%拟合微分方程参数
function myproject
clear;
clc;
treal=[0 1 2 3 4 5 6 7 8 9 10 11 12 13 14];%微分方程的x的值
yreal=[20.4142 22.73032 24.55624 25.88679 26.69019...
27.16334 27.45866 27.64919 27.57615 27.70635...
27.77303 27.86195 27.91911 27.95721 27.98897]';%微分方程y的值
%遗传算法
options = gaoptimset('Generations',1000,'StallGenLimit',300,...
'StallTimeLimit',50,'TolFun',1e-12,'TolCon',1e-12);
[k0,fva,reason,output,final_pop]=ga(@my_funtest,3,options);
options = gaoptimset('InitialPopulation',final_pop,'Generations',1000,'StallGenLimit',300,...
'StallTimeLimit',50,'TolFun',1e-12,'TolCon',1e-12);
[k1,fva,reason,output,final_pop2]=ga(@my_funtest,3,options);
k=k0;
%输出参数
fprintf('\n\n遗传算法的初始估计数值:\n');
fprintf('\n\t参数 a0 = %.9f',k1(1));
fprintf('\n\t参数 b0 = %.9f',k1(2));
fprintf('\n\t参数 c0 = %.9f',k1(3));
%图形结果打印
[t,ycal]=ode45(@myfun,treal,yreal(1),[],k);
figure;
plot(treal,yreal,'r',treal,ycal,'b');
end
%构造适应度函数my_funtest
function yfit=my_funtest(k)
tspan=[0 1 2 3 4 5 6 7 8 9 10 11 12 13 14];
yreal=[20.4142 22.73032 24.55624 25.88679 26.69019 ...
27.16334 27.45866 27.64919 27.57615 27.70635 ...
27.77303 27.86195 27.91911 27.95721 27.98897];
y0=20.4142;
[t,ycal]=ode45(@myfun,tspan,y0,[],k);
n=length(yreal);
for l=1:n
ff(l)=(ycal(l)-yreal(l))^2;
end
yfit=sum(ff(l));
end
function dydt = myfun(t,y,k)
beta(1)=k(1);
beta(2)=k(2);
beta(3)=k(3);
dydt = 0.001414/(1/(4.5727*10^4*beta(1)*(1-0.02119*y)^beta(2))+1/beta(3)/45727);
end
微分方程中有三个参数待拟合,在下初学者,实在没有办法了,希望有高手帮忙解答。。。。。