Riesz fractional derivative Elite-guided sine cosine algorithm

Abstract In order to improve the calculation accuracy of the sine cosine algorithm (SCA), Riesz fractional derivative sine cosine algorithm (RFSCA) based on the Riesz fractional derivative mutation strategy is proposed. The new algorithm uses quasi-opposition learning to initialize the population, which can increase the diversity of the population. Based on the approximate formula of Riesz fractional derivative with second-order accuracy, we construct a new mutation approach to update the optimal individual and improve the calculation accuracy of the algorithm. Furthermore, the proposed method is integrated into quasi-opposition learning and opposition-based learning strategies to enhance the ability of global exploration of the population and improve the convergence speed of the algorithm. The new algorithm is tested in two sets of test sets (classical benchmark of 23 problems and standard IEEE CEC 2017). The simulation experiments demonstrate that the proposed algorithm significantly outperforms the latest heuristic-based algorithms in both exploration, exploitation and solution quality. Two engineering questions (Welded beam design, pressure vessel design) are applied to confirm the superior performance of proposed algorithm.