Adaptive mutation-driven search for global minima in 3D coulomb clusters: A new method with preliminary applications

Abstract

A single-string-based evolutionary algorithm that adaptively learns to control the mutation probability (pm) and mutation intensity (Δm) has been developed and used to investigate the ground-state configurations and energetics of 3D clusters of a finite number (N) of ‘point-like’ charged particles. The particles are confined by a harmonic potential that is either isotropic or anisotropic. The energy per particle (EN /N) and its first and second differences are analyzed as functions of confinement anisotropy, to understand the nature of structural transition in these systems.