CN104239716A - Parameter deviation sensitivity-based power grid equipment parameter identification and estimation method - Google Patents

Abstract

The invention provides a parameter deviation sensitivity-based power grid equipment parameter identification and estimation method, which comprises the following steps of acquiring real-time power grid measurement data and a power grid equipment parameter; establishing a weighted least square model in consideration of a power grid equipment parameter deviation; identifying and estimating the power grid equipment parameter; calculating an average value of estimated power grid equipment parameter values according to the estimated power grid equipment parameter values of a plurality of time sections. According to the method, a state estimation and parameter estimation alternate correction method is utilized, the influence of a power grid equipment parameter error on a target function of the least square model is taken into account, and the average value of power grid equipment parameter estimation values is calculated based on the multiple calculation of the time sections, so that the accuracy of a parameter estimation value is improved.

Description

A kind of grid equipment parameter identification based on parameter error sensitivity and method of estimation

Technical field

The invention belongs to dispatching automation of electric power systems technical field, be specifically related to a kind of grid equipment parameter identification based on parameter error sensitivity and method of estimation.

Background technology

Current power system load flow calculates and state estimation is all be based upon on power network switch remote signalling and the correct basis of grid equipment parameter.Grid equipment parameter is mainly derived from equipment theoretical parameter and actual measurement parameter, by manual entry system database.Along with the fast development of electrical network, electrical network scale is day by day huge, electric network composition is day by day complicated, the inaccurate problem of device parameter manifests, there is theoretical parameter and actual parameter is inconsistent, survey measuring error, actual operation parameters changes and the problem such as personnel's typing mistake, all can have a strong impact on the accuracy of each computation result of system.

Traditional power system network parameter error discrimination method mainly contains: one is the method based on sensitivity analysis, and its deficiency is that the threshold value of its identification result to error in measurement and some artificial settings is very responsive; Two is state variables that extended mode is estimated, adopts generalized state estimation to carry out the estimated value of computational grid parameter, there is numerical stability issues, but needs to specify some suspicious network parameters in advance.Therefore study more practical, precision better based on parameter error sensitivity many times discontinuity surface grid equipment parameter identification and method of estimation significant.

Summary of the invention

In order to overcome above-mentioned the deficiencies in the prior art, the invention provides a kind of grid equipment parameter identification based on parameter error sensitivity and method of estimation, the method that utilization state is estimated and parameter estimation is alternately revised, consider the impact of grid equipment parameter error on LEAST SQUARES MODELS FITTING objective function, based on the repeatedly calculating of discontinuity surface time multiple, ask for the mean value of grid equipment parameter estimation result, improve the accuracy of parameter estimation result.

In order to realize foregoing invention object, the present invention takes following technical scheme:

The invention provides a kind of grid equipment parameter identification based on parameter error sensitivity and method of estimation, said method comprising the steps of:

Step 1: obtain electrical network real-time measurement data and grid equipment parameter;

Step 2: set up the weighted least-squares method model considering grid equipment parameter error;

Step 3: identification and estimation are carried out to grid equipment parameter;

Step 4: according to the grid equipment estimates of parameters of discontinuity surface time multiple, ask for its mean value.

In described step 1, from power network schedule automation data acquisition analysis system and electric power system model database, obtain electrical network real-time measurement data and grid equipment parameter respectively; Described electrical network real-time measurement data comprise physics busbar voltage amplitude, circuit active power, circuit reactive power, transformer active power, transformer reactive power, unit injects active power, unit injects reactive power, load injects active power, load injection reactive power, transverter inject active power, transverter injects reactive power, capacity reactance device injects active power, electric current reactor injects reactive power, zero injection active power and zero injects reactive power;

Described grid equipment parameter comprises line impedance parameter, line admittance parameter, transformer impedance parameter and load tap changer position.

Described step 2 comprises the following steps:

Step 2-1: the state estimation setting up electric system measures equation;

Step 2-2: the weighted least-squares method model considering grid equipment parameter error.

In described step 2-1, according to the electrical network real-time measurement data obtained and grid equipment parameter, the state estimation setting up following electric system measures equation:

z＝h(x,ε)+v

Wherein, z is electrical network real-time measurement data; X is electric network state variable, comprises and calculates busbar voltage amplitude and phase angle; ε is grid equipment parameter error, and it is grid equipment parameter ideal value p _twith the deviation of the grid equipment parameter actual value p obtained, i.e. ε=p _t-p, ε are less, show that grid equipment parameter estimation is more accurate; V is electrical network real-time measurement data error; H (x, ε) is the power system network equation under polar coordinate system, specifically has:

$P_i=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\ΔG}}_{\mathrm{ij}})\cos {\mathrm{\θ}}_{\mathrm{ij}}+(B_{\mathrm{ij}}+{\mathrm{\ΔB}}_{\mathrm{ij}})\sin {\mathrm{\θ}}_{\mathrm{ij}})$

$Q_i=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\ΔG}}_{\mathrm{ij}})\sin {\mathrm{\θ}}_{\mathrm{ij}}-(B_{\mathrm{ij}}+{\mathrm{\ΔB}}_{\mathrm{ij}})\cos {\mathrm{\θ}}_{\mathrm{ij}})$

$P_{\mathrm{ij}}=V_i^2(g+\mathrm{\Δg})-V_i{V}_j(g+\mathrm{\Δg})\cos {\mathrm{\θ}}_{\mathrm{ij}}-V_i{V}_j(b+\mathrm{\Δb})\sin {\mathrm{\θ}}_{\mathrm{ij}}$

$Q_{\mathrm{ij}}=-V_i^2(b+\mathrm{\Δb}+y_c+\mathrm{\Δ}{y}_c)-V_i{V}_j(g+\mathrm{\Δg})\sin {\mathrm{\θ}}_{\mathrm{ij}}+V_i{V}_j(b+\mathrm{\Δb})\cos {\mathrm{\θ}}_{\mathrm{ij}}$

Wherein, P _iand Q _ibe respectively the injection active power and injection reactive power that calculate bus i; P _ijand Q _ijbe respectively real-time measurement active power and the real-time measurement reactive power of grid branch; N is the sum calculating bus; V _iand V _jbe respectively the voltage magnitude that grid branch two ends calculate bus i and calculate bus j, θ _ijfor grid branch first and last section phase angle difference, there is θ _ij=θ _i-θ _j; During i=j, G _ijand B _ijbe respectively the self-conductance that calculates bus i and from susceptance; During i ≠ j, G _ijand B _ijbe respectively the transconductance between calculating bus i and calculating bus j and mutual susceptance, g, b and y _cbe respectively the conductance of grid branch, susceptance and ground capacitance; Δ G _ij, Δ B _ij, Δ g, Δ b and Δ y _cthe parameter error relevant to ε.

In the weighted least-squares model of the grid equipment parameter error of described step 2-2, objective function is expressed as:

minJ＝[z-h(x,ε)] ^TR ^-1[z-h(x,ε)]+ε ^Tε

Wherein, J is target function value, R ^-1be the weight matrix relevant to the error in measurement of electrical network real-time measurement data, it is diagonal matrix, and l diagonal element is σ _lfor error in measurement standard deviation;

X and ε when making target function value J reach minimum is the electric network state variable and grid equipment parameter error that will try to achieve, so, to x and ε differentiate respectively, have:

$\left\{\begin{array}{c}\frac{\∂J}{\∂x}=-2H_x^T{R}^{-1}v=0\\ \frac{\∂J}{\∂\mathrm{\ϵ}}=-2H_{\mathrm{\ϵ}}^T{R}^{-1}v+2\mathrm{\ϵ}=0\end{array}\right.$

Wherein, v=z-h (x, ε)=z-h (x ₀, ε ₀)-H _xΔ x; Δ x is electric network state variable error, and Δ x=x-x ₀, x ₀for initial electric network state variable; ε ₀for initial grid equipment parameter error; H _xfor electric system measures the Jacobian matrix to electric network state variable, h _εfor electric system measures the Jacobian matrix to grid equipment parameter error,

So x and ε is expressed as:

$\left\{\begin{array}{c}\mathrm{\Δx}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x,\mathrm{\ϵ}))\\ \mathrm{\ϵ}=H_{\mathrm{\ϵ}}^T{R}^{-1}v=H_{\mathrm{\ϵ}}^T{R}^{-1}(z-h(x,\mathrm{\ϵ}))\end{array}\right.$

Wherein, v is electrical network real-time measurement data error.

In described step 3, adopt inside and outside double-deck process of iteration to carry out identification and estimation to grid equipment parameter, wherein internal layer iteration is that electric network state variable estimates iteration, and external iteration is that grid equipment parameter error estimates iteration.

Described step 3 specifically comprises the following steps:

Step 3-1: set the electric network state variable of the m-1 time internal layer iteration as x ^(m-1), the grid equipment parameter error of (n-1)th external iteration is ε ^(n-1);

Step 3-2: the electric network state variable modified value Δ x obtaining the m time iteration according to following formula ^(m), have:

${\mathrm{\Δx}}^{\left(m\right)}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x^{(m-1)},{\mathrm{\ϵ}}^{(n-1)}))$

Electric network state variable then after the m time iteration is x ^(m)=x ^(m-1)+ Δ x ^(m);

Step 3-3: if then make iterations m=m+1, repeated execution of steps 3-2; If then perform step 3-4; Wherein w represents the element numbers in electric network state variable, β ₁it is the internal layer iteration convergence threshold value selected by accuracy requirement;

Step 3-4: calculate grid equipment parameter error ε according to the following formula ⁽ⁿ⁾, have:

${\mathrm{\ϵ}}^{\left(n\right)}=H_{\mathrm{\ϵ}}^T{R}^{-1}v=H_{\mathrm{\ϵ}}^T{R}^{-1}(z-h(x^{\left(m\right)},{\mathrm{\ϵ}}^{(n-1)}))$

In order to prevent the excessive calculating caused of grid equipment parameter error inaccurate, to grid equipment parameter error ε ⁽ⁿ⁾repair the modifying factor s being multiplied by and being less than 1, grid equipment parameter actual value p is modified to p ⁽ⁿ⁾=p ^(n-1)+ s ε ⁽ⁿ⁾;

Step 3-5: if | ε ⁽ⁿ⁾| > β ₂, then make iterations n=n+1, return step 3-2; If | ε ⁽ⁿ⁾| < β ₂, then now ε ⁽ⁿ⁾be suspicious grid equipment parameter error value, grid equipment estimates of parameters is p ⁽ⁿ⁾=p ^(n-1)+ ε ⁽ⁿ⁾, wherein β ₂it is the external iteration convergence thresholding selected by accuracy requirement.

In described step 3-4, in order to characterize the dispersion degree of grid equipment parameter error ε, the variance D (ε) of grid equipment parameter error is expressed as:

$\begin{array}{c}D\left(\mathrm{\&epsiv;}\right)=D\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\right)\\ =E\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;})){(z-h(x,\mathrm{\&epsiv;}))}^T{R}^{-1}{H}_{\mathrm{\&epsiv;}}\right)\\ =H_{\mathrm{\&epsiv;}}^T{R}^{-1}\left(H_x{\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T\right)R^{-1}{H}_{\mathrm{\&epsiv;}}\end{array}$

Then in grid equipment parameter error ε, ε _wthe grid equipment parameter that the maximum element of/D (ε) is corresponding is suspicious wrong parameter, and w represents the element numbers in grid equipment parameter error.

In described step 4, in order to reduce grid equipment parameter error, the grid equipment estimates of parameters of discontinuity surface when obtaining multiple, repeated execution of steps, grid equipment estimates of parameters, then averages have:

$\stackrel{\&OverBar;}{p}=\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{p}_k$

Wherein, K is the time discontinuity surface number obtained, p _kfor the grid equipment estimates of parameters that discontinuity surface gained during kth arrives.

Compared with prior art, beneficial effect of the present invention is:

1. the method that the present invention adopts state estimation internal layer iteration and grid equipment parameter estimation external iteration to replace solves, and reduces the error of measurement to the impact of parameter result of calculation;

2. the method that utilization state is estimated and parameter estimation is alternately revised, considers the impact of grid equipment parameter error on LEAST SQUARES MODELS FITTING objective function, by the optimization process of discontinuity surface of many times, improves the accuracy of parameter estimation result.

Accompanying drawing explanation

Fig. 1 is grid equipment parameter identification based on parameter error sensitivity and method of estimation process flow diagram.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

As Fig. 1, the invention provides a kind of grid equipment parameter identification based on parameter error sensitivity and method of estimation, said method comprising the steps of:

Step 1: obtain electrical network real-time measurement data and grid equipment parameter;

Step 2: set up the weighted least-squares method model considering grid equipment parameter error;

Step 3: identification and estimation are carried out to grid equipment parameter;

Step 4: according to the grid equipment estimates of parameters of discontinuity surface time multiple, ask for its mean value.

In described step 1, from power network schedule automation data acquisition analysis system and electric power system model database, obtain electrical network real-time measurement data and grid equipment parameter respectively; Described electrical network real-time measurement data comprise physics busbar voltage amplitude, circuit active power, circuit reactive power, transformer active power, transformer reactive power, unit injects active power, unit injects reactive power, load injects active power, load injection reactive power, transverter inject active power, transverter injects reactive power, capacity reactance device injects active power, electric current reactor injects reactive power, zero injection active power and zero injects reactive power;

Described grid equipment parameter comprises line impedance parameter, line admittance parameter, transformer impedance parameter and load tap changer position.

Described step 2 comprises the following steps:

Step 2-1: the state estimation setting up electric system measures equation;

Step 2-2: the weighted least-squares method model considering grid equipment parameter error.

In described step 2-1, according to the electrical network real-time measurement data obtained and grid equipment parameter, the state estimation setting up following electric system measures equation:

z＝h(x,ε)+v

Wherein, z is electrical network real-time measurement data; X is electric network state variable, comprises and calculates busbar voltage amplitude and phase angle; ε is grid equipment parameter error, and it is grid equipment parameter ideal value p _twith the deviation of the grid equipment parameter actual value p obtained, i.e. ε=p _t-p, ε are less, show that grid equipment parameter estimation is more accurate; V is electrical network real-time measurement data error; H (x, ε) is the power system network equation under polar coordinate system, specifically has:

$P_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\&Delta;G}}_{\mathrm{ij}})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}+(B_{\mathrm{ij}}+{\mathrm{\&Delta;B}}_{\mathrm{ij}})\sin {\mathrm{\&theta;}}_{\mathrm{ij}})$

$Q_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\&Delta;G}}_{\mathrm{ij}})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}-(B_{\mathrm{ij}}+{\mathrm{\&Delta;B}}_{\mathrm{ij}})\cos {\mathrm{\&theta;}}_{\mathrm{ij}})$

$P_{\mathrm{ij}}=V_i^2(g+\mathrm{\&Delta;g})-V_i{V}_j(g+\mathrm{\&Delta;g})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}-V_i{V}_j(b+\mathrm{\&Delta;b})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}$

$Q_{\mathrm{ij}}=-V_i^2(b+\mathrm{\&Delta;b}+y_c+\mathrm{\&Delta;}{y}_c)-V_i{V}_j(g+\mathrm{\&Delta;g})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}+V_i{V}_j(b+\mathrm{\&Delta;b})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}$

Wherein, P _iand Q _ibe respectively the injection active power and injection reactive power that calculate bus i; P _ijand Q _ijbe respectively real-time measurement active power and the real-time measurement reactive power of grid branch; N is the sum calculating bus; V _iand V _jbe respectively the voltage magnitude that grid branch two ends calculate bus i and calculate bus j, θ _ijfor grid branch first and last section phase angle difference, there is θ _ij=θ _i-θ _j; During i=j, G _ijand B _ijbe respectively the self-conductance that calculates bus i and from susceptance; During i ≠ j, G _ijand B _ijbe respectively the transconductance between calculating bus i and calculating bus j and mutual susceptance, g, b and y _cbe respectively the conductance of grid branch, susceptance and ground capacitance; Δ G _ij, Δ B _ij, Δ g, Δ b and Δ y _cthe parameter error relevant to ε.

In the weighted least-squares model of the grid equipment parameter error of described step 2-2, objective function is expressed as:

minJ＝[z-h(x,ε)] ^TR ^-1[z-h(x,ε)]+ε ^Tε

Wherein, J is target function value, R ^-1be the weight matrix relevant to the error in measurement of electrical network real-time measurement data, it is diagonal matrix, and l diagonal element is σ _lfor error in measurement standard deviation;

X and ε when making target function value J reach minimum is the electric network state variable and grid equipment parameter error that will try to achieve, so, to x and ε differentiate respectively, have:

$\left\{\begin{array}{c}\frac{\&PartialD;J}{\&PartialD;x}=-2H_x^T{R}^{-1}v=0\\ \frac{\&PartialD;J}{\&PartialD;\mathrm{\&epsiv;}}=-2H_{\mathrm{\&epsiv;}}^T{R}^{-1}v+2\mathrm{\&epsiv;}=0\end{array}\right.$

Wherein, v=z-h (x, ε)=z-h (x ₀, ε ₀)-H _xΔ x; Δ x is electric network state variable error, and Δ x=x-x ₀, x ₀for initial electric network state variable; ε ₀for initial grid equipment parameter error; H _xfor electric system measures the Jacobian matrix to electric network state variable, h _εfor electric system measures the Jacobian matrix to grid equipment parameter error,

So x and ε is expressed as:

$\left\{\begin{array}{c}\mathrm{\&Delta;x}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\\ \mathrm{\&epsiv;}=H_{\mathrm{\&epsiv;}}^T{R}^{-1}v=H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\end{array}\right.$

Wherein, v is electrical network real-time measurement data error.

In described step 3, adopt inside and outside double-deck process of iteration to carry out identification and estimation to grid equipment parameter, wherein internal layer iteration is that electric network state variable estimates iteration, and external iteration is that grid equipment parameter error estimates iteration.

Described step 3 specifically comprises the following steps:

Step 3-1: set the electric network state variable of the m-1 time internal layer iteration as x ^(m-1), the grid equipment parameter error of (n-1)th external iteration is ε ^(n-1);

Step 3-2: the electric network state variable modified value Δ x obtaining the m time iteration according to following formula ^(m), have:

${\mathrm{\&Delta;x}}^{\left(m\right)}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x^{(m-1)},{\mathrm{\&epsiv;}}^{(n-1)}))$

Electric network state variable then after the m time iteration is x ^(m)=x ^(m-1)+ Δ x ^(m);

Step 3-3: if then make iterations m=m+1, repeated execution of steps 3-2; If then perform step 3-4; Wherein w represents the element numbers in electric network state variable, β ₁it is the internal layer iteration convergence threshold value selected by accuracy requirement;

Step 3-4: calculate grid equipment parameter error ε according to the following formula ⁽ⁿ⁾, have:

${\mathrm{\&epsiv;}}^{\left(n\right)}=H_{\mathrm{\&epsiv;}}^T{R}^{-1}v=H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x^{\left(m\right)},{\mathrm{\&epsiv;}}^{(n-1)}))$

In order to prevent the excessive calculating caused of grid equipment parameter error inaccurate, to grid equipment parameter error ε ⁽ⁿ⁾repair the modifying factor s being multiplied by and being less than 1, grid equipment parameter actual value p is modified to p ⁽ⁿ⁾=p ^(n-1)+ s ε ⁽ⁿ⁾;

Step 3-5: if | ε ⁽ⁿ⁾| > β ₂, then make iterations n=n+1, return step 3-2; If | ε ⁽ⁿ⁾| < β ₂, then now ε ⁽ⁿ⁾be suspicious grid equipment parameter error value, grid equipment estimates of parameters is p ⁽ⁿ⁾=p ^(n-1)+ ε ⁽ⁿ⁾, wherein β ₂it is the external iteration convergence thresholding selected by accuracy requirement.

In described step 3-4, in order to characterize the dispersion degree of grid equipment parameter error ε, the variance D (ε) of grid equipment parameter error is expressed as:

$\begin{array}{c}D\left(\mathrm{\&epsiv;}\right)=D\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\right)\\ =E\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;})){(z-h(x,\mathrm{\&epsiv;}))}^T{R}^{-1}{H}_{\mathrm{\&epsiv;}}\right)\\ =H_{\mathrm{\&epsiv;}}^T{R}^{-1}\left(H_x{\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T\right)R^{-1}{H}_{\mathrm{\&epsiv;}}\end{array}$

Then in grid equipment parameter error ε, ε _wthe grid equipment parameter that the maximum element of/D (ε) is corresponding is suspicious wrong parameter, and w represents the element numbers in grid equipment parameter error.

In described step 4, in order to reduce grid equipment parameter error, the grid equipment estimates of parameters of discontinuity surface when obtaining multiple, repeated execution of steps, grid equipment estimates of parameters, then averages have:

$\stackrel{\&OverBar;}{p}=\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{p}_k$

Wherein, K is the time discontinuity surface number obtained, p _kfor the grid equipment estimates of parameters that discontinuity surface gained during kth arrives.

Finally should be noted that: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit; those of ordinary skill in the field still can modify to the specific embodiment of the present invention with reference to above-described embodiment or equivalent replacement; these do not depart from any amendment of spirit and scope of the invention or equivalent replacement, are all applying within the claims of the present invention awaited the reply.

Claims (9)

1., based on the identification of grid equipment parameter and the method for estimation of parameter error sensitivity, it is characterized in that: said method comprising the steps of:

Step 1: obtain electrical network real-time measurement data and grid equipment parameter;

Step 2: set up the weighted least-squares method model considering grid equipment parameter error;

Step 3: identification and estimation are carried out to grid equipment parameter;

Step 4: according to the grid equipment estimates of parameters of discontinuity surface time multiple, ask for its mean value.

2. the grid equipment parameter identification based on parameter error sensitivity according to claim 1 and method of estimation, it is characterized in that: in described step 1, from power network schedule automation data acquisition analysis system and electric power system model database, obtain electrical network real-time measurement data and grid equipment parameter respectively; Described electrical network real-time measurement data comprise physics busbar voltage amplitude, circuit active power, circuit reactive power, transformer active power, transformer reactive power, unit injects active power, unit injects reactive power, load injects active power, load injection reactive power, transverter inject active power, transverter injects reactive power, capacity reactance device injects active power, electric current reactor injects reactive power, zero injection active power and zero injects reactive power;

Described grid equipment parameter comprises line impedance parameter, line admittance parameter, transformer impedance parameter and load tap changer position.

3. the grid equipment parameter identification based on parameter error sensitivity according to claim 1 and method of estimation, is characterized in that: described step 2 comprises the following steps:

Step 2-1: the state estimation setting up electric system measures equation;

Step 2-2: the weighted least-squares method model considering grid equipment parameter error.

4. the grid equipment parameter identification based on parameter error sensitivity according to claim 3 and method of estimation, it is characterized in that: in described step 2-1, according to the electrical network real-time measurement data obtained and grid equipment parameter, the state estimation setting up following electric system measures equation:

z＝h(x,ε)+v

Wherein, z is electrical network real-time measurement data; X is electric network state variable, comprises and calculates busbar voltage amplitude and phase angle; ε is grid equipment parameter error, and it is the deviation of the grid equipment parameter actual value p of grid equipment parameter ideal value pt and acquisition, i.e. ε=p _t-p, ε are less, show that grid equipment parameter estimation is more accurate; V is electrical network real-time measurement data error; H (x, ε) is the power system network equation under polar coordinate system, specifically has:

$P_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\&Delta;G}}_{\mathrm{ij}})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}+(B_{\mathrm{ij}}+{\mathrm{\&Delta;B}}_{\mathrm{ij}})\sin {\mathrm{\&theta;}}_{\mathrm{ij}})$

$Q_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{V}_i{V}_j((G_{\mathrm{ij}}+{\mathrm{\&Delta;G}}_{\mathrm{ij}})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}-(B_{\mathrm{ij}}+{\mathrm{\&Delta;B}}_{\mathrm{ij}})\cos {\mathrm{\&theta;}}_{\mathrm{ij}})$

$P_{\mathrm{ij}}=V_i^2(g+\mathrm{\&Delta;g})-V_i{V}_j(g+\mathrm{\&Delta;g})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}-V_i{V}_j(b+\mathrm{\&Delta;b})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}$

$Q_{\mathrm{ij}}=-V_i^2(b+\mathrm{\&Delta;b}+y_c+\mathrm{\&Delta;}{y}_c)-V_i{V}_j(g+\mathrm{\&Delta;g})\sin {\mathrm{\&theta;}}_{\mathrm{ij}}+V_i{V}_j(b+\mathrm{\&Delta;b})\cos {\mathrm{\&theta;}}_{\mathrm{ij}}$

Wherein, P _iand Q _ibe respectively the injection active power and injection reactive power that calculate bus i; P _ijand Q _ijbe respectively real-time measurement active power and the real-time measurement reactive power of grid branch; N is the sum calculating bus; V _iand V _jbe respectively the voltage magnitude that grid branch two ends calculate bus i and calculate bus j, θ _ijfor grid branch first and last section phase angle difference, there is θ _ij=θ _i-θ _j; During i=j, G _ijand B _ijbe respectively the self-conductance that calculates bus i and from susceptance; During i ≠ j, G _ijand B _ijbe respectively the transconductance between calculating bus i and calculating bus j and mutual susceptance, g, b and y _cbe respectively the conductance of grid branch, susceptance and ground capacitance; Δ G _ij, Δ B _ij, Δ g, Δ b and Δ y _cthe parameter error relevant to ε.

5. the grid equipment parameter identification based on parameter error sensitivity according to claim 3 and method of estimation, is characterized in that: in the weighted least-squares model of the grid equipment parameter error of described step 2-2, objective function is expressed as:

minJ＝[z-h(x,ε)] ^TR ^-1[z-h(x,ε)]+ε ^Tε

Wherein, J is target function value, R ^-1be the weight matrix relevant to the error in measurement of electrical network real-time measurement data, it is diagonal matrix, and l diagonal element is σ _lfor error in measurement standard deviation;

X and ε when making target function value J reach minimum is the electric network state variable and grid equipment parameter error that will try to achieve, so, to x and ε differentiate respectively, have:

$\left\{\begin{array}{c}\frac{\&PartialD;J}{\&PartialD;x}=-2H_x^T{R}^{-1}v=0\\ \frac{\&PartialD;J}{\&PartialD;\mathrm{\&epsiv;}}=-2H_{\mathrm{\&epsiv;}}^T{R}^{-1}v+2\mathrm{\&epsiv;}=0\end{array}\right.$

Wherein, v=z-h (x, ε)=z-h (x ₀, ε ₀)-H _xΔ x; Δ x is electric network state variable error, and Δ x=x-x ₀, x ₀for initial electric network state variable; ε ₀for initial grid equipment parameter error; H _xfor electric system measures the Jacobian matrix to electric network state variable, h _εfor electric system measures the Jacobian matrix to grid equipment parameter error,

So x and ε is expressed as:

$\left\{\begin{array}{c}\mathrm{\&Delta;x}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\\ \mathrm{\&epsiv;}=H_{\mathrm{\&epsiv;}}^T{R}^{-1}v=H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\end{array}\right.$

Wherein, v is electrical network real-time measurement data error.

6. the grid equipment parameter identification based on parameter error sensitivity according to claim 1 and method of estimation, it is characterized in that: in described step 3, inside and outside double-deck process of iteration is adopted to carry out identification and estimation to grid equipment parameter, wherein internal layer iteration is that electric network state variable estimates iteration, and external iteration is that grid equipment parameter error estimates iteration.

7. the grid equipment parameter identification based on parameter error sensitivity according to claim 1 or 6 and method of estimation, is characterized in that: described step 3 specifically comprises the following steps:

Step 3-1: set the electric network state variable of the m-1 time internal layer iteration as x ^(m-1), the grid equipment parameter error of (n-1)th external iteration is ε ^(n-1);

Step 3-2: the electric network state variable modified value Δ x obtaining the m time iteration according to following formula ^(m), have:

${\mathrm{\&Delta;x}}^{\left(m\right)}={\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T{R}^{-1}(z-h(x^{(m-1)},{\mathrm{\&epsiv;}}^{(n-1)}))$

Electric network state variable then after the m time iteration is x ^(m)=x ^(m-1)+ Δ x ^(m);

Step 3-3: if then make iterations m=m+1, repeated execution of steps 3-2; If then perform step 3-4; Wherein w represents the element numbers in electric network state variable, β ₁it is the internal layer iteration convergence threshold value selected by accuracy requirement;

Step 3-4: calculate grid equipment parameter error ε according to the following formula ⁽ⁿ⁾, have:

${\mathrm{\&epsiv;}}^{\left(n\right)}=H_{\mathrm{\&epsiv;}}^T{R}^{-1}v=H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x^{\left(m\right)},{\mathrm{\&epsiv;}}^{(n-1)}))$

In order to prevent the excessive calculating caused of grid equipment parameter error inaccurate, to grid equipment parameter error ε ⁽ⁿ⁾repair the modifying factor s being multiplied by and being less than 1, grid equipment parameter actual value p is modified to p ⁽ⁿ⁾=p ^(n-1)+ s ε ⁽ⁿ⁾;

Step 3-5: if | ε ⁽ⁿ⁾| > β ₂, then make iterations n=n+1, return step 3-2; If | ε ⁽ⁿ⁾| < β ₂, then now ε ⁽ⁿ⁾be suspicious grid equipment parameter error value, grid equipment estimates of parameters is p ⁽ⁿ⁾=p ^(n-1)+ ε ⁽ⁿ⁾, wherein β ₂it is the external iteration convergence thresholding selected by accuracy requirement.

8. the grid equipment parameter identification based on parameter error sensitivity according to claim 7 and method of estimation, it is characterized in that: in described step 3-4, in order to characterize the dispersion degree of grid equipment parameter error ε, the variance D (ε) of grid equipment parameter error is expressed as:

$\begin{array}{c}D\left(\mathrm{\&epsiv;}\right)=D\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;}))\right)\\ =E\left(H_{\mathrm{\&epsiv;}}^T{R}^{-1}(z-h(x,\mathrm{\&epsiv;})){(z-h(x,\mathrm{\&epsiv;}))}^T{R}^{-1}{H}_{\mathrm{\&epsiv;}}\right)\\ =H_{\mathrm{\&epsiv;}}^T{R}^{-1}\left(H_x{\left(H_x^T{R}^{-1}{H}_x\right)}^{-1}{H}_x^T\right)R^{-1}{H}_{\mathrm{\&epsiv;}}\end{array}$

Then in grid equipment parameter error ε, ε _wthe grid equipment parameter that the maximum element of/D (ε) is corresponding is suspicious wrong parameter, and w represents the element numbers in grid equipment parameter error.

9. the grid equipment parameter identification based on parameter error sensitivity according to claim 1 and method of estimation, it is characterized in that: in described step 4, in order to reduce grid equipment parameter error, the grid equipment estimates of parameters of discontinuity surface when obtaining multiple, repeated execution of steps, grid equipment estimates of parameters, then averages have:

$\stackrel{\&OverBar;}{p}=\frac{1}{K}\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{p}_k$

Wherein, K is the time discontinuity surface number obtained, p _kfor the grid equipment estimates of parameters that discontinuity surface gained during kth arrives.