int function return ans in 'int' itself . I got the expression J and L in terms of g but the integration of expression U1 return as in int function. Thanks in advance

5 views (last 30 days)
clc
clear
syms x g
assume(g >= 0)
Ef=427000*10^6; % in MPa
Em=89000*10^6;
Ec=184000*10^6;
mu=0.2825;
ao=0.001;
a=0.00189;
w=0.00512;
x1=0;
%x1=0.001
%R=.0725;
S=198*10^6;
%tow=20;
m1=0.6147+17.1844*(a^2/w^2)+8.7822*(a^6/w^6);
m2=0.2502+3.2889*(a^2/w^2)+70.0444*(a^6/w^6);
%m1=2.7552;
%m2=0.7889;
H=(1+m1*((g-x)/g)+m2*((g-x)/g)^2);
H1=(1+m1*((g-x1)/g)+m2*((g-x1)/g)^2);
% for remote stress S=198 MPa
c=S*((w/(w-ao))+6*w*ao*((0.5*(w-ao)-(x-ao))/(w-ao)^3));
j=int(((S*H)/(sqrt(g-x))),x,[0,ao]);
L=int(((S-c)*H)/(sqrt(g-x)),x, [ao,a]);
R=(H1*j+H1*L);
R1=(R/(sqrt(g-x1)))
U1=int(R1,g,0, a);

Answers (2)

Walter Roberson
Walter Roberson on 25 Sep 2021
Edited: Walter Roberson on 25 Sep 2021
clc
clear
syms x g
assume(g >= 0)
Ef=427000*10^6; % in MPa
Em=89000*10^6;
Ec=184000*10^6;
mu=0.2825;
ao=0.001;
a=0.00189;
w=0.00512;
x1=0;
%x1=0.001
%R=.0725;
S=198*10^6;
%tow=20;
m1=0.6147+17.1844*(a^2/w^2)+8.7822*(a^6/w^6);
m2=0.2502+3.2889*(a^2/w^2)+70.0444*(a^6/w^6);
%m1=2.7552;
%m2=0.7889;
H=(1+m1*((g-x)/g)+m2*((g-x)/g)^2);
H1=(1+m1*((g-x1)/g)+m2*((g-x1)/g)^2);
% for remote stress S=198 MPa
c=S*((w/(w-ao))+6*w*ao*((0.5*(w-ao)-(x-ao))/(w-ao)^3));
j=int(((S*H)/(sqrt(g-x))),x,[0,ao]);
L=int(((S-c)*H)/(sqrt(g-x)),x, [ao,a]);
R=(H1*j+H1*L);
R1=(R/(sqrt(g-x1)))
R1 = 
sR1 = simplify(R1)
sR1 = 
limit(sR1, g, 0)
ans = 
NaN
If you look at the pretty() output, you will see that there is a division by . The numerator has g to a variety of powers. The result is an expression whos limit cannot be taken at 0, and therefore cannot be integrated.
digits(100)
vpa(subs(sR1, g, sym(10)^(-20)))
ans = 
vpa(subs(sR1, g, sym(10)^(-50)))
ans = 
You can see from those last couple of evaluations that the real part is steady near 0, but that the imaginary part is exploding. I suspect that the denominator has a root in the range of integration.
  40 Comments
Walter Roberson
Walter Roberson on 9 Oct 2021
Before R2016b, to do piecewise() you can use
symLIST = @(varargin)feval(symengine,'DOM_LIST',varargin{:});
and then
feval(symengine, 'piecewise', symLIST(Condition1,Result1), symLIST(Condition2,Result2))
Walter Roberson
Walter Roberson on 9 Oct 2021
inmy case u(x) and c(x) dependent on each other
No they do not. u(x) depends upon c(x) but c(x) does not depend upon u(x) . An iterative approach is not called for.

Sign in to comment.


arvind thakur
arvind thakur on 9 Oct 2021
here is the c(x), ignore previous c(x) and all other value and equation is same.

Products


Release

R2019a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!