Equilibrium microcanonical annealing for first-order phase transitions.

Area of Science:

• Computational Physics
• Statistical Mechanics
• Phase Transitions

Background:

• Simulating equilibrium systems in the microcanonical ensemble is computationally challenging.
• Traditional methods may struggle with complex systems and phase transitions.
• Annealing techniques offer potential solutions for exploring complex energy landscapes.

Purpose of the Study:

• To present a novel framework for microcanonical ensemble simulations.
• To evaluate the performance of different equilibrium annealing algorithms.
• To investigate thermal first-order transitions in a specific model system.

Main Methods:

• Development of a framework for annealing in an energy ceiling.
• Implementation of equilibrium simulated annealing, population annealing, and hybrid algorithms.
• Application of these algorithms to the 20-state, two-dimensional Potts model.

Main Results:

• All tested equilibrium microcanonical annealing algorithms performed well at the first-order transition.
• Equilibrium simulated annealing demonstrated the highest efficiency for the studied system sizes.
• The framework successfully facilitated simulations of equilibrium systems.

Conclusions:

• The proposed framework provides an effective approach for microcanonical ensemble simulations.
• Equilibrium simulated annealing is a highly efficient method for studying first-order phase transitions.
• The developed algorithms offer valuable tools for computational physics research.