find the values of a variable for which the determinant of the matrix become zero?

7 次查看(过去 30 天)
a = 2e-3;
X = 150-3;
h = 20e-3;
b = 20e-3;
A = b*h;
l = 500e-3;
I = (b*(h^3))/12;
p = 2700;
E= 69e9;
v12 = 0.33;
j = 0.016;
Kt = 5.375588477512108e+05;
L = ((w^2*p*A*l^4)/(E*I))^(1/4);
Q = [1 0 1 0 0 0 0 0;...
0 1 0 1 0 0 0 0;...
cosh((L*X)/l) sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l) -cosh((L*X)/l) -sinh( (L*X)/l) -cos((L*X)/l) -sin((L*X)/l) ;...
sinh((L*X)/l) cosh((L*X)/l) sin((L*X)/l) -cos((L*X)/l) -sinh((L*X)/l) -cosh((L*X)/l) -sin((L*X)/l) cos((L*X)/l) ;...
cosh((L*X)/l) sinh((L*X)/l) -cos((L*X)/l) -sin((L*X)/l) -cosh((L*X)/l) -sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l);...
0 0 0 0 cosh(L) sinh(L) -cos(L) -sin(L);...
0 0 0 0 sinh(L) cosh(L) sin(L) -cos(L);...
cosh((L*X)/l)+((Kt*l)/(E*I*L))*sin((L*X)/l) sinh((L*X)/l)+((Kt*l)/(E*I*L))*cosh((L*X)/l) -cos((L*X)/l)-((Kt*l)/(E*I*L))*sin((L*X)/l) -sin((L*X)/l)+((Kt*l)/(E*I*L))*cos((L*X)/l) ((Kt*l)/(E*I*L))*sinh((L*X)/l) ((Kt*l)/(E*I*L))*cosh((L*X)/l) ((Kt*l)/(E*I*L))*sin((L*X)/l) ((Kt*l)/(E*I*L))*cos((L*X)/l)];
det(Q)
A = det(Q) ==0;
solve (A,L)
here I am finding the natural frequencies of beam.
I need to find w values for which the determinant of Q become zero, there will be n number of w values.
please help me to find the solution
Thank you.

采纳的回答

Walter Roberson
Walter Roberson 2021-9-16
syms w real
a = 2e-3;
X = 150-3;
h = 20e-3;
b = 20e-3;
A = b*h;
l = 500e-3;
I = (b*(h^3))/12;
p = 2700;
E= 69e9;
v12 = 0.33;
j = 0.016;
Kt = 5.375588477512108e+05;
L = ((w^2*p*A*l^4)/(E*I))^(1/4);
Q = [1 0 1 0 0 0 0 0;...
0 1 0 1 0 0 0 0;...
cosh((L*X)/l) sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l) -cosh((L*X)/l) -sinh( (L*X)/l) -cos((L*X)/l) -sin((L*X)/l) ;...
sinh((L*X)/l) cosh((L*X)/l) sin((L*X)/l) -cos((L*X)/l) -sinh((L*X)/l) -cosh((L*X)/l) -sin((L*X)/l) cos((L*X)/l) ;...
cosh((L*X)/l) sinh((L*X)/l) -cos((L*X)/l) -sin((L*X)/l) -cosh((L*X)/l) -sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l);...
0 0 0 0 cosh(L) sinh(L) -cos(L) -sin(L);...
0 0 0 0 sinh(L) cosh(L) sin(L) -cos(L);...
cosh((L*X)/l)+((Kt*l)/(E*I*L))*sin((L*X)/l) sinh((L*X)/l)+((Kt*l)/(E*I*L))*cosh((L*X)/l) -cos((L*X)/l)-((Kt*l)/(E*I*L))*sin((L*X)/l) -sin((L*X)/l)+((Kt*l)/(E*I*L))*cos((L*X)/l) ((Kt*l)/(E*I*L))*sinh((L*X)/l) ((Kt*l)/(E*I*L))*cosh((L*X)/l) ((Kt*l)/(E*I*L))*sin((L*X)/l) ((Kt*l)/(E*I*L))*cos((L*X)/l)];
detQ = det(Q)
detQ = 
sol3 = vpasolve(detQ, -1)
sol3 = 
subs(detQ, w, sol3)
ans = 
fplot(detQ, [-1e-2 1e-2])
sol1 = vpasolve(detQ, -1000)
sol1 = 
subs(detQ, w, sol1)
ans = 
0.0
vpa(subs(detQ, w, -1000))
ans = 
3.8828224224878183459368954124443e+1083
vpa(subs(diff(detQ,w),w,sol1))
ans = 
1.7469061892324894642718882345277e+1084
There might be additional solutions, but you can see that the slope is impossibly steep... I am not convinced that the solutin near -1000 is real: if you bump the digits up to 50, -1000.0 shows as the solution but substituting that back in shows a distinct non-zero value.
  7 个评论
Mani S
Mani S 2021-9-22
Respected walter Roberson, there were some mistakes,I have modified matrix. please try this time
syms w
a = 2e-3;
X = 98.4e-3;
h = 10e-3;
b = 20e-3;
A = b*h;
l = 820e-3;
I = (b*(h^3))/12;
p = 2700;
E= 72e9;
v12 = 0.33;
L = ((w^2*p*A*l^4)/(E*I))^(1/4);
j = 0.016;
%j = 1.8624*(a /h)^2 - 3.95*(a/h)^3 + 16.375*(a/h)^4 - 37.226*(a/h)^5 + 76.81*(a/h)^6 - 126.9*(a/h)^7+ 172*(a/h)^8 - 143.97*(a/h)^9 + 66.56*(a/h)^10
Kt = ((E*I)/(6*(1-v12^2)*h))*(1/j)
Q = [1 0 1 0 0 0 0 0;...
0 1 0 1 0 0 0 0;...
cosh((L*X)/l) sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l) -cosh((L*X)/l) -sinh( (L*X)/l) -cos((L*X)/l) -sin((L*X)/l) ;...
sinh((L*X)/l) cosh((L*X)/l) sin((L*X)/l) -cos((L*X)/l) -sinh((L*X)/l) -cosh((L*X)/l) -sin((L*X)/l) cos((L*X)/l) ;...
cosh((L*X)/l) sinh((L*X)/l) -cos((L*X)/l) -sin((L*X)/l) -cosh((L*X)/l) -sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l);...
0 0 0 0 cosh(L) sinh(L) -cos(L) -sin(L);...
0 0 0 0 sinh(L) cosh(L) sin(L) -cos(L);...
cosh((L*X)/l)+((Kt*l)/(E*I*L))*sin((L*X)/l) sinh((L*X)/l)+((Kt*l)/(E*I*L))*cosh((L*X)/l) -cos((L*X)/l)-((Kt*l)/(E*I*L))*sin((L*X)/l) -sin((L*X)/l)+((Kt*l)/(E*I*L))*cos((L*X)/l) -((Kt*l)/(E*I*L))*sinh((L*X)/l) -((Kt*l)/(E*I*L))*cosh((L*X)/l) ((Kt*l)/(E*I*L))*sin((L*X)/l) -((Kt*l)/(E*I*L))*cos((L*X)/l)];
det(Q)
A = det(Q) ==0;
solve (A,w)
the solutions to this problem are .
w1= 12.164Hz,
w2= 76.5471Hz,
w3= 214.6394Hz,
w4=420.4588Hz.
Walter Roberson
Walter Roberson 2021-9-22
No, none of the
format long g
syms w
a = 2e-3;
X = 98.4e-3;
h = 10e-3;
b = 20e-3;
A = b*h;
l = 820e-3;
I = (b*(h^3))/12;
p = 2700;
E= 72e9;
v12 = 0.33;
L = ((w^2*p*A*l^4)/(E*I))^(1/4);
j = 0.016;
%j = 1.8624*(a /h)^2 - 3.95*(a/h)^3 + 16.375*(a/h)^4 - 37.226*(a/h)^5 + 76.81*(a/h)^6 - 126.9*(a/h)^7+ 172*(a/h)^8 - 143.97*(a/h)^9 + 66.56*(a/h)^10
Kt = ((E*I)/(6*(1-v12^2)*h))*(1/j)
Kt =
140276.06329256
Q = [1 0 1 0 0 0 0 0;...
0 1 0 1 0 0 0 0;...
cosh((L*X)/l) sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l) -cosh((L*X)/l) -sinh( (L*X)/l) -cos((L*X)/l) -sin((L*X)/l) ;...
sinh((L*X)/l) cosh((L*X)/l) sin((L*X)/l) -cos((L*X)/l) -sinh((L*X)/l) -cosh((L*X)/l) -sin((L*X)/l) cos((L*X)/l) ;...
cosh((L*X)/l) sinh((L*X)/l) -cos((L*X)/l) -sin((L*X)/l) -cosh((L*X)/l) -sinh((L*X)/l) cos((L*X)/l) sin((L*X)/l);...
0 0 0 0 cosh(L) sinh(L) -cos(L) -sin(L);...
0 0 0 0 sinh(L) cosh(L) sin(L) -cos(L);...
cosh((L*X)/l)+((Kt*l)/(E*I*L))*sin((L*X)/l) sinh((L*X)/l)+((Kt*l)/(E*I*L))*cosh((L*X)/l) -cos((L*X)/l)-((Kt*l)/(E*I*L))*sin((L*X)/l) -sin((L*X)/l)+((Kt*l)/(E*I*L))*cos((L*X)/l) -((Kt*l)/(E*I*L))*sinh((L*X)/l) -((Kt*l)/(E*I*L))*cosh((L*X)/l) ((Kt*l)/(E*I*L))*sin((L*X)/l) -((Kt*l)/(E*I*L))*cos((L*X)/l)];
det(Q)
ans = 
A = det(Q);
%solve (A,w)
expects = [12.164, 76.5471, 214.6394, 420.4588] .';
digits(50)
for K = 1 : length(expects)
W = expects(K);
got(K,1) = subs(A, w, W);
sol = vpasolve(A, w, W);
if isempty(sol)
nearby(K,1) = nan;
val_nearby(K,1) = nan;
else
nearby(K,1) = sol;
val_nearby(K,1) = subs(A, w, sol);
end
end
double([expects, got, nearby, val_nearby])
ans = 4×4
12.164 -20185.0170574006 77.9685880444843 -1.44749574417154e-53 76.5471 -289.096868327054 77.9685880444843 -1.44749574417154e-53 214.6394 25085.7641362872 77.9685880444843 -1.44749574417154e-53 420.4588 22421.5359853579 491.06556340064 3.07614015113951e-53
fplot([A], expects(3)+[-1/100 1/100]); %ylim([-100 100])
testlocs = linspace(expects(3)-1/10, expects(3)+1/10, 50);
vals_at_test = double(subs(A, w, testlocs));
plot(testlocs, vals_at_test, '-*');
fplot(A, [1 2500])
Actual zeros: near 78, 491, 1300, 2600, possibly others

请先登录,再进行评论。

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Creating and Concatenating Matrices 的更多信息

Community Treasure Hunt

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

Start Hunting!

Translated by