Quantum particle swarm algorithm for Many-objective optimization problem

Many-objective optimization problems widely exist in real world. However, there is lack of effective solutions to solve this problem because they contain more than three conflicting objectives. Based on quantum particle swarm optimization algorithm, this paper presents an efficient many-objective particle swarm optimization algorithm. In this algorithm, an improved quantum update method is introduced to update the particles´ positions, a selection strategy based on the global difference order is proposed to update the global best position of particle, and a congestion sorting strategies is applied to update the external repository. By optimizing ZDT and DTLZ series functions, and comparing with representative algorithms such as TV-MOPSO, results indicate that the proposed algorithm is effective for solving many-objective optimization problems.