CN104463374A - Method and system for optimal configuration of distributed power source - Google Patents

Abstract

The invention provides a method for optimal configuration of a distributed power source. The method comprises the steps that a probability density function of the distributed power source and a probability density function of a load are determined; an uncertainty mathematic model of the distributed power source and the load is built according to the probability density function of the distributed power source and the probability density function of the load; an objective function and a constraint condition for optimal configuration of the distributed power source are determined, and a distributed power source optimal configuration model relevant to uncertainty is built; the distributed power source optimal configuration model relevant to uncertainty is solved according to a preset method, and a Pareto optimal solution is obtained; a standardization satisfaction degree value corresponding to the Pareto optimal solution is calculated, and an optical configuration scheme of the distributed power source is determined according to the standardization satisfaction degree value. Uncertainty in optical configuration of the distributed power source can be effectively handled, the correlation of all objective functions is fully considered, and the strong pertinence of the optimal scheme is effectively avoided.

Description

The method and system that a kind of distributed power source is distributed rationally

Technical field

The application relates to distributed generation technology field, particularly a kind of distributed power source method and system of distributing rationally.

Background technology

Along with the development of technology, people require more and more higher to the method that Distributed Generation in Distribution System is distributed rationally.

In the method that existing Distributed Generation in Distribution System is distributed rationally, generally all have ignored the uncertain impact that distributed power source is distributed rationally, and the model set up is more single, poor accuracy.

Therefore, how can carry out distributing rationally of distributed power source is accurately the current technical issues that need to address of those skilled in the art.

Summary of the invention

Technical problems to be solved in this application are to provide the method and system that a kind of distributed power source is distributed rationally, solve in prior art and generally all have ignored the uncertain impact that distributed power source is distributed rationally, and the model set up is more single, the problem of poor accuracy.

Its concrete scheme is as follows:

The method that distributed power source is distributed rationally, is characterized in that,

Determine the probability density function of distributed power source and the probability density function of load;

According to the probability density function of described distributed power source and the probability density function of described load, set up the uncertain mathematics model of distributed power source and load;

Determine the objective function that distributed power source is distributed rationally and constraint condition, set up and take into account probabilistic distributed power source Optimal Allocation Model;

Utilize the Monte Carlo simulation based on Latin Hypercube Sampling to embed multi-objective particle swarm algorithm to take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution;

The standardization calculating described Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.

Above-mentioned method, preferably,

Described utilization embeds multi-objective particle swarm algorithm based on the Monte Carlo simulation of Latin Hypercube Sampling and takes into account probabilistic distributed power source Optimal Allocation Model solve described, obtains Pareto optimum solution, comprising:

Obtain raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale;

Particle is encoded, the speed of particle described in initialization and position;

The Pareto optimum solution adopting outside elite to store to search in iteration, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set;

Using described objective function as the fitness function of multi-objective particle swarm algorithm, adopt the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draw the fitness value of particle;

Based on Pareto dominance relation, determine the individual extreme value of described particle;

According to the global extremum of the method determination population preset;

The inertia weight of described particle, Studying factors, speed and position is upgraded according to the update method preset;

Described iterations t=t+1 is set, repeats described uncertain Load flow calculation, describedly determine individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T;

Export the described Pareto optimum solution that described outside elite concentrates reservation.

Above-mentioned method, preferably,

The described standardization calculating Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value and determines described distributed power source configuration scheme, comprising according to described standardization:

Fuzzy membership function is utilized to represent the satisfaction that in each described Pareto optimum solution, each objective function is corresponding;

According to the computing method preset, the standardization calculating described Pareto optimum solution is satisfied with angle value;

Determine that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

Above-mentioned method, preferably,

Described based on Pareto dominance relation, determine the individual extreme value of described particle, comprising:

Definition Pareto dominance relation;

According to described Pareto dominance relation and Pareto optimum solution, determine the individual extreme value of described particle.

Above-mentioned method, preferably,

The global extremum of the population belonging to the described method determination particle according to presetting, comprising:

Using the candidate collection of described outside elite's collection as described population global extremum, based on described Pareto dominance relation, described outside elite's collection is upgraded;

The described outside elite calculated after upgrading concentrates the crowding distance of each Pareto optimum solution;

Maintain the capacity of described outside elite's collection according to described crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less;

According to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.

The system that distributed power source is distributed rationally, this system comprises:

First determining unit, for the probability density function of the probability density function and load of determining distributed power source;

First sets up unit, for according to the probability density function of described distributed power source and the probability density function of described load, sets up the uncertain mathematics model of distributed power source and load;

Second determining unit, for determining the objective function that distributed power source is distributed rationally and constraint condition, setting up and taking into account probabilistic distributed power source Optimal Allocation Model;

Solve unit, embed multi-objective particle swarm algorithm for utilizing the Monte Carlo simulation based on Latin Hypercube Sampling and take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution;

First computing unit, is satisfied with angle value for the standardization calculating described Pareto optimum solution corresponding, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.

Above-mentioned system, preferably, described in solve unit and comprise:

Acquiring unit, for obtaining raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale;

Coding unit, for encoding to particle, the speed of particle described in initialization and position;

Setup unit, the Pareto optimum solution searched in iteration for adopting outside elite to store, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set;

Second computing unit, for using described objective function as the fitness function of multi-objective particle swarm algorithm, adopts the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draws the fitness value of particle;

3rd determining unit, for based on Pareto dominance relation, determines the individual extreme value of described particle;

4th determining unit, for the global extremum according to the method determination population preset;

First updating block, for upgrading the inertia weight of described particle, Studying factors, speed and position according to the update method preset;

Setting unit, for arranging described iterations t=t+1, repeating described uncertain Load flow calculation, describedly determining individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T;

Output unit, concentrates the described Pareto optimum solution of reservation for exporting described outside elite.

Above-mentioned system, preferably, described first computing unit comprises:

Representing unit, representing for utilizing fuzzy membership function the satisfaction that in each described Pareto optimum solution, each objective function is corresponding;

3rd computing unit, for according to preset computing method, the standardization calculating each described Pareto optimum solution is satisfied with angle value;

5th determining unit, for determining that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

Above-mentioned system, preferably, described 3rd determining unit comprises:

Definition unit, for defining Pareto dominance relation;

6th determining unit, for according to described Pareto dominance relation and Pareto optimum solution, determines the individual extreme value of described particle.

Above-mentioned system, preferably, described 4th determining unit comprises:

Second updating block, for using the candidate collection of described outside elite's collection as described population global extremum, upgrades described outside elite's collection based on described Pareto dominance relation;

4th computing unit, concentrates the crowding distance of each Pareto optimum solution for calculating the described outside elite after renewal;

Maintain unit, for maintaining the capacity of described outside elite's collection according to described crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less;

Choose unit, for according to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.

In the method that a kind of distributed power source that the application provides is distributed rationally, determine the probability density function of distributed power source and the probability density function of load; According to the probability density function of described distributed power source and the probability density function of described load, set up the uncertain mathematics model of distributed power source and load; Determine the objective function that distributed power source is distributed rationally and constraint condition, set up and take into account probabilistic distributed power source Optimal Allocation Model; Take into account probabilistic distributed power source Optimal Allocation Model according to the method preset solve described, obtain Pareto optimum solution; The standardization calculating described Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.Distributed power source Optimal Configuration Method of the present invention, establish consider distributed power source and distribute rationally economy, security and environmental impact Model for Multi-Objective Optimization, actual conditions can be reflected more all sidedly, there is higher accuracy; Monte Carlo simulation based on Latin Hypercube Sampling is embedded multi-objective particle swarm algorithm and carries out model solution, can not only effectively process distributed power source distribute rationally in uncertainty, also take into full account the correlativity between each objective function, effectively prevent the problem that prioritization scheme is with strong points.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme in the embodiment of the present application, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the application, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

Fig. 1 is the process flow diagram of the embodiment of the method that a kind of distributed power source of the application is distributed rationally;

Fig. 2 is the structural representation of the system embodiment that a kind of distributed power source of the application is distributed rationally.

Embodiment

Core of the present invention is to provide the method and system that a kind of distributed power source is distributed rationally, solve in prior art and generally all have ignored uncertain impact of distributing rationally distributed power source, and the model set up is more single, the problem of poor accuracy.

Below in conjunction with the accompanying drawing in the embodiment of the present application, be clearly and completely described the technical scheme in the embodiment of the present application, obviously, described embodiment is only some embodiments of the present application, instead of whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not making the every other embodiment obtained under creative work prerequisite, all belong to the scope of the application's protection.

With reference to figure 1, show the process flow diagram of the embodiment of the method that a kind of distributed power source of the application is distributed rationally, can comprise the following steps:

Step S101: determine the probability density function of distributed power source and the probability density function of load.

Step S102: according to the probability density function of described distributed power source and the probability density function of described load, set up the uncertain mathematics model of distributed power source and load.

The model of aerogenerator to be exerted oneself model two parts composition primarily of Wind speed model and blower fan, and in the model of matching wind speed profile, what be most widely used is Two-parameter Weibull distribution model, and its probability density function f (v) is:

$f\left(v\right)=\frac{k}{c}{\left(\frac{v}{c}\right)}^{k-1}{e}^{-{\left(\frac{v}{c}\right)}^k}---\left(1\right)$

Wherein, v is wind speed, k and c be respectively wind speed Weibull distribute in form parameter and scale parameter, can be similar to by the average of wind speed statistics and standard deviation and calculate.

Relational expression between blower fan output power P and wind speed v is:

Wherein, P _rfor the output rating of blower fan, v _ifor incision wind speed, v _ofor cut-out wind speed, v _rfor wind rating.By the relational expression between the output power of blower fan and wind speed, the output power model of blower fan just can be obtained.

Load has time variation, and general statistics can obtain the ambiguous model of system loading.The Load Probability density function f (P of Normal Distribution _l) be:

$f\left(P_L\right)=\frac{1}{\sqrt{2\mathrm{\π}}{\mathrm{\σ}}_L}{e}^{-{(P_L-P_{\mathrm{Lm}})}^2/2{{\mathrm{\σ}}_L}^2}---\left(3\right)$

Wherein, P _lfor the actual active power of load, P _lmfor the average of load active power fluctuation, σ _lfor the variance of load active power fluctuation, can by the average of load statistics and standard deviation approximate representation.

The output function of blower fan and the uncertain mathematics model of load can be set up by method above.

Step S103: determine the objective function that distributed power source is distributed rationally and constraint condition, sets up and takes into account probabilistic distributed power source Optimal Allocation Model.

With the via net loss P of power distribution network _loss, node voltage deviation delta U, pollutant emission expense C _emifor objective function, set up and take into account probabilistic Distributed Generation in Distribution System Optimal Allocation Model.

In distributed power source Optimal Allocation Model, the calculating formula of objective function is:

$\left\{\begin{array}{c}{P}_{\mathrm{loss}}=\underset{l=1}{\overset{N_{\mathrm{br}}}{\mathrm{\Σ}}}\left(R_l\frac{P_l^2+Q_l^2}{{U_l}^2}\right)\\ \mathrm{\ΔU}=\underset{i=1}{\overset{N_{\mathrm{node}}}{\mathrm{\Σ}}}{\left(\frac{U_i-U_N}{\mathrm{\Δ}{U}_{i\max }}\right)}^2\\ C_{\mathrm{emi}}=\underset{g=1}{\overset{N_{\mathrm{DG}}}{\mathrm{\Σ}}}\left[\underset{h=1}{\overset{N_{\mathrm{em}}}{\mathrm{\Σ}}}\left(C_h{\mathrm{\γ}}_{\mathrm{gh}}\right)P_{\mathrm{DGg}}\right]+\underset{h=1}{\overset{N_{\mathrm{em}}}{\mathrm{\Σ}}}\left(C_h{\mathrm{\γ}}_{\mathrm{grigh}}\right)P_{\mathrm{gridt}}\end{array}\right.---\left(4\right)$

Wherein, l=1,2 ..., N _br, N _brfor branch road sum, R _lfor the resistance of branch road l, P _l, Q _lbe respectively active power and reactive power that branch road l end flows through, U _lfor the voltage of branch road l end; I=1,2 ..., N _node, N _nodefor node total number, U _iand U _nbe respectively virtual voltage and the rated voltage of node i, Δ U _imaxfor the maximum permissible voltage deviation of node i; G=1,2 ..., N _dG, N _dGfor the total number of units of the distributed power source accessed in power distribution network, h=1,2 ..., N _em, N _emfor discharged pollutant type sum, C _hfor processing the expense of every kilogram of h pollutant, γ _ghbe the emission factor of g distributed power source h pollutant when exporting every kilowatt of eletctric energy (KWE), P _dGgbe the active power of g distributed power source, γ _gridhfor the emission factor of h pollutant when bulk power grid exports every kilowatt of eletctric energy (KWE), P _gridtfor the active power of bulk power grid.

In distributed power source Optimal Allocation Model, the calculating formula of constraint condition is:

$\left\{\begin{array}{c}{P}_{\mathrm{Gi}}-P_{\mathrm{Di}}=U_i\underset{j=1}{\overset{N_{\mathrm{node}}}{\mathrm{\Σ}}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})\\ Q_{\mathrm{Gi}}-Q_{\mathrm{Di}}=U_i\underset{j=1}{\overset{N_{\mathrm{node}}}{\mathrm{\Σ}}}{U}_j(G_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}})\\ P\{U_i^{\min }\≤U_i\≤U_i^{\max }\}\≥\mathrm{\α}\\ P\{P_l\≤P_l^{\max }\}\≥\mathrm{\β}\\ P_{\mathrm{\ΣDG}}\≤\mathrm{\η}{P}_{\mathrm{\ΣL}}\\ N_{\mathrm{DGSe}}\≤N_{\mathrm{DGSe}}^{\max }\end{array}\right.---\left(5\right)$

Wherein, j=1,2 ..., N _node, G _ij, B _ij, δ _ijbe respectively the conductance between node i, j, susceptance and phase angle difference, P _gi, Q _gibe respectively active power and the reactive power of node i place power supply, P _di, Q _dibe respectively active power and the reactive power of node i place load, P{} represents the probability meeting constraints condition of opportunity, be respectively the voltage upper and lower limit of node i, α is the confidence level of node voltage, for the active power of branch road l allows maximal value, β is the confidence level of branch power, P _{Σ DG}for distributed power source total volume, P _{Σ L}for distribution network load total volume, η is distributed power source access ratio, e=1,2 ..., N _eq, N _eqfor the node total number of distributed power source to be installed, N _dGSefor the number of units of access distributed power source in node e place to be installed, for node e place to be installed allows the maximum number of units of access distributed power source.

Distributed power source access power distribution network, is conducive to reducing environmental pollution, reduces power distribution network via net loss, improves power distribution network global voltage level.The present invention considers economy that distributed power source distributes rationally, security and environmental impact, propose the via net loss of power distribution network, node voltage deviation, these three objective functions of pollutant emission expense, as the evaluation index that distributed power source is distributed rationally, consider the constraint conditions such as the permeability of power-balance, voltage limits, branch road through-put power and distributed power source.

Step S104: utilize the Monte Carlo simulation based on Latin Hypercube Sampling to embed multi-objective particle swarm algorithm and take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution.

Step S105: the standardization calculating described Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value according to described standardization and determines described distributed power source configuration scheme.

In the present embodiment, in step S104, utilize the Monte Carlo simulation based on Latin Hypercube Sampling to embed multi-objective particle swarm algorithm to take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution, can comprise the following steps:

Step S201: obtain raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale.

Step S202: particle is encoded, the speed of particle described in initialization and position, the Pareto optimum solution adopting outside elite to store to search in iteration, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set.

Described particle to be encoded to:

$P_{\mathrm{DGe}}=\mathrm{round}\left(N_{\mathrm{DGSe}}^{\max }{x}_e\right)P_{\mathrm{DGS}}---\left(6\right)$

$X={[P_{\mathrm{DG}1},P_{\mathrm{DG}2},\·\·\·,P_{{\mathrm{DG}}_e},\·\·\·,P_{{\mathrm{DGN}}_{\mathrm{eq}}}]}^T---\left(7\right)$

Wherein, P _dGeconnect by node e place to be installed the capacity of distributed power source, round () represents and rounds the number in bracket, x _efor the random number between distributed power source node e to be installed place [0,1], P _dGSfor the reference capacity of separate unit distributed power source; X is the position of any one particle in multi-objective particle swarm algorithm.

The position of random initializtion particle and speed, the individual extreme value P of particle _bestwith global extremum G _bestall be set to the initial position of particle.

Step S203: using described objective function as the fitness function of multi-objective particle swarm algorithm, adopts the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draws the fitness value of particle.

Latin Hypercube Sampling is a kind of stratified sampling method, and by sampling and sorting, two parts form; Compared with simple random sampling, Latin Hypercube Sampling is when sample size is identical, and sampled value can cover the input distributed area of stochastic variable completely, sampling efficiency is high, sampling robustness is good; The Monte Carlo simulation approach combined with Latin Hypercube Sampling is by the sampling repeatedly to uncertain factor, uncertain Load flow calculation problem is solved with a series of deterministic parameters calculation, both remain the advantage that Monte Carlo method is flexible, adaptability is good, overcome again the shortcoming that its sampling number is many, calculated amount is large;

Concrete computation process based on the Monte Carlo simulation approach of Latin Hypercube Sampling is as follows: first, by carrying out Latin Hypercube Sampling to the uncertain mathematics model of blower fan and load, obtaining blower fan and exerting oneself and the sample matrix of load; Secondly, to each column data in sample matrix, on the basis considering constraint condition, carry out the determinacy Load flow calculation of distribution network, draw the expectation value of correlated variables; Finally, according to gained expectation value calculating target function value, and as the fitness value of particle.

Step S204: based on Pareto dominance relation, determines the individual extreme value of described particle.

In the application, described based on Pareto dominance relation, determine the individual extreme value of described particle, comprising:

Step a: definition Pareto dominance relation.

Described Pareto dominance relation is: concentrate in a certain solution of multi-objective optimization question, decision variable A and B, if for arbitrary λ=1, and 2 ..., z meets f _λ(A)≤f _λ(B) and at least there is 1 λ and make f _λ(A)≤f _λ(B) set up, then claim A to arrange B, if there is not dominance relation between A and B, then claim A and B not arrange mutually.

In the application, Pareto optimum solution is: if there are not other in the whole set of feasible solution of multi-objective optimization question to separate X ^*, make X ^*domination A, then claim A to be Pareto optimum solution.

Step b: the individual extreme value determining described particle.

Based on described Pareto dominance relation, determine the individual extreme value P of particle _best: if P is not arranged in current particle position _best, then P is kept _bestconstant; If current particle position domination P _best, then P _bestbe updated to current particle position; If current particle position and P _bestdo not arrange mutually, then retain P by the probability of 50% _best.

Step S205: according to the global extremum of the method determination population preset.

In the application, the global extremum of the population belonging to the described method determination particle according to presetting, comprising:

Step a: using the candidate collection of described outside elite's collection as described population global extremum, based on described Pareto dominance relation, described outside elite's collection is upgraded.

Using outside elite's collection as population global extremum G _bestcandidate collection, based on Pareto dominance relation, outside elite collection is upgraded: if not the particle concentrated by elite of the particle that outside elite concentrates arrange, then this particle does not enter elite's collection; If not some particle that particle that outside elite concentrates domination elite concentrates, then reject elite and concentrate those by the particle arranged, and the particle that non-outside elite concentrates is joined elite concentrate; If not this particle all without dominance relation, is then joined elite and concentrates by all particles that outside elite concentrates this particle and elite to concentrate.

Step b: the described outside elite calculated after upgrading concentrates the crowding distance of each Pareto optimum solution.

Step c: the capacity maintaining described outside elite's collection according to described crowding distance, retains the particle that crowding distance is larger, reject the particle that crowding distance is less.

Steps d: according to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.

The crowding distance that the outside elite after upgrading concentrates each Pareto optimum solution is calculated by formula (8), the capacity of outside elite collection is maintained according to crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less, and concentrate random selecting particle by probability P from outside elite, using the global extremum G of this particle position as population _best.

$D_{\mathrm{\ξ}}=\underset{m=1}{\overset{N_{\mathrm{ob}}}{\mathrm{\Σ}}}|f_{{\mathrm{\ξ}}_l}\left(m\right)-f_{{\mathrm{\ξ}}_r}\left(m\right)|---\left(8\right)$

Wherein, D _ξfor the crowding distance of particle ξ, ξ=1,2 ..., N _sum, N _sumrepresent the number of particle, m=1,2 ..., N _ob, N _obfor the number of objective function, ξ _l, ξ _rbe respectively two particles that particle ξ is adjacent, be respectively particle ξ _l, ξ _rm target function value, the crowding distance of border particle is infinitely great.

Step S206: upgrade the inertia weight of described particle, Studying factors, speed and position according to the update method preset.

Respectively by formula (9), (10), (11) the more inertia weight of new particle, Studying factors, speed and position:

ω＝ω _max-(ω _max-ω _min)*t/T (9)

$\left\{\begin{array}{c}{c}_1=c_{1\mathrm{ini}}+(c_{1\mathrm{fin}}-c_{1\mathrm{ini}})*t/T\\ c_2=c_{2\mathrm{ini}}+(c_{2\mathrm{fin}}-c_{2\mathrm{ini}})*t/T\end{array}\right.---\left(10\right)$

$\left\{\begin{array}{c}{v}_{\mathrm{\ξd}}(t+1)=\mathrm{\ω}{v}_{\mathrm{\ξd}}\left(t\right)+c_1{\mathrm{Rand}}_1\left(\right)(p_{\mathrm{best\ξd}}\left(t\right)-x_{\mathrm{\ξd}}\left(t\right))+c_2{\mathrm{Rand}}_2\left(\right)(g_{\mathrm{bestd}}\left(t\right)-x_{\mathrm{\ξd}}\left(t\right))\\ x_{\mathrm{\ξd}}(t+1)=x_{\mathrm{\ξd}}\left(t\right)+v_{\mathrm{\ξd}}(t+1)\end{array}\right.---\left(11\right)$

Wherein, t=1,2 ..., T, T are total iterations, and ω is inertia weight, ω _maxfor the maximal value of inertia weight, ω _minfor the minimum value of inertia weight, c ₁, c ₂for Studying factors, c _1ini, c _2inibe respectively c ₁and c ₂initial value, c _1fin, c _2finbe respectively c ₁and c ₂iteration final value; D=1,2 ..., N _dim, N _dimfor the total dimension of particle, v _{ξ d}(t+1) be the speed of particle ξ d dimension in the t+1 time iteration, x _{ξ d}(t+1) be the position of particle ξ d dimension in the t+1 time iteration, v _{ξ d}t () is for particle ξ is _tthe speed of d dimension in secondary iteration, x _{ξ d}t () is for particle ξ is _tthe position of d dimension in secondary iteration, p _{best ξ d}t () is for particle ξ is _tthe individual extreme value of d dimension in secondary iteration, g _bestd( _t) be whole population d dimension global extremum in the t time iteration; Rand ₁(), Rand ₂() is the random number of (0,1) interval distribution.

In multi-objective particle swarm algorithm, the selection of parameter has a great impact convergence, and inertia weight and choosing of Studying factors determine algorithm ability of searching optimum and local search ability; The present invention adopts inertia weight linear decrease method and becomes Studying factors method and the value of inertia weight and Studying factors is constantly changed along with iterative process, thus overcomes algorithm and be easily absorbed in the defects such as Premature Convergence, local optimal searching ability be poor.

Step S207: described iterations t=t+1 is set.

Step S208: judge whether to meet t > T, if not, then repeats described uncertain Load flow calculation, describedly determines individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T.

Step S209: export the described Pareto optimum solution that described outside elite concentrates reservation.

Export the Pareto optimum solution that outside elite concentrates reservation, use f _εm () represents m target function value of ε Pareto optimum solution, ε=1,2 ..., N _best, N _bestfor outside elite concentrates the number of Pareto optimum solution.

In the present embodiment, in step S105, the described standardization calculating Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value and determines described distributed power source configuration scheme, comprising according to described standardization:

Step S301: utilize fuzzy membership function to represent the satisfaction that in each described Pareto optimum solution, each objective function is corresponding.

The satisfaction that in each Pareto optimum solution, each objective function is corresponding is represented by formula (12) fuzzy membership function:

${\mathrm{\&mu;}}_{\mathrm{\&epsiv;}}\left(m\right)=\left\{\begin{array}{cc}1,& f_{\mathrm{\&epsiv;}}\left(m\right)\&le;f_{\mathrm{\&epsiv;}}\left(m\right)\\ \frac{f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\max }-f_{\mathrm{\&epsiv;}}\left(m\right)}{f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\max }-f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\min }},& f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\min }<f_{\mathrm{\&epsiv;}}\left(m\right)<f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\max }\\ 0,& f_{\mathrm{\&epsiv;}}\left(m\right)\&GreaterEqual;f_{\mathrm{\&epsiv;}}{\left(m\right)}_{\max }\end{array}\right.---\left(12\right)$

Wherein, μ _ε(m) satisfaction corresponding to m objective function of ε Pareto optimum solution, f _ε(m) _maxand f _ε(m) _minbe respectively f _εthe maximal value of (m) and minimum value.

Taking into account that probabilistic Distributed Generation in Distribution System distributes rationally is a multi-objective optimization question, the span of each objective function and dimension are often inconsistent, therefore, adopt fuzzy membership function to be satisfied with angle value to each objective function in each Pareto optimum solution to calculate.μ _εm the value of () is interval in [0,1], μ _εm ()=0 represents completely dissatisfied to m target function value of ε Pareto optimum solution; μ _εm ()=1 represents m the target function value satisfaction completely to ε Pareto optimum solution.

Step S302: according to the computing method preset, the standardization calculating described Pareto optimum solution is satisfied with angle value.

Step S303: determine that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

The standardization calculating each Pareto optimum solution by formula (13) is satisfied with angle value:

${\mathrm{\&mu;}}_{\mathrm{\&epsiv;}}=\frac{\underset{m=1}{\overset{N_{\mathrm{ob}}}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{\&epsiv;}}\left(m\right)}{\underset{\mathrm{\&epsiv;}=1}{\overset{N_{\mathrm{best}}}{\mathrm{\&Sigma;}}}\underset{m=1}{\overset{N_{\mathrm{ob}}}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{\&epsiv;}}\left(m\right)}---\left(13\right)$

Wherein, μ _εbe that the standardization of ε Pareto optimum solution is satisfied with angle value.For each Pareto optimum solution, carry out the standardization being satisfied with angle value, then be satisfied with according to standardization the selection that angle value is optimized allocation plan.

Standardization is satisfied with the distributed power source configuration scheme corresponding to the maximum solution of angle value, is the preferred plan taken into account probabilistic Distributed Generation in Distribution System and distribute rationally.

In the method that a kind of distributed power source that the application provides is distributed rationally, determine the probability density function of distributed power source and the probability density function of load; According to the probability density function of described distributed power source and the probability density function of described load, set up the uncertain mathematics model of distributed power source and load; Determine the objective function that distributed power source is distributed rationally and constraint condition, set up and take into account probabilistic distributed power source Optimal Allocation Model; Take into account probabilistic distributed power source Optimal Allocation Model according to the method preset solve described, obtain Pareto optimum solution; The standardization calculating described Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.Distributed power source Optimal Configuration Method of the present invention, establish consider distributed power source and distribute rationally economy, security and environmental impact Model for Multi-Objective Optimization, actual conditions can be reflected more all sidedly, there is higher accuracy; Monte Carlo simulation based on Latin Hypercube Sampling is embedded multi-objective particle swarm algorithm and carries out model solution, can not only effectively process distributed power source distribute rationally in uncertainty, also take into full account the correlativity between each objective function, effectively prevent the problem that prioritization scheme is with strong points.

Corresponding with the method that the embodiment of the method that a kind of distributed power source of above-mentioned the application is distributed rationally provides, see Fig. 2, present invention also provides the system embodiment that a kind of distributed power source is distributed rationally, in the present embodiment, this system comprises:

First determining unit 401, for the probability density function of the probability density function and load of determining distributed power source.

First sets up unit 402, for according to the probability density function of described distributed power source and the probability density function of described load, sets up the uncertain mathematics model of distributed power source and load.

Second determining unit 403, for determining the objective function that distributed power source is distributed rationally and constraint condition, setting up and taking into account probabilistic distributed power source Optimal Allocation Model.

Solve unit 404, embed multi-objective particle swarm algorithm for utilizing the Monte Carlo simulation based on Latin Hypercube Sampling and take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution.

First computing unit 405, is satisfied with angle value for the standardization calculating described Pareto optimum solution corresponding, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.

In this application, solve unit 404 described in comprise:

Acquiring unit 501, for obtaining raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale.

Coding unit 502, for encoding to particle, the speed of particle described in initialization and position.

Setup unit 503, the Pareto optimum solution searched in iteration for adopting outside elite to store, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set.

Second computing unit 504, for using described objective function as the fitness function of multi-objective particle swarm algorithm, adopts the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draws the fitness value of particle.

3rd determining unit 505, for based on Pareto dominance relation, determines the individual extreme value of described particle.

4th determining unit 506, for the global extremum according to the method determination population preset.

First updating block 507, for upgrading the inertia weight of described particle, Studying factors, speed and position according to the update method preset.

Setting unit 508, for arranging described iterations t=t+1, repeating described uncertain Load flow calculation, describedly determining individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T.

Output unit 509, concentrates the described Pareto optimum solution of reservation for exporting described outside elite.

In the application, described first computing unit 405 comprises:

Representing unit 601, representing for utilizing fuzzy membership function the satisfaction that in each described Pareto optimum solution, each objective function is corresponding.

3rd computing unit 602, for according to preset computing method, the standardization calculating each described Pareto optimum solution is satisfied with angle value.

5th determining unit 603, for determining that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

In the application, described 3rd determining unit 505 comprises:

Definition unit, for defining Pareto dominance relation.

6th determining unit, for according to described Pareto dominance relation and Pareto optimum solution, determines the individual extreme value of described particle.

In the application, described 4th determining unit 506 comprises:

Second updating block, for using the candidate collection of described outside elite's collection as described population global extremum, upgrades described outside elite's collection based on described Pareto dominance relation.

4th computing unit, concentrates the crowding distance of each Pareto optimum solution for calculating the described outside elite after renewal.

Maintain unit, for maintaining the capacity of described outside elite's collection according to described crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less.

Choose unit, for according to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.

It should be noted that, each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.For device class embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.

Finally, also it should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.

For convenience of description, various unit is divided into describe respectively with function when describing above device.Certainly, the function of each unit can be realized in same or multiple software and/or hardware when implementing the application.

As seen through the above description of the embodiments, those skilled in the art can be well understood to the mode that the application can add required general hardware platform by software and realizes.Based on such understanding, the technical scheme of the application can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product can be stored in storage medium, as ROM/RAM, magnetic disc, CD etc., comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform the method described in some part of each embodiment of the application or embodiment.

The method and system that a kind of distributed power source provided the application is above distributed rationally are described in detail, apply specific case herein to set forth the principle of the application and embodiment, the explanation of above embodiment is just for helping method and the core concept thereof of understanding the application; Meanwhile, for one of ordinary skill in the art, according to the thought of the application, all will change in specific embodiments and applications, in sum, this description should not be construed as the restriction to the application.

Claims (10)

1. a distributed power source method of distributing rationally, it is characterized in that, the method comprises:

Determine the probability density function of distributed power source and the probability density function of load;

According to the probability density function of described distributed power source and the probability density function of described load, set up the uncertain mathematics model of distributed power source and load;

Determine the objective function that distributed power source is distributed rationally and constraint condition, set up and take into account probabilistic distributed power source Optimal Allocation Model;

Utilize the Monte Carlo simulation based on Latin Hypercube Sampling to embed multi-objective particle swarm algorithm to take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution;

The standardization calculating described Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.

2. method according to claim 1, it is characterized in that, described utilization embeds multi-objective particle swarm algorithm based on the Monte Carlo simulation of Latin Hypercube Sampling and takes into account probabilistic distributed power source Optimal Allocation Model solve described, obtains Pareto optimum solution, comprising:

Obtain raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale;

Particle is encoded, the speed of particle described in initialization and position;

The Pareto optimum solution adopting outside elite to store to search in iteration, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set;

Using described objective function as the fitness function of multi-objective particle swarm algorithm, adopt the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draw the fitness value of particle;

Based on Pareto dominance relation, determine the individual extreme value of described particle;

According to the global extremum of the method determination population preset;

The inertia weight of described particle, Studying factors, speed and position is upgraded according to the update method preset;

Described iterations t=t+1 is set, repeats described uncertain Load flow calculation, describedly determine individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T;

Export the described Pareto optimum solution that described outside elite concentrates reservation.

3. method according to claim 1, is characterized in that, the described standardization calculating Pareto optimum solution corresponding is satisfied with angle value, is satisfied with angle value and determines described distributed power source configuration scheme, comprising according to described standardization:

Fuzzy membership function is utilized to represent the satisfaction that in each described Pareto optimum solution, each objective function is corresponding;

According to the computing method preset, the standardization calculating described Pareto optimum solution is satisfied with angle value;

Determine that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

4. method according to claim 2, is characterized in that, described based on Pareto dominance relation, determines the individual extreme value of described particle, comprising:

Definition Pareto dominance relation;

According to described Pareto dominance relation and Pareto optimum solution, determine the individual extreme value of described particle.

5. method according to claim 2, is characterized in that, the global extremum of the population belonging to the described method determination particle according to presetting, comprising:

Using the candidate collection of described outside elite's collection as described population global extremum, based on described Pareto dominance relation, described outside elite's collection is upgraded;

The described outside elite calculated after upgrading concentrates the crowding distance of each Pareto optimum solution;

Maintain the capacity of described outside elite's collection according to described crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less;

According to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.

6. a distributed power source system of distributing rationally, it is characterized in that, this system comprises:

First determining unit, for the probability density function of the probability density function and load of determining distributed power source;

First sets up unit, for according to the probability density function of described distributed power source and the probability density function of described load, sets up the uncertain mathematics model of distributed power source and load;

Second determining unit, for determining the objective function that distributed power source is distributed rationally and constraint condition, setting up and taking into account probabilistic distributed power source Optimal Allocation Model;

Solve unit, embed multi-objective particle swarm algorithm for utilizing the Monte Carlo simulation based on Latin Hypercube Sampling and take into account probabilistic distributed power source Optimal Allocation Model solve described, obtain Pareto optimum solution;

First computing unit, is satisfied with angle value for the standardization calculating described Pareto optimum solution corresponding, is satisfied with angle value determines described distributed power source configuration scheme according to described standardization.

7. system according to claim 6, is characterized in that, described in solve unit and comprise:

Acquiring unit, for obtaining raw data, described raw data comprises distribution network network parameter, multi-objective particle swarm algorithm parameter and Latin Hypercube Sampling scale;

Coding unit, for encoding to particle, the speed of particle described in initialization and position;

Setup unit, the Pareto optimum solution searched in iteration for adopting outside elite to store, sets the size of outside elite collection, and outside elite's collection described in initialization, iterations t=1 is set;

Second computing unit, for using described objective function as the fitness function of multi-objective particle swarm algorithm, adopts the Monte Carlo simulation approach based on Latin Hypercube Sampling to carry out uncertain Load flow calculation, draws the fitness value of particle;

3rd determining unit, for based on Pareto dominance relation, determines the individual extreme value of described particle;

4th determining unit, for the global extremum according to the method determination population preset;

First updating block, for upgrading the inertia weight of described particle, Studying factors, speed and position according to the update method preset;

Setting unit, for arranging described iterations t=t+1, repeating described uncertain Load flow calculation, describedly determining individual extreme value, the described process determining global extremum and described renewal, until described iterations t=T;

Output unit, concentrates the described Pareto optimum solution of reservation for exporting described outside elite.

8. system according to claim 6, is characterized in that, described first computing unit comprises:

Representing unit, representing for utilizing fuzzy membership function the satisfaction that in each described Pareto optimum solution, each objective function is corresponding;

3rd computing unit, for according to preset computing method, the standardization calculating described Pareto optimum solution is satisfied with angle value;

5th determining unit, for determining that the described standardization distributed power source configuration scheme be satisfied with corresponding to the maximum solution of angle value is the preferred plan that described distributed power source is distributed rationally.

9. system according to claim 7, is characterized in that, described 3rd determining unit comprises:

Definition unit, for defining Pareto dominance relation;

6th determining unit, for according to described Pareto dominance relation and Pareto optimum solution, determines the individual extreme value of described particle.

10. method according to claim 7, is characterized in that, described 4th determining unit comprises:

Second updating block, for using the candidate collection of described outside elite's collection as described population global extremum, upgrades described outside elite's collection based on described Pareto dominance relation;

4th computing unit, concentrates the crowding distance of each Pareto optimum solution for calculating the described outside elite after renewal;

Maintain unit, for maintaining the capacity of described outside elite's collection according to described crowding distance, retain the particle that crowding distance is larger, reject the particle that crowding distance is less;

Choose unit, for according to the probability preset, concentrate random selecting particle, using the global extremum of the position of described particle as described population from described outside elite.