0
$\begingroup$

The question asks to find y given y(x) is a differentiable function satisfying:

$\frac{dy}{dx}$=-$2xy^4$ , $y(0)$=$\frac{1}{3}$ and $y(x)>0$ , and explain why it is unique.

I assume the steps are to integrate $\frac{dy}{dx}$ to get a function for y(x), then use $y(0)=1/3$ to find the constant and then set $y(x)>0$ to find y but i keep getting stuck.

A differential equation solution gives $y(x)$=$\frac{1}{\sqrt[3]{c_1+3x^2}}$ , which I can use $y(0)$ to find $c_1$ but then I'm not sure how to get y from that since the equation doesn't have any y's in it.

A similar question I found on this site showed to rearrange and integrate each side like $\int$$\frac{dy}{y^4}$=$\int$$-2xdx$ which gives $-\frac{1}{3y^3}+c_1$=$-x^2+c_2$ , but then I don't know how to get $y(x)$ from that frunction, it seems further from the solution than my first try.

$\endgroup$
0
$\begingroup$

That's just a simple manipulation: \begin{equation*} \begin{split} -\frac{1}{3y^3}+c_1=-x^2+c_2 & \iff -\frac{1}{3y^3} = -x^2 + c_2-c_1 \\ & \iff \frac{1}{3y^3} = x^2 +c_1-c_2 \\ & \iff y^3 = \frac{1}{3x^2 + 3(c_1-c_2)} \\ & \iff y = \frac{1}{\sqrt[3]{3x^2+3(c_1-c_2)}}. \end{split} \end{equation*} Now it's just note that $3(c_1-c_2)$ is constant, say $k$, and we obtain $$y=\frac{1}{\sqrt[3]{3x^2+k}}.$$

$\endgroup$
  • $\begingroup$ Is that all I need as the complete answer? No need to use the given properties/conditions? $\endgroup$ – S. Flemming Oct 5 '17 at 22:53
  • $\begingroup$ The given differential equation is separable, and the method to solve separable first order differential equations is as you've mentioned in your question. What I have done is manipulate your partial solution to get the solution you said don't know how to get it. Am I missing something? $\endgroup$ – Rodrigo Dias Oct 5 '17 at 22:57
  • $\begingroup$ What did you mean by "(...) set $y(x) > 0$ to find $y$ (...)" ? As I see it, $y$ is just a function, not an incognite. $\endgroup$ – Rodrigo Dias Oct 5 '17 at 23:01
  • $\begingroup$ Judging by the conditions the question gave I assumed I would end up with an equation y(x) in terms of y and x with a constant that I would need to compute using y(0)=1/3, and then set that equation <0 to get y<some function of x. $\endgroup$ – S. Flemming Oct 5 '17 at 23:01
  • $\begingroup$ I'm thinking that y(x) and y are two different things, are they not? $\endgroup$ – S. Flemming Oct 5 '17 at 23:06

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.