Hiiii having problem in coding optimazation function
이전 댓글 표시
hiii i have problem in defining objective function for optimization. my objective function is
y=sum(abs(Voc-Vbc));
where Voc is given data and Vbc is obtained by
Vbc=(((Q./C1)+(Ic*R2)).*exp(-tc./(R2.*C1)))+Vo-(Ic*(R1+R2));
in this Q tc and Ic are constant and R1 C1 R2 and Vo is obtained from
R1 = ((a(1)+a(2)*cr+a(3)*(cr^2))*exp(-a(4).*SOC))+(a(5)+a(6)*cr+a(7)*(cr^2));
R2 = ((a(8)+a(9)*cr+a(10)*(cr^2))*exp(-a(11).*SOC))+(a(12)+a(13)*cr+a(14)*(cr^2));
C1 = (-(a(15)+a(16)*cr+a(17)*(cr^2))*exp(-a(18).*SOC))+(a(19)+a(20)*cr+a(21)*(cr^2));
Vo = ((a(22)+a(23)*cr+a(24)*(cr^2))*exp(-a(25).*SOC))+(a(26)+a(27).*SOC+a(28).*(SOC.^2)+a(29).*(SOC.^3))-(a(30)*cr)+(a(31)*(cr^2));
Here SOC is an array and cr is constant a is array with (1, 31) dim . this a matrix is variable for our optimization.
in my program i want R1 C1 R2 and Vo always remain positive. how i can define it in matlab so this values always remain positive.
댓글 수: 2
Alan Weiss
2015년 3월 13일
Please format your question using the {} Code button so that we can read it.
Alan Weiss
MATLAB mathematical toolbox documentation
venu sangwan
2015년 3월 15일
답변 (1개)
Alan Weiss
2015년 3월 16일
It is very hard to understand you, so my comments might be off base. I am not sure that I understand what your control variables are. Control variables are the ones that you change to try to optimize something.
Suppose that your control variables are R1, C1, R2, and V0. Then you write your objective function for the lsqnonlin function as
function y = myfun(x,Vbc)
R1 = x(1);
C1 = x(2);
R2 = x(3);
V0 = x(4)
% Now write the calculation of Voc
% End the calculation with
y = Voc - Vbc
and pass your function as
fun = @(x)myfun(x,Vbc); % gets the objective function
solution = lsqnonlin(fun,x0) % you need an initial guess x0
I hope this helps,
Alan Weiss
MATLAB mathematical toolbox documentation
카테고리
도움말 센터 및 File Exchange에서 Solver Outputs and Iterative Display에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!