| 查看: 219 | 回复: 1 | ||
[求助]
感谢热心人啊，急切求助，万分感谢!
|
|
%%%%%3<t<8时,tau=0.7;其余时间，tau=0，下面模型是离散化的方法得到的 %%%%%%%%从理论上应该收敛到0，但是从0到586都是收敛到0的，586之后就出现一段震荡，这是怎么回事？？？ clc clear T=0.01; N=1000;%%%%%%%%%%%%%%%%%%%%%%%%不知道为何后面有一段是发散的，但是现在前面那段收敛的图形与常数单时滞的图像还是比较吻合的 x1=zeros(1,N); x2=zeros(1,N); x3=zeros(1,N); x1(1)=1; x2(1)=1; x3(1)=1; for i=1:300 tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=301:799%%%%%%%%%在此范围内有时滞存在 tau=0.7; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=800:N tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end yy1=x1(1:588); %yy2=x2(1:570); %yy3=x3(1:570); figure (100) plot(yy1,'b-.'); %hold on; %plot(yy2,'k'); %hold on; %plot(yy3); %%%%%3<t<8时,tau=0.7;其余时间，tau=0 %%%%%%%%从理论上应该收敛到0，但是从0到586都是收敛到0的，586之后就出现一段震荡，这是怎么回事？？？ clc clear T=0.01; N=1000;%%%%%%%%%%%%%%%%%%%%%%%%不知道为何后面有一段是发散的，但是现在前面那段收敛的图形与常数单时滞的图像还是比较吻合的 x1=zeros(1,N); x2=zeros(1,N); x3=zeros(1,N); x1(1)=1; x2(1)=1; x3(1)=1; for i=1:300 tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=301:799%%%%%%%%%在此范围内有时滞存在 tau=0.7; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=800:N tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end yy1=x1(1:588); %yy2=x2(1:570); %yy3=x3(1:570); figure (100) plot(yy1,'b-.'); %hold on; %plot(yy2,'k'); %hold on; %plot(yy3); %%%注意：将上面直接复制到MATLAB建立M文件会直接出结果，非常感谢啊，谢谢！！！ %%%%%3<t<8时,tau=0.7;其余时间，tau=0 %%%%%%%%从理论上应该收敛到0，但是从0到586都是收敛到0的，586之后就出现一段震荡，这是怎么回事？？？ clc clear T=0.01; N=1000;%%%%%%%%%%%%%%%%%%%%%%%%不知道为何后面有一段是发散的，但是现在前面那段收敛的图形与常数单时滞的图像还是比较吻合的 x1=zeros(1,N); x2=zeros(1,N); x3=zeros(1,N); x1(1)=1; x2(1)=1; x3(1)=1; for i=1:300 tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=301:799%%%%%%%%%在此范围内有时滞存在 tau=0.7; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end for i=800:N tau=0; x1(i+1)=x1(i)+((-2*i*T-1/2)*x1(i)+x2(i)+sqrt(i*T/2)*x1(i-tau/T)-1/2*exp(i*T)*(x3(i))^3)*T; x2(i+1)=x2(i)+(i*T*x1(i)-(3*T*i+1/2)*x2(i)+sqrt(i*T/6)*x2(i-tau/T))*T; x3(i+1)=x3(i)+(exp(i*T)*x1(i)+sqrt(i*T)*x2(i)-(exp(i*T))*x3(i))*T; end yy1=x1(1:588); %yy2=x2(1:570); %yy3=x3(1:570); figure (100) plot(yy1,'b-.'); %hold on; %plot(yy2,'k'); %hold on; %plot(yy3); |
回复此楼» 猜你喜欢
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有8人回复
-
体制内长辈说体制内绝大部分一辈子在底层，如同你们一样大部分普通教师忙且收入低
已经有13人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有8人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有8人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有11人回复
-
为什么中国大学工科教授们水了那么多所谓的顶会顶刊，但还是做不出宇树机器人？
已经有8人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有5人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有6人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有9人回复
-
售SCI一区文章，我:8 O5 51O 54,科目齐全,可+急
已经有4人回复
2楼2015-11-19 17:53:46
