Qamar Abbas

Differential evolution (DE) is a powerful global optimization algorithm which has been studied intensively by many researchers in recent years. A number of mutation variants have been established for this algorithm. These mutation variants make the DE algorithm more applicable, but random development of these variants has created inconsistencies such as naming and formulation. Hence this study aims to identify inconsistencies and to propose solutions to make them consistent. Most of the inconsistencies exist because of the uncommon nomenclature used for these variants. In this study, a comprehensive study is carried out to identify inconsistencies in the nomenclature of mutation variants that do not match each other. Appropriate and consistent names are proposed for them. The proposed names assigned to conflicting variants are based on the name of the variant, the total number of vectors used to generate the trial vector, and the order of the vectors to form the equation of these mutation variants. To ensure the performance diversity of the consistent set of DE mutation strategies, experimental results are generated using a test suit of benchmark functions.

This paper presents a novel random controlled pool base differential evolution algorithm (RCPDE) where powerful mutation strategy and control parameter pools have been used. The mutation strategy pool contains mutations strategies having diverse parameter values, whereas the control parameter pool contains varying nature pairs of control parameter values. It has also been observed that with the addition of rarely used control parameter values in these pools are highly beneficial to enhance the performance of the DE algorithm. The proposed mutation strategy and control parameter pools improve the solution quality and the convergence speed of DE algorithm. The simulation results of the proposed RCPDE algorithm shows significant performance as compared to other algorithms when tested over a set of multi-dimensional benchmark functions.

Differential evolution (DE) is a powerful global optimization algorithm which has been studied intensively by many researchers in the recent years. A number of variants have been established for the algorithm that makes DE more applicable. However, most of the variants are suffering from the problems of convergence speed and local optima. A novel tournament based parent selection variant of DE algorithm is proposed in this research. The proposed variant enhances searching capability and improves convergence speed of DE algorithm. This paper also presents a novel statistical comparison of existing DE mutation variants which categorizes these variants in terms of their overall performance. Experimental results show that the proposed DE variant has significance performance over other DE mutation variants.