Novel chaotic atomic search optimization algorithm with genetic mutation mechanism for solving engineering design problems

Abstract The Atomic Search Optimization (ASO) algorithm is an efficient metaheuristic algorithm inspired by molecular-based dynamics. Although it effectively solves numerous engineering design optimization problems, the algorithm still has drawbacks, such as premature convergence and an imbalance between exploration and exploitation. This paper proposes an improved Atomic Search Optimization Algorithm (TGD_ASO) with a genetic mutation mechanism, golden sine strategy, and chaotic mapping. Firstly, the Tent chaotic mapping is introduced to improve the population diversity of the algorithm, thereby enhancing the search space. Secondly, the golden sine strategy is utilized to strengthen global search efficiency. After that, we introduce the random variation and crossover operations in differential evolutionary algorithms to perturb the updated positions to help the atoms jump out of the local optimum and improve the population’s diversity and the algorithm’s accuracy. To validate and test the performance of the proposed TGD_ASO algorithm. We analyze the performance of the TGD_ASO algorithm on the CEC2017 benchmarks and six real design issues. In the CEC2017 benchmark test, the algorithm achieved 23 and 22 top positions in the 30 and 50 dimensions, respectively. Its performance significantly surpassed that of the ASO algorithm. Experimental data applied to engineering design problems have shown that the algorithm achieved 5 key outcomes. The algorithm also significantly outperforms metaheuristic algorithms such as ASO in terms of convergence time, accuracy, stability, and optimization ability.