- Do you receive warning and/or error messages? If so the full and exact text of those messages (all the text displayed in orange and/or red in the Command Window) may be useful in determining what's going on and how to avoid the warning and/or error.
- Does it do something different than what you expected? If so, what did it do and what did you expect it to do?
- Did MATLAB crash? If so please send the crash log file (with a description of what you were running or doing in MATLAB when the crash occured) to Technical Support so we can investigate.
Info
This question is closed. Reopen it to edit or answer.
I want to do the optimization but I dont know how to do it ? please help me
2 views (last 30 days)
Show older comments
clear
clc
%A = eye(4);
A=eye(5);
%b = [80; 320; 50;50];
%UB=[80; 320; 50;50];
b = [80*2; 290*2; 200;200;200];
UB=[80*2; 290*2; 200;200;200];
b = [65*2/2/2; 570/2/2; 200;200;220];
UB=[65*2/2/2; 570/2/2; 200;200;220];
%b = [107; 453; 94;163;200];
%UB=[107; 453; 94;163;200];
%lb = [50;200;30;30];
%LB=[50;200;30;30];
lb = [50;220;50;80;50];
LB=[50;220;50;80;50];
lb = [48*2/2/2;215*2/2/2;30;50;20];
LB=[48*2/2/2;215*2/2/2;30;50;20];
%lb = [107;453;94;163;25];
%LB=[107;453;94;163;25];
%pause(5)
%[x,fval,exitflag] = ga(@TMD_nonlineardamploop,4,A,b,[],[],lb)
[x,fval,exitflag] = ga(@TMD_nonlineardamploop,5,A,b,[],[],LB,UB,[],[1,2,3,4,5])
function y = TMD_nonlineardamploop(x)
if ~isvector(x)
error('Input must be a vector')
end
Kxt=x(1);
Kyt=x(2);
Kisayx=x(3)/1000;
Kisayy=x(4)/1000;
Kxt
Kyt
Kisayx
Kisayy
Ly=x(5)/100
m=30.;
Kx=1320; Ky=1320*90.0;
Kx=1320; Ky=1320*1.00;
wx=(Kx/m)^0.5;
%Kyt=13.5;
mt=1.5*2/2/2;
Lx=1.;
%Ly=1.;
Cx=2*0.025*4*(Kx/m)^0.5*m;
Cy=2*0.025*4*(Ky/m)^0.5*m;
%Kisayy=0.065;
%mt is for TMD
Cxt=2*Kisayx*(Kxt/mt)^0.5*mt; Cyt=2*Kisayy*(Kyt/mt)^0.5*mt;
%%stiffness & mass & damping
Beta=0.0;
%6063 7310 6160 6456
%309+15
%1270642
M=[m 0 0 0;0 m 0 0;0 0 mt 0;0 0 0 mt];
K=[Kx+Kxt 0 -Kxt 0;0 Ky+Kyt 0 -Kyt; -Kxt 0 Kxt 0; 0 -Kyt 0 Kyt];
C=[Cx+Cxt 0 -Cxt 0;0 Cy+Cyt 0 -Cyt; -Cxt 0 Cxt 0; 0 -Cyt 0 Cyt];
%eig(K*M^(-1))
wy=(Ky/m)^0.5;
%AAa=0.99;
www=1;
for F_wy=0.65*wx:0.1/2:1.25*wx
for jjjj=1:2000.
time=jjjj*0.02;
AAa=0.65;
%U_dotdotg(jjjj)=AAa*sin(1.*wy*time);
U_dotdotg(jjjj)=AAa*sin(F_wy*time);
end
g=1;
dt=0.02;
%%---------------------------------------------
%these parameters are for inital displacement, acceleration, velocity
x=zeros(length(U_dotdotg),4);
v=zeros(length(U_dotdotg),4);
a=zeros(length(U_dotdotg),4);
%time
time=0:dt:dt*(length(U_dotdotg)-1);
%these parameters are used for Newmark integration
%aa = K*dt^2/2+C*dt;
%bb = K*dt;
%Keff=M+0.5*dt*C+dt^2/4*K;
length(time);
length(U_dotdotg);
for i=1:length(U_dotdotg)-1
i ;
%initial step calcultion
disp(1,:)=x(i,:)+dt*v(i,:)+0.5*dt^2*a(i,:);
velocity(1,:)=v(i,:)+dt*a(i,:);
accel(1,:)=a(i,:);
%initial guess of displacement, velocity, acceleration
DD=disp(i,:)+dt*velocity(i,:)+dt^2/2*accel(i,:);
VV=velocity(i,:)+dt*accel(i,:);
AA=accel(i,:);
criterion=10;
j=1;
%%ccc(1,:)=[0,0];
Keff=M+0.5*dt*C+dt^2/4*K;
eig(K^(-1)*M);
while criterion>0.00005
for ii=1:4
xx=DD(3)-DD(1) ;
yy=DD(4)-DD(2);
VVx=VV(3)-VV(1);
VVy=VV(4)-VV(2);
fx1=Kxt*( Lx+xx -(Lx^2+Lx*xx)/ ((Lx+xx )^2+yy ^2)^0.5)-Kxt*xx ; fx2=Kyt*(xx- (Ly*xx) /((Ly+yy) ^2+xx^2 )^0.5) ;
fxdamp1=Cxt*(VVx*(-yy^2/Lx^2+2*xx*yy^2/Lx^3)+VVy*(yy/Lx-xx*yy/Lx^2-yy^3/Lx^3+yy*xx^2/Lx^3));
fxdamp2=Cyt*(VVx*(+xx^2/Ly^2-2*xx^2*yy/Ly^3)+VVy*(xx/Ly-xx*yy/Ly^2-xx^3/Lx^3+yy^2*xx/Ly^3));
%%fXdamp1=Cxt*(VVx*(Lx+xx)^2+VVy*yy*(Lx+xx))/((Lx+xx)^2+yy^2)-Cxt*VVx;
%%fXdamp2=Cyt*(VVy*(Ly+yy)*xx+VVx*xx^2)/((Ly+yy)^2+xx^2);
%fx1=Kxt*( xx +yy^2/2/Lx -xx*yy^2/Lx^2)-Kxt*xx ; fx2=Kyt*(xx*yy/Ly +xx^3/2/Ly^2 -xx*yy^2/Ly^2) ;
fy1=Kxt*(yy -(Lx*yy) /((Lx+xx)^2+yy^2 )^0.5) ; fy2= Kyt*( Ly+yy -(Ly^2+Ly*yy) /((Ly+yy) ^2+ xx^2)^0.5)-Kyt*yy ;
fydamp1=Cxt*(VVy*(+yy^2/Lx^2-2*yy^2*xx/Lx^3)+VVx*(yy/Lx-yy*xx/Lx^2-yy^3/Ly^3+xx^2*yy/Lx^3));
fydamp2=Cyt*(VVy*(-xx^2/Ly^2+2*yy*xx^2/Ly^3)+VVx*(xx/Ly-yy*xx/Ly^2-xx^3/Ly^3+xx*yy^2/Ly^3));
%%fYdamp1=Cxt*(VVx*(Lx+xx)*yy+VVy*yy^2)/((Lx+xx)^2+yy^2);
fYdamp2=Cyt*(VVy*(Ly+yy)^2+VVx*xx*(Ly+yy))/((Ly+yy)^2+xx^2)-Cyt*VVy;
%fy1= Kxt*(xx*yy/Lx +yy^3/2/Lx^2 -yy*xx^2/Lx^2) ;fy2=Kyt*( yy +xx^2/2/Ly -yy*xx^2/Ly^2)-Kyt*yy;
if ii==1
F_N_S(ii)=-fx1-fx2+(-fxdamp1-fxdamp2);
Fe(i,ii)=-U_dotdotg(i)*g*cos(Beta);
Fe(i+1,ii)=-U_dotdotg(i+1)*g*cos(Beta);
elseif ii==2
F_N_S(ii)=-fy1-fy2+(-fydamp1-fydamp2);
Fe(i,ii)=-U_dotdotg(i)*g*sin(Beta);
Fe(i+1,ii)=-U_dotdotg(i+1)*g*sin(Beta);
elseif ii==3
F_N_S(ii)=fx1+fx2+(fxdamp1+fxdamp2);
Fe(i,ii)=-U_dotdotg(i)*g*cos(Beta);
Fe(i+1,ii)=-U_dotdotg(i+1)*g*cos(Beta);
elseif ii==4
F_N_S(ii)=fy1+fy2+(fydamp1+fydamp2);
Fe(i,ii)=-U_dotdotg(i)*g*sin(Beta);
Fe(i+1,ii)=-U_dotdotg(i+1)*g*sin(Beta);
end
end
%ccc(i+1,:)=[F_N_S(1)+Kxt*xx,F_N_S(2)+Kyt*yy];
%ccc(i+1,:)=[Kxt*( Lx+xx -(Lx^2+Lx*xx)/ ((Lx+xx )^2+yy ^2)^0.5),Kyt*(xx- (Ly*xx) /((Ly+yy) ^2+xx^2 )^0.5)];
%ccc(i+1,:)=[Kxt*(yy -(Lx*yy) /((Lx+xx)^2+yy^2 )^0.5),Kyt*( Ly+yy -(Ly^2+Ly*yy) /((Ly+yy) ^2+ xx^2)^0.5)];
%%ccc(i+1,:)=[fydamp2,fYdamp2];
Fe(i+1,:)=Fe(i+1,:)*M;
deltaPhat=-AA*M +Fe(i+1,:) -VV*C -DD*K -F_N_S;
deltaA=deltaPhat*Keff^(-1);
DD=DD+dt^2*0.25*deltaA;
VV=VV+dt*0.5*deltaA;
AA=AA+deltaA;
criterion=max(abs(deltaA));
j=j+1;
if j==95
criterion=0.0;
end
NORM1(i)=max(abs(deltaPhat));
end
j ;
iterationnum(i)=j;
accel(i+1,:)=AA;
velocity(i+1,:)=VV;
disp(i+1,:)=DD;
end
MMM(www,:)=[max(abs(disp(1500:1999,1))),max(abs(disp(1500:1999,2)))];
C_MMM(www)=www;
C_MMM1(www)=F_wy;
www=www+1;
if mod(www,10)==0
www;
end
%Axt=max(abs(disp(1500:1999,3)-disp(1500:1999,1)));
%Ayt=max(abs(disp(1500:1999,4)-disp(1500:1999,2)));
% AMP(www,:)=[max(abs(disp(1500:1999,1))),max(abs(disp(1500:1999,2))) ,Axt ,Ayt];
end
%plot(C_MMM1,[MMM(:,1),MMM(:,2)]),grid
%load 'disp_linear.txt'
non_linansw=[C_MMM1',MMM(:,1),MMM(:,2)];
%save('non_linansw_1.txt','non_linansw','-ASCII');
%load 'linansw.txt'
%plot(linansw(:,1)/6.63,[linansw(:,2),non_linansw(:,2),linansw(:,3),non_linansw(:,3)]),grid
%plot(AMP(:,1))
%plot(disp(2000:2500,3)-disp(2000:2500,1))
%plot(linansw(:,1)/6.63,[non_linansw(:,2),non_linansw(:,3)]),grid
%plot(time(:),[disp(:,2)-disp(:,4)]),grid
%plot(time(:),[ccc(:,1),ccc(:,2)]),grid
%plot(time(:),[ccc(:,1)]),grid
y =max( max([non_linansw(:,2),non_linansw(:,3)]));
y
end
3 Comments
Rik
on 27 Jun 2023
Edited: Rik
on 27 Jun 2023
You forgot to ask a question; you just dumped uncommented code on us with a "help me". Have a read here and here. It will greatly improve your chances of getting an answer.
I edited your code, moving the part I suspect to be the calling script to the top. I tried running your code, but it took more than the 55 second limit.
Answers (0)
This question is closed.
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!