3
$\begingroup$

I am trying to find the minimum distance from the points of the function $\large{\frac{1}{4xy}}$ to the point $(0, 0, 0)$.

This appears to be a problem of Lagrange in which my condition: $C(x,y,z) = z - \frac{1}{4xy} = 0$, and my function would be $f(x,y,z) = \sqrt{x^2+y^2+z^2}$ or if i'm correct, it would be the same as the minimum value I get from using thee function as $f(x,y,z) = x^2+y^2+z^2$.

If I do this, I would then have:

$$ f(x,y,z,\lambda) = x^2 + y^2 + z^2 + \lambda(z-\frac{1}{4xy}) \\ = x^2 + y^2 + z^2 + \lambda z-\frac{1}{4xy} \lambda $$

Then findind the partial derivatives ($\frac{\partial f}{\partial x}, \frac{\partial f}{\partial y}, \frac{\partial f}{\partial z}, \frac{\partial f}{\partial \lambda}$) and solving for the values of $x, y, z, \lambda$ and the minimum value I get at the end would be my answer.

Would that be the correct solution?

$\endgroup$
2
  • 1
    $\begingroup$ By substitution of the constraint we need to minimize $$\sqrt{x^2+y^2+\frac{1}{4x^2y^2}}$$ and we don't need Lagrange's method. In this case AM-GM is the most effective method. $\endgroup$ – user Jan 31 '18 at 13:23
  • $\begingroup$ Sorry there was a mistake the function to minimize is of course $$\sqrt{x^2+y^2+\frac{1}{16x^2y^2}}$$ $\endgroup$ – user Jan 31 '18 at 14:28
1
$\begingroup$

Note that by substitution of the constraint we need to minimize

$$f(x,y)=\sqrt{x^2+y^2+z^2}=\sqrt{x^2+y^2+\frac{1}{16x^2y^2}}$$

and we don't need Lagrange's method.

In this case AM-GM is the most effective method, indeed

$$\frac{x^2+y^2+\frac{1}{16x^2y^2} }{3}\ge\sqrt[3]{x^2\cdot y^2\cdot \frac{1}{16x^2y^2}}\iff x^2+y^2+\frac{1}{16x^2y^2}\ge3\sqrt[3]{\frac1{16}}=\frac32\sqrt[3]{\frac1{2}}$$

and equality holds when $$x^2=y^2=\frac{1}{16x^2y^2}\implies x=\pm\sqrt[6]\frac1{16}=\pm\sqrt[3]\frac14 \quad y=\pm x$$

Thus the minimum distance is

$$d_{min}=\sqrt{\frac32 \sqrt[3]{\frac1{2}} } =\sqrt{\frac32}\sqrt[6]{\frac1{2}}$$

attained for

$$P_1=\left(\sqrt[3]\frac14,\sqrt[3]\frac14,\frac14\sqrt[3]16\right)\quad P_2=\left(\sqrt[3]\frac14,-\sqrt[3]\frac14,-\frac14\sqrt[3]16\right)$$

$$P_3=\left( -\sqrt[3]\frac14,\sqrt[3]\frac14,-\frac14\sqrt[3]16\right)\quad P_4=\left( -\sqrt[3]\frac14,-\sqrt[3]\frac14,\frac14\sqrt[3]16\right)$$

As an alternative, since $f(x,x)>0$ its minimum coincide with the ninimum of

$$g(x,y)=f^2(x,y)=x^2+y^2+\frac{1}{16x^2y^2}$$

thus for the critical points we have

$$g_x=2 x - \frac1{8 x^3 y^2}=0$$ $$g_y=2 y - \frac1{8 x^2 y^3}=0$$

from which we obtain

$$x=\pm\sqrt[6]\frac1{16}=\pm\sqrt[3]\frac14 \quad y=\pm x$$

$\endgroup$
6
  • $\begingroup$ I have not done minimizing before so this is a bit strange to be. $\endgroup$ – Omari Celestine Jan 31 '18 at 13:33
  • 1
    $\begingroup$ @OmariCelestine You can proceed also with the usual method but you don't need Lagrange at all. And for simplicity you can minimize $f^2$. Note that in general AM-GM inequality are very powerful and important. $\endgroup$ – user Jan 31 '18 at 13:37
  • $\begingroup$ What do you bean by the usual method? $\endgroup$ – Omari Celestine Jan 31 '18 at 16:25
  • $\begingroup$ By partial derivatives, I add some detail on it! $\endgroup$ – user Jan 31 '18 at 16:31
  • $\begingroup$ Exactly as I was proceeding? $\endgroup$ – Omari Celestine Jan 31 '18 at 16:46
1
$\begingroup$

Then you have the following equations:$$2x+\lambda\dfrac{1}{4x^2y}=0\\2y+\lambda\dfrac{1}{4xy^2}=0\\2z+\lambda=0\\z=\dfrac{1}{4xy}$$or$$\lambda=-8x^3y=-8xy^3=-2z=-\dfrac{1}{2xy}$$since $x,y\ne 0$ this leads to $x^2=y^2$ and $16x^4y^2=1$ therefore $16x^6=1$ and the answer is as following:$$x=k_1(\dfrac{1}{4})^{\dfrac{1}{3}}\\y=k_2(\dfrac{1}{4})^{\dfrac{1}{3}}\\z=\dfrac{1}{k_1k_2}(\dfrac{1}{4})^{\dfrac{1}{3}}$$where $$k_1=\pm 1\\k_2=\pm 1$$therefore we have 4 answers.

$\endgroup$

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.