Differential Evolution with Parent Centric Crossover
Millie Pant, Musrrat Ali, V. P. Singh · 2008
Differential evolution (DE) has emerged as a powerful tool for solving optimization problems in the last few years. However, the convergence rate of DE still does not meet all the requirements, and attempts to speed up differential evolution are considered necessary. In order to improve the performance of DE, we propose a modified DE algorithm called DEPCX which uses parent centric approach to manipulate the solution vectors. The performance of DEPCX is evaluated on a test bed of five functions. Numerical results are compared with original differential evolution (DE) and with TDE, another recently modified version of DE. Empirical results indicate that this modification enables the algorithm to get a better transaction between the convergence rate and robustness.