CN104795823A - Power grid minimum newly-increased reactive compensation capacity calculation method - Google Patents

Abstract

The invention relates to a power grid minimum newly-increased reactive compensation capacity calculation method. Based on a power gird load predicated value, the method takes a minimum power grid newly-increased reactive compensation capacity as a target function; meanwhile, reactive compensation minimum capacity out-of-limits of a capacitor and an electric reactor in Power System Voltage and Reactive Power Technology Guides, and constraint conditions including voltage upper and lower limits, line active tides and the like of a transformer substation bus under three operation manners including a highest load, a lowest load and a normal load are considered, and a whole model is decomposed into an upper-layer model capable of being directly solved by a linear planning algorithm and a lower-layer optimization model composed of non-linear constraint conditions through a layered optimization concept; a final optimized result meets the related technology guides through the continuous optimization between the two layers of models, and the voltage operation requirements of an electric power system are met under a plurality of load levels.

Description

The minimum newly-increased reactive compensation capacity computational methods of electrical network based on hierarchy optimization

Technical field

The invention belongs to reactive compensator of electrical network and distribute field rationally, relate to the minimum newly-increased reactive compensation capacity computational methods of a kind of electrical network based on hierarchy optimization thought.

Background technology

The balance locally of reactive power can ensure good quality of voltage, and reactive power lacks and the superfluous safe operation being all unfavorable for ME for maintenance quality and electrical network.

For a long time, people generally concentrate in the bigger than normal and voltage collapse accident that may cause of the low voltage, the network loss that are caused by idle shortage the concern of voltage problem.But along with developing rapidly of electrical network, its feature is also corresponding to change, be mainly reflected in the continuous reinforcement of upper strata main grid structure, the a large amount of uses of cable line in urban distribution network, wind-powered electricity generation, solar energy distributed power supply access electrical network in a large number, and load peak-valley difference strengthens day by day, part Line Flow is lighter, inductive reactive power compensation is not enough, and under load valley phase or little operational mode, voltage is higher even seriously out-of-limit.Cause thus be idlely difficult to in-situ balancing, circuit is idle, and flowing increases, between adjoining area or different voltage layer idle pass through increase, overtension jeopardizes the problem such as equipment and electric power netting safe running.

In recent years, existing multidigit scholar distributes rationally electric network reactive compensation capacity and launches research both at home and abroad, have scholar to propose the abundant intensity of one " power plant average power factor " to the perceptual reactive capability of 220kV and following electrical network to evaluate, and propose a kind of Scheme of Reactive Power Compensation, but do not relate to the selection in reactive capability installing place; Or by the relation between line transmission load and reactive loss, calculate the reactive compensation capacity of wall scroll 500kV circuit under different running method, but be confined to the optimal reactive power allocation of single line.Also scholar is had to adopt Reactive Power Optimization Algorithm for Tower based on catastrophic genetic algorithm, idle work optimization calculating is carried out to the load lowest trough trend section under the little operational mode of area power grid, obtain the concrete compensation scheme comprising compensation point and each point compensation capacity, but do not consider the compensation capacity optimization of capacitor.In addition, on the basis of idle in-situ balancing and layering balance principle, researcher proposes the algorithm based on economic pressure reduction concept, inner link between the active loss caused by analysis transmission reactive power, node voltage, reactive compensation capacity, the reasonable disposition of reactive power compensator is optimized, but is only applicable to the configuration of dynamic reactive compensation device.

Summary of the invention

The object of the invention is, under guarantee electric power netting safe running, to reduce the cost of network reactive-load compensation equipment configuration, give a kind of electrical network based on hierarchy optimization minimum newly-increased reactive compensation capacity computational methods; Its adopt can break through compensation precision that in the past traditional reactive compensation configuration method exists not, the various disadvantages such as voltage bound cannot to be met under different load operational mode, provide the minimum newly-increased reactive compensation capacity computational methods of electrical network, farthest newly-increased needed for each node of reduction electrical network perception is idle and capacitive reactive power compensation capacity, for the saving energy and decreasing loss work of electrical network provides strong theoretical foundation.

For this reason, the present invention adopts following technical scheme:

The minimum newly-increased reactive compensation capacity computational methods of electrical network based on hierarchy optimization, comprise the following steps:

(1) the load or burden without work distribution situation of area power grid is collected; The reactive requirement of all nodes under different load ruuning situation and relevant rudimentary data are determined based on network load prediction;

(2) Optimized model is set up; With the target function that the minimum newly-increased reactive compensation capacity of whole electrical network is model, set up Optimized model with the constraints that substation bus bar voltage bound under PQ node capacitor in " power system voltage and var fire protection technology " and reactor reactive power compensation minimum capacity and maximum load, minimum load and normal duty three kinds of operational modes, circuit effective power flow are model;

(3) based on layering thought, above-mentioned model is solved; Be upper and lower two-layer model according to the type of model constrained condition by whole model decomposition, by the successive ignition Load flow calculation between two-layer model until meet model Prescribed Properties;

(4) the final optimization pass result of the required newly-increased reactive compensation capacity sum of all transformer stations is calculated; When solving result meets the service requirement of the substation bus bar voltage that " power system voltage and var fire protection technology " and dispatching of power netwoks mechanism determine simultaneously, then this result is the minimum newly-increased reactive compensation capacity of whole electrical network.

Further, in described step (1), basic data comprises:

The existing condenser capacity Q of transformer station all PQ node _ciwith reactor capacity Q _li, each node transformer station reactive loss Δ Q _ti, each node transformer station all outlet charge powers sum Q _lCi, transformer station's high-voltage terminal side configuration in node high resistance total capacity Q _ki;

Wherein: i ∈ N _pQ, N _pQthe set of all PQ nodes in expression system, i represents i-th PQ node.

Meanwhile, the predicted load under the highest, normal and minimum load three kinds of operational modes is needed: node voltage value V _x,i; Admittance matrix coefficient correlation G between node i and node j _ij, B _ij; The phase angle difference θ of node i and node j _{x, ij}; The voltage magnitude bound V of node i _iand V _i; The effective power flow P of circuit between node i and node j _{x, ij}; The bound of circuit effective power flow between node i and node j with p _ij.

Wherein: i ∈ N, X ∈ High, the set of all nodes in Normal, Low, N expression system, High, Normal, Low represent the highest respectively, normal and minimum load three kinds of operational modes.

Further, in described step (2), the target function that Optimized model comprises is as follows:

$\min Q_S=\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}{\mathrm{\ΔQ}}_{\mathrm{Ci}}+\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}{\mathrm{\ΔQ}}_{\mathrm{Li}}---\left(1\right)$

Wherein: Q _sfor the newly-increased reactive power compensation total capacity of electrical network; N _pQfor PQ set of node; Δ Q _cifor node i increases condenser capacity newly; Δ Q _lirepresent that node i increases reactor capacity newly.

Further, in described step (2) in " power system voltage and var fire protection technology " PQ node capacitor and reactor reactive power compensation minimum capacity be constrained to Linear Constraints; Substation bus bar voltage bound under maximum load, minimum load and normal duty three kinds of operational modes, circuit effective power flow be constrained to Nonlinear Constraints.

Further, the Linear Constraints setting up Optimized model is specific as follows:

$\left\{\begin{array}{c}{Q}_{\mathrm{Ci}}+{\mathrm{\ΔQ}}_{\mathrm{Ci}}\≥{\mathrm{\ΔQ}}_{\mathrm{Ti}}-Q_{\mathrm{LCi}}+Q_{\mathrm{Ki}}\\ Q_{\mathrm{Li}}+{\mathrm{\ΔQ}}_{\mathrm{Li}}\≥Q_{\mathrm{LCi}}\×0.9\\ i\∈N_{\mathrm{PQ}}\end{array}\right.---\left(2\right)$

Wherein: Q _ciand Q _lirepresent the existing capacitor of node i, reactor capacity respectively; Δ Q _tirepresent the reactive loss of transformer station in node i, be made up of susceptance loss and reactance loss; Q _lCirepresent transformer station's all outlet charge powers sum in node i; Q _kirepresent the high resistance total capacity of transformer station's high-voltage terminal side configuration in node i.

Further, the Nonlinear Constraints setting up Optimized model is specific as follows:

${\mathrm{\ΔP}}_{X,i}=V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,j}(G_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}})$

${\mathrm{\ΔQ}}_{X,i}=Q_{L,i}-Q_{C,i}+V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,j}(G_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}})$

$\underset{\‾}{V_i}\≤V_{X,i}\≤{\stackrel{\‾}{V}}_i$

${\underset{\‾}{P}}_{\mathrm{ij}}\≤P_{X,\mathrm{ij}}\≤{\stackrel{\‾}{P}}_{\mathrm{ij}}$

i∈N,X∈High,Normal,Low

Wherein: Δ P _x,irepresent that node i is gained merit injecting power; Δ Q _x,irepresent the idle injecting power of node i except newly-increased reactive compensation capacity; V _x,irepresent the magnitude of voltage of node i; G _ij, B _ijfor the admittance matrix coefficient correlation between node i and node j; θ _{x, ij}represent the phase angle difference of node i and node j; with v _ibe respectively the voltage magnitude bound of node i; P _{x, ij}represent the effective power flow of circuit between node i and node j; with p _ijrepresent the bound of circuit effective power flow between node i and node j; The set of all nodes in N expression system; High, Normal, Low represent the highest respectively, normal and minimum load three kinds of operational modes, under above-mentioned three kinds of operational modes, and the Nonlinear Constraints shown in above the optimal solution of model all needs to meet.

Further, whole model decomposition is upper according to the type of model constrained condition by described step (3), lower two-layer model, by the successive ignition Load flow calculation between two-layer model until meet model Prescribed Properties and be specially: according to the model set up, target function and Linear Constraints are classified as upper strata Optimized model, Nonlinear Constraints is classified as lower floor's Optimized model, first use linear optimization algorithm directly to upper strata model solution, the optimum results obtained is brought underlying model into and is verified, single website for discontented afc voltage bound constraints is analyzed, and increase or reduce its perception or capacity reactive compensation capacity, Filled function so repeatedly, until meet all constraints.

The beneficial effect that the present invention reaches is: provide a kind of Optimized model for electric network reactive compensation configuration and algorithm, this model is based on network load predicted value, set up the minimum newly-increased reactive compensation capacity target function of electrical network, consider the highest simultaneously, normal and minimum load operation mode Down Highway voltage bound retrains and circuit effective power flow constraints, under guarantee electric power netting safe running, reduce the cost of network reactive-load compensation equipment configuration, it adopts the compensation precision can breaking through in the past traditional reactive compensation configuration method existence inadequate, the various disadvantages such as voltage bound under different load operational mode cannot be met, provide the minimum newly-increased reactive compensation capacity computational methods of electrical network, farthest newly-increased needed for each node of reduction electrical network perception is idle and capacitive reactive power compensation capacity, for the saving energy and decreasing loss work of electrical network provides strong theoretical foundation.

Accompanying drawing explanation

Fig. 1 is the calculation flow chart that hierarchy optimization of the present invention solves the computational methods of the minimum newly-increased reactive compensation capacity of electrical network.

Embodiment

The technological means realized for making the present invention, creation characteristic, reaching object and effect is easy to understand, below in conjunction with accompanying drawing and specific embodiment, technical scheme of the present invention being further elaborated.

With reference to Fig. 1, for 500kV electrical network, the minimum newly-increased reactive compensation capacity computational methods of a kind of electrical network based on hierarchy optimization, comprise the following steps:

(1) obtain the relevant rudimentary data of Jiangsu 500 kV power grid original reactive capability configuration, carry out network load prediction under the highest, normal, minimum load operation condition, calculate transformer reactive loss and the line charging power of all transformer stations;

(2) set up Optimized model, specifically comprise:

(2a) with the target function that the minimum newly-increased reactive compensation capacity of whole electrical network is model:

$\min Q_S=\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}{\mathrm{\ΔQ}}_{\mathrm{Ci}}+\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}{\mathrm{\ΔQ}}_{\mathrm{Li}}---\left(1\right)$

In formula: Q _sfor the newly-increased reactive power compensation total capacity of 500kV electrical network; N _pQfor PQ set of node; Δ Q _cifor node i increases condenser capacity newly; Δ Q _lirepresent that node i increases reactor capacity newly.

(2b) set up the Linear Constraints of Optimized model, comprise PQ node capacitor and the constraint of reactor reactive power compensation minimum capacity in " power system voltage and var fire protection technology ":

$\left\{\begin{array}{c}{Q}_{\mathrm{Ci}}+{\mathrm{\ΔQ}}_{\mathrm{Ci}}\≥{\mathrm{\ΔQ}}_{\mathrm{Ti}}-Q_{\mathrm{LCi}}+Q_{\mathrm{Ki}}\\ Q_{\mathrm{Li}}+{\mathrm{\ΔQ}}_{\mathrm{Li}}\≥Q_{\mathrm{LCi}}\×0.9\\ i\∈N_{\mathrm{PQ}}\end{array}\right.---\left(2\right)$

In formula: Q _ciand Q _lirepresent the existing capacitor of node i, reactor capacity respectively; Δ Q _tirepresent the reactive loss of transformer station in node i, be made up of susceptance loss and reactance loss; Q _lCirepresent transformer station's all 500kV outlets charge power sum in node i; Q _kirepresent the high resistance total capacity of transformer station's high-voltage terminal side configuration in node i.

(2c) set up the Nonlinear Constraints of Optimized model, comprise the constraints such as 500kV substation bus bar voltage bound, circuit effective power flow under maximum load, minimum load and normal duty three kinds of operational modes:

${\mathrm{\ΔP}}_{X,i}=V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,j}(G_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}})$

${\mathrm{\ΔQ}}_{X,i}=Q_{L,i}-Q_{C,i}+V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,j}(G_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}})$

$\underset{\‾}{V_i}\≤V_{X,i}\≤{\stackrel{\‾}{V}}_i$

${\underset{\‾}{P}}_{\mathrm{ij}}\≤P_{X,\mathrm{ij}}\≤{\stackrel{\‾}{P}}_{\mathrm{ij}}$

i∈N,X∈High,Normal,Low

In formula: Δ P _x,irepresent that node i is gained merit injecting power; Δ Q _x,irepresent the idle injecting power of node i except newly-increased reactive compensation capacity; V _x,irepresent the magnitude of voltage of node i; G _ij, B _ijfor the admittance matrix coefficient correlation between node i and node j; θ _{x, ij}represent the phase angle difference of node i and node j; with v _ibe respectively the voltage magnitude bound of node i; P _{x, ij}represent the effective power flow of circuit between node i and node j; with p _ijrepresent the bound of circuit effective power flow between node i and node j; The set of all nodes in N expression system; High, Normal, Low represent the highest respectively, normal and minimum load three kinds of operational modes, under above-mentioned three kinds of operational modes, and the Nonlinear Constraints shown in above the optimal solution of model all needs to meet.

(3) based on layering thought, above-mentioned model is solved: be upper and lower two models according to the type of model constrained condition by whole model decomposition, target function and Linear Constraints are classified as upper strata Optimized model, Nonlinear Constraints are divided into lower floor's Optimized model.First use linear optimization algorithm directly to upper strata model solution, the optimum results obtained is brought underlying model into and is verified, single website for discontented afc voltage bound constraints is analyzed, and increase or reduce its perception or capacity reactive compensation capacity, by the repeatedly Filled function between two-layer model, until substation bus bar voltage all meets the requirement of voltage bound.

(4) export all reactive compensation capacity of substation sums final optimization pass result; When above-mentioned solving result meets the service requirement of 500kV substation bus bar voltage of " power system voltage and var fire protection technology " and dispatching of power netwoks mechanism simultaneously, draw the minimum newly-increased reactive compensation capacity of final whole electrical network, and each reactive compensation capacity of substation configuration capacity in the 500kv electrical network of Jiangsu.

Jiangsu 500kV transformer station quantity in 2012 is 37, and through this method, the newly-increased reactive compensation capacity needed for each transformer station is as shown in the table.

Table 1. Jiangsu 500kV transformer station increases the final optimization pass result of reactive compensation capacity newly

Note: above showing numerical value in bracket is after submodel optimization calculates, newly-increased reactive compensation capacity.

From upper table result of calculation, in Jiangsu Power Grid 37 500kV transformer stations, 31 transformer stations need increase reactor configuration capacity, and wherein change required reactor capacity in Changshu is maximum, is 303.88MVar; 4 transformer stations need to increase capacitor arrangements capacity, and it is maximum that Ai Tang becomes required condenser capacity, is 78.57MVar.

For 2012 the end of the year Jiangsu Power Grid the idle configuration capacity of 500kV grid structure and transformer station, consider Jiangsu Power Grid (2015) development in the recent period, in conjunction with the inspection of the scene of a crime and analysis, propose following 5 500kV transformer stations in southern Jiangsu and install the suggestion of inductive reactive power compensation device additional.Specifically comprise Hui Fengshan and become enlarging 120Mvar reactor, the Maoshan Mountain becomes enlarging 120Mvar reactor, and Tianmu Lake becomes enlarging 60Mvar reactor, and Yushan becomes enlarging 60Mvar reactor, and wood is shown disrespect on and become enlarging 120Mvar reactor.While final optimization pass result ensures that 500kV substation bus bar voltage all rationally runs under the highest, normal and minimum load operation mode, newly-increased reactive compensation capacity is minimum, effectively improves 500kV line voltage lsafety level.

The present invention is directed to the deficiency of existing electric network reactive compensation capacity optimized algorithm, a kind of Optimized model for electric network reactive compensation configuration and algorithm are proposed, model is based on network load predicted value, set up the minimum newly-increased reactive compensation capacity target function of electrical network, consider the highest, normal and minimum load operation mode Down Highway voltage bound constraint and circuit effective power flow constraints simultaneously.Project team is in the existing reactive capability configure base of Jiangsu Power Grid, and project team adopts " the minimum newly-increased reactive compensation capacity computational methods of the electrical network based on hierarchy optimization " to be optimized analysis to Jiangsu 500 kV power grid reactive compensation capacity.

More than show and describe general principle of the present invention and principal character and advantage of the present invention.The technical staff of the industry should understand; the present invention is not restricted to the described embodiments; what describe in above-described embodiment and specification just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.Application claims protection range is defined by appending claims and equivalent thereof.

Claims (7)

1., based on the minimum newly-increased reactive compensation capacity computational methods of electrical network of hierarchy optimization, comprise the following steps:

(1) the load or burden without work distribution situation of area power grid is collected; The reactive requirement of all nodes under different load ruuning situation and relevant rudimentary data are determined based on network load prediction;

(2) Optimized model is set up; With the target function that the minimum newly-increased reactive compensation capacity of whole electrical network is model, set up Optimized model with the constraints that substation bus bar voltage bound under PQ node capacitor in " power system voltage and var fire protection technology " and reactor reactive power compensation minimum capacity and maximum load, minimum load and normal duty three kinds of operational modes, circuit effective power flow are model;

(3) based on layering thought, above-mentioned model is solved; Be upper and lower two-layer model according to the type of model constrained condition by whole model decomposition, by the successive ignition Load flow calculation between two-layer model until meet model Prescribed Properties;

(4) the final optimization pass result of the required newly-increased reactive compensation capacity sum of all transformer stations is calculated; When solving result meets the service requirement of the substation bus bar voltage that " power system voltage and var fire protection technology " and dispatching of power netwoks mechanism determine simultaneously, then this result is the minimum newly-increased reactive compensation capacity of whole electrical network.

2. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 1, it is characterized in that, in described step (1), basic data comprises:

The existing condenser capacity Q of transformer station all PQ node _ciwith reactor capacity Q _li, each node transformer station reactive loss Δ Q _ti, each node transformer station all outlet charge powers sum Q _lCi, transformer station's high-voltage terminal side configuration in node high resistance total capacity Q _ki;

Wherein: i ∈ N _pQ, N _pQthe set of all PQ nodes in expression system, i represents i-th PQ node.

Meanwhile, the predicted load under the highest, normal and minimum load three kinds of operational modes is needed: node voltage value V _x,i; Admittance matrix coefficient correlation G between node i and node j _ij, B _ij; The phase angle difference θ of node i and node j _{x, ij}; The voltage magnitude bound of node i _iwith ; The effective power flow P of circuit between node i and node j _{x, ij}; The bound of circuit effective power flow between node i and node j with

Wherein: i ∈ N, X ∈ High, Normal, Low, X represent operation of power networks state, the set of all nodes in N expression system, and High, Normal, Low represent the highest respectively, normal and minimum load three kinds of operational modes.

3. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 1, it is characterized in that, in described step (2), the target function that Optimized model comprises is as follows:

$\min Q_S=\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}\mathrm{\Δ}{Q}_{\mathrm{Ci}}+\underset{i\∈N_{\mathrm{PQ}}}{\mathrm{\Σ}}\mathrm{\Δ}{Q}_{\mathrm{Li}}---\left(1\right)$

Wherein: Q _sfor the newly-increased reactive power compensation total capacity of electrical network; N _pQfor PQ set of node; Δ Q _cifor node i increases condenser capacity newly; Δ Q _lirepresent that node i increases reactor capacity newly.

4. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 1, it is characterized in that, in described step (2) in " power system voltage and var fire protection technology " PQ node capacitor and reactor reactive power compensation minimum capacity be constrained to Linear Constraints; Substation bus bar voltage bound under maximum load, minimum load and normal duty three kinds of operational modes, circuit effective power flow be constrained to Nonlinear Constraints.

5. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 4, it is characterized in that, the Linear Constraints setting up Optimized model is specific as follows:

$\left\{\begin{array}{c}{Q}_{\mathrm{Ci}}+\mathrm{\Δ}{Q}_{\mathrm{Ci}}\≥\mathrm{\Δ}{Q}_{\mathrm{Ti}}-Q_{\mathrm{LCi}}+Q_{\mathrm{Ki}}\\ Q_{\mathrm{Li}}+\mathrm{\Δ}{Q}_{\mathrm{Li}}\≥Q_{\mathrm{LCi}}\×0.9\\ i\∈N_{\mathrm{PQ}}\end{array}\right.---\left(2\right)$

Wherein: Q _ciand Q _lirepresent the existing capacitor of node i, reactor capacity respectively; Δ Q _tirepresent the reactive loss of transformer station in node i, be made up of susceptance loss and reactance loss; Q _lCirepresent transformer station's all outlet charge powers sum in node i; Q _kirepresent the high resistance total capacity of transformer station's high-voltage terminal side configuration in node i.

6. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 1, it is characterized in that, the Nonlinear Constraints setting up Optimized model is specific as follows:

$\mathrm{\Δ}{P}_{X,i}=V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,j}(G_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}})$

$\mathrm{\Δ}{Q}_{X,i}=Q_{L,i}-Q_{C,i}+V_{X,i}\underset{j\∈i}{\mathrm{\Σ}}{V}_{X,i}(G_{\mathrm{ij}}\sin {\mathrm{\θ}}_{X,\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\θ}}_{X,\mathrm{ij}})$

${\underset{\‾}{V}}_i\≤V_{X,i}\≤{\stackrel{\‾}{V}}_i$

${\underset{\‾}{P}}_{\mathrm{ij}}\≤P_{X,\mathrm{ij}}\≤{\stackrel{\‾}{P}}_{\mathrm{ij}}$

i∈N,X∈High,Normal,Low

Wherein: Δ P _x,irepresent that node i is gained merit injecting power; Δ Q _x,irepresent the idle injecting power of node i except newly-increased reactive compensation capacity; V _x,irepresent the magnitude of voltage of node i; G _ij, B _ijfor the admittance matrix coefficient correlation between node i and node j; θ _{x, ij}represent the phase angle difference of node i and node j; with be respectively the voltage magnitude bound of node i; P _{x, ij}represent the effective power flow of circuit between node i and node j; with represent the bound of circuit effective power flow between node i and node j; The set of all nodes in N expression system; High, Normal, Low represent the highest respectively, normal and minimum load three kinds of operational modes, under above-mentioned three kinds of operational modes, and the Nonlinear Constraints shown in above the optimal solution of model all needs to meet.

7. the minimum newly-increased reactive compensation capacity computational methods of electrical network according to claim 1, it is characterized in that, whole model decomposition is upper according to the type of model constrained condition by described step (3), lower two-layer model, by the successive ignition Load flow calculation between two-layer model until meet model Prescribed Properties and be specially: according to the model set up, target function and Linear Constraints are classified as upper strata Optimized model, Nonlinear Constraints is classified as lower floor's Optimized model, first use linear optimization algorithm directly to upper strata model solution, the optimum results obtained is brought underlying model into and is verified, single website for discontented afc voltage bound constraints is analyzed, and increase or reduce its perception or capacity reactive compensation capacity, Filled function so repeatedly, until meet all constraints.