PEAO: A Cooperative Parallel Enzyme Optimization Algorithm with Adaptive Search Mechanisms

Abstract

Bio-inspired optimization algorithms have become an effective class of techniques for addressing challenging continuous optimization problems. In this work, we introduce the Parallel Enzyme Action Optimization (PEAO) algorithm, a parallel bio-inspired optimization approach that incorporates a multi-strategy communication mechanism among cooperative subpopulations. The population is partitioned into multiple subpopulations that evolve concurrently, promoting a more effective exploration of the search space. Furthermore, PEAO integrates adaptive search factors, local search procedure and communication strategies to improve solution quality while reducing the risk of premature convergence. In addition, a K-means-based population initialization procedure and a convergence-driven stopping criterion based on successive improvements in the best objective-function value are incorporated to reduce unnecessary objective-function evaluations and improve the overall efficiency of the optimization process.