Intelligent water drops algorithm for multimodal spaces

Abstract

This paper presents a new nature inspired Intelligent Water Drops (IWD) based algorithm for finding peaks in continuous multimodal optimization problems. Initially various conceptual similarities were identified between IWD algorithm and Genetic Algorithm(GA). Simultaneously applying IWD-Continuous Optimization(IWD-CO) algorithm and GA on a function in finding the global optima and found IWDCO having faster convergence qualities. By taking this as basis, GA has been replaced with IWD-CO in a recently developed Modified Roaming Optimization(MRO) algorithm and applied to various benchmark functions and found drastic variation in convergence. Results are proving that replacing GA with IWD-CO can be a novel step in evolutionary based multimodal search algorithms.