Indexed text attachment
partial-differential-equations-in-higher-categories.tex
Source for Partial Differential Equations in Higher Categories
View paper · Download original file
Lines 1621–1720 of 3921. Source text is displayed literally.
1621\section{Stokes' theorem, twice}1622 1623\begin{proposition}[Green's identities]\label{prop:green}1624\Scl[source, \S7.1--7.2] For functions on a bounded domain $D$ with smooth boundary,1625\begin{align*}1626\int_Dv\,\Lap u+\int_D\nabla v\cdot\nabla u&=\int_{\pd D}v\,\pd_nu,\\1627\int_D(u\,\Lap v-v\,\Lap u)&=\int_{\pd D}(u\,\pd_nv-v\,\pd_nu).1628\end{align*}1629\end{proposition}1630 1631\noindent Both are Stokes' theorem for the form $v\star\dd u$. On a grid the first is the factorization of1632\Cref{prop:tonti}: the Dirichlet Laplacian is $D^{\T}D$. \chk{G1} The right-hand side of the second is a symplectic1633form on the space of pairs (boundary values, normal derivatives), the same form as in \Cref{thm:lagrangian}.1634 1635\begin{corollary}[The Dirichlet-to-Neumann map is symmetric]\label{cor:dtn}1636\Srf The pairs $(u|_{\pd D},\pd_nu|_{\pd D})$ for $u$ harmonic in $D$ form a Lagrangian subspace of the boundary data.1637It is the graph of the operator $\Lambda_D\colon h\mapsto\pd_nu$, where $u$ is the harmonic function with boundary1638values $h$; hence $\Lambda_D$ is symmetric, and by the first identity non-negative.1639\end{corollary}1640 1641\section{Dirichlet's principle}1642 1643\begin{theorem}\label{thm:dirichlet}1644\Scl[source, \S7.1] Among functions with given boundary values, the harmonic one has the least energy1645$E(w)=\tfrac12\int_D|\nabla w|^2$. More precisely, if $u$ is harmonic and $z$ vanishes on the boundary, then1646$E(u+z)=E(u)+E(z)$.1647\end{theorem}1648 1649\noindent The identity is the Pythagorean theorem for the energy inner product: harmonic functions are orthogonal to1650functions vanishing on the boundary. \chk{G2} So the solution operator of the Dirichlet problem is an orthogonal1651projection. The same holds in any subspace, which is the Rayleigh--Ritz or Galerkin method \src{\S7.1, Exercise16527}: the approximate solution in a trial space is the projection of the true one, and its error is orthogonal to1653the trial space. \chk{G7} Finite elements are this statement with a particular trial space (\Cref{ch:numerics}).1654 1655\section{The kernel of the inverse}1656 1657\begin{definition}[source, \S7.3]1658The \term{Green's function} of $D$ with pole $x_0$ is the function $G(x,x_0)$ that is harmonic in $D$ away from1659$x_0$, vanishes on $\pd D$, and differs from $-1/(4\pi|x-x_0|)$ by a function harmonic near $x_0$ (in three1660dimensions).1661\end{definition}1662 1663\begin{theorem}\label{thm:greenfn}1664\Scl[source, \S7.3] \emph{(a)} The solution of $\Lap u=f$ in $D$, $u=h$ on $\pd D$ is1665$u(x_0)=\int_{\pd D}h\,\pd_nG+\int_DfG$. \emph{(b)} $G(x,x_0)=G(x_0,x)$. \emph{(c)} $G\le0$ in $D$, and1666$\pd_nG\ge0$ on $\pd D$ with total mass one.1667\end{theorem}1668 1669\noindent In operator terms, $-G$ is the kernel of the inverse of $-\Lap$ with Dirichlet conditions; (b) says the1670inverse is symmetric because the operator is; (c) says the inverse preserves positivity, which is the maximum1671principle, and identifies $\pd_nG$ with the harmonic measure of \Cref{thm:poisson}. On eight grid points the1672inverse of the Dirichlet Laplacian is the symmetric positive matrix $\min(i,j)(9-\max(i,j))/9$, the discrete form1673of the kernel $x(1-y)$ for $x<y$ on the unit interval. \chk{G3, G8}1674 1675\section{Half-space and ball}1676 1677\begin{theorem}\label{thm:images-green}1678\Scl[source, \S7.4] For the half-space $\{z>0\}$, $G(x,x_0)=-\frac1{4\pi|x-x_0|}+\frac1{4\pi|x-x_0^*|}$ with1679$x_0^*$ the mirror image of $x_0$. For the ball of radius $a$,1680$G(x,x_0)=-\frac1{4\pi|x-x_0|}+\frac{a}{|x_0|}\frac1{4\pi|x-x_0^*|}$ with $x_0^*=a^2x_0/|x_0|^2$, and1681$\pd_nG=\dfrac{a^2-|x_0|^2}{4\pi a|x-x_0|^3}$ on the sphere.1682\end{theorem}1683 1684\noindent The first is \Cref{thm:images} for the reflection in a plane with the sign character. The second is the1685same after a conformal change: inversion in the sphere exchanges the ball with its exterior, and the factor1686$a/|x_0|$ is the weight of the Kelvin transform of \Cref{thm:conformal}(c). Both kernels vanish on the boundary and1687have normal derivatives of total mass one; the Poisson kernel of the ball returns the value $r_0$ of the harmonic1688function $z$ at height $r_0$. \chk{G4, G5}1689 1690\begin{center}1691\begin{tikzpicture}1692\begin{scope}1693\fill[fwd!10] (-2.2,0) rectangle (2.4,1.9);1694\draw[very thick] (-2.2,0)--(2.4,0);1695\fill[fwd] (0,1.1) circle (2pt) node[above left]{\scriptsize $x_0\ (+)$};1696\fill[bwd] (0,-1.1) circle (2pt) node[below left]{\scriptsize $x_0^*\ (-)$};1697\draw[densely dashed,ink2] (0,1.1)--(0,-1.1);1698\fill (1.5,0) circle (1.4pt) node[below right]{\scriptsize $x$};1699\draw[fwd] (0,1.1)--(1.5,0);1700\draw[bwd] (0,-1.1)--(1.5,0);1701\node at (0.1,-2.0) {\scriptsize half-space: equal distances on the wall};1702\end{scope}1703\begin{scope}[xshift=6.6cm]1704\fill[fwd!10] (0,0) circle (1.4);1705\draw[very thick] (0,0) circle (1.4);1706\draw[densely dashed,ink2] (0,0)--(3.1,0);1707\fill (0,0) circle (1pt) node[below left]{\scriptsize $0$};1708\fill[fwd] (0.7,0) circle (2pt) node[below]{\scriptsize $x_0$};1709\fill[bwd] (2.8,0) circle (2pt) node[below]{\scriptsize $x_0^*$};1710\fill (0.7,1.2124) circle (1.4pt) node[above]{\scriptsize $x$};1711\draw[fwd] (0.7,0)--(0.7,1.2124);1712\draw[bwd] (2.8,0)--(0.7,1.2124);1713\node at (1.2,-2.0) {\scriptsize ball: $|x-x_0^*|/|x-x_0|=a/|x_0|$ on the sphere};1714\end{scope}1715\end{tikzpicture}1716\figcap{The two image constructions. In the half-space the image is the mirror point with the opposite charge. In the ball of radius $a$ it is the inverse point $x_0^*=a^2x_0/|x_0|^2$ with charge $-a/|x_0|$: on the sphere the ratio of the two distances is constant (here $a=1.4$, $|x_0|=0.7$, ratio $2$), which is what makes the sum vanish there.}1717\end{center}1718 1719\section{Gluing}1720