Ask your own question, for FREE!
Mathematics 6 Online
OpenStudy (anonymous):

When \[\mathcal{F}=\int \frac{1}{2}\alpha | \nabla C(\vec{x}) |^2 + \beta \Psi(C(\vec{x})) d\vec{x}\], with alpha and beta some certain constants and psi a certain function of C, what is \[\frac{\partial \mathcal{F}}{\partial C}\]

OpenStudy (anonymous):

Ok as the equation interpreter has died: When \[\mathcal{F}=\int \frac{1}{2}\alpha | \nabla C(\vec{x}) |^2 + \beta \Psi(C(\vec{x})) d\vec{x}\], with alpha and beta some certain constants and psi a certain function of C, what is \[\frac{\partial \mathcal{F}}{\partial C}\]

OpenStudy (anonymous):

Do you think the chain rule could help?

OpenStudy (anonymous):

\[ \frac{d\mathcal{F}}{d\vec{x}} =\frac{1}{2}\alpha | \nabla C(\vec{x}) |^2 + \beta \Psi(C(\vec{x})) \]

OpenStudy (anonymous):

I would suppose the chain rule is something like: \[ \frac{d\mathcal{F}}{d\vec{x}} = \frac{\partial \mathcal{F}}{\partial C} \cdot \huge ? \]

OpenStudy (anonymous):

\frac{dC}{d\vec{x}}

OpenStudy (anonymous):

Ah yes \[\frac{dC}{d\vec{x}}\]

OpenStudy (anonymous):

So just divide the integrand by that and you got it? I dunno how else to simplify it.

OpenStudy (anonymous):

Ah of course, as \[\frac{\partial \mathcal{F}}{\partial \vec{x}}\] is the integrand

OpenStudy (anonymous):

Though I'm a bit concerned because I don't remember the chain rule for the multivariable case.

OpenStudy (anonymous):

Well I have the answer at hand but the road to it is a bit unclear, would it help?

OpenStudy (anonymous):

Sure.

OpenStudy (anonymous):

\[\frac{\partial \mathcal{F}}{\partial C} = \beta \Psi'(C) - \alpha \nabla^2C\]

OpenStudy (anonymous):

So I guess the chain rule on the part with alpha would: \[\nabla \cdot \left(\frac{1}{2} \alpha |\nabla C|^2 \right) = \alpha \nabla \cdot \left(\nabla C\right)\]

OpenStudy (anonymous):

I'm just not sure how to 'divide the integrand by' \[\frac{dC}{d\vec{x}}\]

OpenStudy (anonymous):

Oh, it's just division by a function, but in this case it is clearly not what they're looking for anyway.

OpenStudy (anonymous):

Do you know Leibniz integral rule? http://mathworld.wolfram.com/LeibnizIntegralRule.html

OpenStudy (anonymous):

I do now :)

OpenStudy (anonymous):

Ah, that seems promising.

OpenStudy (anonymous):

Though I have an indefinite integral

OpenStudy (anonymous):

It seems that using the variational derivative: \[F(C(x)) = \int f(C,\nabla C)\, dx\] Then the Euler-Lagrange equation: \[\frac{\partial \mathcal{F}(C)}{\partial C} = \frac{\partial f(C,\nabla C)}{\partial C} - \nabla \cdot \frac{\partial f(C, \nabla C)}{\partial \nabla C}\] Then: \[\frac{\partial \mathcal{F}(C)}{\partial C} = \beta \Psi'(C) - \nabla \cdot \left(\alpha \nabla C\right)\] where \[\frac{\partial f(C, \nabla C)}{\partial \nabla C}=\frac{\partial}{\partial \nabla C} \left(\frac{1}{2}\alpha |\nabla C|^2\right)\]

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!