💥💥💞💞欢迎来到本博客❤️❤️💥💥
🏆博主优势:🌞🌞🌞博客内容尽量做到思维缜密,逻辑清晰,为了方便读者。
⛳️座右铭:行百里者,半于九十。
📋📋📋本文目录如下:🎁🎁🎁
目录
💥1 概述
直流/直流变换器(HVDC)多端高压直流输电系统是一种新型的电力输电技术,它具有高效、稳定、灵活等优点,因此受到了广泛关注。在这种系统中,为了实现最优的功率分配和输电效率,需要进行最优功率流(OPF)研究。而半定规划(SDP)是一种用于求解非线性优化问题的方法,可以有效地应用于HVDC多端系统的OPF问题求解中。
在HVDC多端系统中,存在多个直流/直流变换器和多个相互连接的直流输电线路,因此系统的复杂度较高。为了实现最优的功率分配和输电效率,需要考虑多个变换器之间的相互影响和线路的功率损耗等因素。这就需要一个高效的优化算法来求解OPF问题。
SDP方法可以有效地处理具有复杂约束条件和非线性目标函数的优化问题,因此非常适合用于HVDC多端系统的OPF问题求解。通过将系统的功率流模型转化为半定规划问题,并结合求解器进行求解,可以得到最优的功率分配方案,从而提高系统的输电效率和稳定性。
因此,使用半定规划(SDP)求解最优功率流(OPF)是HVDC多端系统研究中的重要课题,可以为实际工程中的输电系统设计和运行提供重要的参考和指导。同时,也为HVDC多端系统的优化运行和控制提供了新的研究思路和方法。
本文考虑直流/直流变换器的多端高压直流输电系统中使用半定规划(SDP)来进行最优功率流(OPF)的方法。该方法来源于A. Garces, D. Montoya, R. Torres. "Optimal Power Flow in Multiterminal HVDC Systems considering DC/DC converters",
📚2 运行结果
Calling SDPT3 4.0: 172 variables, 47 equality constraints
For improved efficiency, SDPT3 is solving the dual problem.
------------------------------------------------------------
num. of constraints = 47
dim. of sdp var = 10, num. of sdp blk = 2
dim. of socp var = 2, num. of socp blk = 1
dim. of linear var = 130
dim. of free var = 1
6 linear variables from unrestricted variable.
*** convert ublk to lblk
*******************************************************************
SDPT3: Infeasible path-following algorithms
*******************************************************************
version predcorr gam expon scale_data
HKM 1 0.000 1 0
it pstep dstep pinfeas dinfeas gap prim-obj dual-obj cputime
-------------------------------------------------------------------
0|0.000|0.000|9.1e+01|3.3e+02|6.5e+05| 6.789949e+02 0.000000e+00| 0:0:00| chol 1 1
1|0.957|0.845|3.9e+00|5.1e+01|1.0e+05| 1.378562e+03 -1.974896e+02| 0:0:00| chol 1 1
2|1.000|0.952|1.3e-06|2.5e+00|5.9e+03| 7.058639e+02 -9.361565e+00| 0:0:00| chol 1 1
3|0.820|0.868|1.5e-06|3.5e-01|1.3e+03| 5.260220e+02 -1.188702e+00| 0:0:00| chol 1 1
4|0.982|0.954|2.8e-07|2.2e-02|1.6e+02| 1.123898e+02 -7.621943e-01| 0:0:00| chol 1 1
5|0.992|0.425|2.1e-07|1.4e-02|8.2e+01| 5.170660e+01 -8.262528e-01| 0:0:01| chol 1 1
6|1.000|0.387|6.6e-08|8.6e-03|4.7e+01| 2.762801e+01 -8.797233e-01| 0:0:01| chol 1 1
7|1.000|0.527|2.6e-08|4.2e-03|1.8e+01| 8.904206e+00 -9.778464e-01| 0:0:01| chol 1 1
8|0.955|0.702|7.2e-09|1.3e-03|4.4e+00| 1.654725e+00 -1.040556e+00| 0:0:01| chol 1 1
9|1.000|0.514|1.5e-09|6.3e-04|1.1e+00|-4.022400e-01 -1.035286e+00| 0:0:01| chol 1 1
10|0.788|0.290|1.5e-10|4.5e-04|8.0e-01|-4.795658e-01 -9.778514e-01| 0:0:01| chol 1 1
11|1.000|0.249|2.2e-10|3.4e-04|6.3e-01|-5.445901e-01 -9.435774e-01| 0:0:01| chol 1 1
12|1.000|0.434|9.3e-11|1.9e-04|3.3e-01|-6.892845e-01 -8.939429e-01| 0:0:01| chol 1 1
13|1.000|0.550|4.3e-11|8.6e-05|1.2e-01|-7.877809e-01 -8.553636e-01| 0:0:01| chol 1 1
14|1.000|0.351|9.4e-12|5.6e-05|7.7e-02|-8.021517e-01 -8.438502e-01| 0:0:01| chol 1 1
15|1.000|0.618|5.1e-12|2.1e-05|2.7e-02|-8.164122e-01 -8.294243e-01| 0:0:01| chol 1 1
16|0.990|0.924|2.5e-12|1.6e-06|1.9e-03|-8.198496e-01 -8.206746e-01| 0:0:01| chol 1 1
17|0.981|0.986|4.6e-12|1.1e-06|2.8e-05|-8.199469e-01 -8.199574e-01| 0:0:01| chol 1 1
18|1.000|0.980|1.8e-12|1.7e-08|9.9e-07|-8.199484e-01 -8.199490e-01| 0:0:01| chol 1 1
19|1.000|0.987|3.0e-11|6.0e-10|2.9e-08|-8.199488e-01 -8.199488e-01| 0:0:01| chol 1 1
20|1.000|0.986|7.8e-12|1.8e-11|1.1e-09|-8.199488e-01 -8.199488e-01| 0:0:01| chol 1 2
21|1.000|0.986|2.7e-13|7.0e-13|5.1e-11|-8.199488e-01 -8.199488e-01| 0:0:01| chol
linsysolve: Schur complement matrix not positive definite
linsysolve: solution contains NaN or inf
stop: difficulty in computing predictor directions
-------------------------------------------------------------------
number of iterations = 22
primal objective value = -8.19948777e-01
dual objective value = -8.19948777e-01
gap := trace(XZ) = 5.06e-11
relative gap = 1.92e-11
actual relative gap = 2.28e-11
rel. primal infeas (scaled problem) = 2.73e-13
rel. dual " " " = 6.97e-13
rel. primal infeas (unscaled problem) = 0.00e+00
rel. dual " " " = 0.00e+00
norm(X), norm(y), norm(Z) = 5.0e+02, 4.5e+00, 1.0e+01
norm(A), norm(b), norm(C) = 1.2e+03, 1.3e+02, 1.4e+01
Total CPU time (secs) = 0.63
CPU time per iteration = 0.03
termination code = 0
DIMACS: 3.9e-13 0.0e+00 1.7e-12 0.0e+00 2.3e-11 1.9e-11
-------------------------------------------------------------------
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0199488
Rango =
3
---------------- Power Losses ----------------
ans =
0.0131
---------------- Voltages --------------------
1.0000
1.0100
1.0053
0.9907
0.9873
1.0068
0.9896
0.9911
---------------- Line Power Flow -------------
2 -> 6 : 0.300000
6 -> 2 : -0.299057
1 -> 6 : -0.191593
6 -> 1 : 0.192901
6 -> 3 : 0.106157
3 -> 6 : -0.105998
1 -> 3 : -0.373545
3 -> 1 : 0.375534
1 -> 4 : 0.785087
4 -> 1 : -0.777768
4 -> 5 : 0.277768
5 -> 4 : -0.276834
5 -> 7 : -0.123166
7 -> 5 : 0.123443
8 -> 7 : 0.123628
7 -> 8 : -0.123443
---------------- DC/DC Converters -----------
Qp =
0.0068
>>
部分代码:
% We define a new variable X = V'*V
cvx_begin
cvx_precision high
variable X(NumN,NumN) symmetric;
variable P(NumN,1);
variable Q(NumDCDC,1);
variable S(NumN,1);
variable F(2*NumL,1);
variable Qp;
minimize trace(G*X) + Qp;
subject to
X == semidefinite(NumN);
X <= (Vmax^2)*ones(NumN);
X >= (Vmin^2)*ones(NumN);
P - S - Ad'*Q == diag(G*X);
P <= Pmax;
P >= Pmin;
Q <= Qmax;
Q >= -Qmax;
2*S >= abs(Ad')*(gamma*Q.^2 + beta*abs(Q));
X(1,1) == 1;
Qp == sum(S);
F == HH*vec(X);
F >= -Fmax;
F <= Fmax;
cvx_end
Rango = rank(X)
%% Eigenvalue approximation
[V,D] = eig(X);
lm = diag(D);
vm = V(:,NumN);
vm*vm'*lm(NumN);
disp('---------------- Power Losses ----------------');
trace(G*X)
disp('---------------- Voltages --------------------');
Vap = sqrt(lm(NumN))*vm;
disp(Vap);
disp('---------------- Line Power Flow -------------');
kk = 0;
for k = 1:NumL
kk = kk +1;
fprintf('%i%s%i%s%f\n',MT.Lines(k,1),' -> ',MT.Lines(k,2),' : ',F(kk));
kk = kk +1;
fprintf('%i%s%i%s%f\n',MT.Lines(k,2),' -> ',MT.Lines(k,1),' : ',F(kk));
end
disp('---------------- DC/DC Converters -----------');
Qp
🎉3 参考文献
文章中一些内容引自网络,会注明出处或引用为参考文献,难免有未尽之处,如有不妥,请随时联系删除。
A. Garces, D. Montoya and R. Torres, "Optimal power flow in multiterminal HVDC systems considering DC/DC converters," 2016 IEEE 25th International Symposium on Industrial Electronics (ISIE), Santa Clara, CA, USA, 2016, pp. 1212-1217, doi: 10.1109/ISIE.2016.7745067.