Adaptive levy-assisted salp swarm algorithm: Analysis and optimization case studies

The salp swarm algorithm (SSA) is a recent and straightforward swarm intelligent optimizer. It mainly simulates the foraging and navigational behavior of salp in the ocean by forming a salp chain. The salp in the front of the chain guides the moving direction of the population, which makes the algorithm easy to fall into local optimum and lead to premature convergence. In order to tackle this shortcoming, an improved SSA integrated with adaptive weight and levy flight mechanism is proposed, which is called WLSSA. In this research, the adaptive weight is proposed to extend the exploratory scope of conventional SSA throughout the early stages and speeds up the convergence swiftness of the method in the later stages. By random walk of levy flight to explore the solution space, the global exploratory and local exploitation capabilities of the algorithm are more well-adjusted and enhanced. Under the cooperation and concurrent influence of the two mechanisms, the overall performance of the algorithm is significantly boosted in terms of the excellence of solutions. Twenty-three essential classical functions and selected IEEE CEC 2014 test functions are utilized to validate the effectiveness of the proposed WLSSA and compare and analyze the optimization capacity of WLSSA versus six mainstream meta-heuristic algorithms and eight improved advanced algorithms in solving function optimization problems. The results of the test cases confirm the significant improvements of the proposed SSA-based algorithm over the original SSA, and it also shows strong competitiveness compared to the associated technique. Also, to study the potential of WLSSA in treating practical problems in the real world, three constrained engineering cases are considered. Similarly, the comparison results reveal that it is possible to find a better solution using the proposed WLSSA to the same problem compared to the existing methods.