CN106253308A - A kind of var Optimization Method in Network Distribution - Google Patents

Abstract

The present invention relates to a kind of var Optimization Method in Network Distribution, described method comprises determining that the history object function value matrix of power distribution network each idle work optimization object function；Determine the entropy weight that described power distribution network each idle work optimization object function is corresponding；According to described power distribution network each idle work optimization object function and the entropy weight of correspondence thereof, set up GA for reactive power optimization general objective function；Determine that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme according to described GA for reactive power optimization general objective function；The method that the present invention provides, it is possible to by the analyzing and processing to history schemes multiple in data base, determine the weight of each object function in idle work optimization, is converted into multiple-objection optimization single object optimization, thus selects the idle work optimization scheme of optimum.

Description

A kind of var Optimization Method in Network Distribution

Technical field

The present invention relates to reactive power optimization of power system and control technical field, be specifically related to a kind of GA for reactive power optimization side Method.

Background technology

Idle work optimization, it is simply that when the structural parameters of system and load condition give timing, excellent by some control variable Change, can find on the premise of meeting all appointment constraints, make some or multiple performance indications of system reach Reactive-power control means time optimum.Reactive Power Optimazation Problem is that the branch gradually differentiated from the development of optimal load flow is asked Topic.Electrical network carries out in power system idle work optimization to control voltage levvl and reduce active loss.

Have accumulated substantial amounts of historical data in distribution runtime database, corresponding each historical juncture have recorded corresponding nothing Merit prioritization scheme, can instruct the idle work optimization of future time instance by the analyzing and processing of historical data.In general, every day is same The load of time point meets normal distribution from the point of view of certain time scope, then select the optimization time from historical data base The history prioritization scheme that point is corresponding, therefrom selects the scheme of optimum and instructs the idle work optimization of future time instance to be feasible.Power distribution network Idle work optimization is a multi-objective optimization question, and object function frequently includes network loss, variation, power factor etc., but many How weight between target determines is a difficult problem.

Summary of the invention

The present invention provides a kind of var Optimization Method in Network Distribution, its objective is by history schemes multiple in data base Analyzing and processing, determines the weight of each object function in idle work optimization, multiple-objection optimization is converted into single object optimization, thus selects Optimum idle work optimization scheme.

It is an object of the invention to use following technical proposals to realize:

A kind of var Optimization Method in Network Distribution, it thes improvement is that, including:

Determine the history object function value matrix of power distribution network each idle work optimization object function；

Determine the entropy weight that described power distribution network each idle work optimization object function is corresponding；

According to described power distribution network each idle work optimization object function and the entropy weight of correspondence thereof, set up GA for reactive power optimization catalogue Scalar functions；

Determine that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme according to described GA for reactive power optimization general objective function.

Preferably, described power distribution network each idle work optimization object function includes: network loss object function, node voltage side-play amount mesh Scalar functions, power factor object function and voltage stability margin object function.

Preferably, as the following formula (1) determines the history target function value matrix A of power distribution network each idle work optimization object function:

$A=\left[\begin{array}{ccccc}{f}_{11}& ...& f_{1i}& ...& f_{1m}\\ ...& ...& ...& ...& ...\\ f_{j1}& ...& f_{ji}& ...& f_{jm}\\ ...& ...& ...& ...& ...\\ f_{n1}& ...& f_{ni}& ...& f_{nm}\end{array}\right]---\left(1\right)$

In formula (1), f_jiFor jth idle work optimization object function idle work optimization in i-th history idle work optimization scheme Target function value, i ∈ [1, m], j ∈ [1, n], m are history idle work optimization scheme sum, and n is GA for reactive power optimization target letter Number sum.

Preferably, the described entropy weight determining that described power distribution network each idle work optimization object function is corresponding, including:

It is each that history object function value matrix according to described power distribution network each idle work optimization object function obtains described power distribution network The normalized matrix R={r of the history object function value matrix of idle work optimization object function_ji, wherein, r_jiIdle excellent for jth Change the standard value of object function idle work optimization target function value in i-th history idle work optimization scheme, i ∈ [1, m], j ∈ [1, n], m is history idle work optimization scheme sum, and n is GA for reactive power optimization object function sum；

Determine the entropy of described power distribution network each idle work optimization object function；

Determine the entropy weight of described power distribution network each idle work optimization object function.

Further, as the following formula (2) determine that described jth idle work optimization object function is in i-th history idle work optimization side Standard value r of the idle work optimization target function value in case_ji:

$r_{ji}=\frac{\underset{i}{max}\left\{f_{ji}\right\}-f_{ji}}{\underset{i}{max}\left\{f_{ji}\right\}-\underset{i}{\min }\left\{f_{ji}\right\}}---\left(2\right)$

In formula (2), f_jiFor jth idle work optimization object function idle work optimization in i-th history idle work optimization scheme Target function value.

Further, as the following formula (3) determine the entropy H of power distribution network jth idle work optimization object function_j:

$H_j=-k\underset{i=1}{\overset{m}{\Σ}}{r}_{ji}{\mathrm{lnr}}_{ji}---\left(3\right)$

In formula (3), k is constant.

Further, as the following formula (4) determine entropy weight w of power distribution network jth idle work optimization object function_j:

$w_j=\frac{1-H_j}{n-\underset{j=1}{\overset{n}{\Σ}}{H}_j}---\left(4\right).$

Preferably, as the following formula (5) set up GA for reactive power optimization general objective function:

$\min f=\underset{j}{\overset{n}{\Σ}}{\mathrm{\ω}}_j{f}_j---\left(5\right)$

In formula (5), f_jFor power distribution network jth idle work optimization object function, ω_jFor power distribution network jth idle work optimization target letter The entropy weight of number, j ∈ [1, n], n are GA for reactive power optimization object function sum.

Preferably, described according to described GA for reactive power optimization general objective function determine power distribution network OPTIMAL REACTIVE POWER control optimize Scheme, including:

In m history idle work optimization scheme, power distribution network each idle work optimization object function in i-th history idle work optimization scheme The value of described GA for reactive power optimization general objective function corresponding to value be minimum, the most described i-th history idle work optimization scheme The value of middle power distribution network each idle work optimization object function is that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme, and wherein, i ∈ [1, m], m are History idle work optimization scheme sum.

Beneficial effects of the present invention:

The technical scheme that the present invention provides, from historical data, is incorporated into entropy weight thought in multi-objective reactive optimization, Entropy according to each object function determines its corresponding weight, and multi-objective reactive optimization is converted into single-object problem, point Do not calculate the target function value that each history scheme is corresponding, and to choose history scheme corresponding to object function minima be optimum control Scheme.Compared with traditional method, the inventive method, from the angle of data analysis Yu modeling, can determine each target objectively The weight of function, and computational efficiency is high.

Accompanying drawing explanation

Fig. 1 is the flow chart of a kind of var Optimization Method in Network Distribution of the present invention.

Detailed description of the invention

Below in conjunction with the accompanying drawings the detailed description of the invention of the present invention is elaborated.

For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is The a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art The all other embodiments obtained under not making creative work premise, broadly fall into the scope of protection of the invention.

A kind of var Optimization Method in Network Distribution that the present invention provides, as it is shown in figure 1, include:

The 101. history object function value matrixs determining power distribution network each idle work optimization object function；

102. determine the entropy weight that described power distribution network each idle work optimization object function is corresponding；

103., according to described power distribution network each idle work optimization object function and the entropy weight of correspondence thereof, set up GA for reactive power optimization General objective function；

According to described GA for reactive power optimization general objective function, 104. determine that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme.

Wherein, each idle work optimization of power distribution network described in prior art object function includes: network loss object function, node voltage Side-play amount object function, power factor object function and voltage stability margin object function etc..

Such as: network loss object function f₁With node voltage side-play amount object function f₂Formula be:

$\left\{\begin{array}{c}{f}_1=\underset{i,j\∈N}{\mathrm{\Σ}}{G}_{ij}(V_i^2+V_j^2-2V_i{V}_j{\mathrm{cos\θ}}_{ij})\\ f_2=\underset{i=1}{\overset{N}{\Σ}}{\left(\frac{V_i-V_i^B}{{\mathrm{\ΔV}}_i^{max}}\right)}^2\end{array}\right.$

In formula, G_ijFor the conductance between node i, j；V_i、V_jRepresent the voltage magnitude of node i, j respectively；θ_ijFor node i, j Between phase difference of voltage；N is the set of system node composition；Desired voltage for node i；Maximum for node i Allow variation；

Concrete, in described step 101, (1) determines the history target of power distribution network each idle work optimization object function as the following formula Functional value matrix A:

$A=\left[\begin{array}{ccccc}{f}_{11}& ...& f_{1i}& ...& f_{1m}\\ ...& ...& ...& ...& ...\\ f_{j1}& ...& f_{ji}& ...& f_{jm}\\ ...& ...& ...& ...& ...\\ f_{n1}& ...& f_{ni}& ...& f_{nm}\end{array}\right]---\left(1\right)$

In formula (1), f_jiFor jth idle work optimization object function idle work optimization in i-th history idle work optimization scheme Target function value, i ∈ [1, m], j ∈ [1, n], m are history idle work optimization scheme sum, and n is GA for reactive power optimization target letter Number sum.

Described step 102, uses entropy weight thought to determine the weight of each target in multi-objective reactive optimization, including:

It is each that history object function value matrix according to described power distribution network each idle work optimization object function obtains described power distribution network The normalized matrix R={r of the history object function value matrix of idle work optimization object function_ji, wherein, r_jiIdle excellent for jth Change the standard value of object function idle work optimization target function value in i-th history idle work optimization scheme, i ∈ [1, m], j ∈ [1, n], m is history idle work optimization scheme sum, and n is GA for reactive power optimization object function sum；

Determine the entropy of described power distribution network each idle work optimization object function；

Determine the entropy weight of described power distribution network each idle work optimization object function.

Further, as the following formula (2) determine that described jth idle work optimization object function is in i-th history idle work optimization side Standard value r of the idle work optimization target function value in case_ji:

$r_{ji}=\frac{\underset{i}{max}\left\{f_{ji}\right\}-f_{ji}}{\underset{i}{max}\left\{f_{ji}\right\}-\underset{i}{\min }\left\{f_{ji}\right\}}---\left(2\right)$

In formula (2), f_jiFor jth idle work optimization object function idle work optimization in i-th history idle work optimization scheme Target function value.

(3) determine the entropy H of power distribution network jth idle work optimization object function as the following formula_j:

$H_j=-k\underset{i=1}{\overset{m}{\Σ}}{r}_{ji}\ln r_{ji}---\left(3\right)$

In formula (3), k is constant.

Entropy is probabilistic measuring, and entropy is the least, shows that the quantity of information in corresponding evaluation index is the most effective, because of This, (4) determine entropy weight w of power distribution network jth idle work optimization object function as the following formula_j:

$w_j=\frac{1-H_j}{n-\underset{j=1}{\overset{n}{\Σ}}{H}_j}---\left(4\right).$

In described step 103, (5) set up GA for reactive power optimization general objective function as the following formula:

$\min f=\underset{j}{\overset{n}{\Σ}}{\mathrm{\ω}}_j{f}_j---\left(5\right)$

In formula (5), f_jFor power distribution network jth idle work optimization object function, ω_jFor power distribution network jth idle work optimization target letter The entropy weight of number, j ∈ [1, n], n are GA for reactive power optimization object function sum.

In described step 104, described determine power distribution network OPTIMAL REACTIVE POWER according to described GA for reactive power optimization general objective function Control prioritization scheme, including:

In m history idle work optimization scheme, power distribution network each idle work optimization object function in i-th history idle work optimization scheme The value of described GA for reactive power optimization general objective function corresponding to value be minimum, the most described i-th history idle work optimization scheme The value of middle power distribution network each idle work optimization object function is that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme, and wherein, i ∈ [1, m], m are History idle work optimization scheme sum.

Finally should be noted that: above example is only in order to illustrate that technical scheme is not intended to limit, to the greatest extent The present invention has been described in detail by pipe with reference to above-described embodiment, and those of ordinary skill in the field are it is understood that still The detailed description of the invention of the present invention can be modified or equivalent, and any without departing from spirit and scope of the invention Amendment or equivalent, it all should be contained within the claims of the present invention.

Claims (9)

1. a var Optimization Method in Network Distribution, it is characterised in that described method includes:

Determine the history object function value matrix of power distribution network each idle work optimization object function；

Determine the entropy weight that described power distribution network each idle work optimization object function is corresponding；

According to described power distribution network each idle work optimization object function and the entropy weight of correspondence thereof, set up the GA for reactive power optimization catalogue offer of tender Number；

Determine that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme according to described GA for reactive power optimization general objective function.

2. the method for claim 1, it is characterised in that described power distribution network each idle work optimization object function includes: network loss Object function, node voltage side-play amount object function, power factor object function and voltage stability margin object function.

3. the method for claim 1, it is characterised in that (1) determines power distribution network each idle work optimization object function as the following formula History target function value matrix A:

$A=\left[\begin{array}{ccccc}{f}_{11}& ...& f_{1i}& ...& f_{1m}\\ ...& ...& ...& ...& ...\\ f_{j1}& ...& f_{ji}& ...& f_{jm}\\ ...& ...& ...& ...& ...\\ f_{n1}& ...& f_{ni}& ...& f_{nm}\end{array}\right]---\left(1\right)$

In formula (1), f_jiFor jth idle work optimization object function idle work optimization target in i-th history idle work optimization scheme Functional value, i ∈ [1, m], j ∈ [1, n], m are history idle work optimization scheme sum, and n is that GA for reactive power optimization object function is total Number.

4. the method for claim 1, it is characterised in that described determine described power distribution network each idle work optimization object function pair The entropy weight answered, including:

It is each idle that history object function value matrix according to described power distribution network each idle work optimization object function obtains described power distribution network The normalized matrix R={r of the history object function value matrix of optimization object function_ji, wherein, r_jiFor jth idle work optimization mesh The standard value of scalar functions idle work optimization target function value in i-th history idle work optimization scheme, i ∈ [1, m], j ∈ [1, N], m is history idle work optimization scheme sum, and n is GA for reactive power optimization object function sum；

Determine the entropy of described power distribution network each idle work optimization object function；

Determine the entropy weight of described power distribution network each idle work optimization object function.

5. method as claimed in claim 4, it is characterised in that (2) determine described jth idle work optimization object function as the following formula Standard value r of the idle work optimization target function value in i-th history idle work optimization scheme_ji:

$r_{ji}=\frac{\underset{i}{max}\left\{f_{ji}\right\}-f_{ji}}{\underset{i}{\max }\left\{f_{ji}\right\}-\underset{i}{min}\left\{f_{ji}\right\}}---\left(2\right)$

In formula (2), f_jiFor jth idle work optimization object function idle work optimization target in i-th history idle work optimization scheme Functional value.

6. method as claimed in claim 4, it is characterised in that (3) determine power distribution network jth idle work optimization target letter as the following formula The entropy H of number_j:

$H_j=-k\underset{i=1}{\overset{m}{\Σ}}{r}_{ji}\ln r_{ji}---\left(3\right)$

In formula (3), k is constant.

7. method as claimed in claim 4, it is characterised in that (4) determine power distribution network jth idle work optimization target letter as the following formula Entropy weight w of number_j:

$w_j=\frac{1-H_j}{n-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{H}_j}---\left(4\right).$

8. the method for claim 1, it is characterised in that (5) set up GA for reactive power optimization general objective function as the following formula:

$\min f=\underset{j}{\overset{n}{\Σ}}{\mathrm{\ω}}_j{f}_j---\left(5\right)$

In formula (5), f_jFor power distribution network jth idle work optimization object function, ω_jFor power distribution network jth idle work optimization object function Entropy weight, j ∈ [1, n], n are GA for reactive power optimization object function sum.

9. the method for claim 1, it is characterised in that described true according to described GA for reactive power optimization general objective function Determine power distribution network OPTIMAL REACTIVE POWER and control prioritization scheme, including:

In m history idle work optimization scheme, the value of power distribution network each idle work optimization object function in i-th history idle work optimization scheme The value of corresponding described GA for reactive power optimization general objective function is minimum, joins in the most described i-th history idle work optimization scheme The value of electrical network each idle work optimization object function is that power distribution network OPTIMAL REACTIVE POWER controls prioritization scheme, and wherein, i ∈ [1, m], m are history Idle work optimization scheme sum.