Getting particular solution for harmonic oscillator

212 Views Asked by At

I have $\ddot u + \omega^2\cdot u = \frac{K(t)}{m}$ whereas $u$ is the "distance" of a mass $m$.

Now I need to find a particular solution to it for the case that $K(t)=K_0$, so it is a constant. I somehow can't solve it.

My first confusion is about the exersice itself. Do they actually mean the general solution for the non-homogenic harm. osc. or "just" the particular solution?

Anyway, that's what I'd do:

First let's find the homogenic solution, I won't include much details since it's a really good known solution.

Ansatz: $u(t)=e^{\lambda t}$ so we get $\lambda_{1,2}=\pm i \omega$

Now: $u_h(t)=Ae^{\omega i t} + Be^{-\omega i t}$

or by using the addition theorem of cos/sin we can also write it down as:

$u_h(t)=C\cdot \cos(\omega t - \varphi)$

For getting the particular solution I'd use the homogenic solution as a ansatz but consider the integration constant as a function of t.

Set $k:=\omega t -\varphi$

Ansatz: $u_p(t)=C(t)\cos(k)$

Now: $\dot u = C'\cos(k) - C\omega\sin(k)$

$\ddot u = C''\cos(k) - C'\omega\sin(k) - C'\omega\sin(k) - C\omega^2\cos(k)$

We put it inot the DGL and get:

$C''\omega \cos(k) - 2C'\omega\sin(k) - 2C\omega^2\cos(k)=\frac{K_0}{m}$

I see tha I can either get rid of cosinus or sinus by just choosing k accordingly. (we can do that, since $K_0=const$) but in both cases, I don't really get the desired result.

Can someone please give me a hint on how to solve this? :)

The solution should be: $u_{sp}(t)=\frac{K_0}{m\omega^2}$ (no idea what sp stands for)

1

There are 1 best solutions below

0
On BEST ANSWER

When the right hand side is a constant, try $u(t)=u_0$, then $\ddot{u}(t)=0$, so $$\omega^2 u_0=\frac{K_0}{m}$$ implies $$u_0=\frac{K_0}{m\omega^2}$$ That means $u(t)=\frac{K_0}{m\omega^2}$ solves the DE.