A Novel Quantum Vacuum Fluctuations Optimization Algorithm for Global Optimization

Abstract

This paper introduces the Quantum Vacuum Fluctuations Optimization Algorithm (QVFOA), a novel metaheuristic inspired by the spontaneous emergence of virtual particles in empty space. The algorithm mimics quantum fluctuations through elite‑guided exploration, memory‑based diversity maintenance, and a multi‑scale local search inspired by wave refraction. The proposed method is evaluated on a set of 10 two‑dimensional benchmark functions of diverse nature (unimodal, multimodal, and valley‑shaped), including the challenging Schubert, Rastrigin, and Easom functions. Over 50 independent runs per function, QVFOA achieves the global optimum with 100% success rate (within   tolerance) on all functions, attaining an accuracy of   and zero standard deviation on most of them. The results demonstrate the algorithm's robustness, precision, and excellent balance between exploration and exploitation. A qualitative comparison with well‑known algorithms (PSO, GA, DE, SA) highlights the superiority of QVFOA on multimodal and valley‑shaped landscapes. A comprehensive parameter sensitivity analysis is provided to guide practical applications and confirm the algorithm's robustness to parameter variations within safe domains. The source code and complete results are publicly available (see Section 8).