Advancements in the Sine Cosine Algorithm

Abstract

AbstractIn the last few decades, the development and advancement of meta-heuristic algorithms have become the focus of the research community as these algorithms face various challenges like, balance between exploration and exploitation, tuning of parameters, getting trapped in local optima, and very slow convergence rate. Sine cosine algorithm (SCA) also faces similar kinds of challenges and sometimes fails to perform effectively in finding the global optimal solution. Sine and cosine are trigonometric operators with a 90$$^\circ $$ phase shift from each other. The range of sine and cosine functions lies in the range $$[-1,1]$$. Sine and cosine functions in the position update equation of SCA help solutions to perform search procedure. However, in some situations, SCA promotes similar solutions in the search space, which results in the loss of diversity in the population, and the search process is susceptible to trapping in the region of local optimum [1]. Motivated by these challenges, SCA has been modified to improve its capability and efficiency in several ways. Several strategies have been employed to alter the basic version of SCA [2], aiming to enhance its effectiveness and optimization capabilities. In this chapter, we will discuss about these modifications and strategies, which have been incorporated into the sine cosine algorithm (SCA) in past few years. Apart from this, we will briefly describe the applications of the modified versions of SCA.