0
$\begingroup$

The problem I am working on is this:

Find ${\nabla}f(a,b)$ for the differential function $f(x,y)$ given the directional derivatives:

$D_{(i+j)/2}f(a,b) = 3\sqrt2$ , $D_{(3i-4j)/5}f(a,b) = 5$

I've done some reading on Directional Derivatives to see if I could then solve them, but I can't make head or tails of this problem. Could someone guide me through this or point me in the right direction?

Thanks in advance.

| cite | improve this question | | | | |
$\endgroup$
  • $\begingroup$ Perhaps the formula $D_af=(\nabla f)\cdot a$ might help. $\endgroup$ – user137731 Mar 23 '15 at 21:34
  • $\begingroup$ The thing that is confusing me is the $(i+j)/2$ part. Where do I use that? Oh sorry, didn't see the $a$. Okay... so I dot product the function with what exactly? $\endgroup$ – Kommander Kitten Mar 23 '15 at 21:36
  • $\begingroup$ That's your $a$ in my formula. $\endgroup$ – user137731 Mar 23 '15 at 21:36
1
$\begingroup$

Hints: $\nabla f(a.b) = (D_i f(a,b), D_j f(a,b))$, write $i$ and $j$ in terms of $(i+j)/2$ and $(3i-4j)/5$, and: $$i = \alpha \cdot \frac{i+j}{2} + \beta \cdot\frac{3i-4j}{5} \implies D_i f(a,b) = \alpha\cdot(3\sqrt{2})+\beta\cdot (5),$$and similarly for the other one.


I'll do the first one for you. Write: $$(1,0) = \alpha \left(\frac{1}{2},\frac{1}{2}\right)+\beta \left(\frac{3}{5},-\frac{4}{5}\right) \implies \begin{cases} \frac{\alpha}{2}+\frac{3\beta}{5} = 1 \\ \frac{\alpha}{2} - \frac{4\beta}{5} = 0\end{cases},$$so that $\beta = 5/7$ and $\alpha = 4/7$. Hence: $$D_i f(a,b) = D_{\frac{4}{7}\left(\frac{i+j}{2}\right)+\frac{5}{7}\left(\frac{3i-4j}{5}\right)}f(a,b) = \frac{4}{7}D_{\frac{i+j}{2}}f(a,b)+\frac{5}{7}D_{\frac{3i-4j}{5}} f(a,b).$$

| cite | improve this answer | | | | |
$\endgroup$
  • $\begingroup$ I don't understand what you mean by "write $i$ and $j$ in terms of $(i+j)/2$ and $(3i-4j)/5$ $\endgroup$ – Kommander Kitten Mar 23 '15 at 21:48
  • $\begingroup$ If I told you to write $(1,0)$ as a combination of $(1/2,1/2)$ and $(3/5,-4/5)$ would you understand? There is more than one possible combination, but the final outcome will be the same. $\endgroup$ – Ivo Terek Mar 23 '15 at 21:49
  • $\begingroup$ A bit clearer, but still unsure of myself. Could you explain further? I feel like I'm missing something fairly obvious. $\endgroup$ – Kommander Kitten Mar 23 '15 at 21:54
  • $\begingroup$ I'll add an edit. By the way, I just noticed that we will have a unique combination, I looked too fast. $\endgroup$ – Ivo Terek Mar 23 '15 at 21:56
  • $\begingroup$ @Kommander see if you can do the second one $\endgroup$ – Ivo Terek Mar 23 '15 at 22:01

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.