Weighted minimum $\alpha$-Green energy problems

TL;DR

This paper studies a weighted energy minimization problem involving the $ ext{α}$-Green kernel on a domain, providing conditions for solutions, characterizations, and convergence results based on the kernel's perfectness.

Contribution

It introduces new existence criteria, support descriptions, and convergence theorems for weighted minimum $ ext{α}$-Green energy problems, leveraging the kernel's perfectness.

Findings

Established necessary and sufficient conditions for solution existence.

Provided detailed descriptions of the solution support.

Proved convergence theorems for set approximations.

Abstract

For the $\alpha$-Green kernel $g^\alpha_D$ on a domain $D\subset\mathbb R^n$, $n\geqslant2$, associated with the $\alpha$-Riesz kernel $|x-y|^{\alpha-n}$, where $\alpha\in(0,n)$ and $\alpha\leqslant2$, and a relatively closed set $F\subset D$, we investigate the problem on minimizing the Gauss functional \[\int g^\alpha_D(x,y)\,d(\mu\otimes\mu)(x,y)-2\int g^\alpha_D(x,y)\,d(\vartheta\otimes\mu)(x,y),\] $\vartheta$ being a given positive (Radon) measure concentrated on $D\setminus F$, and $\mu$ ranging over all probability measures of finite energy, supported in $D$ by $F$. For suitable $\vartheta$, we find necessary and/or sufficient conditions for the existence of the solution to the problem, give a description of its support, provide various alternative characterizations, and prove convergence theorems when $F$ is approximated by partially ordered families of sets. The analysis performed is substantially based on the perfectness of the $\alpha$-Green kernel, discovered by Fuglede and Zorii (Ann. Acad. Sci. Fenn. Math., 2018).