Improved convergence performance of advanced crow search algorithm and application to optimal design of large-scale truss dome structures

Due to its small number of parameters, the crow search (CS) algorithm is easy to apply to engineering problems and solves optimization problems in various fields. The CS algorithm is not harmoniously used with the exploitation and exploration performance, and the convergence performance varies greatly depending on the initial population. An advanced CS (ACS) algorithm with a new method was proposed to improve the convergence performance of the CS algorithm, and it was proved that the convergence performance was improved by using benchmark functions and engineering example problems. However, the convergence performance according to the probability used in the proposed method for harmonizing the global search and the local search among ACS algorithms was not accurately compared and mentioned, and it was proposed to use a value of 0.5. Therefore, this paper examines the probability that the best convergence performance can be derived to harmonize global and local searches. In addition, it solves the engineering optimization problem of building structures by applying it to the optimal design of large-scale truss dome structures with natural frequencies as constraints. As a result, the minimum weight was derived from all truss dome structures when the probability proposed in this paper was used.