Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I know that a solution to a pde I am interested is in the form: $ u(x,y)=(x-y)^{5}\frac{\partial^{4}}{\partial x^{2} \partial y^{2}}\left(\frac{F(x)-G(y)}{x-y}\right) $ where $F$ and $G$ are arbitrary functions to be determined.

For my case I know $u(x,y_{0})$ and $u(x_{0},y)$ explicitly. What is the best way to compute $F$ and $G$ in this case?

share|cite|improve this question

Let $U^{xx}(x):=\int_{x_0}^{x}dx'\int_{x_0}^{x'}dx''\frac{u(x'',y_0)}{(x''-y_0)^5}$ and $U^{yy}(y):=\int_{y_0}^{y}dy'\int_{y_0}^{y'}dy''\frac{u(x_0,y'')}{(x_0 -y'')^5}$. Because $u(x,y_0)$ and $u(x_0,y)$ are known explicitly, also $U^{xx}(x)$ and $U^{yy}(y)$ are known explicitly, at least in principle. From

$U^{xx}(x)=\int_{x_0}^{x}dx'\left[\frac{\partial^{3}}{\partial x \partial y^{2}} \left(\frac{F(x)-G(y)}{x-y}\right) |_{y=y_0} \right]_{x_0}^{x'}$

$U^{yy}(y)=\int_{y_0}^{y}dy'\left[\frac{\partial^{3}}{\partial y\partial x^{2}} \left(\frac{G(y)-F(x)}{y-x}\right) |_{x=x_0} \right]_{y_0}^{y'}$

it follows that




$a=\frac{\partial}{\partial x} \left(\frac{F(x)}{(x-y_0)^3}\right)|_{x=x_0}+\frac{\partial^{2}}{\partial y^{2}} \left(\frac{G(y)}{(x-y)^2}\right)|_{y=y_0}$

$b=\frac{\partial^{2}}{\partial y^{2}} \left(\frac{F(x_0)-G(y)}{x_0-y}\right) |_{y=y_0}$

$c=\frac{\partial}{\partial y} \left(\frac{G(y)}{(y-x_0)^3}\right)|_{y=y_0}+\frac{\partial^{2}}{\partial x^{2}} \left(\frac{F(x)}{(y-x)^2}\right)|_{x=x_0}$

$d=\frac{\partial^{2}}{\partial x^{2}} \left(\frac{G(y_0)-F(x)}{y_0-x}\right) |_{x=x_0}$

Note that $a$, $b$, $c$ and $d$ are actual linear combinations of $F(x_0)$, $F'(x_0)$, $F''(x_0)$, $G(y_0)$, $G'(y_0)$ and $G''(y_0)$. By differentiating the equations for $U^{xx}(x)$ and $U^{yy}(y)$ two times, and plugging $x_0$ and $y_0$ into all obtained equations, we seem to obtain 6 different linear equations for $F(x_0)$, $F'(x_0)$, $F''(x_0)$, $G(y_0)$, $G'(y_0)$ and $G''(y_0)$. But two of these equations are just the original equations, so we just have 5 different equations. However, this is no problem, because only $F(x)-G(y)$ is important, so $F$ and $G$ are only determined up to a constant.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.