A similarity operator is proposed to construct the network among the super individuals. The cluster division
algorithm is used to divide the network to find the niche. In order to judge the evolutionary degree, the convergence
operator is introduced. The multi-position mutation policy is used to avoid the premature convergence and maintain the
diversity of the population, according to the schema existing in the niche. The convergence of the algorithm is proved
theoretically. Simulation results show that the improved algorithm can effectively avoid premature and greatly improve the
global efficiency, and the overall performance exceeds the deterministic crowding genetic algorithm, which is convenient in
parallel implementation respectively.