CN106251043A - Multizone economic environment method for optimizing scheduling and device - Google Patents

Abstract

The embodiment of the invention discloses multizone economic environment method for optimizing scheduling and the device of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution, the method is iterated by the first computing operator, the second computing operator and the continuous first parent population of the 3rd computing operator, the second parent population and the 3rd parent population, it is thus possible to obtain the particle with number of different types, the i.e. population of solution, thus improve quanta particle swarm optimization in prior art and be easily trapped into the shortcoming of local optimum, improve the effect of multizone economic environment scheduling.

Description

Multizone economic environment method for optimizing scheduling and device

Technical field

The invention relates to intelligent power grid technology field, more particularly relate to a kind of multizone economic environment scheduling excellent Change method and apparatus.

Background technology

The scheduling of power system economic environment is the main contents of EMS, power system under some specific environments Economic environment scheduling is equal to generation schedule on conceptual category, and generation schedule includes generator combination, Hydro-Thermal Systems plan, exchange Plan, repair schedule and fuel planning etc.；By the cycle, it has: ultra-short term plan, i.e. Automatic Generation Control, Short Term Generation Schedules, This day or the plan in week；Mid-term generation schedule, the plan of front-month to year and correction；Long-term plan, i.e. several years are to the meter of many decades Draw, including power source developing plan and network Development planning etc..

In recent years, along with day by day highlighting of problem of environmental pollution, each thermal power plant while pursuing power benefit more Consider the reduction of discharging of waste gas, and formulate dusty gas discharge statute of limitation one after another.Power system economic environment is dispatched also by passing The single goal economic load dispatching of system has turned to environmental economy scheduling (Environmental Economic load more Dispatch, EED).EED problem is a multi-objective optimization question, must be to expense and row on the premise of meeting various constraint Put that the two mutually retrains, the target conflicted is optimized.Simple single area is frequently not for actual power system But multizone, regional is got up by interconnection interconnection, so the target of multizone economic environment scheduling is at satisfied electricity Under the constraint such as power demand, the operation characteristic of electromotor, tie-line power transmission, seek the generating capacity of power system and each region Between exchange of electric power, thus give electric power system dispatching personnel provide a series of decision-making to carry out economic environment scheduling.

Current quantum particle swarm (Quantum-behaved Particle Swarm Optimization Algorithm, QPSO) algorithm is used for solving multizone economic environment scheduling problem, but owing to quantum particle swarm itself has in optimizing Journey is easily trapped into the shortcoming of local optimum, Pareto forward position narrow distribution that quanta particle swarm optimization is found out and skewness, The solution lack of diversity arrived, it is impossible to allow policymaker make the most rational decision-making and judge.

Summary of the invention

In view of this, the invention provides a kind of multizone economic environment method for optimizing scheduling and device, existing to overcome In technology, quanta particle swarm optimization itself has the shortcoming being easily trapped into local optimum at searching process, and quanta particle swarm optimization is looked for The Pareto forward position narrow distribution gone out and skewness, the solution lack of diversity obtained, it is impossible to allow policymaker make more The problem judged for rational decision-making.

For achieving the above object, the present invention provides following technical scheme:

A kind of multizone economic environment scheduling of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Optimization method, including:

S1, set up the economic environment Scheduling Optimization Model of multizone, the economic environment Scheduling Optimization Model of described multizone Including two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows:

$F=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\Σ}}\underset{r=1,r\≠i}{\overset{N}{\Σ}}1/2\·t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{\mathrm{\α}}_{ij}+{\mathrm{\β}}_{ij}{P}_{ij}+{\mathrm{\γ}}_{ij}{P}_{ij}^2+{\mathrm{\ξ}}_{ij}\exp \left({\mathrm{\λ}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is i-th district respectively The cost coefficient of the jth platform electromotor in territory, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij (P_ij) it is the discharge function of the jth platform electromotor of ith zone, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth of ith zone respectively The emission factor of platform electromotor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt is The actual power that the jth platform electromotor in i region is sent；I, r and j are the positive integer more than or equal to 1；

Described inequality constraints condition is respectively as follows:

Wherein,The respectively jth platform electromotor of ith zone can send Peak power and minimum power；

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the most whole more than or equal to 1 Number；

Wherein, T_ir,min T_ir,maxIt is the minimum power from ith zone through-put power to r region and maximum respectively Power；

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\Σ}}{P}_{ij}=P_{Di}+\underset{r=1,r\≠i}{\overset{N}{\Σ}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt is to be transferred to the r district from ith zone by interconnection The power in territory；

S2, economic environment Scheduling Optimization Model based on described multizone initialize P in said two object function_ijValue And T_irValue, it is thus achieved that comprising the initial population of Z particle, each described particle includes meet said two object function one group P_ijAnd T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing operator and the 3rd computing operator, Using described initial population as the first parent population of iteration first, Z is the positive integer more than or equal to 2；

S3, each particle in described first parent population is performed the first computing operator, it is thus achieved that by Z grain molecular One progeny population, merges described F1 population and described first parent population, it is thus achieved that first merges population, to described first Merging each particle in population and carry out selecting operation, the molecular population of Z grain that will select, as performing described second fortune Calculate the second parent population of operator；

S4, each particle in described lower second parent population is performed the second computing operator, it is thus achieved that molecular by Z grain Second filial generation population, merges described second filial generation population and described second parent population, it is thus achieved that second merges population, to described the Two merge each particle in population carries out selecting operation, and the molecular population of Z grain that will select, as performing the described 3rd 3rd parent population of computing operator；

S5, each particle in described 3rd parent population is performed the 3rd computing operator, it is thus achieved that by Z grain molecular the Three progeny populations, merge described F3 population and described 3rd parent population, it is thus achieved that the 3rd merges population, to the described 3rd Merge each particle in population to carry out selecting operation, the molecular population of Z grain that will select, hold during as non-iteration first First parent population of the described first computing operator of row；

If S6 current iteration number of times is more than described maximum iteration time, then the first father that output maximum iteration time is corresponding For population, if described current iteration number of times is less than or equal to described maximum iteration time, return step S3.

A kind of multizone economic environment scheduling of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Optimize device, including:

Building module, for setting up the economic environment Scheduling Optimization Model of multizone, the economic environment of described multizone is adjusted Degree Optimized model includes two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows:

$F=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\Σ}}\underset{r=1,r\≠i}{\overset{N}{\Σ}}1/2\·t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{\mathrm{\α}}_{ij}+{\mathrm{\β}}_{ij}{P}_{ij}+{\mathrm{\γ}}_{ij}{P}_{ij}^2+{\mathrm{\ξ}}_{ij}\exp \left({\mathrm{\λ}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is i-th district respectively The cost coefficient of the jth platform electromotor in territory, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij (P_ij) it is the discharge function of the jth platform electromotor of ith zone, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth of ith zone respectively The emission factor of platform electromotor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt is The actual power that the jth platform electromotor in i region is sent；I, r and j are the positive integer more than or equal to 1；

Described inequality constraints condition is respectively as follows:

Wherein,The jth platform electromotor being respectively ith zone can send Peak power and minimum power；

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the most whole more than or equal to 1 Number；

Wherein, T_ir,min T_ir,maxIt is the minimum power from ith zone through-put power to r region and maximum respectively Power；

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\Σ}}{P}_{ij}=P_{Di}+\underset{r=1,r\≠i}{\overset{N}{\Σ}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt is to be transferred to the r district from ith zone by interconnection The power in territory；

First acquisition module, initializes said two mesh for economic environment Scheduling Optimization Model based on described multizone P in scalar functions_ijValue and T_irValue, it is thus achieved that comprising the initial population of Z particle, each described particle includes meeting described two One group of P of individual object function_ijAnd T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing are calculated Son and the 3rd computing operator, using described initial population as the first parent population of iteration first, Z is the most whole more than or equal to 2 Number；

Second acquisition module, for performing the first computing operator by each particle in described first parent population, it is thus achieved that by Z Individual grain molecular F1 population, merges described F1 population and described first parent population, it is thus achieved that first merges Population, merges each particle in population and carries out selecting operation, the molecular population of Z grain that will select, make described first For performing the second parent population of described second computing operator；

3rd acquisition module, is used for each particle in described lower second parent population is performed the second computing operator, it is thus achieved that By Z grain molecular second filial generation population, merge described second filial generation population and described second parent population, it is thus achieved that second closes And population, merge each particle in population to described second and carry out selecting to operate, the molecular population of Z grain that will select, As the 3rd parent population performing described 3rd computing operator；

4th acquisition module, for performing the 3rd computing operator by each particle in described 3rd parent population, it is thus achieved that by Z Individual grain molecular F3 population, merges described F3 population and described 3rd parent population, it is thus achieved that the 3rd merges Population, merges each particle in population and carries out selecting operation, the molecular population of Z grain that will select, make the described 3rd The first parent population of described first computing operator is performed during for non-iteration first；

Return module, if for current iteration number of times more than described maximum iteration time, then export described greatest iteration time The first parent populations that number is corresponding, if described current iteration number of times is less than or equal to described maximum iteration time, return described the Two acquisition modules.

Understand via above-mentioned technical scheme, compared with prior art, embodiments provide a kind of multizone warp Ji environment method for optimizing scheduling, by the first computing operator, the second computing operator and continuous first parent of the 3rd computing operator Population, the second parent population and the 3rd parent population are iterated such that it is able to obtain the particle with number of different types, i.e. The population of solution, thus improve quanta particle swarm optimization in prior art and be easily trapped into the shortcoming of local optimum, improve The effect of multizone economic environment scheduling.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to The accompanying drawing provided obtains other accompanying drawing.

A kind of multiple target variation quantum particle swarm sorted based on quick noninferior solution that Fig. 1 provides for the embodiment of the present application is calculated The schematic flow sheet of the multizone economic environment method for optimizing scheduling of method；

The multiple target sorted based on the quick noninferior solution variation quanta particle swarm optimization that Fig. 2 provides for the embodiment of the present application Multizone economic environment method for optimizing scheduling embodiment performs to select the method flow diagram of a kind of implementation of operation；

The each region of one that Fig. 3 provides for the embodiment of the present application is got in touch with and power load distributing schematic diagram；

The Pareto forward position utilizing NSVAQPSO algorithm with utilizing other algorithm contrast that Fig. 4 provides for the embodiment of the present application Figure.

Detailed description of the invention

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise Embodiment, broadly falls into the scope of protection of the invention.

Refer to Fig. 1, a kind of multiple target variation quantum sorted based on quick noninferior solution provided for the embodiment of the present application The schematic flow sheet of the multizone economic environment method for optimizing scheduling of particle cluster algorithm, the method includes:

Step S1: set up the economic environment Scheduling Optimization Model of multizone, the economic environment optimizing scheduling of described multizone Model includes two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows: economic goal function F and emissions object line number E:

$F=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\Σ}}\underset{r=1,r\≠i}{\overset{N}{\Σ}}1/2\·t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{Mi}{\Σ}}{\mathrm{\α}}_{ij}+{\mathrm{\β}}_{ij}{P}_{ij}+{\mathrm{\γ}}_{ij}{P}_{ij}^2+{\mathrm{\ξ}}_{ij}\exp \left({\mathrm{\λ}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is i-th district respectively The cost coefficient of the jth platform electromotor in territory, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij (P_ij) it is the discharge function of the jth platform electromotor of ith zone, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth of ith zone respectively The emission factor of platform electromotor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt is The actual power that the jth platform electromotor in i region is sent；I, r and j are the positive integer more than or equal to 1.

Consider that the economic goal function F of electromotor valve point effect can also be:

Wherein, e_ij、f_ijIt is the cost coefficient of the jth platform electromotor of ith zone respectively, needs for contrasting other algorithms, present case The valve point effect that do not consider electromotor consistent with other existing research, therefore e_ij、f_ijIt is zero, so will notWrite in economic goal function.

When setting up object function, need to consider the power system multizone that generator power limits, area power balances Economic environment scheduling problem, considers the restriction of tie-line power transmission between region, such that it is able to obtain inequality constraints simultaneously Condition and equality constraint.

Described inequality constraints condition is respectively as follows:

Wherein,The respectively jth platform electromotor of ith zone can send Peak power and minimum power.

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the most whole more than or equal to 1 Number.

Wherein, T_ir,minT_ir,maxIt is the minimum power from ith zone through-put power to r region and maximum work respectively Rate.Set in the embodiment of the present application, if power is to be transferred to the r region, then T from ith zone_irIt is positive number, otherwise then T_irFor negative.By T_irIt is referred to as dominant eigenvalues.

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\Σ}}{P}_{ij}=P_{Di}+\underset{r=1,r\≠i}{\overset{N}{\Σ}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt is to be transferred to the r district from ith zone by interconnection The power in territory.

Step S2: economic environment Scheduling Optimization Model based on described multizone initializes P in said two object function_ij Value and T_irValue, it is thus achieved that comprising the initial population of Z particle, each described particle includes meeting said two object function One group of P_ijAnd T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing operator and the 3rd computing Operator, using described initial population as the first parent population of the first computing operator during iteration first, Z more than or equal to 2 is just Integer.

When carrying out initialization of population, need to choose Population Size Z, then according to the power upper limit of each electromotor, each Individual electromotor lower limit, the dominant eigenvalues upper limit, dominant eigenvalues lower limit carry out stochastic generation and meet two object functions The performance number of electromotor and dominant eigenvalues value, thus form a number of initial population.Each particle is a kind of solution Certainly scheme, population is exactly the set including multiple solution.

Step S3: each particle in described first parent population is performed the first computing operator, it is thus achieved that be made up of Z particle F1 population, merge described F1 population and described first parent population, it is thus achieved that first merge population, to described First merges each particle in population carries out selecting operation, and the molecular population of Z grain that will select, as performing the second fortune Calculate the second parent population of operator.

Optionally, can also include before step S3: judge that whether current iteration number of times is that iterations is (such as first 1), the most described first parent population is merged population as described first；If it is not, then perform the operation of step S3.

When initial population is merged population as first, it is thus achieved that the second parent population in the particle that comprises plant with initial In Qun, particle is the same, and simply put in order difference.

Step S4: each particle in described lower second parent population is performed the second computing operator, it is thus achieved that by Z particle group The second filial generation population become, merges described second filial generation population and described second parent population, it is thus achieved that second merges population, to institute Stating each particle in the second merging population and carry out selecting operation, the molecular population of Z grain that will select, as performing the 3rd 3rd parent population of computing operator.

Step S5: each particle in described 3rd parent population is performed the 3rd computing operator, it is thus achieved that be made up of Z particle F3 population, merge described F3 population and described 3rd parent population, it is thus achieved that the 3rd merge population, to described 3rd merges each particle in population carries out selecting operation, and the molecular population of Z grain that will select, as performing the first fortune Calculate the first parent population of operator.

Step S6: if current iteration number of times is more than described maximum iteration time, then output maximum iteration time corresponding the One parent population, if described current iteration number of times is less than or equal to described maximum iteration time, returns step S3.

The multi-region of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution that the embodiment of the present invention provides In the economic environment method for optimizing scheduling of territory, by the first computing operator, the second computing operator and the 3rd computing operator continuous One parent population, the second parent population and the 3rd parent population are iterated such that it is able to obtain and have number of different types The population of particle, i.e. solution, thus improve quanta particle swarm optimization in prior art and be easily trapped into lacking of local optimum Point, improves the effect of multizone economic environment scheduling.

By the first computing operator, the second computing operator and continuous first father of the 3rd computing operator in the embodiment of the present application It is iterated for population, the second parent population and the 3rd parent population so that by original particle optimal solution and overall situation particle The competitive strategy of excellent solution changes Pareto optimality forward position competitive strategy into so that quantum particle swarm is easily trapped into Pareto local optimum The shortcoming solved is improved so that the particle tried to achieve, closer to Pareto globally optimal solution, is distributed more in Pareto disaggregation The uniform loss being simultaneously effectively prevented Pareto optimal solution, improves the effect of multizone economic environment scheduling.

The embodiment of the present application additionally provides a kind of multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Multizone economic environment method for optimizing scheduling in a kind of realization side method of step S2, the method includes:

Determine the one group of P meeting said two object function that each particle includes_ijAnd T_ijNumber sum be D, i.e.Described number D is referred to as space dimensionality.

Determine that the number that described initial population comprises particle is Z.

Determine the first computing operator, the second computing operator and the 3rd computing operator.

P in random initializtion said two object function in D dimension space_ijValue and T_irValue, it is thus achieved that include Z grain Initial population A of son, wherein, i-th particle is A_i=[P_i1,P_i2,...P_iD], i ∈ (1, Z).

According to initial population A described in below equation stochastic generation.

$\left\{\begin{array}{c}{P}_{ij}=P_{ij}^{\min }+(P_{ij}^{\max }-P_{ij}^{\min }).rand(0,1),i=1,2,3,\mathrm{...},M,j=1,2,3,\mathrm{...},Mi\\ T_{ir}=T_{ir}^{\min }+(T_{ir}^{\max }-T_{ir}^{\min }).rand(0,1),i=1,2,3,\mathrm{...},N,r=1,2,3,\mathrm{...},N,i\≠r\end{array}\right.$

Rand (0,1) is the random number of 0～1.

The embodiment of the present application additionally provides a kind of multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Multizone economic environment method for optimizing scheduling in a kind of realization side method of step S3, the method includes:

Obtain described first parent population.

Each particle in described first parent population is performed the operation of the first computing operator: son1 (i :)=r1* Parent1 (i :)+(1-r1) * parentbest1, wherein r1 is the random number of 0～1, parent1 (i :) represent described first I-th particle in parent population, son1 (i :) represents i-th particle in described F1 population, son1 (i :) for institute State described F1 population obtained after the i-th particle in the first parent population performs the most described first computing operator In i-th particle, parent1 is described first parent population, parentbest1 be described first parent population is positioned at non- Arbitrary particle in inferior solution the highest grade non-bad layer and the maximum Noninferior Solution Set of crowding distance, crowding distance indicates particle at sky Between distribution congestion state.

Merge described F1 population and described first parent population, it is thus achieved that first merges population, closes described first And each particle carries out selecting operation in population, the molecular population of Z grain that will select, as performing the second computing operator The second parent population.

The embodiment of the present application additionally provides a kind of multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Multizone economic environment method for optimizing scheduling in a kind of realization side method of step S4, the method includes:

Obtain described second parent population.

Generate one 0～the random number R 2 of 1, the most described random number R 2 and crossover probability set in advance probability2；

If R2 > probability2, for the described second computing operator of each particle execution in described second parent population:

Son2 (i :) and=parent2 (i :)+(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

If R2 < probability2, for the described second computing operator of each particle execution in described second parent population:

Son (i :) and=parent2 (i :)-(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

Wherein, parent2 (i :) represents i-th particle in described second parent population, and son2 (i :) represent described I-th particle in two progeny populations, maxgen is described maximum iteration time, and m is current iterations, son2 (i :) be I-th particle parent (i :) in described second parent population is performed obtained by after the most described second computing operator I-th particle in described second filial generation population, mparentbest is to be positioned at noninferior solution grade in described second parent population Optimum averaged particles in the Noninferior Solution Set of high non-bad layer and crowding distance maximum.

Merge described second filial generation population and described second filial generation parent population, it is thus achieved that second merges population, to described second Merge each particle in population and carry out selecting operation, the molecular population of Z grain that will select, calculate as performing the 3rd computing 3rd parent population of son.

The embodiment of the present application additionally provides a kind of multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Multizone economic environment method for optimizing scheduling in a kind of realization side method of step S5, the method includes:

Obtain described 3rd parent population.

Generate one 0～the random number R 3 of 1, the most described random number R 3 and predetermined crossover probability probability3。

If R3 > probability3, then described 3rd computing is performed for each particle in described 3rd parent population and calculate Son:

S (m)=1-rand () ^ ((1-m/maxgen) ^3)；

Low=parent3 (i :)-s (m) * (parent3 (i :)-field (2 :))；

High=parent3 (i :)+s (m) * (field (1 :)-parent3 (i :))；

Son3 (i :)=rand () * (high-low)+low；

If R3≤probability3, then by described 3rd parent population directly as performing first during non-iteration first First parent population of computing operator.

Wherein, parent3 (i :) represents the i-th particle of described 3rd parent population, and son3 (i :) represent described The i-th particle of three progeny populations, rand () is the random number of 0～1, son (i :) it is in described 3rd parent population I particle parent3 (i :) performs i-th in the described F3 population obtained by after the most described 3rd computing operator Individual particle, maxgen is described maximum iteration time, and m is current iterations, field (2 :) it is each space dimensionality of particle Lower limit, field (1 :) is the upper limit of each space dimensionality of particle.

Merge described F3 population and described third generation parent population, it is thus achieved that the 3rd merges population, to the described 3rd Merging each particle in population to carry out selecting operation, the molecular population of Z grain that will select, as described non-iteration first First parent population of Shi Zhihang the first computing operator.

Refer to Fig. 2, the multiple target variation quanta particle sorted based on quick noninferior solution provided for the embodiment of the present application The multizone economic environment method for optimizing scheduling embodiment of group's algorithm performs to select the method stream of a kind of implementation of operation Cheng Tu, the method includes:

Step S201: determine penalty coefficient f and penalty coefficient g.

Step S202: particle each in merging population is substituted into below equation respectively, calculates each particle in described merging population Respectively corresponding fitness value (F ', E '), described merging population is the first merging population, second merges population or the 3rd and merge kind Group.

$F^{,}=\underset{i=1}{\overset{N}{\&Sigma;}}(\underset{j=1}{\overset{M_i}{\&Sigma;}}{F}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{t}_{ir}{T}_{ir}+|\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir}-P_{Di}|\&times;f);$

$E^{,}=\underset{i=1}{\overset{N}{\&Sigma;}}(\underset{j=1}{\overset{M_i}{\&Sigma;}}{E}_{ij}+|\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir}-P_{Di}|\&times;g).$

Step S203: each particle carried out quick noninferior solution row according to the respectively corresponding fitness value of each particle (F ', E ') Sequence, in order in described merging population each particle be divided into multiple non-bad layer, each non-bad layer comprise one or more have identical non- The particle of inferior solution grade.

Noninferior solution sequence refers to that each particle being combined in population according to noninferior solution level is carried out point according to noninferior solution grade Layer, it is thus achieved that multiple non-bad layers, the particle being positioned at same noninferior solution has identical noninferior solution grade, and noninferior solution sequence serves finger Draw search to advance to Pratuo optimal solution set direction.Noninferior solution sequence is the noninferior solution classification process of a circulation:

First, compare merging each particle is the most corresponding in population fitness value (F ', E '), non-according to fitness value Pecking order is ranked up.

Owing to fitness value includes two values, when sequence, any value can be compared, it is assumed that merge population and include 3 grains Son, the fitness value of particle 1 correspondence is (1,2), and the fitness value of particle 2 correspondence is (2,1), the fitness value of particle 3 correspondence For (4,3), then particle 1 and particle 2 are positioned at same non-bad layer, because particle 1 is not arranged by residual particles 2,3, i.e. planting Can not find a particle in Qun and meet two fitness values the most all ratio little particles of particle 1 itself, then particle 1 is called non- Join solution, it is possible to be referred to as noninferior solution, in like manner, satisfied two fitness also be can not find simultaneously than 2, particle for particle 2 in this population Body wants little particle, therefore particle 2 is also noninferior solution, it will be apparent that, 3 two fitness values of particle are all big than particle 2 and particle 1, I.e. can look for the particle less than two fitness values of particle 3 itself in this population, therefore particle 3 solves (inferior solution) for domination, The disaggregation that in current population, all noninferior solution set are constituted is referred to as non-bad layer, therefore in this illustrates, particle 1 particle 2 constitutes non-bad Layer 1, particle 3 is positioned at non-bad layer 2.

Again, after obtaining the first non-bad layer, from merging, population peels off the particle that the first non-bad layer is corresponding, can obtain surplus Remaining population, finds out Noninferior Solution Set corresponding to residue population for residue population by same method and constitutes non-bad layer 2, by that analogy Until it is complete to merge population layering.

Can be by the first non-bad layer F1 (the first non-bad layer F1 is to merge the non-bad layer that in population, noninferior solution the highest grade) All particles give noninferior solution sequence i_rank1(wherein i_rank1For the noninferior solution sequence of particle i, the i of the first non-bad layer_rank1Value is 1, should The non-all particle of bad layer is all endowed noninferior solution sequence i_rank1, the i of the second non-bad layer_rank2Value is 2, and this non-all particle of bad layer is equal It is endowed noninferior solution sequence i_rank2, by that analogy), and peel off from merging population, then proceed to find out residue grain in merging population First non-bad layer of son is designated as merging the second non-bad layer F2 of population, i.e. obtains the Noninferior Solution Set of residual particles, the second non-bad layer The all particles of F2 kind are endowed noninferior solution sequence i_rank2, according to this rule, particle all of in merging population is all layered complete, with Particle in one non-bad layer has identical noninferior solution sequence i_rank。

Step S204: for each non-bad layer, calculate the crowding distance of each particle, by crowded according to its correspondence of each particle Distance is ranked up from big to small, obtains population of sorting, and described crowding distance indicates the particle congestion state in spatial distribution.

In order to have identical i_rankParticle in carry out selectivity sequence, introduce crowding distance concept.I-th The crowding distance of particle is that (after each particle substitutes into two object functions, can obtain fitness value, space is by 2Z grain in space The fitness value composition that son is corresponding) upper distance between i+1 particle and the i-th-1 particle adjacent with i-th particle, For its calculation procedure of each space it is:

The particle of same non-bad layer is initialized distance.Make L [i]_d=0 (wherein L [i]_dRepresent gathering around of any i-th particle Squeeze distance).

To same non-bad layer particle, by m-th fitness value ascending sort, (in the embodiment of the present application, fitness value includes F Value and E value, be referred to as the 1st fitness value by F value here, E value is referred to as the 2nd fitness value, if also increasing other kinds of During object function, such as, increase object function G, the most also include the 3rd fitness value G), m ∈ (1, Q), wherein Q is object function Number (the embodiment of the present application includes two object functions, economic goal function F and emissions object function E, i.e. Q=2, if Adding other kinds of object function, the value of Q can change therewith).After making to merge the middle sequence of population (including 2Z particle) It is positioned at the particle on edge (be i.e. ordered as 1 and be ordered as the particle of 2Z) and there is selection advantage, will be located in the grain on edge The crowding distance of son is entered as Inf (Inf is the infinitely great number in MATLAB).The grain of centre it is positioned at after being combined in population sequence Son, seeks crowding distance L [i]_dm=(L [i+1]_m—L[i-1]_m)/(f_mmax—f_mmin), (m ∈ (1, Q)), wherein f_mmaxRefer to m The maximum of individual fitness value, f_mminRefer to the minima of m-th fitness value, L [i]_dmRefer to according to m-th fitness value liter After sequence sequence, the crowding distance of the i-th particle obtained.To other fitness values also repeat the above steps, thus phase will be obtained The crowding distance of the i-th particle answered (such as, carry out ascending sort, through above-mentioned steps by second fitness value E, it is thus achieved that The crowding distance L [i] that i-th particle is corresponding_d2).The final crowding distance of i-th particle is L [i]_d=L [i]_d1+ ... ,+L [i]_dQ, in the embodiment of the present application, Q=2, certainly, after the number increasing object function, Q-value also can increase accordingly.

By the particle that prioritizing selection crowding distance is bigger, result of calculation can be made more uniform at spatial distribution ratio, to maintain Merge the multiformity of particle in population.

Step S205: obtain front Z particle from described sequence population, by Z particle composition front in described sequence population Population, perform the parent population of computing operator during next iteration, described parent population is described first parent population, described Second parent population or described 3rd parent population, described computing operator is described first corresponding with described first parent population The institute that described second computing operator that computing operator is corresponding with described second parent population is corresponding with described 3rd parent population State the 3rd computing operator.

By elitism strategy in the embodiment of the present application, the most preferentially choose merging population (merge population be the first merging population, the Two merge population or the 3rd merges population) in, the particle that affiliated non-bad layer is forward.Such as merge population include the first non-bad layer, Second non-bad layer and the 3rd non-bad layer, it is assumed that Z is 3, if the first non-bad layer has two particles, it is clear that the two particle all can be selected Select, if the second non-bad layer has two particles, then need to select a particle in the second non-bad layer, now can choose second non- The particle that in bad layer, crowding distance is bigger.

Elitism strategy i.e. retains the excellent particle merged in population, and by Z excellent particle one population of composition, by this kind Group is as progeny population (progeny population is F1 population, second filial generation population or F3 population), to prevent handkerchief from tiring out The loss of torr optimal solution.

The embodiment of the present application additionally provides a kind of multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution Multizone economic environment optimizing scheduling device, this device includes: build module, the first acquisition module, the second acquisition module, the Three acquisition modules, the 4th acquisition module, return module, wherein:

Building module, for setting up the economic environment Scheduling Optimization Model of multizone, the economic environment of described multizone is adjusted Degree Optimized model includes two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows:

$F=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\&Sigma;}}\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}1/2\&CenterDot;t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{\mathrm{\&alpha;}}_{ij}+{\mathrm{\&beta;}}_{ij}{P}_{ij}+{\mathrm{\&gamma;}}_{ij}{P}_{ij}^2+{\mathrm{\&xi;}}_{ij}\exp \left({\mathrm{\&lambda;}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is i-th district respectively The cost coefficient of the jth platform electromotor in territory, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij (P_ij) it is the discharge function of the jth platform electromotor of ith zone, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth of ith zone respectively The emission factor of platform electromotor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt is The actual power that the jth platform electromotor in i region is sent；I, r and j are the positive integer more than or equal to 1；

Described inequality constraints condition is respectively as follows:

Wherein,The respectively jth platform electromotor of ith zone can send Peak power and minimum power；

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the most whole more than or equal to 1 Number, r is the positive integer more than or equal to 1；

Wherein, T_ir,min T_ir,maxIt is the minimum power from ith zone through-put power to r region and maximum respectively Power；

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}=P_{Di}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt is to be transferred to the r district from ith zone by interconnection The power in territory.

First acquisition module, initializes said two mesh for economic environment Scheduling Optimization Model based on described multizone P in scalar functions_ijValue and T_irValue, it is thus achieved that comprising the initial population of Z particle, each described particle includes meeting described two One group of P of individual object function_ijAnd T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing are calculated Son and the 3rd computing operator, using described initial population as the first parent population of iteration first, Z is the most whole more than or equal to 2 Number.

Second acquisition module, for performing the first computing operator by each particle in described first parent population, it is thus achieved that by Z Individual grain molecular F1 population, merges described F1 population and described first parent population, it is thus achieved that first merges Population, merges each particle in population and carries out selecting operation, the molecular population of Z grain that will select, make described first For performing the second parent population of the second computing operator；

3rd acquisition module, is used for each particle in described lower second parent population is performed the second computing operator, it is thus achieved that By Z grain molecular second filial generation population, merge described second filial generation population and described second parent population, it is thus achieved that second closes And population, merge each particle in population to described second and carry out selecting to operate, the molecular population of Z grain that will select, As the 3rd parent population performing the 3rd computing operator；

4th acquisition module, for performing the 3rd computing operator by each particle in described 3rd parent population, it is thus achieved that by Z Individual grain molecular F3 population, merges described F3 population and described 3rd parent population, it is thus achieved that the 3rd merges Population, merges each particle in population and carries out selecting operation, the molecular population of Z grain that will select, make the described 3rd The first parent population of the first computing operator is performed during for non-iteration first；

Return module, if for current iteration number of times more than described maximum iteration time, then export described greatest iteration time Number corresponding the first parent population, if described current iteration number of times is less than or equal to described maximum iteration time, return second Acquisition module.

The multizone economic environment scheduling of the above-mentioned multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution The first acquisition module optimized in device may include that

First determines unit, for determining the one group of P meeting said two object function that each particle includes_ijAnd T_ij Number sum be D, i.e.Described number D is referred to as space dimensionality.

Second determines unit, for determining that the number that described initial population comprises particle is Z.

3rd determines unit, is used for determining the first computing operator, the second computing operator and the 3rd computing operator；

First acquiring unit, for P in random initializtion said two object function in D dimension space_ijValue and T_ir's Value, it is thus achieved that including initial population A of Z particle, wherein, i-th particle is A_i=[P_i1,P_i2,...P_iD], i ∈ (1, Z).

First signal generating unit, for according to initial population A described in below equation stochastic generation.

$\left\{\begin{array}{c}{P}_{ij}=P_{ij}^{\min }+(P_{ij}^{\max }-P_{ij}^{\min }).rand(0,1),i=1,2,3,\mathrm{...},M,j=1,2,3,\mathrm{...},Mi\\ T_{ir}=T_{ir}^{\min }+(T_{ir}^{\max }-T_{ir}^{\min }).rand(0,1),i=1,2,3,\mathrm{...},M,r=1,2,3,\mathrm{...},M,i\&NotEqual;r\end{array}\right.$

Rand (0,1) is the random number of 0～1.

The multizone economic environment scheduling of the above-mentioned multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution The second acquisition module optimized in device may include that

Second acquisition unit, is used for obtaining described first parent population.

Performance element, for each particle in described first parent population being performed the operation of the first computing operator: Son1 (i :) and=r1*parent1 (i :)+(1-r1) * parentbest1, wherein r1 is the random number of 0～1, parent1 (i :) represents i-th particle in described first parent population, and son1 (i :) represent i-th grain in described F1 population Son, and son1 (i :) it is to obtained after the i-th particle the most described first computing operator of execution in described first parent population Described F1 population in i-th particle, parent1 is described first parent population, and parentbest1 is described One parent population is positioned at arbitrary particle in the Noninferior Solution Set of noninferior solution the highest grade non-bad layer and crowding distance maximum, crowded Distance indicates the particle congestion state in spatial distribution.

3rd acquiring unit, is used for merging described F1 population and described first parent population, it is thus achieved that first merges Population, merges each particle in population and carries out selecting operation, the molecular population of Z grain that will select, make described first For performing the second parent population of the second computing operator.

The multizone economic environment scheduling of the above-mentioned multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution The 3rd acquisition module optimized in device may include that

4th acquiring unit, is used for obtaining described second parent population.

Second signal generating unit, for generating one 0～the random number R 2 of 1, relatively described random number R 2 is with set in advance Crossover probability probability2；

If R2 > probability2, for the described second computing operator of each particle execution in described second parent population:

Son2 (i :) and=parent2 (i :)+(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

If R2 < probability2, for the described second computing operator of each particle execution in described second parent population:

Son (i :) and=parent2 (i :)-(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

Wherein, parent2 (i :) represents i-th particle in described second parent population, and son2 (i :) represent described I-th particle in two progeny populations, maxgen is described maximum iteration time, and m is current iterations, son2 (i :) be I-th particle parent (i :) in described second parent population is performed obtained by after the most described second computing operator I-th particle in described second filial generation population, mparentbest is to be positioned at noninferior solution grade in described second parent population Optimum averaged particles in the Noninferior Solution Set of high non-bad layer and crowding distance maximum.

5th acquiring unit, is used for merging described second filial generation population and described second filial generation parent population, it is thus achieved that second closes And population, merge each particle in population to described second and carry out selecting to operate, the molecular population of Z grain that will select, As the 3rd parent population performing the 3rd computing operator.

The multizone economic environment scheduling of the above-mentioned multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution The 4th acquisition module optimized in device may include that

6th acquiring unit, is used for obtaining described 3rd parent population；

3rd signal generating unit, for generating one 0～the random number R 3 of 1, relatively described random number R 3 is with predetermined Crossover probability probability3；

If R3 > probability3, then described 3rd computing is performed for each particle in described 3rd parent population and calculate Son:

S (m)=1-rand () ^ ((1-m/maxgen) ^3)；

Low=parent3 (i :)-s (m) * (parent3 (i :)-field (2 :))；

High=parent3 (i :)+s (m) * (field (1 :)-parent3 (i :))；

Son3 (i :)=rand () * (high-low)+low；

If R3≤probability3, then described 3rd parent population is described directly as performing during non-iteration first First parent population of the first computing operator.

Wherein, parent3 (i :) represents the i-th particle of described 3rd parent population, and son3 (i :) represent described The i-th particle of three progeny populations, rand () is the random number of 0～1, son (i :) it is in described 3rd parent population I particle parent3 (i :) performs i-th in the described F3 population obtained by after the most described 3rd computing operator Individual particle, maxgen is described maximum iteration time, and m is current iterations, field (2 :) it is each space dimensionality of particle Lower limit, field (1 :) is the upper limit of each space dimensionality of particle；

7th acquiring unit, is used for merging described F3 population and described third generation parent population, it is thus achieved that the 3rd closes And population, merge each particle in population to the described 3rd and carry out selecting to operate, the molecular population of Z grain that will select, The first parent population of described first computing operator is performed during as non-iteration first.

Many for the proposed by the invention multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution of checking Regional economy environment method for optimizing scheduling or the effectiveness of device and superiority, say below by the mode of parameter and curve Bright:

Preset as follows: the quantity of electromotor is 16, carry out subregion according to such as Fig. 3 mode.Include four as shown in Figure 3 Region, four regions are respectively region 1, region 2, region 3 and region 4, and region 1 includes that 4 units, i.e. 4 electromotors (are sent out Motor 1, electromotor 2, electromotor 3 and electromotor 4), the burden requirement of 4 electromotors is 400MW；Region 2 includes 4 units, I.e. 4 electromotors (electromotor 5, electromotor 6, electromotor 7 and electromotor 8), the burden requirement of 4 electromotors is 200MW；District Territory 3 includes 4 units, i.e. 4 electromotors (electromotor 9, electromotor 10, electromotor 11 and electromotor 12), bearing of 4 electromotors Lotus requires as 350MW；Region 4 includes 4 units, i.e. 4 electromotors (electromotor 13, electromotor 14, electromotor 15 and generatings Machine 16), the burden requirement of 4 electromotors is 300MW；The code T 12 of the interconnection in four regions, T21, T13, T31, T14, T41、T23、T32、T24、T42、T43、T34.The power of each interconnection is as shown in table 2.

Assuming Population Size Z=100, maximum iteration time is maxgen=5000, and the penalty coefficient f in four regions is respectively Being 2.5,3.5,2.5 and 1.5, the penalty coefficient g in four regions is respectively 12,17,17 and 12, crossover probability probability2 =0.8, probability3=0.816, the parameter of electromotor such as table 1,

Interconnection parameter such as table 2.

1 16 generator parameters of table

Wherein, a_ij、b_ij、c_ijIt is the cost coefficient of the jth platform electromotor of ith zone respectively, α_ij、β_ij、γ_ij、ξ_ij、 λ_ijIt is the emission factor of the jth platform electromotor of ith zone respectively.Such as, for electromotor 1, a_ij、b_ij、c_ijIt is respectively a₁₁、b₁₁、c₁₁；α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is respectively α₁₁、β₁₁、γ₁₁、ξ₁₁、λ₁₁, other are similar to, repeat no more.

Table 2 dominant eigenvalues parameter

Interconnection is numbered	Tir,min	Tir,max	tir
				T12/T21	-100	100	1
T13/T31	-100	100	1
				T14/T41	-100	100	1
T23/T32	-100	100	1
				T24/T42	-100	100	1

Wherein, T_ir,min T_ir,maxIt is the minimum power from ith zone through-put power to r region and maximum respectively Power, such as T12 refer to that the minimum power from region 1 through-put power to region 2 is-100, and peak power is 100；T21 refers to Being-100 from the minimum power of region 2 through-put power to region 1, peak power is 100, the t that T12 is corresponding_ir, refer to from region 1 The transmission cost coefficient of interconnection between region 2, wherein T12=T21, other are similar to, repeat no more here.

In the present example, use sort based on quick noninferior solution multiple target variation quanta particle swarm optimization (NSVAQPSO, Non-dominated Sorting Aviation Quantum-behaved Particle Swarm Optimization) Multizone economic environment method for optimizing scheduling or device, it is thus achieved that optimum compromise to solve scheduling result as shown in table 3, in order to prove The superiority of NSVAQPSO algorithm, compares Fig. 4 by the optimum results of NSVAQPSO algorithm and other intelligent optimization algorithms Shown in.

The contrast that table 3 NSVAQPSO algorithm solves with the compromise of other algorithm optimum

PSO:Particle Swarm Optimization, particle cluster algorithm；QPSO:Quantum-behaved Particle Swarm Optimization, quanta particle swarm optimization；DEQPSO:Difference Quantum-behaved Particle Swarm Optimization, difference quanta particle swarm optimization；VAQPSO:Aviation Quantum- Behaved Particle Swarm Optimization makes a variation quanta particle swarm optimization.

Optimum compromise solution refers to final population (the i.e. described maximum of last output in step S6 returned after iteration terminates The first parent population that iterations is corresponding) in, according to fuzzy membership principle in prior art, filter out from final population The particle come is exactly that optimum compromise solves.

The unbalanced power degree that the optimum compromise of table 4 solves

Owing to, in multizone Economic Dispatch Problem, the workload demand in each region is a definite value (4 districts as shown in Figure 3 The workload demand in territory), when final population (that i.e. in step S6, the maximum iteration time of last output is corresponding in some region One parent population) in, it is positioned at the workload demand in all generator powers in this region and dominant eigenvalues sum and this region not Strict equal time, both subtract each other and just have a deviation value, and this deviation value is unbalanced power degree.

The contrast of table 5 algorithms of different extremum

Final population (the i.e. maximum iteration time pair of last output in step S6 that extremum i.e. returns after iteration terminates The the first parent population answered) in all particles economic goal function F value in minima and emissions object function E value in Minima.

By table 3 above, table 4 and Fig. 3, Fig. 4 it can be seen that use NSVAQPSO algorithm, to 16 electromotors, 4 region electricity Force system emulates, and the Pareto optimality compromise solution electromotor scheduling result obtained by emulation and unbalanced power degree are such as Shown in table 3 and table 4.NSVAQPSO algorithm and other mentioned algorithm are by rejecting unbalanced power degree as can be seen from Table 3 Exceed in the Pareto forward position of each partition load and total load 0.04%, utilize the Pareto optimality that fuzzy membership principle obtains Compromise majorization of solutions result is expense 7582.228336 $/h, discharge 9142.364138t/h, it is clear that other being better than in table are calculated The optimum results of method.Still further aspect, the general power degree of unbalancedness of NSVAQPSO algorithmic dispatching result and regional power Degree of unbalancedness is as shown in table 4, it can be seen that NSVAQPSO performance in terms of power-balance is the most satisfactory.Therefore, it can Say that NSVAQPSO algorithm performs better than in solution multi-region power system economic environment scheduling problem.

In addition can be seen that the fitness value extremum obtained by NSVAQPSO algorithm is less by table 5 and Fig. 4, this also anticipates It is wider that taste the decision region that obtained Pareto optimality forward position provided.Can be seen that from the Pareto forward position of each algorithm of Fig. 4, Under the same conditions, the Pareto optimality forward position convergence effect of NSVAQPSO algorithm is more preferable, and this is owing to NSVAQPSO algorithm exists Traditional NSQPSO algorithm adds the first computing operator, the second computing operator and the 3rd computing operator so that be absorbed in local In optimum merging population, some particles has an opportunity to break away from local optimum, so that merging population preferably obtain global optimum Pareto forward position.

Finally, in addition it is also necessary to explanation, in this article, the relational terms of such as first and second or the like be used merely to by One entity or operation separate with another entity or operating space, and not necessarily require or imply these entities or operation Between exist any this reality relation or order.And, term " includes ", " comprising " or its any other variant meaning Containing comprising of nonexcludability, so that include that the process of a series of key element, method, article or equipment not only include that A little key elements, but also include other key elements being not expressly set out, or also include for this process, method, article or The key element that equipment is intrinsic.In the case of there is no more restriction, statement " including ... " key element limited, do not arrange Except there is also other identical element in including the process of described key element, method, article or equipment.

In this specification, each embodiment uses the mode gone forward one by one to describe, and what each embodiment stressed is and other The difference of embodiment, between each embodiment, identical similar portion sees mutually.

Described above to the disclosed embodiments, makes professional and technical personnel in the field be capable of or uses the application. Multiple amendment to these embodiments will be apparent from for those skilled in the art, as defined herein General Principle can realize in the case of without departing from spirit herein or scope in other embodiments.Therefore, the application It is not intended to be limited to the embodiments shown herein, and is to fit to and principles disclosed herein and features of novelty phase one The widest scope caused.

Claims (11)

1. the multizone economic environment scheduling of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution is excellent Change method, it is characterised in that including:

S1, setting up the economic environment Scheduling Optimization Model of multizone, the economic environment Scheduling Optimization Model of described multizone includes Two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows:

$F=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{r=1,r\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}1/2\&CenterDot;t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{ij}+{\mathrm{\&beta;}}_{ij}{P}_{ij}+{\mathrm{\&gamma;}}_{ij}{P}_{ij}^2+{\mathrm{\&xi;}}_{ij}\exp \left({\mathrm{\&lambda;}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is the of ith zone respectively The cost coefficient of j platform electromotor, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij(P_ij) it is i-th The discharge function of the jth platform electromotor in individual region, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth platform electromotor of ith zone respectively Emission factor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt it is ith zone The actual power that jth platform electromotor is sent；I, r and j are the positive integer more than or equal to 1；

Described inequality constraints condition is respectively as follows:

Wherein,It is respectively the maximum that the jth platform electromotor of ith zone can send Power and minimum power；

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the positive integer more than or equal to 1；

Wherein, T_ir,minT_ir,maxIt is the minimum power from ith zone through-put power to r region and peak power respectively；

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}=P_{Di}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt it is the merit being transferred to r region from ith zone by interconnection Rate；

S2, economic environment Scheduling Optimization Model based on described multizone initialize P in said two object function_ijValue and T_ir Value, it is thus achieved that comprising the initial population of Z particle, each described particle includes the one group of P meeting said two object function_ijWith T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing operator and the 3rd computing operator, by institute Stating the initial population the first parent population as iteration first, Z is the positive integer more than or equal to 2；

S3, each particle in described first parent population is performed the first computing operator, it is thus achieved that by Z molecular first son of grain For population, merge described F1 population and described first parent population, it is thus achieved that first merges population, merges described first In population, each particle carries out selecting operation, the molecular population of Z grain that will select, and calculates as performing described second computing Second parent population of son；

S4, each particle in described lower second parent population is performed the second computing operator, it is thus achieved that by Z grain molecular second Progeny population, merges described second filial generation population and described second parent population, it is thus achieved that second merges population, closes described second And each particle carries out selecting operation in population, the molecular population of Z grain that will select, as performing described 3rd computing 3rd parent population of operator；

S5, each particle in described 3rd parent population is performed the 3rd computing operator, it is thus achieved that sub by Z grain the molecular 3rd For population, merge described F3 population and described 3rd parent population, it is thus achieved that the 3rd merges population, merges the described 3rd In population, each particle carries out selecting operation, and the molecular population of Z grain that will select, as performing institute during non-iteration first State the first parent population of the first computing operator；

If S6 current iteration number of times is more than described maximum iteration time, then the first parent kind that output maximum iteration time is corresponding Group, if described current iteration number of times is less than or equal to described maximum iteration time, returns step S3.

The multizone warp of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 1 Ji environment method for optimizing scheduling, it is characterised in that described step S2 includes:

S21, determine the one group of P meeting said two object function that each particle includes_ijAnd T_ijNumber sum be D i.e.Described number D is referred to as space dimensionality；

S22, determine that the number that described initial population comprises particle is Z；

S23, determine the first computing operator, the second computing operator and the 3rd computing operator；

S24, in D dimension space P in random initializtion said two object function_ijValue and T_irValue, it is thus achieved that include Z particle Initial population A, wherein, i-th particle is A_i=[P_i1,P_i2,...P_iD], i ∈ (1, Z)；

S25, according to initial population A described in below equation stochastic generation:

$\left\{\begin{array}{c}{P}_{ij}=P_{ij}^{\min }+\left(P_{ij}^{\max }-P_{ij}^{\min }\right).rand\left(0,1\right),i=1,2,3,...,N,j=1,2,3,...,Mi\\ T_{ir}=T_{ir}^{\min }+\left(T_{ir}^{\max }-T_{ir}^{\min }\right).rand\left(0,1\right),i=1,2,3,...,N,r=1,2,3,...,N,i\&NotEqual;r\end{array}\right.$

Rand (0,1) is the random number of 0～1.

The multizone warp of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 1 Ji environment method for optimizing scheduling, it is characterised in that described step S3 includes:

S31, obtain described first parent population；

S32, each particle in described first parent population is performed the operation of the first computing operator: son1 (i :)=r1* Parent1 (i :)+(1-r1) * parentbest1, wherein r1 is the random number of 0～1, parent1 (i :) represent described first I-th particle in parent population, son1 (i :) represents i-th particle in described F1 population, son1 (i :) for institute State described F1 population obtained after the i-th particle in the first parent population performs the most described first computing operator In i-th particle, parent1 is described first parent population, parentbest1 be described first parent population is positioned at non- Arbitrary particle in inferior solution the highest grade non-bad layer and the maximum Noninferior Solution Set of crowding distance, crowding distance indicates particle at sky Between distribution congestion state；

S33, merge described F1 population and described first parent population, it is thus achieved that first merges population, closes described first And each particle carries out selecting operation in population, the molecular population of Z grain that will select, as performing described second computing Second parent population of operator.

4. according to the arbitrary described multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution of claims 1 to 3 Multizone economic environment method for optimizing scheduling, it is characterised in that described step S4 includes:

S41, obtain described second parent population；

S42, generation one 0～the random number R 2 of 1, the most described random number R 2 and crossover probability set in advance probability2；

If R2 > probability2, for the described second computing operator of each particle execution in described second parent population:

Son2 (i :) and=parent2 (i :)+(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

If R2 < probability2, for the described second computing operator of each particle execution in described second parent population:

Son (i :) and=parent2 (i :)-(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

Wherein, parent2 (i :) represents i-th particle in described second parent population, and son2 (i :) represent described second son For i-th particle in population, maxgen is described maximum iteration time, and m is current iterations, and son2 (i :) for institute That states that i-th particle parent in the second parent population (i :) performs obtained by after the most described second computing operator is described I-th particle in second filial generation population, mparentbest is that be positioned at noninferior solution the highest grade in described second parent population Optimum averaged particles in the Noninferior Solution Set of non-bad layer and crowding distance maximum；

S43, merge described second filial generation population and described second filial generation parent population, it is thus achieved that second merges population, to described second Merging each particle in population and carry out selecting operation, the molecular population of Z grain that will select, as performing described 3rd fortune Calculate the 3rd parent population of operator.

5. according to the arbitrary described multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution of claims 1 to 3 Multizone economic environment method for optimizing scheduling, it is characterised in that described step S5 includes:

S51, obtain described 3rd parent population；

S52, generation one 0～the random number R 3 of 1, the most described random number R 3 and predetermined crossover probability probability3；

If R3 > probability3, then for the described 3rd computing operator of each particle execution in described 3rd parent population:

S (m)=1-rand () ^ ((1-m/maxgen) ^3)；

Low=parent3 (i :)-s (m) * (parent3 (i :)-field (2 :))；

High=parent3 (i :)+s (m) * (field (1 :)-parent3 (i :))；

Son3 (i :)=rand () * (high-low)+low；

If R3≤probability3, then by described 3rd parent population directly as non-iteration first time perform described first First parent population of computing operator；

Wherein, parent3 (i :) represents the i-th particle of described 3rd parent population, and son3 (i :) represent described 3rd son For the i-th particle of population, rand () is the random number of 0～1, and son (i :) it is to the i-th in described 3rd parent population Particle parent3 (i :) performs the i-th grain in the described F3 population obtained by after the most described 3rd computing operator Son, maxgen is described maximum iteration time, and m is current iterations, field (2 :) it is under each space dimensionality of particle Limit, field (1 :) is the upper limit of each space dimensionality of particle；

S53, merge described F3 population and described third generation parent population, it is thus achieved that the 3rd merges population, to the described 3rd Merging each particle in population to carry out selecting operation, the molecular population of Z grain that will select, as described non-iteration first First parent population of the first computing operator described in Shi Zhihang.

The multizone warp of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 1 Ji environment method for optimizing scheduling, it is characterised in that described execution selects operation to include:

Determine penalty coefficient f and penalty coefficient g；

Particle each in merging population is substituted into below equation respectively, calculates the adaptation that in described merging population, each particle is the most corresponding Angle value (F ', E '), described merging population is the first merging population, the second merging population or the 3rd merging population；

$F^{,}=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\left(\underset{j=1}{\overset{M_i}{\mathrm{\&Sigma;}}}{F}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}{t}_{ir}{T}_{ir}+|\underset{j=1}{\overset{M_i}{\mathrm{\&Sigma;}}}{P}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}{T}_{ir}-P_{Di}|\&times;f\right);$

$E^{,}=\underset{i=1}{\overset{N}{\&Sigma;}}(\underset{j=1}{\overset{M_i}{\&Sigma;}}{E}_{ij}+|\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir}-P_{Di}|\&times;g);$

According to the respectively corresponding fitness value of each particle (F ', E '), each particle is carried out quick noninferior solution sequence, in order to described conjunction And each particle is divided into multiple non-bad layer, each non-bad layer to comprise one or more grain with identical noninferior solution grade in population Son；

For each non-bad layer, calculate the crowding distance of each particle, by each particle according to its correspondence crowding distance from big to small Being ranked up, obtain population of sorting, described crowding distance indicates the particle congestion state in spatial distribution；

From described sequence population, obtain front Z particle, by front Z the molecular population of grain in described sequence population, as under Performing the parent population of computing operator during an iteration, described parent population is described first parent population, described second parent Population or described 3rd parent population, described first computing that described computing operator is corresponding with described first parent population is calculated The son described second computing operator corresponding with described second parent population corresponding with described 3rd parent population the described 3rd Computing operator.

7. the multizone economic environment scheduling of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution is excellent Gasifying device, it is characterised in that including:

Building module, for setting up the economic environment Scheduling Optimization Model of multizone, the economic environment scheduling of described multizone is excellent Change model and include two object functions, two inequality constraints conditions and equality constraint, wherein:

Two target function types are respectively as follows:

$F=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{F}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{Mi}{\&Sigma;}}{a}_{ij}+b_{ij}{P}_{ij}+c_{ij}{P}_{ij}^2+\underset{i=1}{\overset{N}{\&Sigma;}}\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}1/2\&CenterDot;t_{ir}\left|T_{ir}\right|$

$E=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{E}_{ij}\left(P_{ij}\right)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{Mi}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{ij}+{\mathrm{\&beta;}}_{ij}{P}_{ij}+{\mathrm{\&gamma;}}_{ij}{P}_{ij}^2+{\mathrm{\&xi;}}_{ij}\exp \left({\mathrm{\&lambda;}}_{ij}{P}_{ij}\right)$

Wherein F_ij(P_ij) it is the cost function of the jth platform electromotor of ith zone, a_ij、b_ij、c_ijIt is the of ith zone respectively The cost coefficient of j platform electromotor, t_irIt is the i-th region to the transmission cost coefficient of interconnection, E between r region_ij(P_ij) it is i-th The discharge function of the jth platform electromotor in individual region, α_ij、β_ij、γ_ij、ξ_ij、λ_ijIt is the jth platform electromotor of ith zone respectively Emission factor, N is the quantity in the region of described multizone, M_iIt is the electromotor quantity of ith zone, P_ijIt it is ith zone The actual power that jth platform electromotor is sent；I, r and j are the positive integer more than or equal to 1；

Described inequality constraints condition is respectively as follows:

Wherein,It is respectively the maximum that the jth platform electromotor of ith zone can send Power and minimum power；

T_ir,min≤T_ir≤T_ir,max, i=1,2 ...., N, r=1,2 ... ..N, i ≠ r, N are the positive integer more than or equal to 1；

Wherein, T_ir,minT_ir,maxIt is the minimum power from ith zone through-put power to r region and peak power respectively；

Described equality constraint includes:

$\underset{j=1}{\overset{M_i}{\&Sigma;}}{P}_{ij}=P_{Di}+\underset{r=1,r\&NotEqual;i}{\overset{N}{\&Sigma;}}{T}_{ir},r=1,2,\mathrm{....},N;$

Wherein, P_DiIt is the burden requirement of ith zone, T_irIt it is the merit being transferred to r region from ith zone by interconnection Rate；

First acquisition module, initializes said two target letter for economic environment Scheduling Optimization Model based on described multizone P in number_ijValue and T_irValue, it is thus achieved that comprising the initial population of Z particle, each described particle includes meeting said two mesh One group of P of scalar functions_ijAnd T_irValue, and pre-set maximum iteration time, the first computing operator, the second computing operator and 3rd computing operator, using described initial population as the first parent population of iteration first, Z is the positive integer more than or equal to 2；

Second acquisition module, for performing the first computing operator by each particle in described first parent population, it is thus achieved that by Z grain Molecular F1 population, merges described F1 population and described first parent population, it is thus achieved that first merges population, Merging each particle in population to carry out selecting operation to described first, the molecular population of Z grain that will select, as execution Second parent population of described second computing operator；

3rd acquisition module, for performing the second computing operator by each particle in described lower second parent population, it is thus achieved that by Z The molecular second filial generation population of grain, merges described second filial generation population and described second parent population, it is thus achieved that the second merging kind Group, merges each particle in population and carries out selecting to operate described second, the molecular population of Z grain that will select, as Perform the 3rd parent population of described 3rd computing operator；

4th acquisition module, for performing the 3rd computing operator by each particle in described 3rd parent population, it is thus achieved that by Z grain Molecular F3 population, merges described F3 population and described 3rd parent population, it is thus achieved that the 3rd merges population, Merging each particle in population to carry out selecting operation to the described 3rd, the molecular population of Z grain that will select, as non-head The first parent population of described first computing operator is performed during secondary iteration；

Return module, if for current iteration number of times more than described maximum iteration time, then exporting described maximum iteration time pair The the first parent population answered, if described current iteration number of times is less than or equal to described maximum iteration time, returns described second and obtains Delivery block.

The multizone warp of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 7 Ji environment optimizing scheduling device, it is characterised in that described first acquisition module includes:

First determines unit, for determining the one group of P meeting said two object function that each particle includes_ijAnd T_ijNumber Sum is D, i.e.Described number D is referred to as space dimensionality；

Second determines unit, for determining that the number that described initial population comprises particle is Z；

3rd determines unit, is used for determining the first computing operator, the second computing operator and the 3rd computing operator；

First acquiring unit, for P in random initializtion said two object function in D dimension space_ijValue and T_irValue, obtain Must include initial population A of Z particle, wherein, i-th particle is A_i=[P_i1,P_i2,...P_iD], i ∈ (1, Z)；

First signal generating unit, for according to initial population A described in below equation stochastic generation:

$\left\{\begin{array}{c}{P}_{ij}=P_{ij}^{\min }+\left(P_{ij}^{\max }-P_{ij}^{\min }\right).rand\left(0,1\right),i=1,2,3,...,N,j=1,2,3,...,Mi\\ T_{ir}=T_{ir}^{\min }+\left(T_{ir}^{\max }-T_{ir}^{\min }\right).rand\left(0,1\right),i=1,2,3,...,N,r=1,2,3,...,N,i\&NotEqual;r\end{array}\right.$

Rand (0,1) is the random number of 0～1.

The multizone warp of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 7 Ji environment optimizing scheduling device, it is characterised in that described second acquisition module includes:

Second acquisition unit, is used for obtaining described first parent population；

Performance element, for performing the operation of the first computing operator: son1 to each particle in described first parent population (i :) and=r1*parent1 (i :)+(1-r1) * parentbest1, wherein r1 is the random number of 0～1, parent1 (i :) table Show i-th particle in described first parent population, and son1 (i :) represent i-th particle in described F1 population, son1 (i :) is for described in performing the i-th particle in described first parent population obtained by after the most described first computing operator I-th particle in F1 population, parent1 is described first parent population, and parentbest1 is described first parent Population is positioned at arbitrary particle in the Noninferior Solution Set of noninferior solution the highest grade non-bad layer and crowding distance maximum, crowding distance table Understand the particle congestion state in spatial distribution；

3rd acquiring unit, is used for merging described F1 population and described first parent population, it is thus achieved that the first parent population, Carrying out each particle in described first parent population selecting operation, the molecular population of Z grain that will select, as execution Second parent population of described second computing operator.

The multizone of the multiple target variation quanta particle swarm optimization sorted based on quick noninferior solution the most according to claim 7 Economic environment optimizing scheduling device, it is characterised in that described 3rd acquisition module includes:

4th acquiring unit, is used for obtaining described second parent population；

Second signal generating unit, for generating one 0～the random number R 2 of 1, relatively described random number R 2 is intersected with set in advance Probability p robability2；

If R2 > probability2, for the described second computing operator of each particle execution in described second parent population:

Son2 (i :) and=parent2 (i :)+(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

If R2 < probability2, for the described second computing operator of each particle execution in described second parent population:

Son (i :) and=parent2 (i :)-(0.5+log10 ((1+2*abs (maxgen-m)/maxgen))) * abs (mparentbest2-parent2 (i :)) * log (1/u)；

Wherein, parent2 (i :) represents i-th particle in described second parent population, and son2 (i :) represent described second son For i-th particle in population, maxgen is described maximum iteration time, and m is current iterations, and son2 (i :) for institute That states that i-th particle parent in the second parent population (i :) performs obtained by after the most described second computing operator is described I-th particle in second filial generation population, mparentbest is that be positioned at noninferior solution the highest grade in described second parent population Optimum averaged particles in the Noninferior Solution Set of non-bad layer and crowding distance maximum；

5th acquiring unit, is used for merging described second filial generation population and described second filial generation parent population, it is thus achieved that the second merging kind Group, merges each particle in population and carries out selecting to operate described second, the molecular population of Z grain that will select, as Perform the 3rd parent population of described 3rd computing operator.

The multizone of the 11. multiple target variation quanta particle swarm optimizations sorted based on quick noninferior solution according to claim 7 Economic environment optimizing scheduling device, it is characterised in that described 4th acquisition module includes:

6th acquiring unit, is used for obtaining described 3rd parent population；

3rd signal generating unit, for generating one 0～the random number R 3 of 1, relatively described random number R 3 is intersected with predetermined Probability p robability3；

If R3 > probability3, then for the described 3rd computing operator of each particle execution in described 3rd parent population:

S (m)=1-rand () ^ ((1-m/maxgen) ^3)；

Low=parent3 (i :)-s (m) * (parent3 (i :)-field (2 :))；

High=parent3 (i :)+s (m) * (field (1 :)-parent3 (i :))；

Son3 (i :)=rand () * (high-low)+low；

If R3≤probability3, then by described 3rd parent population directly as non-iteration first time perform described first First parent population of computing operator；

Wherein, parent3 (i :) represents the i-th particle of described 3rd parent population, and son3 (i :) represent described 3rd son For the i-th particle of population, rand () is the random number of 0～1, and son (i :) it is to the i-th in described 3rd parent population Particle parent3 (i :) performs the i-th grain in the described F3 population obtained by after the most described 3rd computing operator Son, maxgen is described maximum iteration time, and m is current iterations, field (2 :) it is under each space dimensionality of particle Limit, field (1 :) is the upper limit of each space dimensionality of particle；

7th acquiring unit, is used for merging described F3 population and described third generation parent population, it is thus achieved that the 3rd merging kind Group, merges each particle in population and carries out selecting to operate the described 3rd, the molecular population of Z grain that will select, as The first parent population of described first computing operator is performed during non-iteration first.