Suppose $U$ is an orthogonal matrix, and $D$ is a diagonal matrix. Let $I$ denote the identity matrix. Let $k$ be a positive integer.
I think the following holds:
$$(I - UDU^T)^k = U(I - D)^kU^T$$
But I got a little lost while writing out the steps \begin{align*} (I - UDU^T)^k &= (UIU^T - UDU^T)^k\\ &= (U(I - D)U^T)^k\\ &= ? \end{align*}
What exactly are $(U^T)^k$ and $U^k$?
I am not sure why you are asking about $(U^T)^k$ and $U^k$, but recall that it does not generally hold that $(ABC)^k = A^kB^kC^k$; we would need to have more information about $A,B,C$ (for instance, that they commute).
We can prove the result inductively. Note that $$ \begin{align} (U(I - D)U^T)^k &= (U(I - D)U^T)^{k-1}U(I - D)U^T \\ &= [(U(I - D)U^T)^{k-1}]U(I - D)U^T \\ & = \color{red}{[U(I - D)^{k-1}U^T]}U(I - D)U^T \\ & = U(I - D)^{k-1}\color{red}{[U^TU]}(I - D)U^T \\ & = U\color{red}{(I - D)^{k-1}(I - D)}U^T = U(I - D)^kU^T. \end{align} $$