An Improved Clonal Algorithm in Multiobjective Optimization

Abstract

In this paper, we develop a novel clonal algorithm for multiobjective optimization (NCMO) which is improved from three approaches, i.e., dynamic mutation probability, dynamic simulated binary crossover (D-SBX) operator and hybrid mutation operator combining with Gaussian and polynomial mutations (GP-HM operator). Among them, the GP-HM operator is controlled by the dynamic mutation probability. These approaches adopt a cooling schedule, reducing the parameters gradually to a minimal threshold. By this means, they can enhance exploratory capabilities, and keep a desirable balance between global search and local search, so as to accelerate the convergence speed to the true Pareto-optimal front. When comparing NCMO with various state-of-the-art multiobjective optimization algorithms developed recently, simulation results show that NCMO evidently has better performance.