I've come across a type of homogeneous 2nd order ode that I haven't seen before and isn't covered in the book I have. I've also tried googling it but haven't been able to find anything similar since I don't know the name for this type of problem. Any help in how to solve it would be much appreciated. Question is below.

Find general solutions with the replacement $x= e^t$ to the following differential equations, in each of which a function $y(x)$ is defined for $x>0$. Here, $y^{\prime}$ denotes the derivative of first order with respect to $x$ for a function $y(x)$.

1.)$$x^2y^{\prime\prime}-xy^{\prime}-x-3y=0$$ 2.)$$x^3y^{\prime\prime\prime}+6x^2y^{\prime\prime}+4xy^{\prime}-4y=0$$


Since you need to express as a function of $t$ using $x=e^t$ (that is to say $t=\log(x)$), start with $$\frac{dy}{dx}=\frac{dy}{dt}\times \frac{dt}{dx}=\frac 1 x\frac{dy}{dt}=e^{-t}\frac{dy}{dt}$$ Deriving a second time $$\frac{d^2y}{dx^2}=\frac d{dt}\big(\frac{dy}{dx}\big)\frac{dt}{dx}=\frac d{dt}\big(e^ {-t}\frac{dy}{dt}\big)\frac{dt}{dx}=e^{-t} \left(\frac{d^2y}{dt^2}-\frac{dy}{dt}\right)e^{-t}=e^{-2t} \left(\frac{d^2y}{dt^2}-\frac{dy}{dt}\right)$$ So, the differential equation $$x^2y^{\prime\prime}-xy^{\prime}-x-3y=0$$ write $$e^{2t}e^{-2t} \left(\frac{d^2y}{dt^2}-\frac{dy}{dt}\right)-e^te^{-t}\frac{dy}{dt}-e^t-3t=0$$ that is to say $$\frac{d^2y}{dt^2}-2\frac{dy}{dt}-3y=e^t$$ which looks simple.

Continue the same way to get $\frac{d^3y}{dx^3}$ for the second problem.

  • $\begingroup$ thanks for your answer I was reading some of the material linked to before is it possible to do this using the cauchy-euler method also ? $\endgroup$ – dmnte Oct 15 '15 at 8:51
  • $\begingroup$ can anyone confirm that the answer for the second question is $y = Ae^x+Bx^-2x+Cxe^-2x$. I found a method where you just integrate $y=e^t$ and plug that into the original question. Can that be used for any question ? $\endgroup$ – dmnte Oct 19 '15 at 11:04

The key point point about these differential equations is that the coefficient of each derivative of order $n$ is $x^n$. I know that the second order ODE the you wrote is named Euler differential equation. You can use the special property I mentioned to find the general solution of the corresponding homogeneous ODE namely ${y_h}$ and then find a particular solution ${y_p}$ for the non-homogeneous ODE. Finally, summing the two solutions ${y_h} + {y_p}$ will give you the general solution of the ODE. I will work out the first example, leaving the second for you. we have

$${x^2}y'' - xy' - 3y = x$$

fist consider the corresponding homogeneous ODE

$${x^2}y'' - xy' - 3y = 0$$

Now we guess that $y(x) = {x^r}$ can be a solution. This is motivated by the special property of this ODE that was pointed out. Now put it into the homogeneous ODE to get

$$\left( {{r^2} - r - 3} \right){x^r} = 0$$

so $x^r$ can be a solution if and only if

$${r^2} - r - 3 = 0\,\,\,\,\,\, \to \,\,\,\,\,\,\,{r_{1,2}} = {{1 \pm \sqrt {13} } \over 2}$$

so we have

$${y_h} = {c_1}{x^{{r_1}}} + {c_2}{x^{{r_2}}}$$

for finding a particular solution since the RHS is a polynomial we consider

$${y_p} = Ax + B$$

putting into the non-homogeneous ODE leads to

$$0 - Ax - 3\left( {Ax + B} \right) = - 4Ax - 3B = x\,\,\,\,\,\,\, \to \,\,\,\,\,\,\left\{ \matrix{ - 4A = 1 \hfill \cr - 3B = 0 \hfill \cr} \right.\,\,\,\,\, \to \,\,\,\,\,\,\left\{ \matrix{ A = - {1 \over 4} \hfill \cr B = 0 \hfill \cr} \right.$$

and hence

$${y_p} = - {1 \over 4}x$$

and the general solution of the whole problem will be

$$y = {y_h} + {y_p} = {c_1}{x^{{r_1}}} + {c_2}{x^{{r_2}}} - {1 \over 4}x$$

  • $\begingroup$ thanks for your reply, isnt it r^2-2r-3 $\endgroup$ – dmnte Oct 19 '15 at 4:18
  • $\begingroup$ @dmnte: Sorry for the delay. The system didn't notify me and I just saw this by accident! :) I don't think so! It is $r^2-r-3=0$. :) $\endgroup$ – Hosein Rahnama Oct 26 '15 at 15:17

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.