A Comparative Study of Differential Evolution (DE) Algorithm, and Genetic Algorithm (GA) on Optimization of Opencast Machinery Noise

Numerous extensions to evolutionary algorithms (EAs) have been proposed during the last decades offering improved performance on engineering problems. In this paper, the algorithmic concepts of the Differential Evolution (DE) and Genetic Algorithm (GA) algorithms have been analyzed.The aim of the paper is to identify which one of them is more suitable to solve the optimization problem, depending on the problem's features and also to identify the variant with the best performance, regardless of the features of the problem to be solved. Eight variants of DE and GA were implemented and tested on machinery noise optimization problem in an opencast mine. Optimization problem solving successes of the mentioned algorithms have also been compared statistically by testing over machinery noise problem using ISO 9613-2 prediction function. Empirical results reveal that the run-time complexity and the required function-evaluation number for acquiring global optimization by the DE algorithm is superior to GA. Numerical experiments and comparisons from the study indicate that the DE/RAND/2 variant of DE algorithm outperforms and is superior to GA in terms of robustness, success rate, precise results and convergence rate. Therefore, DE/RAND/2 variant of DE algorithm shows a potential approach for machinery noise optimization related problems.