Solve this 3rd Order Differential Equation

eric2921
Messages
8
Reaction score
0

Homework Statement



d^3x/dt^3 - d^2x/dt^2 = 3e^t - sin(t)

Homework Equations



not sure. maybe sin(t)=.5*i*e^(-i*t)-.5*i*e^(i*t)

The Attempt at a Solution



I'm pretty sure we start by finding the complementary solution by setting the right side equal to 0. So we have:

D^3 x - D^2 x = 0
D^2 [D-1] x = 0

then we solve for D and get 0, 0 and 1 for solutions.
this gives us:

xcomp. = [c1+t c2] e^(0t) + c3 e^t
xcomp. = c1 + c2 t + c3 e^t

now its my understanding that i need to find particular solution because:

x = xcomp. + xpart.

So i was told that the solutions to the particular solution are 1, i and -i. I don't understand why that is though but if that's the case then i think:

xpart. = t e^t + A cos(t) + B cos(t)

and

x= c1 + c2 t + c3 e^t + t e^t + A cos(t) + B cos(t)

I'm stuck here, I'm pretty sure it's correct up to this point although i could be wrong. I don't know to find A or B though which i think all i have left to do. Thanks for any help!
 
Physics news on Phys.org
Hi,

your template is correct: xpart. = t e^t + A cos(t) + B sin(t). a little thing you forgot is a constant in front of t e^t --> K t e^t

the next step is to "plug in" only the xpart. solution into the original equation and find the constants K, A, B. then you are done.

P.S. this is the form of the full answer: x(t) = C1 + C2 t + C3 e^t + K t e^t + A cos(t) + B sin(t)
** you can find C1, C2, C3 if you have boundary conditions (starting values).

good luck.
 
Last edited:
Why is xpart = t e^t + A cos(t) + B cos(t) ?

shouldnt it be:
xpart = t e^t + A cos(t) + B sin(t) or am I just missing something?
 
You can integrate twice directly to reduce this third order down to a first order
 
eric2921 said:

Homework Statement



d^3x/dt^3 - d^2x/dt^2 = 3e^t - sin(t)

Homework Equations



not sure. maybe sin(t)=.5*i*e^(-i*t)-.5*i*e^(i*t)

The Attempt at a Solution



I'm pretty sure we start by finding the complementary solution by setting the right side equal to 0. So we have:

D^3 x - D^2 x = 0
D^2 [D-1] x = 0

then we solve for D and get 0, 0 and 1 for solutions.
this gives us:

xcomp. = [c1+t c2] e^(0t) + c3 e^t
xcomp. = c1 + c2 t + c3 e^t

now its my understanding that i need to find particular solution because:

x = xcomp. + xpart.
Good so far.
So i was told that the solutions to the particular solution are 1, i and -i. I don't understand why that is though but if that's the case then i think:

xpart. = t e^t + A cos(t) + B cos(t)

and

x= c1 + c2 t + c3 e^t + t e^t + A cos(t) + B cos(t)
Almost.

The function sin t can be written in terms of complex exponentials:

\sin t = \frac{e^{it} - e^{-it}}{2i}

so the righthand side can be written as

e^t - \frac{1}{2i} e^{it} + \frac{1}{2i} e^{-it}

You should be able to see that

(D+i)(D-i)(D-1) [3 et - sin t] = 0

Those are the 1, i, and -i roots they're referring to. It follows then that the complete solution x(t) satisfies the homogeneous equation

(D+i)(D-i)(D-1) D2(D-1)x(t) = 0

If you write down the solution to this equation, you'd get

x(t) = c_1 + c_2 t + c_3 e^t + A t e^t + B e^{it} + C e^{-it}

or equivalently

x(t) = c_1 + c_2 t + c_3 e^t + A t e^t + B \cos t + C \sin t

The first three terms are the homogeneous solution you found earlier. The last three terms make up xp(t). In practice, you don't go through this long procedure. You just look at the terms of the forcing function and deduce what terms will arise in the particular solution (taking into account what type of terms already appear in the complementary solution). But it's good to know why this works.
I'm stuck here, I'm pretty sure it's correct up to this point although i could be wrong. I don't know to find A or B though which i think all i have left to do. Thanks for any help!
Now you want to plug x_p(t) = A t e^t + B \cos t + C \sin t into the original differential equation, then match coefficients for the various terms. From the three resulting equations, you can solve for A, B, and C.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top