidance system embodiment of electric power data network of the present invention.As shown in Figure 5, the congested avoidance system of electric power data network of the present embodiment comprises comparison module 201, update module 202, adjusting module 203, discard module 204, wherein:

Comparison module 201, for obtaining the length of nodal cache queue, compares this length and the length threshold value preset;

Update module 202, for the Congestion Level SPCC of the comparative result determination node according to described length and length threshold value;

Adjusting module 203, for when described Congestion Level SPCC reaches default congestion threshold, carries out position adjustment to the energy communication service in described nodal cache queue by the service priority mark preset;

Discard module 204, for abandoning the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route.

Accordingly, according to the scheme in the present embodiment, the Congestion Level SPCC of node and the priority of energy communication service due to reasonable contemplation, efficiently avoid node congestion, meet the communication requirement of energy communication service.

Wherein, the priority of energy communication service can be that the priority to energy communication service divides according to singal reporting code time delay and bandwidth, this implementation is preferably applied to that electric power data network is congested evades, there is good application prospect, high-priority service can be made preferentially to be served, ensure that the QoS of high-priority service.

Wherein in an embodiment, described service priority mark can be configured in the classification control field of network control message, and specific implementation can as previously mentioned, not repeat them here.

Wherein in an embodiment, as shown in Figure 6, the congested avoidance system of electric power data network of the present invention can also comprise routing selecting module 205,

Routing selecting module 205 for being alternative particle with source node corresponding to described low priority traffice to the path of destination node, with $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ As fitness function, utilize particle swarm optimization algorithm, at described source node place for this low priority traffice reselects route;

Wherein, x is path, f _dx () represents the propagation delay time function on the x of path, f _bx () represents path x available bandwidth function.

Wherein in an embodiment, as shown in Figure 7, routing selecting module 205 can comprise initialization unit 2051, fitness value determining unit 2052, selected cell 2053, judging unit 2054, wherein:

Initialization unit 2051 is LoopCount for the total iterations of initialization, current iteration number of times t=0, and population scale is N, population X (t)=X (0);

Fitness value determining unit 2052, for determining the fitness value of each particle of population X (t), obtains the fitness value fitness (X of N number of particle _i(t)), i=1,2, L, N;

Selected cell 2053, for the particle position of selecting fitness value in population maximum as the current globally optimal solution P of population _g, select the particle i particle position that fitness value is maximum to current iteration number of times as the history optimal solution P of particle i _i;

Judging unit 2054, for judging current globally optimal solution P _gwhether meet Optimization route condition, if so, then terminate and export optimal solution; If not, then upgrade population X (t) according to the history optimal solution of current globally optimal solution and each particle, obtain new population X (t+1).

Specific implementation process can as previously mentioned, not repeat them here.

Relative to prior art, tool of the present invention has the following advantages and beneficial effect:

First by carrying out position adjustment to the data in heavy congestion node and moderate congested node buffer queue by service priority, high-priority service preferentially being served, ensure that the QoS of high-priority service; Secondly, by abandoning the low priority traffice after being positioned at heavy congestion threshold value, the congestion state of heavy congestion node is alleviated; Finally, with time delay and bandwidth for target function, utilize particle swarm optimization algorithm, reselect route at the source node place of low priority traffice for it, also ensure for low priority traffice provides higher QoS.

The above embodiment only have expressed several execution mode of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection range of patent of the present invention should be as the criterion with claims.

Claims (10)

1. the congested bypassing method of electric power data network, is characterized in that, comprise the steps:

Obtain the length of nodal cache queue, this length and the length threshold value preset are compared;

According to the Congestion Level SPCC of the comparative result determination node of described length and length threshold value;

When described Congestion Level SPCC reaches default congestion threshold, by the service priority mark preset, position adjustment is carried out to the energy communication service in described nodal cache queue;

Abandon the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route.

2. the congested bypassing method of electric power data network according to claim 1, is characterized in that, according to singal reporting code time delay and bandwidth, the priority to energy communication service divides.

3. the congested bypassing method of electric power data network according to claim 1, is characterized in that, described service priority mark is configured in the classification control field of network control message.

4. according to the congested bypassing method of the electric power data network of one of claims 1 to 3, it is characterized in that, is alternative particle with source node corresponding to described low priority traffice to the path of destination node, with $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ As fitness function, utilize particle swarm optimization algorithm, be that described low priority traffice reselects route at described source node place, wherein, x represents path, f _dx () represents the propagation delay time function on the x of path, f _bx () represents path x available bandwidth function;

Wherein, f _dx the computing formula of () is:

$f_D\left(x\right)=\left\{\begin{array}{cc}\frac{{\mathrm{Delay}}_{\max }-\mathrm{Delay}\left(x\right)}{{\mathrm{Delay}}_{\max }}& \mathrm{Delay}\left(x\right)<{\mathrm{Delay}}_{\max }\\ 0& \mathrm{Delay}\left(x\right)\&GreaterEqual;{\mathrm{Delay}}_{\max }\end{array}\right.$

In formula, Delay _maxfor the tolerable time delay maximum of energy communication service;

F _bthe computing formula of (x):

$f_B\left(x\right)=\frac{\mathrm{Bandwidth}\left(x\right)-{\mathrm{Bandwidth}}_{\min }}{{\mathrm{Bandwidth}}_{\max }-{\mathrm{Bandwidth}}_{\min }}$

In formula, Bandwidth (x) is the available bandwidth of path x, Bandwidth _maxfor the available bandwidth maximum in x, Bandwidth _minfor the available bandwidth minimum value in x.

5. the congested bypassing method of electric power data network according to claim 4, is characterized in that, the described source node corresponding with described low priority traffice is alternative particle to the path of destination node, with $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ As fitness function, utilizing particle swarm optimization algorithm, is that described low priority traffice reselects route and comprises the steps: at described source node place

The total iterations of initialization is LoopCount, current iteration number of times t=0, and population scale is N, population X (t)=X (0);

According to $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ Determine the fitness value of each particle of population X (t), obtain the fitness value fitness (X of N number of particle _i(t)), i=1,2 ..., N;

Select fitness value is maximum in population particle position as the current globally optimal solution P of population _g, select the particle i particle that fitness value is maximum to current iteration number of times as the history optimal solution P of particle i _i;

Judge current globally optimal solution P _gwhether meet Optimization route condition, if so, then process ends export optimal solution; If not, then upgrade population X (t) according to the history optimal solution of current globally optimal solution and each particle, obtain new population X (t+1).

6. the congested avoidance system of electric power data network, is characterized in that, comprising:

Comparison module, for obtaining the length of nodal cache queue, compares this length and the length threshold value preset;

Update module, for the Congestion Level SPCC of the comparative result determination node according to described length and length threshold value;

Adjusting module, for when described Congestion Level SPCC reaches default congestion threshold, carries out position adjustment to the energy communication service in described nodal cache queue by the service priority mark preset;

Discard module, for abandoning the low priority traffice being positioned at tail of the queue in the nodal cache queue after adjustment, and the source node of notice correspondence reselects route.

7. the congested avoidance system of electric power data network according to claim 6, is characterized in that, according to singal reporting code time delay and bandwidth, the priority to energy communication service divides.

8. the congested avoidance system of electric power data network according to claim 6, is characterized in that, described service priority mark is configured in the classification control field of network control message.

9., according to the congested avoidance system of the electric power data network of one of claim 6 to 8, it is characterized in that, also comprise:

Routing selecting module, for being alternative particle with source node corresponding to described low priority traffice to the path of destination node, with $\mathrm{fitness}\left(x\right)=\frac{f_D\left(x\right)\&CenterDot;f_B\left(x\right)}{1-f_D\left(x\right)-f_B\left(x\right)+2\&CenterDot;f_D\left(x\right)\&CenterDot;f_B\left(x\right)}$ As fitness function, utilize particle swarm optimization algorithm, at described source node place for this low priority traffice reselects route;

Wherein, x is path, f _dx () represents the propagation delay time function on the x of path, f _bx () represents path x available bandwidth function;

Wherein, f _dx the computing formula of () is:

$f_D\left(x\right)=\left\{\begin{array}{cc}\frac{{\mathrm{Delay}}_{\max }-\mathrm{Delay}\left(x\right)}{{\mathrm{Delay}}_{\max }}& \mathrm{Delay}\left(x\right)<{\mathrm{Delay}}_{\max }\\ 0& \mathrm{Delay}\left(x\right)\&GreaterEqual;{\mathrm{Delay}}_{\max }\end{array}\right.$

In formula, Delay _maxfor the tolerable time delay maximum of energy communication service;

F _bthe computing formula of (x):

$f_B\left(x\right)=\frac{\mathrm{Bandwidth}\left(x\right)-{\mathrm{Bandwidth}}_{\min }}{{\mathrm{Bandwidth}}_{\max }-{\mathrm{Bandwidth}}_{\min }}$

In formula, Bandwidth (x) is the available bandwidth of path x, Bandwidth _maxfor the available bandwidth maximum in x, Bandwidth _minfor the available bandwidth minimum value in x.

10. the congested avoidance system of electric power data network according to claim 9, is characterized in that, described routing selecting module comprises:

Initialization unit is LoopCount for the total iterations of initialization, current iteration number of times t=0, and population scale is N, population X (t)=X (0);

Fitness value determining unit, for determining the fitness value of each particle of population X (t), obtains the fitness value fitness (X of N number of particle _i(t)), i=1,2 ..., N;

Selected cell, for the particle position of selecting fitness value in population maximum as the current globally optimal solution P of population _g, select the particle i particle position that fitness value is maximum to current iteration number of times as the history optimal solution P of particle i _i;

Judging unit, for judging current globally optimal solution P _gwhether meet Optimization route condition, if so, then process ends export optimal solution; If not, then upgrade population X (t) according to the history optimal solution of current globally optimal solution and each particle, obtain new population X (t+1).