CN105846443A - Power grid reactive compensation optimization configuration system and method based on multi-knapsack problem solution - Google Patents

Abstract

The invention discloses a power grid reactive compensation optimization configuration system and method based on a multi-knapsack problem solution. The power grid reactive compensation optimization configuration method comprises steps of establishing an optimization model in which the total value of articles inside the knapsack is up to a target function through abstracting a transformer station having reactive vacancy as an article and viewing a transformer station having reactive opulence as a knapsack, adopting a heredity algorithm to solve a reactive balance relation of the full power grid, and stipulating a reactive compensation increase scheme according to a nearby arrangement principle. The analysis for the whole process is automatically finished through a data input module, a reactive vacancy calculation module, a reactive balance analysis module, a compensation increase distribution analysis module and a data output module, and does not need to be intervened by people. The power grid reactive compensation optimization configuration system and method improve the efficiency of power grid reactive planning, improve the reactive voltage optimization operation level of the power grid, and have a good application prospect.

Description

The electric network reactive compensation Optimizing Configuration System solved based on many knapsack problems and method

Technical field

The present invention relates to a kind of electric network reactive compensation solved based on many knapsack problems distribute rationally System and method, belongs to electric network reactive-load planning technology field.

Background technology

Idle planning is the important component part of Electric Power Network Planning, and rational idle planning can not only be protected The quality of voltage of card electrical network, moreover it is possible to effectively reduce active loss, for improving the stability of electrical network Particularly voltage stability is most important.

Being improved by electric power proportion along with large-capacity city receiving end electrical network is outer, quality of voltage problem is also Become increasingly conspicuous, have that load density is high, power supply distance is short, by electricity than great, user's voltage matter Amount and power supply reliability require high.Additionally, due to environmental constraints, cable in urban distribution network Circuit uses more extensive, and cable run charge power is high, the voltage power-less problem caused It is also required to cause concern.

According in power system voltage and var fire protection technology (Ministry of Energy SD325 89) Regulation:

(1) Main Function that 330kV and above transformer station capacitive reactive power compensate is When compensation main transformer reactive loss and transmission line of electricity transmission capacity are bigger, the idle of electrical network lacks Volume.Capacitive reactive power compensation capacity should configure according to the 10% of main transformer capacity～20%, or Determine after calculating.

(2) 330kV and above high-voltage shunt reactor (include the little electricity of neutral point Anti-) Main Function be to limit power-frequency overvoltage and reduce secondary arc current, recover voltage and flat The charge power of weighing apparatus extra high voltage network, the capacity of high-voltage shunt reactor should be wanted according to above-mentioned Ask and determine.The effect of main transformer low-pressure side shunt reactor group mainly compensates EHV transmission The residue charge power of circuit, its capacity should determine according to electric network composition and the needs of operation.

At present, the reactive-load compensation planning application of actual electric network generally uses expertise analysis, work Work amount is big and the longest, wastes time and energy, therefore, and the practical side of research reactive-load compensation configuration Method contributes to Operation Mode Optimization, work efficiency that raising mode arranges and automatization level, right The safe and stable operation of electrical network has great significance.

Summary of the invention

The reactive-load compensation planning application that the invention aims to overcome electrical network in prior art leads to Frequently with expertise analysis, workload is big and the longest, the problem wasted time and energy.

In order to achieve the above object, the technical solution adopted in the present invention is:

The electric network reactive compensation Optimizing Configuration System solved based on many knapsack problems, it is characterised in that: Including the data input module being sequentially connected with, no-power vacancy computing module, analysis of reactive power balance mould Block, compensation increase partition analysis module and data outputting module,

Described data input module, for obtaining the substation data of electrical network, packet purse rope frame Topological structure, line impedance parameter, capacity of reactive power compensation device parameter；

Described no-power vacancy computing module, for calculating the no-power vacancy of transformer station, described idle Vacancy includes perception no-power vacancy and capacitive reactive power vacancy；

Described analysis of reactive power balance module, for by abstract for reactive-load compensation balanced arrangement problem be many Knapsack Solve problems, sets up many knapsack problems solving model, and uses genetic algorithm to many knapsacks Problem solves, and obtains the reactive balance result of optimum；

Described compensation increases partition analysis module, needs to increase for being calculated the reactive balance result of optimum The reactive compensation capacity joined, the optimum reactive balance knot obtained according to analysis of reactive power balance module Really, calculate the residual capacity of each knapsack, for being not put into the remaining articles of any knapsack, look into Look for whether the knapsack being directly connected has residual capacity, the no-power vacancy after deduction residual capacity For the reactive compensation capacity joined need to be increased；

Described data outputting module, for according to compensating the analysis result increasing partition analysis module, adopting Determine that final compensation increases complexion amount and exports by the most regular principle.

The aforesaid electric network reactive compensation Optimizing Configuration System solved based on many knapsack problems, it is special Levy and be: described compensation increases partition analysis module, for being not put into the remaining articles of any knapsack, Searching whether the knapsack being directly connected has a residual capacity, idle after deduction residual capacity lacks Volume is to increase the reactive compensation capacity joined, if a knapsack is connected with two or more remaining articles, Then in the remaining articles big relative to this knapsack value, preferentially deduct residual capacity.

A kind of electric network reactive compensation Optimal Configuration Method solved based on many knapsack problems, its feature It is: the method operates in the above-mentioned electric network reactive compensation optimization solved based on many knapsack problems In configuration system, comprise the following steps,

Step (A), obtains the network frame topology structure of transformer station, circuit by data input module Impedance parameter, capacity of reactive power compensation device parameter；

Step (B), the data obtained according to step (A), by no-power vacancy computing module Calculating the no-power vacancy of transformer station, described no-power vacancy includes perception no-power vacancy and capacitive reactive power Vacancy；

Step (C), by analysis of reactive power balance module, takes out reactive-load compensation balanced arrangement problem As for many knapsacks Solve problems, setting up many knapsack problems solving model；

Step (D), by analysis of reactive power balance module, uses genetic algorithm to step (C) Many knapsack problems solving model solves, and obtains the reactive balance result of optimum；

Step (E), according to optimum reactive balance result, increases partition analysis module meter by compensating Calculating the residual capacity of each knapsack, the reactive balance result obtaining optimum need to increase the reactive-load compensation appearance joined Amount；

Step (F), need to increase, according to the reactive balance result obtaining optimum of step (E), the nothing joined Merit compensation capacity, uses the most regular principle to determine final compensation by data outputting module Increase complexion amount and export.

The aforesaid electric network reactive compensation Optimal Configuration Method solved based on many knapsack problems, it is special Levy and be: step (B), calculated the no-power vacancy of transformer station by no-power vacancy computing module, Comprise the following steps,

(B1) according to the 500kV transformer station idle full compensation principle of perception, each transformer station is determined Inductive reactive power compensation configuration needs, the reactor in conjunction with each transformer station configures, and obtains each power transformation The perceptual no-power vacancy stood, computing formula such as following formula,

ΔQ_{In_i}=Q_i-Q_Li

Wherein, Δ Q_{In_i}For the idle configuration vacancy of the perception needed for transformer station i；Q_iFor transformer station i The half of line charge power, Q_LiFor the transformer station i idle configuration capacity of existing perception；

(B2) according to the principle of compensation 500kV transformer station maximum capacitive no-power vacancy, right by seat Each transformer station carries out main transformer and circuit maximum reactive loss and calculates, obtain the capacitive of each transformer station without Merit configuration vacancy, computing formula such as following formula,

ΔQ_{Ca_i}=Q_Loss+ Q_L+Q_x-Q_c

Wherein, Δ Q_{Ca_i}Capacitive reactive power for transformer station i configures vacancy；Q_LossNothing for transformer station Merit is lost；Q_LFor transformer station's all outlets reactive loss；Q_xCircuit for transformer station's configuration is high Anti-capacity；Q_cFor transformer station's existing capacitive reactive power configuration capacity.

The aforesaid electric network reactive compensation Optimal Configuration Method solved based on many knapsack problems, it is special Levy and be: step (C), by abstract for reactive-load compensation balanced arrangement problem for many knapsacks Solve problems, Set up many knapsack problems solving model, comprise the following steps,

(C1) for transformer station i, its no-power vacancy is Δ Q_i, as Δ Q_iThe website of ＞ 0 represents scarce Idle, it is defined as article, Item Weight is Δ Q_i；As Δ Q_iThe website of ＜ 0 represents idle more than needed, Being defined as knapsack, knapsack capacity is-Δ Q_i；

(C2) for the article i being joined directly together with knapsack j, its value pi relative to knapsack j_j The inverse of line taking road per unit reactance；The article i not being joined directly together with knapsack j, its relative to Value pi of knapsack j_jIt is set to 0；

(C3) turning to object function maxf (x) with knapsack loaded article Maximum Value, foundation is many Knapsack problem solving model is as follows,

$\begin{array}{c}\max f\left(x\right)=\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{m}{\Σ}}{p}_{ij}{x}_{ij}\\ s.t.\left\{\begin{array}{ccc}\underset{j=1}{\overset{m}{\Σ}}{x}_{ij}\≤1& i=1,2,...,n& \\ \underset{i=1}{\overset{n}{\Σ}}{w}_i{x}_{ij}\≤b_j& j=1,2,...,m& \\ x_{ij}\∈\{0,1\}& i=1,2,...,n& j=1,2,\mathrm{...},m\end{array}\right.\end{array}$

Wherein, n is article number, i.e. there is transformer station's number of no-power vacancy；M is knapsack Number, the most idle transformer station's number more than needed；pi_jFor the article i value relative to knapsack j；w_iFor The weight of article i, takes w_i=Δ Q_i；b_jFor the capacity of knapsack j, take b_j=-Δ Q_j；First about Bundle conditionI=1,2 ..., n, represent that 1 article at most can only be placed into a knapsack； Second constraintsJ=1,2 ..., m represents the Item Weight sum in knapsack Not can exceed that knapsack capacity；3rd constraints x_ij∈ 0,1} i=1,2 ..., n j=1,2 ..., m Represent x_ij0 or 1, x can only be taken_ijFor characterizing whether article i puts into knapsack j, it is to represent thing when 0 Product i does not puts into knapsack j, is to represent when 1 that article i puts into knapsack j.

The aforesaid electric network reactive compensation Optimal Configuration Method solved based on many knapsack problems, it is special Levy and be: step (D), use genetic algorithm to step (C) many knapsack problems solving model Solving, obtain the reactive balance result of optimum, wherein, individual E coded system is as follows:

E=(e₁,e₂,...,e_n)

Wherein, element e_iCharacterize the knapsack numbering that article i is put into, and e_i∈ 0,1,2 ..., m}, I=1,2 ..., n, m are knapsack number, and n is article number, chosen, the something lost that intersects, make a variation After passing operation, obtain optimum individual E_op, E_opCorresponding to optimum reactive balance result.

The invention has the beneficial effects as follows: the electric network reactive-load solved based on many knapsack problems of the present invention Compensation optimizing configuration system and method, by abstract for the transformer station that there is no-power vacancy for article, deposit It is considered as knapsack in idle transformer station more than needed, sets up and be up to mesh with knapsack loaded article total value The Optimized model of scalar functions, then use genetic algorithm for solving to obtain the reactive balance relation of full electric network, Formulating reactive-load compensation finally according to the most regular principle and increase formula case, whole process analysis is by number According to input module, no-power vacancy computing module, analysis of reactive power balance module, compensate increasing partition analysis Module and data outputting module are automatically performed, it is not necessary to human intervention, improve electric network reactive-load planning Efficiency, be favorably improved the reactive Voltage Optimum operation level of electrical network, there is good application Prospect.

Accompanying drawing explanation

Fig. 1 is that the electric network reactive compensation solved based on many knapsack problems of the present invention is distributed rationally The system block diagram of system.

Fig. 2 is that the electric network reactive compensation solved based on many knapsack problems of the present invention is distributed rationally The flow chart of system.

Fig. 3 is the schematic diagram of one embodiment of the invention.

Detailed description of the invention

Below in conjunction with Figure of description, the present invention will be further described.Following example are only For technical scheme is clearly described, and can not limit the present invention's with this Protection domain.

As it is shown in figure 1, the electric network reactive compensation optimization solved based on many knapsack problems of invention is joined Put system, including the data input module 1 being sequentially connected with, no-power vacancy computing module 2, nothing Merit equilibrium analysis module 3, compensation increase partition analysis module 4 and data outputting module 5,

Described data input module 1, for obtaining the substation data of electrical network, packet purse rope Frame topological structure, line impedance parameter, capacity of reactive power compensation device parameter；

Described no-power vacancy computing module 2, for calculating the no-power vacancy of transformer station, described nothing Merit vacancy includes perception no-power vacancy and capacitive reactive power vacancy；

Described analysis of reactive power balance module 3, for by abstract for reactive-load compensation balanced arrangement problem being Many knapsacks Solve problems, sets up many knapsack problems solving model, and uses genetic algorithm to many back ofs the body Bag problem solves, and obtains the reactive balance result of optimum；

Described compensation increases partition analysis module 4, needs for being calculated the reactive balance result of optimum Increase the reactive compensation capacity joined, put down according to the idle of optimum that analysis of reactive power balance module 3 obtains Weighing apparatus result, calculates the residual capacity of each knapsack, for being not put into the remaining articles of any knapsack, Searching whether the knapsack being directly connected has a residual capacity, idle after deduction residual capacity lacks Volume is to increase the reactive compensation capacity joined；If a knapsack is connected with two or more remaining articles, Then in the remaining articles big relative to this knapsack value, preferentially deduct residual capacity；

Described data outputting module 5, for increasing the analysis result of partition analysis module 4 according to compensation, Final compensation increases complexion amount and exports to use the most regular principle to determine.

As in figure 2 it is shown, the electric network reactive compensation optimization solved based on many knapsack problems of the present invention Collocation method, the electric network reactive compensation optimization solved based on many knapsack problems that the method operates in In configuration system, comprise the following steps,

Step (A), obtains the network frame topology structure of transformer station, line by data input module 1 Road impedance parameter, capacity of reactive power compensation device parameter；

Step (B), the data obtained according to step (A), by no-power vacancy computing module 2 Calculating the no-power vacancy of transformer station, described no-power vacancy includes perception no-power vacancy and capacitive reactive power Vacancy, calculates the no-power vacancy of transformer station by no-power vacancy computing module 2, including following step Suddenly,

(B1) according to the 500kV transformer station idle full compensation principle of perception, each transformer station is determined Inductive reactive power compensation configuration needs, the reactor in conjunction with each transformer station configures, and obtains each power transformation The perceptual no-power vacancy stood, computing formula such as following formula,

ΔQ_{In_i}=Q_i-Q_Li

Wherein, Δ Q_{In_i}For the idle configuration vacancy of the perception needed for transformer station i；Q_iFor transformer station i The half of line charge power, Q_LiFor the transformer station i idle configuration capacity of existing perception；

(B2) according to the principle of compensation 500kV transformer station maximum capacitive no-power vacancy, right by seat Each transformer station carries out main transformer and circuit maximum reactive loss and calculates, obtain the capacitive of each transformer station without Merit configuration vacancy, computing formula such as following formula,

ΔQ_{Ca_i}=Q_Loss+Q_L+Q_x-Q_c

Wherein, Δ Q_{Ca_i}Capacitive reactive power for transformer station i configures vacancy；Q_LossNothing for transformer station Merit is lost；Q_LFor transformer station's all outlets reactive loss；Q_xCircuit for transformer station's configuration is high Anti-capacity；Q_cFor transformer station's existing capacitive reactive power configuration capacity；

Step (C), by analysis of reactive power balance module 3, by reactive-load compensation balanced arrangement problem Abstract for many knapsacks Solve problems, set up many knapsack problems solving model, comprise the following steps,

(C1) for transformer station i, its no-power vacancy is Δ Q_i, as Δ Q_iThe website of ＞ 0 represents scarce Idle, it is defined as article, Item Weight is Δ Q_i；As Δ Q_iThe website of ＜ 0 represents idle more than needed, Being defined as knapsack, knapsack capacity is-Δ Q_i；

(C2) for the article i being joined directly together with knapsack j, its value pi relative to knapsack j_j The inverse of line taking road per unit reactance；The article i not being joined directly together with knapsack j, its relative to Value pi of knapsack j_jIt is set to 0；

(C3) turning to object function maxf (x) with knapsack loaded article Maximum Value, foundation is many Knapsack problem solving model is as follows,

$\begin{array}{c}\max f\left(x\right)=\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{m}{\Σ}}{p}_{ij}{x}_{ij}\\ s.t.\left\{\begin{array}{ccc}\underset{j=1}{\overset{m}{\Σ}}{x}_{ij}\≤1& i=1,2,...,n& \\ \underset{i=1}{\overset{n}{\Σ}}{w}_i{x}_{ij}\≤b_j& j=1,2,...,m& \\ x_{ij}\∈\{0,1\}& i=1,2,...,n& j=1,2,\mathrm{...},m\end{array}\right.\end{array}$

Wherein, n is article number, i.e. there is transformer station's number of no-power vacancy；M is knapsack Number, the most idle transformer station's number more than needed；pi_jFor the article i value relative to knapsack j；w_iFor The weight of article i, takes w_i=Δ Q_i；b_jFor the capacity of knapsack j, take b_j=-Δ Q_j；First about Bundle conditionI=1,2 ..., n, represent that 1 article at most can only be placed into a knapsack； Second constraintsJ=1,2 ..., m represents the Item Weight sum in knapsack Not can exceed that knapsack capacity；3rd constraints x_ij∈ 0,1} i=1,2 ..., n j=1,2 ..., m Represent x_ij0 or 1, x can only be taken_ijFor characterizing whether article i puts into knapsack j, it is to represent thing when 0 Product i does not puts into knapsack j, is to represent when 1 that article i puts into knapsack j；

Step (D), by analysis of reactive power balance module 3, uses genetic algorithm to step (C) Many knapsack problems solving model solves, and obtains the reactive balance result of optimum, genetic algorithm Individual E coded system as follows:

E=(e₁,e₂,...,e_n)

Wherein, element e_iCharacterize the knapsack numbering that article i is put into, and e_i∈ 0,1,2 ..., m}, I=1,2 ..., n, m are knapsack number, and n is article number, chosen, the something lost that intersects, make a variation After passing operation, obtain optimum individual E_op, E_opCorresponding to optimum reactive balance result；

Step (E), according to optimum reactive balance result, increases partition analysis module 4 by compensating Calculating the residual capacity of each knapsack, the reactive balance result obtaining optimum need to increase the reactive-load compensation joined Capacity；

Step (F), need to increase, according to the reactive balance result obtaining optimum of step (E), the nothing joined Merit compensation capacity, uses the most regular principle to determine final benefit by data outputting module 5 Repay increasing complexion amount and export.

The electric network reactive compensation side of distributing rationally solved based on many knapsack problems according to the present invention Method, the reactive power compensation ability configuration process of concrete introduction topological structure of electric as shown in Figure 3,

(1) for 500kV transformer station, every half stood according to 500kV line length Line charging power is calculated, primarily determines that each power transformation according to the idle full compensation principle of perception Stand inductive reactive power compensation configuration needs, in conjunction with the reactor configuring condition at this station, obtain each change The perceptual no-power vacancy in power station, computing formula is as follows:

ΔQ_{In_i}=Q_i-Q_Li

Wherein, Δ Q_{In_i}For the idle configuration vacancy of the perception needed for transformer station i；Q_iEnter for transformer station i The half of outlet charge power, Q_LiFor the transformer station i idle configuration capacity of existing perception；

According to the principle of compensation 500kV transformer station maximum capacitive no-power vacancy, by seat to 500kV Transformer station carries out main transformer and circuit maximum reactive loss calculates, and obtains the capacitive reactive power of each transformer station Configuration vacancy, computing formula is as follows:

ΔQ_{Ca_i}=Q_Loss+Q_L+Q_x-Q_c

Wherein, Δ Q_{Ca_i}Capacitive reactive power for transformer station i configures vacancy；Q_LossNothing for transformer station Merit is lost；Q_LFor transformer station's all 500kV outlet reactive loss, (line loss deducts circuit Charge power)；Q_xThe high anti-capacity of circuit for transformer station's configuration；Q_cFor the existing appearance of transformer station The idle configuration capacity of property；

As a example by perception is idle, as it is shown on figure 3, calculate the no-power vacancy of each node respectively For: Δ Q₁、ΔQ₂、ΔQ₃、ΔQ₄、ΔQ₅、ΔQ₆。

(2) by analysis of reactive power balance module 3 by abstract for reactive-load compensation balanced arrangement problem it is Many knapsacks Solve problems, for transformer station i, its no-power vacancy is Δ Q_i, as Δ Q_iThe website of ＞ 0 Representing scarce idle, be defined as article, Item Weight is Δ Q_i；As Δ Q_iThe website of ＜ 0 indicate without Merit is had more than needed, and is defined as knapsack, and knapsack capacity is-Δ Q_i；Here obtain node 1 and node 5 is Knapsack, capacity is respectively-Δ Q₁With-Δ Q₅；Node 2, node 3, node 4 and node 6 are Article, weight is respectively Δ Q₂、ΔQ₃、ΔQ₄With Δ Q₆；

(3) by analysis of reactive power balance module 3, for the article i being joined directly together with knapsack j, Its value p relative to knapsack j_ijThe inverse of line taking road per unit reactance；It is not connected with knapsack j Article i, its value p relative to knapsack j_ijBeing set to 0, each article are relative to the value of knapsack It is listed in the table below 1:

The each article of table 1 are relative to the value of knapsack

	Article 1 (node 2)	Article 2 (node 3)	Article 3 (node 4)	Article 4 (node 6)
					Knapsack 1 (node 1)	1/x1	0	1/x3	0
Knapsack 2 (node 5)	0	1/x4	0	1/x6

With knapsack loaded article Maximum Value as optimization aim, set up many knapsack problems solving model As follows:

$\begin{array}{c}\max f\left(x\right)=\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{m}{\Σ}}{p}_{ij}{x}_{ij}\\ s.t.\left\{\begin{array}{ccc}\underset{j=1}{\overset{m}{\Σ}}{x}_{ij}\≤1& i=1,2,...,n& \\ \underset{i=1}{\overset{n}{\Σ}}{w}_i{x}_{ij}\≤b_j& j=1,2,...,m& \\ x_{ij}\∈\{0,1\}& i=1,2,...,n& j=1,2,\mathrm{...},m\end{array}\right.\end{array}$

Wherein, n=4 is article number, i.e. there is transformer station's number of no-power vacancy；M=2 is Knapsack number, the most idle transformer station's number more than needed；p_ijFor the article i valency relative to knapsack j Value；w_iFor the weight of article i, take w_i=Δ Q_i；b_jFor the capacity of knapsack j, take b_j=-Δ Q_j； First constraints represents that 1 article at most can only be placed into a knapsack；Second constraint Condition represents that the Item Weight sum in knapsack not can exceed that knapsack capacity；3rd constraints Represent x_ij0 or 1 can only be taken, be when 0, to represent that article i does not puts into knapsack j, be to represent thing when 1 Product i puts into knapsack j.

(4) by analysis of reactive power balance module 3, use genetic algorithm that many knapsack problems are asked Solving model to solve, individual E coded system is as follows:

E=(e₁,e₂,...,e_n)

Wherein, element e_i∈ 0,1,2 ..., m}, i=1,2 ..., n, chosen, intersect, make a variation heredity behaviour After work, obtain optimum individual E_op=(1,2,1,0), i.e. corresponding to optimum reactive balance result, tool Body is that the no-power vacancy of node 2 and node 4 is balanced by node 1, the no-power vacancy of node 3 by Node 5 balances, and node 6 vacancy is not balanced in optimization temporarily.

(5) according to the optimized results of analysis of reactive power balance module 3, the residue of each knapsack is calculated Capacity, for being not put into the remaining articles of any knapsack, searches the knapsack being directly connected Residual capacity, the no-power vacancy after deduction residual capacity whether is had to be and need to increase the reactive-load compensation joined Capacity, if a knapsack is connected with two or more remaining articles, then relative to this knapsack valency The remaining articles that value is big is preferentially deducted residual capacity；

(6) by data outputting module 5 according to compensating the analysis result increasing partition analysis module 4, Final compensation increases complexion amount and exports to use the most regular principle to determine.

In sum, the electric network reactive compensation optimization solved based on many knapsack problems of the present invention is joined Put system and method, by abstract for the transformer station that there is no-power vacancy for article, exist idle more than needed Transformer station be considered as knapsack, set up and be up to the excellent of object function with knapsack loaded article total value Change model, then use genetic algorithm for solving to obtain the reactive balance relation of full electric network, finally according to The most regular principle is formulated reactive-load compensation and is increased formula case, and whole process analysis inputs mould by data Block, no-power vacancy computing module, analysis of reactive power balance module, compensation increase partition analysis module sum It is automatically performed according to output module, it is not necessary to human intervention, improves the efficiency of electric network reactive-load planning, It is favorably improved the reactive Voltage Optimum operation level of electrical network, has a good application prospect.

The ultimate principle of the present invention, principal character and advantage have more than been shown and described.The industry Skilled person will appreciate that, the present invention is not restricted to the described embodiments, above-described embodiment and The principle that the present invention is simply described described in description, without departing from spirit and scope of the invention On the premise of, the present invention also has various changes and modifications, and these changes and improvements both fall within to be wanted In seeking the scope of the invention of protection.Claimed scope by appending claims and Its equivalent defines.

Claims (6)

1. the electric network reactive compensation Optimizing Configuration System solved based on many knapsack problems, its feature Be: the data input module (1) that includes being sequentially connected with, no-power vacancy computing module (2), Analysis of reactive power balance module (3), compensation increase partition analysis module (4) and data outputting module (5),

Described data input module (1), for obtaining the substation data of electrical network, data include Network frame topology structure, line impedance parameter, capacity of reactive power compensation device parameter；

Described no-power vacancy computing module (2) is for calculating the no-power vacancy of transformer station, described No-power vacancy includes perception no-power vacancy and capacitive reactive power vacancy；

Described analysis of reactive power balance module (3), for by abstract for reactive-load compensation balanced arrangement problem For many knapsacks Solve problems, set up many knapsack problems solving model, and use genetic algorithm to many Knapsack problem solves, and obtains the reactive balance result of optimum；

Described compensation increases partition analysis module (4), for being calculated the reactive balance result of optimum The reactive compensation capacity joined need to be increased, according to analysis of reactive power balance module (3) obtain optimum Reactive balance result, calculates the residual capacity of each knapsack, remains for being not put into any knapsack Excess product, search whether the knapsack being directly connected has residual capacity, after deduction residual capacity No-power vacancy be to increase the reactive compensation capacity joined；

Described data outputting module (5), for according to compensating dividing of increasing partition analysis module (4) Analysis result, final compensation increases complexion amount and exports to use the most regular principle to determine.

The electric network reactive compensation solved based on many knapsack problems the most according to claim 1 Optimizing Configuration System, it is characterised in that: described compensate increase partition analysis module (4), for not by Put into the remaining articles of any knapsack, search whether the knapsack being directly connected has residual capacity, No-power vacancy after deduction residual capacity is to increase the reactive compensation capacity joined, if knapsack with Two or more remaining articles are connected, then be worth in big remaining articles excellent relative to this knapsack First deduct residual capacity.

3. the electric network reactive compensation Optimal Configuration Method solved based on many knapsack problems, its feature Be: the method operate in the electrical network solved based on many knapsack problems described in claim 1 without In merit compensation optimizing configuration system, comprise the following steps,

Step (A), by data input module (1) obtain transformer station network frame topology structure, Line impedance parameter, capacity of reactive power compensation device parameter；

Step (B), the data obtained according to step (A), by no-power vacancy computing module (2) calculating the no-power vacancy of transformer station, described no-power vacancy includes perception no-power vacancy and appearance Property no-power vacancy；

Step (C), by analysis of reactive power balance module (3), asks reactive-load compensation balanced arrangement Inscribe abstract for many knapsacks Solve problems, set up many knapsack problems solving model；

Step (D), by analysis of reactive power balance module (3), uses genetic algorithm to step (C) Many knapsack problems solving model solves, and obtains the reactive balance result of optimum；

Step (E), according to optimum reactive balance result, increases partition analysis module (4) by compensating Calculating the residual capacity of each knapsack, the reactive balance result obtaining optimum need to increase the reactive-load compensation joined Capacity；

Step (F), need to increase, according to the reactive balance result obtaining optimum of step (E), the nothing joined Merit compensation capacity, uses the most regular principle to determine finally by data outputting module (5) Compensation increase complexion amount exporting.

The electric network reactive compensation solved based on many knapsack problems the most according to claim 3 is excellent Change collocation method, it is characterised in that: step (B), by no-power vacancy computing module (2) Calculate the no-power vacancy of transformer station, comprise the following steps,

(B1) according to the 500kV transformer station idle full compensation principle of perception, each transformer station is determined Inductive reactive power compensation configuration needs, the reactor in conjunction with each transformer station configures, and obtains each power transformation The perceptual no-power vacancy stood, computing formula such as following formula,

ΔQ_{In_i}=Q_i-Q_Li

Wherein, Δ Q_{In_i}For the idle configuration vacancy of the perception needed for transformer station i；Q_iFor transformer station i The half of line charge power, Q_LiFor the transformer station i idle configuration capacity of existing perception；

(B2) according to the principle of compensation 500kV transformer station maximum capacitive no-power vacancy, right by seat Each transformer station carries out main transformer and circuit maximum reactive loss and calculates, obtain the capacitive of each transformer station without Merit configuration vacancy, computing formula such as following formula,

ΔQ_{Ca_i}=Q_Loss+Q_L+Q_x-Q_c

Wherein, Δ Q_{Ca_i}Capacitive reactive power for transformer station i configures vacancy；Q_LossNothing for transformer station Merit is lost；Q_LFor transformer station's all outlets reactive loss；Q_xCircuit for transformer station's configuration is high Anti-capacity；Q_cFor transformer station's existing capacitive reactive power configuration capacity.

The electric network reactive compensation solved based on many knapsack problems the most according to claim 3 is excellent Change collocation method, it is characterised in that: step (C), by abstract for reactive-load compensation balanced arrangement problem For many knapsacks Solve problems, set up many knapsack problems solving model, comprise the following steps,

(C1) for transformer station i, its no-power vacancy is Δ Q_i, as Δ Q_iThe website of ＞ 0 represents scarce Idle, it is defined as article, Item Weight is Δ Q_i；As Δ Q_iThe website of ＜ 0 represents idle more than needed, Being defined as knapsack, knapsack capacity is-Δ Q_i；

(C2) for the article i being joined directly together with knapsack j, its value pi relative to knapsack j_j The inverse of line taking road per unit reactance；The article i not being joined directly together with knapsack j, its relative to Value p of knapsack j_ijIt is set to 0；

(C3) turning to object function maxf (x) with knapsack loaded article Maximum Value, foundation is many Knapsack problem solving model is as follows,

$\max f\left(x\right)=\underset{j=1}{\overset{m}{\Σ}}\underset{i=1}{\overset{n}{\Σ}}{p}_{ij}{x}_{ij}$

$s.t.\left\{\begin{array}{ccc}\underset{j=1}{\overset{m}{\Σ}}{x}_{ij}\≤1& i=1,2,...,n& \\ \underset{i=1}{\overset{m}{\Σ}}{w}_i{x}_{ij}\≤b_j& j=1,2,...,m& \\ x_{ij}\∈\{0,1\}& i=1,2,...,n& j=1,2,...,m\end{array}\right.$

Wherein, n is article number, i.e. there is transformer station's number of no-power vacancy；M is knapsack Number, the most idle transformer station's number more than needed；p_ijFor the article i value relative to knapsack j；w_iFor The weight of article i, takes w_i=Δ Q_i；b_jFor the capacity of knapsack j, take b_j=-Δ Q_j；First about Bundle conditionRepresent that 1 article at most can only be placed into a knapsack； Second constraintsRepresent the Item Weight sum in knapsack Not can exceed that knapsack capacity；3rd constraints x_ij∈ 0,1} i=1,2 ..., n j=1,2 ..., m Represent x_ij0 or 1, x can only be taken_ijFor characterizing whether article i puts into knapsack j, it is to represent thing when 0 Product i does not puts into knapsack j, is to represent when 1 that article i puts into knapsack j.

The electric network reactive compensation solved based on many knapsack problems the most according to claim 3 Optimal Configuration Method, it is characterised in that: step (D), use genetic algorithm to step (C) Many knapsack problems solving model solves, and obtains the reactive balance result of optimum, wherein, individual Body E coded system is as follows:

E=(e₁,e₂,...,e_n)

Wherein, element e_iCharacterize the knapsack numbering that article i is put into, and e_i∈ 0,1,2 ..., m}, I=1,2 ..., n, m are knapsack number, n is article number, chosen, intersect, make a variation After genetic manipulation, obtain optimum individual E_op, E_opCorresponding to optimum reactive balance result.