Multi-Objective Evolutionary Algorithm Based on Improved Clonal Selection


Abstract

Evolutionary algorithm is widely used to search Pareto-optimal set. The paper proposes a kind of multi-objective evolutionary algorithm based on improved clonal selection (MOEAICS), which incorporates improved clonal selection algorithm (ICSA) into multi-objective evolutionary algorithms to replace genetic operators such as crossover and mutation in traditional evolutionary algorithms. ICSA characterized by 1) excavation of excellent gene fragments in antibody population and generating a memory antibody set, 2) packaging operation of excellent gene fragment, and 3) replacing low affinity antibody with high affinity antibody with probability from mutation antibody population during updating memory antibody population. The testing results show that MOEAICS is capable of maintaining population diversity, improving search efficiency, and accelerating convergence.