CN101325336A - Method for matching multi-region electric network swim based on belt restriction state estimation - Google Patents

Abstract

The invention relates to a multiregion power network current matching method based on the estimation with the constrained state, and belongs to the automation field of electric power system management. The method includes the following steps: according to the power network real time data, superior and subordinate power network dispatching centers generate internal power network models and current distribution of the region, and build corresponding external power network equivalent models; the models and the current data of the subordinate power network are sent to the superior power network dispatching center which connects and combines the superior and the subordinate power network models into a total power network model through connecting lines or transformers; and the total power network current distribution can be obtained by applying the estimation method with the constrained state to the total power network model. The method can ensure that the current distribution of the superior power network part in the total power network model which is obtained after the combination of the multiregion power network models remains the same, and the boundary current mismatches of the superior and the subordinate power networks are distributed to the subordinate power network. The method ensures that the interconnected network on-line safety analysis and the results of simulation computation are accurate, thereby having important significance for ensuring the safe operation of the power networks.

Description

A kind of method for matching multi-region electric network swim based on the belt restraining state estimation

Technical field

The present invention relates to a kind of method for matching multi-region electric network swim, belong to dispatching automation of electric power systems and grid simulation technical field based on the belt restraining state estimation.

Background technology

Electrical network is formed by a plurality of regional power grids are interconnected, is an integral body, and the influence to electrical network after accident takes place is overall.In interconnected network, the electrical network of each control centre's administration generally is a part of big electrical network, the EMS system of control centre only partly carries out modeling to the administration electrical network, its external electrical network is replaced with the ready-made equivalent electrical network of off-line, precision is very poor, also can't adapt to the development and change of electrical network, be the difficult problem in the traditional E MS practicability.

In order to address this problem, by the model exchange of upper and lower level electrical network between vertically, by the merging of model, online foundation overall situation electric network model.Existing method as measurement, is set up the state estimation model of overall electrical network to the trend estimated value of each subregion electrical network, thereby the trend that obtains overall electrical network coupling distributes.But this method is distributed to the trend amount of mismatch between each subregion electrical network in each subregion, can not guarantee higher level's electrical network or need the trend of the electrical network of labor to be distributed in model to remain unchanged before and after merging, therefore can not practical requirement.

Summary of the invention

The objective of the invention is to propose a kind of method for matching multi-region electric network swim based on the belt restraining state estimation, each regional power grid is set up local separately electric network model in each control centre, utilize the high-speed data communication network transmission that has built up between the current grid dispatching center to the higher level control centre, the higher level control centre is by the merging of model, form overall electric network model, make the online merging of multi-region electric network tide model in the electric power system, and guarantee to merge higher level's electric network swim and higher level's real-time power network state consistency in the overall electric network model in back, higher level's electric network swim is remained unchanged before and after merging.

The method for matching multi-region electric network swim based on the belt restraining state estimation that the present invention proposes may further comprise the steps:

(1-1) higher level's grid dispatching center is according to the real time data of administration electrical network, automatically generating one's respective area internal electric network model and trend distributes, set up the external electrical network Equivalent Model that is connected with this internal electric network model simultaneously, internal electric network model, external electrical network Equivalent Model are interconnected to form higher level's electric network model by interconnection or transformer;

(1-2) each subregion grid dispatching center of subordinate is according to the real time data of administration electrical network, automatically generating one's respective area internal electric network model and trend distributes, set up the external electrical network Equivalent Model that is connected with this internal electric network model simultaneously, internal electric network model, external electrical network Equivalent Model are interconnected to form subordinate's electric network model by interconnection or transformer, and by wide area network the flow data of this subordinate's electric network model and subordinate's electrical network are sent to higher level's grid dispatching center;

(1-3) higher level's grid dispatching center interconnects higher level's electric network model and each the subordinate's electric network model that receives by interconnection or transformer, external electrical network Equivalent Model during connection in deletion higher level's electric network model and each subordinate's electric network model is merged into overall electric network model;

(1-4) utilize the belt restraining method for estimating state, estimate the trend of above-mentioned overall electric network model.

In the said method, utilize the belt restraining method for estimating state, estimate the process of the trend of above-mentioned overall electric network model, specifically may further comprise the steps:

(2-1) construct a weighting square law state estimation model that comprises trend equality constraint and node injection inequality constraints for overall electrical network:

$\min J(V,\mathrm{\θ})=\frac{1}{2}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{w}_i{(Z_i-h_i(V,\mathrm{\θ}))}^2$

Satisfy following constraints:

$P_{\mathrm{ij}}=P_{\mathrm{ij}}^0,$ i∈A，j∈A

$Q_{\mathrm{ij}}=Q_{\mathrm{ij}}^0,$ i∈A，j∈A

P _j ＜P _j＜P _j，j∈A

Q _j ＜Q _j＜Q _j，j∈A

Wherein, Z _iBe the state estimation value before upper and lower level electric network model merges, comprise the active power P of power network line or transformer _IjAnd reactive power Q _Ij, the voltage magnitude V of bus _iAnd the meritorious injecting power P of bus _iWith idle injecting power Q _i, h _i(V θ) is measurement equation, w _iBe the weight coefficient in the weighting square law state estimation, A is the set of node of higher level's electrical network, and A is the set of node of all subordinate's electrical networks, P _Ij, P _Ij ⁰Be respectively interconnection between the upper and lower level electrical network after the merging of higher level's grid side active power and merge before active power, Q _Ij, Q _Ij ⁰Be interconnection between the upper and lower level electrical network after the merging of higher level's grid side reactive power and merge before reactive power, P _j, P _j , P _jBe respectively injection active power, active power lower limit and the active power upper limit of the grid nodes j of subordinate, Q _j, Q _j , Q _jBe respectively injection reactive power, reactive power lower limit and the reactive power upper limit of the grid nodes j of subordinate, V and θ are respectively the voltage magnitude and the phase angles of all nodes of electrical network;

Above-mentioned measurement equation h _i(V θ) is defined as:

The measurement equation of circuit or transformer is:

$\left\{\begin{array}{c}{P}_{\mathrm{ij}}=V_i^2{g}_{\mathrm{ij}}-V_i{V}_j(g_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}}-b_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}})\\ Q_{\mathrm{ij}}=-V_i^2(b_{\mathrm{ij}}+y_c)-V_i{V}_j(g_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}}-b_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}})\end{array}\right.$

In the following formula, P _IjBe the active power of circuit or transformer, Q _IjIt is the reactive power of circuit or transformer;

The voltage measurement equation of any bus i of overall situation electrical network:

V _i＝V _i

The injection measurement equation of any bus i of overall situation electrical network:

$\left\{\begin{array}{c}{P}_i=V_i^2{G}_{\mathrm{ii}}+\underset{j\∈i}{\overset{j\≠i}{\mathrm{\Σ}}}{V}_i{V}_j(G_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}})\\ Q_i=-V_i^2{B}_{\mathrm{ii}}+\underset{j\∈i}{\overset{j\≠i}{\mathrm{\Σ}}}{V}_i{V}_j(G_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}})\end{array}\right.$

In the following formula, P _i, Q _iBe arbitrarily meritorious injecting power and the idle injecting power of generator or load i in the overall electrical network;

(2-2) adopt interior point method to find the solution above-mentioned weighting square law state estimation model;

(2-3) according to the voltage magnitude V and the phase angle theta of above-mentioned overall all nodes of electrical network of finding the solution, the trend that obtains overall electrical network distributes.

The method for matching multi-region electric network swim that the present invention proposes based on the belt restraining state estimation, be characterized in: the tide model of this rank, one's respective area electrical network is set up and safeguarded to each subregion grid dispatching center of higher level's grid dispatching center and subordinate respectively, to meet the existing management mode of classification, subregion scheduling.Subordinate's grid dispatching center need send to the higher level control centre with this area under one's jurisdiction real-time power network tide model, after the higher level collects the control centre tide model of each subregion electrical network of subordinate, each subregion electric network model of each subordinate and higher level's electric network model are merged, form overall electric network model.Decontrol the constraint that part measures, the overall electric network model that is combined into calculating usefulness carries out state estimation, and the trend that obtains upper and lower level electrical network coupling distributes.Whole process is finished automatically, is fit to safety on line analysis, scheduling decision and the analog simulation of extensive interconnected network.Therefore the inventive method has the following advantages:

1, grid dispatching center at different levels only need be safeguarded electric network tide model separately, do not increase maintenance workload;

2, by the gathering and state estimation of each subregion electric network model, realize the online merging of multi-region electric network tide model automatically, the trend that forms global registration distributes;

3, higher level's electrical network part is consistent with actual conditions in the overall electric network tide model in merging back, has guaranteed interconnected network safety analysis, control decision and analog simulation precision, thereby has guaranteed the safe operation of electrical network.

4, the important generator and the load of subordinate's electrical network big variation can not occur in the overall electric network tide model in merging back after merging, makes exerting oneself of all generators remain on quota limit simultaneously.

Description of drawings

Fig. 1 is the online merging method schematic diagram of electric network tide model of the present invention;

Wherein, Fig. 1 (a) is subordinate's electric network tide model, and Fig. 1 (b) is higher level's electric network tide model, and Fig. 1 (c) is the overall electric network tide model after merging.

Fig. 2 is the overall electric network tide model schematic diagram that utilizes after the inventive method merges.

Among Fig. 1 and Fig. 2, the 1st, the one's respective area internal electric network model that subordinate's grid dispatching center generates automatically, the 2nd, connect the internal electric network model, the interconnection of external electrical network Equivalent Model or transformer, the 3rd, the external electrical network Equivalent Model that is connected with this internal electric network model, the 4th, the external electrical network Equivalent Model that is connected with this internal electric network model, the 5th, connect the internal electric network model, the interconnection of external electrical network Equivalent Model or transformer, the 6th, the one's respective area internal electric network model that higher level's grid dispatching center generates automatically, the 7th, higher level's electric network model, the 8th, the trend of interconnection or system interconnection transformer higher level grid side measures, the 9th, system interconnection transformer, the 10th, interconnector, the 11st, subordinate's electrical network, the 12nd, the trend of interconnection or system interconnection transformer subordinate grid side measures, and the 13rd, the injection trend of subordinate's electrical network median generatrix measures.

Embodiment

Below in conjunction with accompanying drawing and one embodiment of the present of invention, introduce method for matching multi-region electric network swim in detail based on the belt restraining state estimation.

Specific implementation method of the present invention may further comprise the steps as shown in Figure 1:

Step 1, shown in Fig. 1 (b), higher level's grid dispatching center is according to administration real-time power network data, utilize known Power Network Status Estimation method, automatically generating one's respective area internal electric network model 6 and trend distributes, electric network model 6 and trend distribute and comprise topological structure of electric, the active power of grid generator, reactive power or machine end busbar voltage, the active power of network load and reactive power, the active power of electrical network busbar voltage and circuit and reactive power, and the active power of transformer, reactive power, load tap changer gear etc.; Set up the external electrical network Equivalent Model 4 that is connected with this internal electric network simultaneously, described inside and outside model passes through the in esse interconnection 5 continuous higher level of formation electric network models between the two;

Step 2, shown in Fig. 1 (a), each subregion grid dispatching center of subordinate is according to administration real-time power network data, utilize known method for estimating state to generate one's respective area internal electric network model 1 and trend distribution automatically, electric network model 1 and trend distribute and comprise topological structure of electric, the active power of grid generator, reactive power or machine end busbar voltage, the active power of network load and reactive power, the active power of electrical network busbar voltage and circuit and reactive power, and the active power of transformer, reactive power, load tap changer gear etc., set up the external electrical network Equivalent Model 3 that is connected with this internal electric network simultaneously, in described, external model links to each other by in esse interconnection 2 between the two and forms subordinate's electric network model, and by wide area network described subordinate electric network model and flow data is sent to higher level's grid dispatching center;

Step 3, higher level's grid dispatching center receives subordinate's electric network model and the flow data that above-mentioned each subregion grid dispatching center is sent, at first reject external network equivalent model (dotted line) part in the electric network model separately, then, described higher level's electric network model and each subordinate's electric network model are docked (interconnections 2 by the interconnection between them 5, same group of interconnection of 5 expressions, here adopt higher level's grid side 5), just can with on, subordinate's electric network model is merged into the overall electric network model that calculates usefulness, shown in Fig. 1 (c), this overall electric network model removes model among Fig. 1 (a) and Fig. 1 (b) respectively behind the dotted portion zone to be docking together by interconnection and obtains as can be seen.

Step 4, the estimated result separately of upper and lower level electrical network is measured as equipment, and carry out the whole network by the belt restraining method for estimating state and estimate.The method for estimating state of this belt restraining can estimate that the whole network trend that obtains new upper and lower level electrical network coupling distributes, and can guarantee that higher level's electric network state remains unchanged before and after estimating.Concrete steps are as follows:

(1) at first construct one and contain interconnection between upper and lower level electrical network and system interconnection transformer device trend equality constraint and node and inject the weighting square law state estimation model of inequality constraints, just obtain one group of electric network state variable (the voltage magnitude V and the phase angle theta that comprise all nodes of electrical network) in the Optimization Model that satisfies under the constraints:

$\min J(V,\mathrm{\θ})=\frac{1}{2}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{w}_i{(Z_i-h_i(V,\mathrm{\θ}))}^2$

Satisfy following constraint:

$P_{\mathrm{ij}}=P_{\mathrm{ij}}^0,$ i∈A，j∈A

$Q_{\mathrm{ij}}=Q_{\mathrm{ij}}^0,$ i∈A，j∈A(1)

P _j ＜P _j＜P _j，j∈A

Q _j ＜Q _j＜Q _j，j∈A

Wherein, Z _iBe the state estimation value before upper and lower level electric network model merges, comprise the active power P of power network line or transformer _IjAnd reactive power Q _Ij, the voltage magnitude V of bus _iAnd the meritorious injecting power P of bus _iWith idle injecting power Q _i, h _i(V θ) is measurement equation, w _iBe the weight coefficient in the weighting square law state estimation, A is the set of node of higher level's electrical network, and A is the set of node of all subordinate's electrical networks, P _Ij, P _Ij ⁰Be respectively interconnection between the upper and lower level electrical network after the merging of higher level's grid side active power and merge before active power, Q _Ij, Q _Ij ⁰Be interconnection between the upper and lower level electrical network after the merging of higher level's grid side reactive power and merge before reactive power, P _j, P _j , P _jBe respectively injection active power, active power lower limit and the active power upper limit of the grid nodes j of subordinate, Q _j, Q _j , Q _jBe respectively injection reactive power, reactive power lower limit and the reactive power upper limit of the grid nodes j of subordinate, V and θ are respectively the voltage magnitude and the phase angles of all nodes of electrical network;

Measurement equation h _i(V θ) is defined as follows:

The measurement equation of circuit or transformer:

$\left\{\begin{array}{c}{P}_{\mathrm{ij}}=V_i^2{g}_{\mathrm{ij}}-V_i{V}_j(g_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}}-b_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}})\\ Q_{\mathrm{ij}}=-V_i^2(b_{\mathrm{ij}}+y_c)-V_i{V}_j(g_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}}-b_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}})\end{array}\right.---\left(2\right)$

In the following formula, P _IjBe the active power of circuit or transformer, Q _IjIt is the reactive power of circuit or transformer;

The voltage measurement equation of any bus i of overall situation electrical network:

V _i＝V _i (3)

The injection measurement equation of any bus i of overall situation electrical network:

$\left\{\begin{array}{c}{P}_i=V_i^2{G}_{\mathrm{ii}}+\underset{j\∈i}{\overset{j\≠i}{\mathrm{\Σ}}}{V}_i{V}_j(G_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}})\\ Q_i=-V_i^2{B}_{\mathrm{ii}}+\underset{j\∈i}{\overset{j\≠i}{\mathrm{\Σ}}}{V}_i{V}_j(G_{\mathrm{ij}}\sin {\mathrm{\θ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\θ}}_{\mathrm{ij}})\end{array}\right.---\left(4\right)$

In the following formula, P _i, Q _iBe arbitrarily meritorious injecting power and the idle injecting power of generator or load i in the overall electrical network;

(2) adopt interior point method to find the solution the state estimation problem of the belt restraining of formula (1) representative

Can be write as following Non-Linear Programming standard type for formula (1):

minimize f(x)

subject?to h(x)＝0 (5)

g≤g(x)≤g

Wherein: x ∈ R ⁽ⁿ⁾, h (x) ≡ [h ₁(x) ..., h _m(x)] ^T, g (x) ≡ [g ₁(x) ..., g _r(x)] ^T

Introduce slack variable (l, u) ∈ R ^(r), the Non-Linear Programming form that (5) formula is transformed into band equality constraint and simple variable constraint is as follows:

minimize f(x)

subjectto?h(x)＝0

g(x)-l- g＝0 (6)

g(x)+u-g＝0

(l，u)≥0

The Lagrangian of definition (6) formula is:

$L(x,l,u;y,z,w;\tilde{z},\tilde{w})\≡f\left(x\right){-y}^{T}h\left(x\right)-z^T(g\left(x\right)-l-\underset{\‾}{g})$

$-w^T(g\left(x\right)+u-\stackrel{\‾}{g})-{\tilde{z}}^{T}l-{\tilde{w}}^{T}u---\left(7\right)$

Here, y ∈ R ^(m)With $(z,w,\tilde{z},\tilde{w})\∈R^{\left(r\right)}$ It is Lagrange multiplier.

According to the Karush-Kuhn-Tucker theorem, optimal solution must satisfy following KKT system:

${\▿}_{x}L\≡{\▿}_{x}f\left(x\right)-{\▿}_{x}h\left(x\right)y-{\▿}_{x}g\left(x\right)(z+w)=0---\left(8\right)$

${\▿}_{y}L\≡h\left(x\right)=0---\left(9\right)$

${\▿}_{z}L\≡g\left(x\right)-l-\underset{\‾}{g}=0---\left(10\right)$

${\▿}_{w}L\≡g\left(x\right)+u-\stackrel{\‾}{g}=0---\left(11\right)$

${\▿}_{l}L\≡z-\tilde{z}=0---\left(12\right)$

${\▿}_{u}L\≡w+\tilde{w}=0---\left(13\right)$

$L\tilde{Z}e=0---\left(14\right)$

$U\tilde{W}e=0---\left(15\right)$

$(l,u)\≥0,(\tilde{z},\tilde{w})\≥0,y\≠0---\left(16\right)$

Wherein, (L, U, Z, W) ∈ R ^{(r * r)}According to the convention in the Mathematical Planning, the expression element is l, u, and z, the diagonal matrix of w, e ≡ [1 ..., 1] ^T∈ R ^(r)

(12) (13) formula and complementarity condition (14) (15) formula are merged, again according to disturbance KKT theory, introduce disturbance parameter μ＞0, the complementarity condition after lax the merging replaces with (12)-(15):

${\▿}_l^{\mathrm{\μ}}L\≡\mathrm{LZe}-\mathrm{\μe}=0---\left(17\right)$

${\▿}_u^{\mathrm{\μ}}L\≡\mathrm{UWe}+\mathrm{\μe}=0---\left(18\right)$

Directly use Newton method and find the solution above-mentioned disturbance KKT system (8)-(11), (17)-(18) obtain following update the system:

$({\▿}_x^{2}f\left(x\right)-{\▿}_x^{2}h\left(x\right)y-{\▿}_x^{2}g\left(x\right)(z+w))\mathrm{\Δx}---\left(19\right)$

$-{\▿}_{x}h\left(x\right)\mathrm{\Δy}-{\▿}_{x}g\left(x\right)(\mathrm{\Δz}+\mathrm{\Δw})=-{\▿}_x{L}_0$

${\▿}_{x}h{\left(x\right)}^T\mathrm{\Δx}=-{\▿}_y{L}_0---\left(20\right)$

${\▿}_{x}g{\left(x\right)}^T\mathrm{\Δx}-\mathrm{\Δl}=-{\▿}_z{L}_0---\left(21\right)$

${\▿}_{x}g{\left(x\right)}^T\mathrm{\Δx}+\mathrm{\Δu}=-{\▿}_w{L}_0---\left(22\right)$

$\mathrm{Z\Δl}+\mathrm{L\Δz}=-{\▿}_l^{\mathrm{\μ}}{L}_0---\left(23\right)$

$\mathrm{W\Δu}+\mathrm{U\Δw}=-{\▿}_u^{\mathrm{\μ}}{L}_0---\left(24\right)$

Wherein, $({\▿}_x{L}_0,{\▿}_y{L}_0,{\▿}_z{L}_0,{\▿}_w{L}_0,{\▿}_l^{\mathrm{\μ}}{L}_0,{\▿}_u^{\mathrm{\μ}}{L}_0)$ It is the residual error of disturbance KKT system.

With

Be f (x), the extra large gloomy matrix of h (x) and g (x).

Abbreviation (19-24), cancellation Δ l, Δ u, Δ z, Δ w, it is as follows to obtain final brief KKT system:

Wherein:

$(-{\▿}_x^{2}f\left(x\right)+{\▿}_x^{2}h\left(x\right)y+{\▿}_x^{2}g\left(x\right)(z+w))=H_g+H_h---\left(26\right)$

$J\left(x\right)\≡{\▿}_{x}h{\left(x\right)}^T---\left(27\right)$

$-{\▿}_{x}g\left(x\right)\left((U^{-1}-L^{-1}\right)\mathrm{\μe}---\left(28\right)$

$+L^{-1}Z(g\left(x\right)-l-\underset{\‾}{g})-U^{-1}W(g\left(x\right)+u-\stackrel{\‾}{g}))$

And after the cancellation, Δ l, Δ u, Δ z, the expression formula of Δ w:

$\mathrm{\Δl}={\▿}_x{g\left(x\right)}^T\mathrm{\Δx}+(g\left(x\right)-l-\underset{\‾}{g})---\left(29\right)$

$\mathrm{\Δu}=-{\▿}_{x}g{\left(x\right)}^T\mathrm{\Δx}-(g\left(x\right)+u-\stackrel{\‾}{g})---\left(30\right)$

$\mathrm{\Δz}=-L^{-1}Z{\▿}_{x}g{\left(x\right)}^T\mathrm{\Δx}-L^{-1}((\mathrm{LZe}-\mathrm{\μe})+Z(g\left(x\right)-l-\underset{\‾}{g}))---\left(31\right)$

$\mathrm{\Δw}=U^{-1}W{\▿}_x{g\left(x\right)}^T\mathrm{\Δx}-U^{-1}((\mathrm{UWe}+\mathrm{\μe})-W(g\left(x\right)+u-\stackrel{\‾}{g}))---\left(32\right)$

(26) in the formula, H () is made up of two parts: H _hBe f (x), the linear combination of h (x) and the extra large gloomy matrix of g (x); H _gBe called obstacle matrix (Barrier Matrix).H _gItem prevents that inequality constraints g (x) from violating their the bilateral limit.

Point iteration form is as follows in the consequent path trace method:

(0) initialization.Put k=0, K _Max, Center Parameter (centering parameter) σ ∈ (0,1], convergence precision ε=10 ^-6Select [l, u] ^T＞0, [z＞0, w＜0, y=0] ^T, here, k, K _MaxBe respectively iteration count and the maximum iterations that allows;

(1) if k＜K _Max, continue; Otherwise, print and do not restrain;

(2) calculate complementary gap (complementary gap) C _Gap:

$C_{\mathrm{Gap}}\≡\underset{i=1}{\overset{r}{\mathrm{\Σ}}}(l_i{z}_i-u_i{w}_i)---\left(33\right)$

If C _Gap＜ε stops, the output optimal solution;

(3) find the solution update the system (25)-(32), obtain correction [Δ x, Δ y], [Δ l, Δ u] and [Δ z, Δ w];

(4) ratio check, determine the maximum step-length in original and the dual spaces:

${\mathrm{step}}_P=0.9995\min \{\underset{i}{\min }(\frac{-l_i}{\mathrm{\&Delta;}{l}_i}:\mathrm{\&Delta;}{l}_i<0;\frac{-u_i}{\mathrm{\&Delta;}{u}_i}:\mathrm{\&Delta;}{u}_i<0),1\}---\left(34\right)$

${\mathrm{step}}_D=0.9995\min \{\underset{i}{\min }(\frac{-z_i}{\mathrm{\&Delta;}{z}_i}:\mathrm{\&Delta;}{z}_i<0;\frac{-w_i}{\mathrm{\&Delta;}{w}_i}:\mathrm{\&Delta;}{w}_i>0),1\}---\left(35\right)$

i＝1，2，…，r

(5) upgrade original and dual variable:

$\left[\begin{array}{c}x\\ l\\ u\end{array}\right]=\left[\begin{array}{c}x\\ l\\ u\end{array}\right]+{\mathrm{step}}_P\left[\begin{array}{c}\mathrm{\&Delta;x}\\ \mathrm{\&Delta;l}\\ \mathrm{\&Delta;u}\end{array}\right]---\left(36\right)$

$\left[\begin{array}{c}y\\ z\\ w\end{array}\right]=\left[\begin{array}{c}y\\ z\\ w\end{array}\right]+{\mathrm{step}}_D\left[\begin{array}{c}\mathrm{\&Delta;y}\\ \mathrm{\&Delta;z}\\ \mathrm{\&Delta;w}\end{array}\right]---\left(37\right)$

(6) k=k+1 returns (1);

(3) obtain the state variable (the voltage magnitude V and the phase angle theta that comprise all nodes of electrical network) of system by previous step, the trend that just can obtain the whole network distributes.Owing to comprised the meritorious and idle equality constraint that remains unchanged of interconnections all between higher level's electrical network and subordinate's electrical network and system interconnection transformer in the formula (1), convergence before the trend of higher level's electrical network is distributed in and merges simultaneously, the trend that therefore merges back higher level's electrical network distributes and can remain unchanged.In addition, subordinate's electrical network distributes in electrical network at the corresponding levels by the principle handle of the weighted least-squares departure with higher level's electrical network, utilize the inequality constraints that node injects in the formula (1) so suitable that important generator and the load of subordinate's electrical network big variation can not occur after merging simultaneously, make exerting oneself of all generators remain on quota limit simultaneously.

Claims (2)

1. Based on the method for matching multi-region electric network swim of belt restraining state estimation, it is characterized in that 1, this method may further comprise the steps:

   (1-1) higher level's grid dispatching center is according to the real time data of administration electrical network, automatically generating one's respective area internal electric network model and trend distributes, set up the external electrical network Equivalent Model that is connected with this internal electric network model simultaneously, internal electric network model, external electrical network Equivalent Model are interconnected to form higher level's electric network model by interconnection or transformer;

   (1-2) each subregion grid dispatching center of subordinate is according to the real time data of administration electrical network, automatically generating one's respective area internal electric network model and trend distributes, set up the external electrical network Equivalent Model that is connected with this internal electric network model simultaneously, internal electric network model, external electrical network Equivalent Model are interconnected to form subordinate's electric network model by interconnection or transformer, and by wide area network the flow data of this subordinate's electric network model and subordinate's electrical network are sent to higher level's grid dispatching center;

   (1-3) higher level's grid dispatching center interconnects higher level's electric network model and each the subordinate's electric network model that receives by interconnection or transformer, external electrical network Equivalent Model during connection in deletion higher level's electric network model and each subordinate's electric network model is merged into overall electric network model;

   (1-4) utilize the belt restraining method for estimating state, estimate the trend of above-mentioned overall electric network model.
2. 2, the method for claim 1 is characterized in that, utilizes the belt restraining method for estimating state in the described step (1-4), estimates the process of the trend of above-mentioned overall electric network model, specifically may further comprise the steps:

   (2-1) construct a weighting square law state estimation model that comprises trend equality constraint and node injection inequality constraints for overall electrical network:

   $\min J(V,\mathrm{\&theta;})=\frac{1}{2}\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{w}_i{(Z_i-h_i(V,\mathrm{\&theta;}))}^2$

   Satisfy following constraints:

   $P_{\mathrm{ij}}=P_{\mathrm{ij}}^0,$ i∈A，j∈A

   $Q_{\mathrm{ij}}=Q_{\mathrm{ij}}^0,$ i∈A，j∈A

   P _j ＜P _j＜P _j，j∈A

   Q _j ＜Q _j＜Q _j，j∈A

   Wherein, Z _iBe the state estimation value before upper and lower level electric network model merges, comprise the active power P of power network line or transformer _IjAnd reactive power Q _Ij, the voltage magnitude V of bus _iAnd the meritorious injecting power P of bus _iWith idle injecting power Q _i, h _i(V θ) is measurement equation, w _iBe the weight coefficient in the weighting square law state estimation, A is the set of node of higher level's electrical network, and A is the set of node of all subordinate's electrical networks, P _Ij, P _Ij ⁰Be respectively interconnection between the upper and lower level electrical network after the merging of higher level's grid side active power and merge before active power, Q _Ij, Q _Ij ⁰Be respectively interconnection between the upper and lower level electrical network after the merging of higher level's grid side reactive power and merge before reactive power, P _j, P _j , P _jBe respectively injection active power, active power lower limit and the active power upper limit of the grid nodes j of subordinate, Q _j, Q _j , Q _jBe respectively injection reactive power, reactive power lower limit and the reactive power upper limit of the grid nodes j of subordinate, V and θ are respectively the voltage magnitude and the phase angles of all nodes of electrical network;

   Above-mentioned measurement equation h _i(V θ) is defined as follows:

   The measurement equation of circuit or transformer:

   $\left\{\begin{array}{c}{P}_{\mathrm{ij}}{=V}_i^2{g}_{\mathrm{ij}}-V_i{V}_j(g_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}-b_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}})\\ Q_{\mathrm{ij}}=-V_i^2(b_{\mathrm{ij}}+y_c)-V_i{V}_j(g_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}-b_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}})\end{array}\right.$

   In the following formula, P _IjBe the active power of circuit or transformer, Q _IjIt is the reactive power of circuit or transformer; The voltage measurement equation of any bus i of overall situation electrical network:

   V _i＝V _i

   The injection measurement equation of any bus i of overall situation electrical network:

   $\left\{\begin{array}{c}{P}_i=V_i^2{G}_{\mathrm{ii}}+\underset{j\&Element;i}{\overset{j\&NotEqual;i}{\mathrm{\&Sigma;}}}{V}_i{V}_j(G_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}})\\ Q_i=-V_i^2{B}_{\mathrm{ii}}+\underset{j\&Element;i}{\overset{j\&NotEqual;i}{\mathrm{\&Sigma;}}}{V}_i{V}_j(G_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}-B_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}})\end{array}\right.$

   In the following formula, P _i, Q _iBe arbitrarily meritorious injecting power and the idle injecting power of generator or load i in the overall electrical network;

   (2-2) adopt interior point method to find the solution above-mentioned weighting square law state estimation model;

   (2-3) according to the voltage magnitude V and the phase angle theta of above-mentioned overall all nodes of electrical network of finding the solution, the trend that obtains overall electrical network distributes.

Images