A new hybrid multi-objective optimization quantum evolutionary algorithm

A hybrid multi-objective optimization quantum evolutionary algorithm is proposed. Considering the characteristics of multi-objective optimization, three update strategies designed specially for the quantum individuals are adopted. One is an improved quantum gate with a self-adaptive tuning rotation angle. In order to prevent premature convergence, values of quantum bits which represent probability amplitudes are limited in a certain range. Another one is a quantum rotation crossover operator based on the proposed quantum gate. The third strategy is to to mutate the quantum individual by a not-operation. MQEA is compared with a classical algorithm in different test functions. Experimental results indicate that MQEA has higher searching efficiency. It can find a set of non-dominated solutions with better performance of convergence and distribution in less time.