A novel version of slime mould algorithm for global optimization and real world engineering problems Enhanced slime mould algorithm

The slime mould algorithm is a stochastic optimization algorithm based on the oscillation mode of nature's slime mould, and it has effective convergence. On the other hand, it gets stuck at the local optimum and struggles to find the global optimum. Location updates of slime moulds are very important in terms of convergence to optimum. In this study, the position updates of the sine cosine algorithm are combined with the slime mould algorithm. In these updates, besides the existing sine cosine algorithm, different types of sine cosine algorithmic transformations are used and the oscillation processes of the slime moulds are also modified. In the mathematical model of the slime mould algorithm, the arctanh function that stacks two random slime moulds in a certain interval has been replaced by a novel modified sigmoid function. The proposed function is presented with its theoretical derivations based on Schwarz lemma. According to experimental results, it has been observed that the exploration and exploitation capabilities of the proposed algorithm are highly effective. In the study, sine cosine trigonometric functions have been used while updating the position in slime mould algorithm. The performance of the presented algorithm has been considered for fifty benchmark functions and has also been tested on cantilever beam design, pressure vessel design, 3-bar truss and speed reducer real world problems. Accordingly, it is possible to conclude that the proposed hybrid algorithm has better ability to escape from local optima with faster convergence than standard sine cosine and slime mould algorithms.(c) 2022 International Association for Mathematics and Computers in Simulation (IMACS). Published by Elsevier B.V. All rights reserved.

Automated summary

The study combines the position updates of the sine cosine algorithm with the slime mould algorithm to improve its convergence and ability to find global optima. The proposed hybrid algorithm, which modifies the oscillation processes of slime moulds, demonstrates highly effective exploration and exploitation capabilities. Experimental results show that it outperforms the standard sine cosine and slime mould algorithms in escaping local optima with faster convergence.