A Dual-Population Cooperative Water Wave Optimization Algorithm with Knowledge-Driven Learning Mechanism and Its Performance Analysis

Abstract The water wave optimization algorithm (WWO), which is inspired by the shallow water wave theory, is a promising meta-heuristic in continuous optimization problems. However, the canonical WWO algorithm has proved to fall into the local optimum as lacking the guidance of the search direction and iteration step sizes in the search process. In this paper, a dual-population cooperative WWO with a knowledge-driven learning mechanism (DCLWWO) is proposed to relieve the drawback. The improved learning mechanism of WWO with the searching operations of different types of the population is combined to strengthen the self-learning and self-adapting capabilities of the algorithm. The wavelength [[EQUATION]] was redefined with a fixed external archive to further control the size of the search step during the iterations. DCLWWO is compared with the existing variants of WWO and several well-established algorithms to substantiate the performance of the algorithm. Meanwhile, DCLWWO is applied to solve the blocking flow shop scheduling problem (BFSP), which has been widely studied in artificial intelligence manufacturing. Experimental results and systematic analysis indicate the effectiveness of DCLWWO in solving complex optimization problems.