• $U$ and $V$ be separable $\mathbb R$-Hilbert spaces
  • $\iota:U\to V$ be a Hilbert-Schmidt embedding
  • $Q:=\iota\iota^\ast$
  • $(e_n)_{n\in\mathbb N}$ be an orthonormal basis of $U$
  • $(\Omega,\mathcal A,\operatorname P)$ be a probability space
  • $(B_n)_{n\in\mathbb N}$ be a family of independent Brownian motions on $(\Omega,\mathcal A,\operatorname P)$ and $$W_n(t):=\sum_{i=1}^dB_i(t)e_i\;\;\;\text{for }n\in\mathbb N\text{ and }t\ge 0\;.$$

It is well-known that $$\left\|W_n(t)-W(t)\right\|_{L^2(\operatorname P,\:V)}\;\xrightarrow{n\to\infty}\;0\;\;\;\text{for all }t\ge 0\tag 1$$ for some $Q$-Wiener process $\left(W\left(t\right)\right)_{t\ge0}$ on $(\Omega,\mathcal A,\operatorname P)$.

Now, imagine that $u=\left(u_1,\ldots,u_d\right):[0,\infty)\times\mathbb R^d\to\mathbb R^d$ is the velocity of a fluid and that $[0,\infty)\ni t\mapsto X_t\in\mathbb R^d$ is the trajectory of a single fluid particle $X_0$.

I want to obtain a SDE for $u$ of type as studied in A Concise Course on Stochastic Partial Differential Equations. As usual, I want to assume that $u$ takes values in $$H:=\overline{\mathfrak D}^{\left\|\;\cdot\;\right\|_{L^2(\Omega,\:\mathbb R^d)}}$$ with $$\mathfrak D:=\left\{\phi\in C_c^\infty(\Omega)^d:\nabla\cdot\phi=0\right\}$$ and $\Omega\subseteq\mathbb R^d$ being open (the "domain" occupied by the fluid).

I want to assume that the trajectory is subject to a random forcing. Having the desired type of SDE for $u$ in mind, I want to choose the forcing to be driven by $W$ (note that we will choose $U$ to be $H$ or a superset of $H$).

So, I want to assume that $${\rm d}X(t)=u(t,X(t)){\rm d}t+\xi(t,X(t)){\rm d}W(t)\;\;\;\text{for all }t\ge 0\tag 2$$ for some $\xi:\Omega\times[0,\infty)\times\mathbb R^d\to\operatorname{HS}(U,\mathbb R^d)$.

By the Itō formula, we obtain $${\rm d}u_i(t,X(t))=\left[\frac{\partial u_i}{\partial t}(t,X(t))+\left(u\left(t,X(t)\right)\cdot\nabla\right)u_i\left(t,X(t)\right)+\frac 12\operatorname{tr}\left[{\xi\left(t,X(t)\right)}^\ast\nabla^2u_i\left(t,X(t)\right)\xi\left(t,X(t)\right)\right]\right]{\rm d}t+\left(\xi\left(t,X(t)\right)\cdot\nabla\right)u_i(t,X_t){\rm d}W(t)\tag 3$$ where $\nabla^2u_i(t,x)$ denotes the Hessian of $u_i$ at $(t,x)\in[0,\infty)\times\mathbb R^d$.

How can we rewrite $(3)$ to obtain a SDE for $u$ of the desired type?


Your Answer

By clicking "Post Your Answer", you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.