easy question : why i get NaN here ?

>> u=1/(1+limit(K/(w*(((J*w+b)*(L*w+R))+K^2)),w,0))

or

>> limit(K/(w*(((J*w+b)*(L*w+R))+K^2)),w,0)

[yahya elyan]

my bad xD

J = 0.01;
b = 0.0737;
K = 0.072;
R = 1.17;
L = 0.024;
syms w

[yahya elyan]

it should = 0 xD

[Walter Roberson]

Those both involve the limit of K/0 . That would not necessarily result in NaN, depending on what is known about K.

If K is 0, then it would be limit 0/0 which would be nan.

[yahya elyan]

but K not = 0

[yahya elyan]

UP

[yahya elyan]

it can be solved by using

>>U=1/(1+limit((K/(H*((J*H+b)*(L*H+R)+K^2))),H,0,'right'));

Answer 1

[yahya elyan]

try using

U=1/(1+limit((K/(H*((J*H+b)*(L*H+R)+K^2))),H,0,'right'));

[madhan ravi]

It is solved but do you know the reason for NaN xD? Because your original question asks “why...?”

[yahya elyan]

yah if i remember well it was about being unable to approach zero for solving limit so we usually took 0+ or 0- value thats a way instead of using lupital xD

Answer 2

[Walter Roberson]

J = 0.01;
b = 0.0737;
K = 0.072;
R = 1.17;
L = 0.024;
syms w

temp = K/(w*(((J*w+b)*(L*w+R))+K^2));
tempc = collect(temp, w);
disp(tempc)

ans =
 
120000/(400*w^3 + 22448*w^2 + 152355*w)

Examing this, we see that as w approaches 0, the controlling expression would be the 152355*w part -- the w^3 and w^2 parts would vanish compared to the w component. So the limit of temp as w goes to 0 would be like

limit( (120000/152355)/w, w, 0)

When w approaches from the right, that is positive divided by +0, which is +infinity

When w approaches from the left, that is positive divided by -0, which is -infinity

Therefore the left and right limits are different, so the limit is not defined.