Solve the following linear ODE $$3t^{2}y'+t^{3}y=cos(t)$$

What i tried

Since this is a linear equation, i used the integrating factor method. First i didide both the LHS and RHS of the equation by $3t^{2}$ so that $y'$ have a coefficient of $1$. Then i find the integrating factor $u(t)=t^{1/3}$. I the multiply the integrating factor of $u(t)$ to both the LHS and RHS of the equation and then integreating both sides of the equation, getting $$yt^{1/3}=\int(1/3)t^{-5/3}cos(t)$$ Im stuck from here onwards though,I tried integration by parts but it seems to lead me to nowhere. Could anyone explain. Thanks

  • $\begingroup$ The integrating factor is exp $( \int \alpha (t) dt)$, where $\alpha$ is the coefficient of $y$. So I think your integrating factor is wrong. $\endgroup$ Jan 24 '15 at 6:32
  • $\begingroup$ My integrating factor is $e^{1/3ln(t)}$ which simplifies to $t^{1/3}$ $\endgroup$
    – ys wong
    Jan 24 '15 at 6:35

You have \begin{align} 3t^2\frac{dy}{dt}+t^3y=\cos\left(t\right).\tag{1} \end{align} Now put the equation in standard form as follows: \begin{align} 3t^2\frac{dy}{dt}+t^3y=\cos\left(t\right)\Leftrightarrow \frac{dy}{dt}+\left(\frac{t}{3}\right)y=\frac{\cos\left(t\right)}{3t^2}\tag{2}. \end{align} Notice that the LHS looks a lot like the product rule of differentiation. Therefore, we must find an integrating factor $\mu\left(t\right)$: \begin{align} &\mu\left(t\right) = \exp\left(\int \frac{t}{3}\:dt\right)=e^{t^2/6},\tag{3} \\ \therefore e^{t^2/6}\frac{dy}{dx}+\left(\frac{t}{3}\right)&e^{t^2/6}y=\frac{d}{dt}\left[\left(e^{t^2/6}\right)\left(y\right)\right]=\frac{e^{t^2/6}\cos\left(t\right)}{3t^2} \tag{4} \\ &\implies \int\frac{d}{dt}\left[\left(e^{t^2/6}\right)\left(y\right)\right]\:dt=\int\frac{e^{t^2/6}\cos\left(t\right)}{3t^2}\:dt\tag{5}\\ &\implies y\left(t\right)=e^{-t^2/6}\int\frac{e^{t^2/6}\cos\left(t\right)}{3t^2}\:dt.\tag{6} \end{align} Now solve for the integral (it's not too bad).

  • $\begingroup$ Is it through integration by parts? $\endgroup$
    – ys wong
    Jan 24 '15 at 6:48
  • 2
    $\begingroup$ I am not too sure that "it's not too bad". $\endgroup$ Jan 24 '15 at 7:49
  • $\begingroup$ Actually this is the difficult part of the whole question. $\endgroup$
    – ys wong
    Jan 24 '15 at 7:56
  • $\begingroup$ Could anyone help me with this? This is where im actually stuck at. $\endgroup$
    – ys wong
    Jan 24 '15 at 14:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.