CN103683329B - Based on the optimization method of the grid-connected DCgenerator motor field difference coefficient of the whole network loss minimization - Google Patents

Abstract

The invention discloses the optimization method of a kind of grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization in power system analysis and control technology field.Comprise: set up the power flow algorithm taking into account the impact of generator excitation difference coefficient; Determine trend constraint equation; Set up the grid-connected DCgenerator motor field difference coefficient optimization object function being target with the whole network loss minimization, and determine the constraints of described optimization object function; According to power flow algorithm and trend constraint equation, ask for the optimal solution of described optimization object function, thus the excitation difference coefficient be optimized.Instant invention overcomes the deficiency that conventional method is determined empirically difference coefficient, not only ensure the stable of voltage, and make the economic benefit of electrical network the highest.

Description

Based on the optimization method of the grid-connected DCgenerator motor field difference coefficient of the whole network loss minimization

Technical field

The invention belongs to power system analysis and control technology field, particularly relate to a kind of optimization method of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization.

Background technology

The limitation that tradition excitation difference coefficient is selected is: to the unit of paired running, for ensureing the stable operation of unit and idle reasonable distribution, require that each unit has at arranged side by side point in principle and just adjust difference and difference coefficient is completely equal, general desirable difference coefficient 3%-5%.But the selection of traditional excitation difference coefficient is just roughly in the scope of 3%-5%, but how much often rule of thumb concrete selection, lack concrete theory calculate.Meanwhile, because traditional excitation difference coefficient is rule of thumb determined, randomness is comparatively large, must guarantee that electrical network is in the economic scene of loss minimization.For this reason, the invention provides a kind of optimization method of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization.

Summary of the invention

The object of the invention is to, providing a kind of optimization method of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization, for determining the excitation difference coefficient making the whole network loss minimization.

To achieve these goals, the technical scheme that the present invention proposes is that a kind of optimization method of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization, is characterized in that described method comprises:

Step 1: set up the power flow algorithm taking into account the impact of generator excitation difference coefficient;

Described power flow algorithm is:

Wherein, Q _gfor the reactive power of generator;

R is generator excitation difference coefficient;

U _g0for generator floating voltage;

U _gfor generator terminal voltage;

Step 2: determine trend constraint equation;

Described trend constraint equation is: $\left\{\begin{array}{c}{U}_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-P_{L0i}=0\\ U_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-Q_{L0i}=0\\ U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}+B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-P_{\mathrm{Gk}}=0\\ U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}-B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-Q_{\mathrm{Gk}}=0\end{array}\right.;$

Wherein, U _ifor the voltage magnitude of load bus i;

U _jfor the voltage magnitude of node j;

G _ijfor the real part of node admittance matrix i-th row jth column element;

B _ijfor the imaginary part of node admittance matrix i-th row jth column element;

δ _ijfor the phase angle δ of load bus i _iwith the phase angle δ of node j _jdifference;

P _l0ifor the active power of load bus i;

Q _l0ifor the reactive power of load bus i;

U _kfor the voltage magnitude of generator node k;

G _kjfor the real part of node admittance matrix row k jth column element;

B _kjfor the imaginary part of node admittance matrix row k jth column element;

δ _kjfor the phase angle δ of generator node k _kwith the phase angle δ of node j _jdifference;

P _gkfor the active power of generator node k;

Q _gkfor the reactive power of generator node k;

represent

I=1,2 ..., m, m are the number of load bus;

K=m+1, m+2 ..., n, m-n are the number of generator node;

J=1,2 .., m, m+1 ..., n, n are the number sum of load bus and generator node;

Step 3: the grid-connected DCgenerator motor field difference coefficient optimization object function that to set up with the whole network loss minimization be target, and determine the constraints of described optimization object function;

Described optimization object function is $\min f\left(P_G\right)=\underset{k=m+1}{\overset{n}{\mathrm{\Σ}}}{P}_{\mathrm{Gk}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{P}_{L0i};$

Wherein, P _gfor the active power column vector of generator;

The constraints of described optimization object function is $\left\{\begin{array}{c}{U}_i^{\min }\≤U_i\≤U_i^{\max }\\ P_{\mathrm{Gk}}^{\min }\≤P_{\mathrm{Gk}}\≤P_{\mathrm{Gk}}^{\max }\\ R_{\min }\≤R_{\mathrm{Gk}}\≤R_{\max }\end{array}\right.;$

Wherein, for the voltage magnitude lower limit of load bus;

for the voltage magnitude upper limit of load bus;

for the active power lower limit of generator node;

for the active power upper limit of generator node;

R _minfor the lower limit of excitation difference coefficient;

R _maxfor the upper limit of excitation difference coefficient;

Step 4: the optimal solution asking for described optimization object function, thus the Excitation Adjustment be optimized difference coefficients R _gk.

The present invention takes into account the Mathematical Modeling of excitation difference coefficient by setting up, the optimization problem of excitation difference coefficient is converted into the optimal solution problem solving Non-Linear Programming, overcomes the deficiency that conventional method is determined empirically difference coefficient; In addition, contemplated by the invention the problem of power system voltage stabilization, keeping the stable of voltage by the voltage swing of bus being limited in certain scope; Meanwhile, the present invention for optimization aim with full loss minimization, can not only keep the stable of voltage by the optimization of excitation difference coefficient, and make the economic benefit of electrical network the highest.

Accompanying drawing explanation

Fig. 1 is the optimization method flow chart of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization;

Fig. 2 is certain province's electrical network reduced graph that embodiment provides.

Embodiment

Below in conjunction with accompanying drawing, preferred embodiment is elaborated.It is emphasized that following explanation is only exemplary, instead of in order to limit the scope of the invention and apply.

Embodiment 1

Fig. 1 is the optimization method flow chart of the grid-connected DCgenerator motor field difference coefficient based on the whole network loss minimization, and as shown in Figure 1, optimization method provided by the invention comprises:

Step 1: establish the power flow algorithm taking into account the impact of generator excitation difference coefficient.

The relation of the reactive power of generator terminal voltage and output is as the formula (1):

$Q_G=\frac{1}{R}(U_{G0}{U}_G-U_G^2)---\left(1\right)$

In formula (1), Q _gfor the reactive power of generator, R is generator excitation difference coefficient, U _g0for generator floating voltage, U _gfor generator terminal voltage.Formula (1) is power flow algorithm.

Step 2: determine trend constraint equation.

To suppose in electrical network total n node, wherein the 1st to a node m load bus, and remaining as generator node, then the trend constraint equation of system is as shown in the formula shown in (2)-(5):

$U_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-P_{L0i}=0---\left(2\right)$

$U_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-Q_{L0i}=0---\left(3\right)$

$U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}+B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-P_{\mathrm{Gk}}=0---\left(4\right)$

$U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}-B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-Q_{\mathrm{Gk}}=0---\left(5\right)$

In formula (2) and (3), U _ifor the voltage magnitude of load bus i, U _jfor the voltage magnitude of node j, G _ijfor the real part of node admittance matrix i-th row jth column element, B _ijfor the imaginary part of node admittance matrix i-th row jth column element, δ _ijfor the difference of the phase angle of load bus i and the phase angle of node j, P _l0ifor the active power of load bus i, Q _l0ifor the reactive power of load bus i, i=1,2 ..., m, m are the number of load bus.

In formula (4) and (5), U _kfor the voltage magnitude of generator node k, U _jfor the voltage magnitude of node j, G _kjfor the real part of node admittance matrix row k jth column element, B _kjfor the imaginary part of node admittance matrix row k jth column element, δ _kjfor the difference of the phase angle of generator node k and the phase angle of node j, P _gkfor the active power of generator node k, Q _gkfor the reactive power of generator node k, k=m+1, m+2 ..., n, m-n are the number of generator node.

In formula (2)-(5), node j is load bus or generator node, represent j=1,2 .., m, m+1 ..., n, n are the number sum of load bus and generator node.

According to power flow algorithm, there is following formula (6):

$Q_{\mathrm{Gk}}=\frac{1}{R_k}(U_{G0k}-U_{\mathrm{Gk}}-U_{\mathrm{Gk}}^2)---\left(6\right)$

In formula (6), Q _gkfor the reactive power of generator node k, R _kfor generator node k excitation difference coefficient, U _g0kfor the floating voltage of generator node k, U _gkfor generator node k set end voltage.

So, formula (6) is substituted into formula (5) can obtain:

$U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}-B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-\frac{1}{R_k}(U_{G0k}{U}_{\mathrm{Gk}}-U_{\mathrm{Gk}}^2)=0---\left(7\right)$

Step 3: the grid-connected DCgenerator motor field difference coefficient optimization object function that to set up with the whole network loss minimization be target, and determine the constraints of described optimization object function.

Optimization object function is:

$\min f\left(P_G\right)=\underset{k=m+1}{\overset{n}{\mathrm{\Σ}}}{P}_{\mathrm{Gk}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{P}_{L0i}---\left(8\right)$

In formula (8), P _gfor the active power column vector of generator.

Constraints is:

$\left\{\begin{array}{c}{U}_i^{\min }\≤U_i\≤U_i^{\max }\\ P_{\mathrm{Gk}}^{\min }\≤P_{\mathrm{Gk}}\≤P_{\mathrm{Gk}}^{\max }\\ R_{\min }\≤R_{\mathrm{Gk}}\≤R_{\max }\end{array}\right.---\left(9\right)$

In formula (9), for the voltage magnitude lower limit of load bus, for the voltage magnitude upper limit of load bus, for the active power lower limit of generator node, for the active power upper limit of generator node, R _minfor the lower limit of excitation difference coefficient, R _maxfor the upper limit of excitation difference coefficient.

Step 4: the optimal solution asking for described target function, thus the Excitation Adjustment be optimized difference coefficients R _gk.

By above-mentioned formula (8), the optimization problem of excitation difference coefficient is converted into the problem solving minimum value by the present invention.And formula (8) and (9) determined Mathematical Modeling are Nonlinear programming Model, therefore by selecting suitable numerical computation method, can try to achieve most suitable excitation difference coefficient, it both met constraints, and target function can be made again to obtain minimum value.

Embodiment 2

Fig. 2 be embodiment provide certain economize electrical network reduced graph, this figure comprises 22 nodes, wherein 1,5,8,20,21 nodes are generator node.

First, determine the value of each variable, comprise the floating voltage U of node admittance matrix, generator node _g0k, load bus active-power P _l0iwith the reactive power Q of load bus _l0i.

Secondly, given generator active power P _gk, Excitation Adjustment difference coefficients R _k, node voltage amplitude U _jwith node voltage phase angle δ _jinitial value.

Finally, utilize the optimization object function that the present invention sets up, in conjunction with constraints, in MATLAB software, solve by numerical solution algorithm iteration, obtain result as shown in the table.

As can be seen from the above table, after optimizing, the difference coefficient of generator 1 is 0.05000, and the difference coefficient of generator 5 is 0.00662, the difference coefficient of generator 8 is 0.05000, the difference coefficient of generator 20 is 0.00525, and the difference coefficient of generator 21 is 0.01035, and minimum network loss is 1.1632.

From result of calculation, it is more reasonable that the active power sent by optimization excitation difference coefficient and generator makes the active power of system and reactive power distribute, thus make the loss minimization of system.Moreover, because reactive power distribution is more reasonable, the voltage's distribiuting of bus is also more reasonable, is more conducive to maintaining the stable of system voltage.

The above; be only the present invention's preferably embodiment, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claim.

Claims (1)

1., based on an optimization method for the grid-connected DCgenerator motor field difference coefficient of the whole network loss minimization, it is characterized in that described method comprises:

Step 1: set up the power flow algorithm taking into account the impact of generator excitation difference coefficient;

Described power flow algorithm is:

Wherein, Q _gfor the reactive power of generator;

R is generator excitation difference coefficient;

U _g0for generator floating voltage;

U _gfor generator terminal voltage;

Step 2: determine trend constraint equation;

Described trend constraint equation is: $\left\{\begin{array}{c}{U}_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}+B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-P_{L0i}=0\\ U_i\underset{\mathrm{j\ϵi}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{ij}}\cos {\mathrm{\δ}}_{\mathrm{ij}}-B_{\mathrm{ij}}\sin {\mathrm{\δ}}_{\mathrm{ij}})-Q_{L0i}=0\\ U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}+B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-P_{\mathrm{Gk}}=0\\ U_k\underset{\mathrm{j\ϵk}}{\mathrm{\Σ}}{U}_j(G_{\mathrm{kj}}\cos {\mathrm{\δ}}_{\mathrm{kj}}-B_{\mathrm{kj}}\sin {\mathrm{\δ}}_{\mathrm{kj}})-Q_{\mathrm{Gk}}=0\end{array}\right.;$

Wherein, U _ifor the voltage magnitude of load bus i;

U _jfor the voltage magnitude of node j;

G _ijfor the real part of node admittance matrix i-th row jth column element;

B _ijfor the imaginary part of node admittance matrix i-th row jth column element;

δ _ijfor the phase angle δ of load bus i _iwith the phase angle δ of node j _jdifference;

P _l0ifor the active power of load bus i;

Q _l0ifor the reactive power of load bus i;

U _kfor the voltage magnitude of generator node k;

G _kjfor the real part of node admittance matrix row k jth column element;

B _kjfor the imaginary part of node admittance matrix row k jth column element;

δ _kjfor the phase angle δ of generator node k _kwith the phase angle δ of node j _jdifference;

P _gkfor the active power of generator node k;

Q _gkfor the reactive power of generator node k;

represent

I=1,2 ..., m, m are the number of load bus;

K=m+1, m+2 ..., n, m-n are the number of generator node;

J=1,2 .., m, m+1 ..., n, n are the number sum of load bus and generator node;

Step 3: the grid-connected DCgenerator motor field difference coefficient optimization object function that to set up with the whole network loss minimization be target, and determine the constraints of described optimization object function;

Described optimization object function is $\min f\left(P_G\right)=\underset{k=m+1}{\overset{n}{\mathrm{\Σ}}}{P}_{\mathrm{Gk}}+\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{P}_{L0i};$

Wherein, P _gfor the active power column vector of generator;

The constraints of described optimization object function is $\left\{\begin{array}{c}{U}_i^{\min }\≤U_i\≤U_i^{\max }\\ P_{\mathrm{Gk}}^{\min }\≤P_{\mathrm{Gk}}\≤P_{\mathrm{Gk}}^{\max }\\ R_{\min }\≤R_{\mathrm{Gk}}\≤R_{\max }\end{array}\right.;$

Wherein, for the voltage magnitude lower limit of load bus;

for the voltage magnitude upper limit of load bus;

for the active power lower limit of generator node;

for the active power upper limit of generator node;

R _minfor the lower limit of excitation difference coefficient;

R _maxfor the upper limit of excitation difference coefficient;

Step 4: the optimal solution asking for described optimization object function, thus the Excitation Adjustment be optimized difference coefficients R _gk.