Improved Quantum-behaved Particle Swarm Optimization Algorithm Based on Differential Evolution Operator and Its Application

The evolution equation of Quantum-behaved Particle Swarm Optimization (QPSO) algorithm was analyzed and then the premature convergence problem in QPSO algorithm was pointed out. The idea of Differential Evolution (DE) was introduced into QPSO algorithm and the improved algorithm was proposed, which was called QPSO-DE. During the search procedure of particle swarm, every dimension of a particle executes the DE operator according to a certain probability. The DE operator in QPSO can increase the randomicity and enhance the particles’ search ability and the ability of obtaining the optimal solutions. At the same time, the cases of trapping into the local minima and stagnancy in QPSO for the reason of diversity loss were decreased. The superior performance of the proposed algorithm was shown by comparing with PSO and QPSO on the benchmark functions and the design of IIR digital filter.